hash stringlengths 32 32 | doc_id stringlengths 7 13 | section stringlengths 3 121 | content stringlengths 0 2.2M |
|---|---|---|---|
434af603c87796b4b286aa062f07354f | 104 096 | 9.1.4.2 Japan and Korea | National Institute of Information and Communications Technology (NICT) in Japan and Electronics and Telecommunications Research Institute (ETRI) in South Korea are working on aligning its THz frequency regulations with international standards to facilitate global interoperability and support the development of next-generation technologies. |
434af603c87796b4b286aa062f07354f | 104 096 | 9.2 Proposed regulation | The 300 - 355 GHz and 650 - 700 GHz bands can be used by SRDs on a non-interference - non-protection basis in accordance with Article 4.4 of the ITU Radio Regulations [i.2]. An example regulation is given in Table 5. ETSI ETSI TR 104 096 V1.1.1 (2025-06) 25 Table 5: Proposed regulation for ground based vehicular Terahertz Imaging applications Frequency range Peak e.i.r.p. dBm Mean Power e.i.r.p. (average over signal repetition time) dBm 300 - 355 GHz 55 50 650 - 700 GHz 55 50 |
434af603c87796b4b286aa062f07354f | 104 096 | 9.3 Agenda Item 1.8 of WRC-27 Conference | The aim of Agenda item 1.8 of WRC-27 Conference is to consider possible additional spectrum allocations to the radiolocation service on a primary basis in the frequency range 231,5 - 275 GHz and possible new identifications for radiolocation service applications in frequency bands within the frequency range 275 - 700 GHz for millimetric and sub-millimetric wave imaging systems, in accordance with Resolution 663 (Rev.WRC-23). Resolution 663 asks for "Studies on possible new additional allocations to the radiolocation service on a primary basis in the frequency range 231,5 ‑ 275 GHz, and possible new identifications for radiolocation service applications in frequency bands within the frequency range 275 - 700 GHz". As can be seen from the agenda item 1.8 and Resolution 663, the studies are asked to investigate possible new identifications for radiolocation service applications in the 275 - 700 GHz. The present document provides the technical characteristics of Short-Range Devices to be operated in terahertz frequencies, but not any radiolocation service application. Therefore, there is no relation between the development of the present document and the studies to be conducted for WRC-27 Agenda Item 1.8. ETSI ETSI TR 104 096 V1.1.1 (2025-06) 26 Annex A: Change history Date Version Information about changes 2024-08-26 0.0.1 First draft for TGUWB #69 review 2024-10-09 0.0.2 Draft for TGUWB Rapporteur's M#1 review 2024-11-07 0.0.3 Draft for TGUWB Rapporteur's M#2 review 2024-11-24 0.0.4 Draft for TGUWB M#70 review 2024-12-18 0.0.5 Stable draft for approval 2025-01-28 0.1.0 Stable draft accepted by TGUWB 2025-02-04 0.1.1 Clean version for approval by ERM ETSI ETSI TR 104 096 V1.1.1 (2025-06) 27 History Document history V1.1.1 June 2025 Publication |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 1 Scope | The present document describes SRD radar equipment operating in 57 - 64 GHz and 76 - 77 GHz for applications upon drones which may require a change in the present regulatory framework for the proposed band. It includes in particular: • Market information. • Technical information regarding equipment type and typical installation. • Regulatory issues. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 2 References | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 2.1 Normative references | Normative references are not applicable in the present document. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 2.2 Informative references | References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding, but are not required for conformance to the present document. [i.1] ECC Report 262: "Studies related to surveillance radar equipment operating in the 76 to 77 GHz range for fixed transport infrastructure". [i.2] ERC/REC 70-03 (7 June 2024): "ERC Recommendation of 1997 relating to the use of Short Range Devices (SRD)". [i.3] ETSI EN 301 091-1 (V2.1.1): "Short Range Devices; Transport and Traffic Telematics (TTT); Radar equipment operating in the 76 GHz to 77 GHz range; Harmonised Standard covering the essential requirements of article 3.2 of Directive 2014/53/EU; Part 1: Ground based vehicular radar ". [i.4] FCC: "FCC Empowers Short-Range Radars in the 60 GHz Band", 47 CFR Part 15 (ET Docket No. 21-363; FCC 23-35; FR ID 153948). [i.5] ETSI TR 103 137 (V1.1.1): "Electromagnetic compatibility and Radio spectrum Matters (ERM); System Reference Document (SRdoc); Surveillance Radar equipment for helicopter application operating in the 76 GHz to 79 GHz frequency range". [i.6] ECC Report 268 (2018-02): "Technical and Regulatory Aspects and the Needs for Spectrum Regulation for Unmanned Aircraft Systems (UAS)". [i.7] ITU-R Report M.2204 (11/2010): "Characteristics and spectrum considerations for sense and avoid systems use on unmanned aircraft systems". [i.8] RTCA DO-366: "Minimum Operational Performance Standards (MOPS), for Air-to-Air Radar for Traffic Surveillance". [i.9] Joint Authorities for Rule making for Unmanned Systems (JARUS) Specific Operations Risk Assessment (SORA) v2.5. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 8 [i.10] ECC Report 222 (2014-09): "The impact of Surveillance Radar equipment operating in the 76 to 79 GHz range for helicopter application on radio systems". [i.11] ERC/REC 74-01: "Unwanted emissions in the spurious domain". [i.12] ETSI EN 303 360 (V1.1.1) (2017-02): "Short Range Devices; Transport and Traffic Telematics (TTT); Radar equipment operating in the 76 GHz to 77 GHz range; Harmonised Standard covering the essential requirements of article 3.2 of Directive 2014/53/EU; Obstacle Detection Radars for Use on Manned Rotorcraft". [i.13] ECC Report 352 (2023-06): "Harmonised conditions and spectrum bands for the operation of governmental Unmanned Aircraft System". [i.14] Arizton: "Healthcare Logistics Market Size, Share, Growth & Competitive Analysis Report By Product (Pharmaceuticals and Medical Devices), By Functionality, By End-User, By Geography - Forecast 2024-2029". [i.15] PWC: "Skies Without Limits v2.0". [i.16] ETSI EN 303 883-1: " Short Range Devices (SRD) and Ultra Wide Band (UWB); Part 1: Measurement techniques for transmitter requirements". [i.17] ECC Report 176 (2012-03): "The impact of non-specific SRDs on radio services in the band 57-66 GHz". [i.18] EASA - Easy Access Rules for Unmanned Aircraft Systems (Regulations (EU) 2019/947). [i.19] RTCA DO-365C: "Minimum Operational Performance Standards (MOPS) for Detect and Avoid (DAA) Systems". [i.20] Top 10 reasons for drone insurance claims in 2023. [i.21] ASTM F3442/F3442M: "Standard Specification for Detect and Avoid System Performance Requirements". [i.22] RTCA SC-228: "Minimum Performance Standards for Uncrewed Aircraft Systems". |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 3 Definition of terms, symbols and abbreviations | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 3.1 Terms | For the purposes of the present document, the following terms apply: governmental use: operations carried out by or on behalf of a public authority for the maintenance of law and order, protection of life and property, disaster relief and emergency response activities or services undertaken in the public interest excluding military operations/activities NOTE: As defined in [i.13]. othership: aircraft other than the ownship (or the ego aircraft) ownship: aircraft which should See and Avoid another ship |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 3.2 Symbols | Void. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 9 |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 3.3 Abbreviations | For the purposes of the present document, the following abbreviations apply: ADS-B Automatic Dependent Surveillance-Broadcast AGL (altitude) Above Ground Level ARC Air Risk Category ATAR Air To Air Radar BVLOS Beyond Visual Line Of Sight CAA Civil Aviation Authority dBsm decibels per square meter EASA European Union Aviation Safety Agency FMCW Frequency Modulated Continuous Wave FoV Field of View GNSS Global Navigation Satellite System GRC Ground Risk Categories JARUS Joint Authorities for Rulemaking of Unmanned Systems MOPS Minimum Operational Performance Standards NMAC Near Mid Air Collision PCB Printed Circuit Board RAS Radio Astronomy Services RCS Radar Cross Section RPAS Remotely Piloted Aircraft Systems RWC Remain Well Clear SAA See And Avoid SoC System on Chip SORA Specific Operations Risk Assessment SWaP-c Size Weight and Power - low cost SWaPc Small Weight and Power, low cost. TMPR Tactical Mitigation Performance Requirement UAS Uncrewed Aerial System UAV Uncrewed Aerial Vehicle UTM Uncrewed Traffic Management VLLF Very Low-Level Flying VTOL Vertical Take-Off and Landing |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 4 Comments on the System Reference Document | No ETSI member raised any comment. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5 Presentation of the system or technology | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.1 Uncrewed aircraft systems | Uncrewed Aircraft Systems (UASs), or drones, have huge potential to transform the productivity of businesses, and to introduce an entirely new way of transporting goods, and for the delivery of services, Figure 1. During a time when organizations are under pressure to be more efficient, innovative, and ambitious in how they operate, drones offer a unique opportunity. Drones can operate autonomously and gather data quickly and accurately from hard-to-reach places. This can make a crucial difference in managing costs, controlling risks, increasing safety, and influencing outcomes. The commercial use cases of drones, include a diverse range of industries, including utilities - for remote surveying of infrastructure -, transport and logistics and agriculture. Delivery of medicines is an area that has recently undergone massive expansion; the global healthcare logistics market alone is currently worth approximately £117 billion and is forecast to expand at a Compound Annual Growth Rate (CAGR) of 7,8 % in the period to 2026 [i.14]. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 10 Figure 1: Drone use cases, from ECC Report 268 [i.6] The PricewaterhouseCoopers (PwC) 'Skies without Limits' study [i.15] into the impact of drones on the UK economy alone forecasts that by 2030 drones will account for a £42 billion increase in UK Gross Domestic Product (GDP). However, there is a key obstacle to this expansion. The PwC report emphasized that maximum commercial growth could only be achieved when BVLOS autonomous operations, out of sight of a ground-based operator, which cannot currently take place due to outstanding regulatory or technical issues, including obstacle detection and avoidance in flight. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.2 Beyond Visual Line of Sight | One of the greatest challenges regarding commercial BVLOS drone operations, is that of risk-limitation in what are referred to as Very Low-Level Flying (VLLF) environments. These environments are often urban and heavily populated. Drones are required to be capable of avoiding other craft in flight, or structures on the ground including buildings, trees, wind turbines, or the landscape. Navigation is required to be resilient; satellite navigation is often occluded during VLLF and is easily interfered with. Full BVLOS operations require autonomy during take-off and landing phases as well as in flight. BVLOS drone operations necessitate that drones can undertake several complex and inter-related tasks throughout their flight missions. They plan a route, navigate to follow that route, detect, and avoid unforeseen obstacles, take-off, and land safely, and communicate flight information into UAV Traffic Management (UTM) systems. Operations are required to be resilient to changes in weather and, with redundancy by design, to single points of failure. Equipment will be required on-board the UAS to support safe and certified UAS flights, BVLOS. Such equipment will enable traffic-sensing, obstacle-avoidance, communication and navigation. The challenge for BVLOS is to replace the functions undertaken by a pilot during crewed flight, with on board UAV sensing, autonomy systems, and intelligence. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 11 Pilots of conventional light aircraft, UAS, and helicopters, have identified that See And Avoid (SAA) systems are required to: provide only relevant information to pilots of obstacles that require corrective action; reliably identify the 3D location; and present prioritized information based upon time. Business requirements, to have these capabilities under the widest range of operating conditions, and aircraft requirements for low Size, Weight, Power, and low lost (SWaP-c) instrumentation, make radar systems an obvious choice. Air To Air Radar (ATAR) for collision avoidance is an important part of equipage to support BVLOS flights and has been recognized by aviation authorities and standards bodies. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.3 See and Avoid requirement | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.3.0 General | The onboard UAS equipage requirements are dependent upon the Air Risk Category (ARC) within which the flight occurs. SORA [i.9] classifies the air risk of a UAS operation into one of four categories, from ARC-a (minimal risk) to ARC-d (high risk). The classification is based upon a flowchart which focusses primarily on encounter types, the airspace ruleset and whether the air environment is either recognized or contains known traffic. Operations in higher-risk airspace (e.g. controlled airspace with commercial aircraft) are assigned higher ARCs and require more robust See and Avoid mitigations. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.3.1 Low air risk categories | Low air risk categories include for example airspace in the absence of manned aviation, and low altitude, Figure 2, Figure 3 and Figure 4. Figure 2: Typical UAS flight profile in lower air risk category: delivery from one ground installation to another (source: Robert Bosch GmbH) ETSI ETSI TR 104 078 V1.1.1 (2025-06) 12 Figure 3: Typical UAS flight profile in lower air risk category: surveillance, or search (source: Robert Bosch GmbH) UAS Flight Inspections i.e. Power Line Inspections and Railway track Surveillance © Cambridge Sensoriis Long linear flights such as those by drones inspecting power lines or railway track surveillance, in changeable weather, require a robust onboard sensor. Micro radar enables a drone to detect objects in the flight path, whether in the air or ground based i.e. trees, power lines, birds, or other aircraft. As well as measuring the relative position of potential obstacles, the radar also directly measures relative velocity - allowing the UAS pilot, specialized in power line or railway track inspections, to gather close-quarter data assessing for corrosion and other maintenance needs. Figure 4: Use case: Drone linear surveys (source: ©Cambridge Sensoriis Ltd.) This leads to a requirement for a Shorter range and low SWaP SAA radar. In this context short range is for detection up to 150 m for objects that could include steel wire of diameter 2 mm and greater. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 13 5.3.2 High air risk categories High air risk categories include for example airspace in the vicinity of crewed and/or passenger carrying aviation, Figure 6. In these cases, airspace safety requirements provide for volumes of airspace around the ownship (that is, the ego aircraft) within which the othership (an aircraft, other than the ego aircraft) should be detected, Figure 5. Figure 5: RWC and NMAC boundaries (source: ©Cambridge Sensoriis Ltd.) The definitions for Remain Well Clear (RWC) and Near Mid Air Collision (NMAC) are as defined in ASTM F3442/F3442M [i.21] and are as follows: • RWC is defined as no incursion less than 2 000 feet (609,6 m) horizontally and 250 feet (76,2 m) vertically; • NMAC is defined as no incursion less than 500 feet (152,4 m) horizontally and 250 feet (30,5 m) vertically. This leads to a requirement for a medium range and low SWaP SAA radar. In this context, longer range ATAR are as defined in RTCA DO-366 [i.8], with radar detection range of 6,7 NM (12,4 km), for a large (circa 10 dBsm radar cross section) othership. And medium range is to 2 500 m for the same RCS. Figure 6: Resilient drone operations (source: ©Cambridge Sensoriis Ltd.) Situational awareness and SAA for crewed/ uncrewed aircraft in urban environments. © Cambridge Sensoriis BVLOS drone operations necessitate that drones can undertake several complex and inter-related tasks throughout their flight missions. They plan a route, navigate to follow that route, See and Avoid unforeseen obstacles, take-off, and land safely, and communicate flight information into UAS Traffic Management (UTM) systems. Operations are required to be resilient to changes in weather and, with redundancy by design, to single points of failure. Micro radars will be one of the key sensors onboard the UAS in order to detect obstacles and provide situational awareness to the UAS pilot/operator, or autonomous operations. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 14 |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.4 See and Avoid regulations for UAS | JARUS SORA [i.9] is the closest basis available to an international framework for Uncrewed Aircraft Systems (UAS) operations, and Annex D describes the 'Tactical Mitigation Performance Requirements' (TMPR) (i.e. the SAA requirements). The most difficult requirement to achieve, known as Air Risk Category (ARC), are ARC-d and ARC-c, see JARUS SORA 2.5 [i.9], Annex D, clause 5.3.2, Table 1. These require a system that meets performance standards as determined by RTCA SC-228 [i.22] or EUROCAE WG-105. The JARUS SORA [i.9] is, with minor adjustments, the Acceptable Means of Compliance (AMC) defined by EASA and the SAA requirements are the same (Easy Access Rules for UAS Implementing Regulation (EU) 2019/947 [i.18], p. 100). These rules require that SAA radar be incorporated into UAS aircraft design, and the following standards are available: • RTCA DO-365C [i.19] ("Minimum Operational Performance Standards (MOPS) for Detect and Avoid (DAA) Systems"). (DAA in this context, is synonymous with See and Avoid for obstacles in flight - SAA.) • RTCA DO-366 [i.8] ("Minimum Operational Performance Standards (MOPS) for Air-to-Air Radar for Traffic Surveillance"). One current problem with RTCA DO-366 [i.8] is that the standard is written around large and costly UAS, which can in turn support larger radar (for ARC-d and some ARC-c). Smaller, sub 25 kg UAS aircraft cannot support either the weight or the cost of such radar. These deficiencies are being addressed with aviation regulators, see clause A.5 - Requirements for modification - and will lead to a requirement for a frequency band of operation for a low SWaP-c radar. Proposed changes to RTCA DO-366 [i.8], for smaller UAS operating at low altitude, lead to a radar detection requirement for light aircraft to 2 500 m. One purpose of the present document is to propose radar operating in 76 - 77 GHz band for medium range SAA, on board small to medium sized UAS. Low SWaP-c radar are typically implemented through system-on-chip devices that include radar transceiver, digital signal processing, and micro-processor(s), Figure 10. These SoCs support bandwidths of up to 4 GHz in many cases, and FMCW devices operating in the 57 - 64 GHz, and 76 - 81 GHz band are readily available. Through careful design, transmit power of a few dBm up to 20 dBm are feasible, and typical receiver noise figures of 12 dB or lower. Such radars are often implemented with a planar phased array antenna, measuring both azimuth and elevation angles of detected objects though phase differences between receiver elements. Field of view requirements for SAA are often greater than the 120° possible from a planar antenna, and so 2 or more radar can be networked to support greater field of view. 5.5 Radar Landing systems As well as resilient See and Avoid technologies to support BVLOS operations, the growth towards drone autonomy requires technologies to provide resilience during the landing phase of flight. Sole reliance upon a drone's GNSS, and in some cases additional camera sensing, is limiting. GNSS spoofing, or restricted visibility towards satellites in built up areas prohibits resilient autonomy for higher air risk and ground risk categories. Camera based sensing is susceptible to poor weather conditions and limiting during nighttime operations. Several use cases require that UAS land upon moving targets, for example drones that fly from ships in support of border force operations. In these cases, GNSS based localization is overly complex, the drone should land upon a position on the ship deck, not upon a position in world coordinates as the vessel is moving. Relative positioning systems are becoming available which are active on both the drone and the ship deck (or ground-based landing pad), Figure 8 and Figure 9. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 15 Ship-to-shore/Shore-to-ship Operations © Cambridge Sensoriis A UAS integrated with a cooperative radar on the landing pad of a vessel enables an automated landing upon landing platform, either on a vessel or fixed ground position. The radars on the UAS and vessel can also help provide the heading, roll, and pitch of the vessel, all vital information to help reduce workload on pilots, support the safe operation and landing of drones in dynamic and challenging conditions, and protect high-value assets. Figure 7: Use case - autonomous drone landing (source: ©Cambridge Sensoriis Ltd.) © Cambridge Sensoriis Figure 8: Landing radar, supporting autonomous landing for a quadcopter (source: ©Cambridge Sensoriis Ltd.) |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 5.6 Other UAS Technologies | UAS are usually equipped with additional sensing and control systems to support flight. These include a flight controller; radio data link to a ground control station; GNSS system; inertial navigation system. Some drones have altimeters measuring height above ground, and Automatic Dependent Surveillance-Broadcast (ADS-B), which reports UAS location as derived from the onboard GNSS. See and Avoid systems, as described in greater detail in the present document, support the detection of objects that are not actively cooperating. All UAS technologies are designed for low size, weight and power. All functions are powered and lifted from the onboard battery that supports the flight mission and excess weight limits flight time and/or payload carrying capacity. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 16 |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 6 Market information in the EU | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 6.1 UAS Market size for Certified and Specific drone operations | Only higher risk UAS operations are expected to require medium and long Range SAA capabilities. These include certified categories of operation, and higher risk activities in the Specific category, Figure 9. Figure 9: Operating categories of Remotely Piloted Aircraft Systems (RPAS) (source: UK Civil Aviation Authority) Note that in this context 'Certified' refers to certification by a competent body for use in conjunction with aircraft (uncrewed or not) to support flight operations, and not certified to be compliant with a standard that relates to use of electromagnetic spectrum. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 6.2 Deployed numbers | The PwC projections are for 76 000 drones in use in the UK by 2030, across all sectors. These would be used in both urban, inter-urban and rural areas. Deployed numbers per area are more difficult to estimate. In areas around drone ports serving urban areas the numbers may be greater, and much lower in rural areas. Drones have to keep a minimum safe distance of hundreds of meters between each other, in 3 dimensions, to operate safely which naturally sets a limit on deployment density. Only professionally or governmental operated UAS in ARC-D and some ARC-C will be suitable for adopting radar technologies, those whose operating time and payload are not significantly affected by the sensor. Assuming 10 % market penetration, some 7 600 UAS radar could be in use nationally (UK) by 2030 based on PwCs projections. Deployment densities of radar on UAS can be expected to be much less than those for Advanced Driver-Assistance Systems (ADAS) radar deployed on vehicles. The projected UAS radar deployments in 2030, are < 1 % of the current number of registered cars and light goods vehicles. ECC Report 222 from 2014 [i.10] studies the impact of Surveillance Radar equipment operating in the 76 - 79 GHz range for helicopter applications on radio systems. It makes the further points in relation to potential automotive and UAS interference: • Both radar types (vehicular and helicopter radar) are likely to use FMCW modulation that mitigates the mutual interference. Here it should be considered that the distance between interferer and victim is assumed to be much larger than in the inter-vehicle situation. • The beam and frequency scanning capabilities of both radar types can reduce the intercept probability even further. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 17 The same report estimates a total number of manned turbine helicopters across Europe of 4 400, and an 80 % market penetration for the collision avoidance radar for manned helicopter flights. These numbers can be expected to increase by 2030, by which time the projected 7 600 UAS borne radar systems could still be greater, but not significantly so. These crewed rotorcrafts are permitted to deploy radar for collision avoidance, and a harmonised standard exists [i.12]. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 6.3 Drone deployment densities | The ITU projected drone densities are modest across all drone categories and particularly for the larger drones (more likely for ARC-d and ARC-c operations), which in turn require medium range SAA. Table 1 shows the projected (by 2030) UAS density across UAS size categories. Table 1: UAS densities vs. Size (Source: ITU [i.7]) UAS Categories Per 10 000 kms2 Large 0,440 Medium 1,950 Small 8,031 TOTAL 10,421 |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 6.4 Drone insurance claims | It is also interesting to note that the within the Top 10 reasons for drone insurance claims in 2023 [i.20] (courtesy of CoverDrone Insurance), the top reason is pilot error, which accounts for over 50 % of all claims reported. Factors such as fatigue, poor communication and distraction can all increase the likelihood of a drone-related incident resulting from pilot error. The third and fourth top reason for insurance claims relate to accidental damage or loss, while the fifth top reason are bird strikes. Radar provides excellent all weather object detection performance in a low SWaPc package. Radar technology will support navigation, collision avoidance, and airspace deconfliction. By providing greater and relevant situational awareness of flying and ground-based obstructions, this reduces the workload on UAS pilots and helps to avoid or greatly reduce accidental damage or loss. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7 Technical information | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.1 Technical Description | Both the See and Avoid radar, and landing radar, are implemented using highly integrated SoC devices, including radar FMCW transmitter, receiver, mixer, associated Power amplifier and Low noise amplifiers, and digital signals processing, Figure 10. Antenna for transmitted and receive are isolated and implemented upon planar PCB to form a phased array. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 18 Figure 10: 76 - 81 GHz Radar SoC architecture (source: ©Cambridge Sensoriis Ltd.) Modulation patterns for the SAA radar are software configurable, with a continuous modulation. Systems are compliant with the technical parameters of ETSI EN 301 091-1 [i.3]. Landing radar have an intermittent modulation schema, and a radar whether on the ship deck or on the UAS is typically modulated 50 % of the time. Because modulation parameters are software configurable, whether a landing radar or a SAA radar is largely determined in firmware. The radar could step from "See and Avoid" mode (purely horizontal) into "landing" (surrounding) mode, through beam steering. The following tables list typical parameters for a small SWaP radar for UAS applications. It is advantageous to deliver multiple capabilities through a single radar hardware set onboard the UAS, thereby reducing UAS payload and maximizing the mission flight time. Radar operating in 76 - 77 GHz Band, as defined in Table 2 and Table 3. Table 2:Technical parameters of See and Avoid at medium range Use Case See and Avoid - Medium Range Frequency Band 76 - 77 GHz (fmcw) Detection Requirement Large Other ship Aircraft, RCS 10 dBsm 2 500 m (Annex A) Occupied Bandwidth Configurable depending on range between 150 MHz and 1 000 MHz FoV 120° horizontal 40° vertical (± 20°) Instrumented Range 3 000 m Mean Power 50 dBm eirp Peak Power 55 dBm eirp Receiver Noise Figure 12 dB or better Duty Cycle Up to 100 % Maximum height AGL 120 m Directionality Forward facing, horizontal ETSI ETSI TR 104 078 V1.1.1 (2025-06) 19 Table 3: Technical parameters of landing and altimeter use-case Use Case 1. Landing zone clear 2. Altimeter Frequency Band 76 - 77 GHz Detection Requirement Minimum target - steel cables > 2 mm diameter Resolution in measurement distance: 50 mm Occupied Bandwidth 1 GHz for 15 cm resolution more desirable, 5 cm and 3 GHz FoV 90 - 120° horizontal 90 - 120° vertical Instrumented Range 150 m Mean Power 50 dBm Peak Power 55 dBm eirp Duty Cycle Up to 100 % Maximum height AGL 120 m Directionality Downward facing, vertical Radar operating in 57 - 64 GHz Band, as defined in Table 4 and Table 5: Table 4: Technical parameters of See and Avoid at short range Use Case See and Avoid - Short range Frequency Band 57 - 64 GHz (ERC/REC 70-03 [i.2] Annex 1, Generic use) Occupied Bandwidth Configurable depending on range between 1 000 MHz and 2 000 MHz FoV 90° horizontal 20° vertical Instrumented Range 100 m For minimum target 4,29 dBsm at 64 GHz (2 mm wire radius, 1 m length, perpendicular onto target) Mean Power 20 dBm eirp Duty Cycle Up to 100 % Maximum height AGL 120 m Directionality Predominately forward facing (moving direction), horizontal but also, a requirement for 360° coverage around the UAS Table 5: Technical parameters of See and Avoid during landing and take-off Use Case See and Avoid - Very short range Landing/take-off Frequency Band 57 - 64 GHz Occupied Bandwidth Configurable depending on range. 3 - 4 GHz Resolution measuring distance: 50 mm FoV 90° horizontal 20° vertical Instrumented Range 20 m; (UAS speed of 5 m/s is considered) Mean Power 20 dBm eirp Duty Cycle Up to 100 % maximum height AGL Up to 120 m Detection requirements Minimum target (cables): 2 mm diameter Directionality Forward facing (moving direction), horizontal, or downward facing, vertical |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.2 Status of technical parameters | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.2.1 Current ITU and European Common Allocations | Current allocation of the candidate bands in the CEPT (European common allocation ECA) is included in Table 6, together with actual usage within the CEPT. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 20 Table 6: Allocations and usage within CEPT Frequency band Allocations - Europe (ECA) Applications 56,9 - 57 GHz Earth Exploration-Satellite (passive), Fixed, Inter-Satellite, Mobile, Space Research (passive) Passive sensors (satellite), Fixed 57 - 58,2 GHz Space Research (passive), Mobile, Inter-Satellite, Fixed, Earth Exploration-Satellite (passive) Fixed, LPR, Passive sensors (satellite), Non-specific SRDs, TLPR, Wideband data transmission systems 58,2 - 59 GHz Earth Exploration-Satellite (passive), Fixed, Space Research (passive), Radio Astronomy TLPR, Non-specific SRDs, Radio astronomy, Passive sensors (satellite), LPR, Wideband data transmission systems, Fixed 59 - 59,3 GHz Radiolocation, Space Research (passive), Inter-Satellite, Mobile, Earth Exploration-Satellite (passive), Fixed Fixed, LPR, Passive sensors (satellite), Non-specific SRDs, TLPR, Wideband data transmission systems 59,3 - 64 GHz Fixed, Mobile, Inter-Satellite, Radiolocation Wideband data transmission systems, TLPR, Non-specific SRDs, LPR, ISM, ITS, Fixed 64 - 65 GHz Mobile Except Aeronautical Mobile, Fixed, Inter-Satellite Fixed, ITS, Radio astronomy, Wideband data transmission systems 75,5 - 76 GHz Broadcasting, Broadcasting-Satellite, Fixed, Fixed-Satellite (Space-To-Earth), Amateur, Amateur-Satellite Amateur, Amateur-satellite, Space research, TLPR, LPR, Fixed 76 - 77,5 GHz Amateur-Satellite, Amateur, Radio Astronomy, Radiolocation, Space Research (Space-To-Earth) TTT, TLPR, Railway applications, LPR, SRR, GBSAR, Amateur-satellite, Amateur, Radiolocation (civil), Radio astronomy |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.2.2 Sharing and compatibility studies already available | ECC Report 176 [i.17] from 2012 considered the impact of non-specific SRDs on radio services in the band 57 - 66 GHz. This report was the basis for the entry in Annex 1 of ERC/REC 70-03 [i.2] from 57 - 64 GHz. The potential of interference from non-specific SRDs to the Fixed Service in the frequency range 64 - 66 GHz and the lack of information relating to the deployment of Fixed Service links in this frequency range, led to the proposal not to include 64 - 66 GHz in the proposed extension of Annex 1 of ERC/REC 70-03 [i.2] for non-specific SRDs. For the protection of Fixed Service, a maximum output power limit of 10 dBm was proposed. In 2014 ECC Report 222 [i.10] was published, following compatibility studies performed on the impact of airborne surveillance radar in the 76 - 79 GHz frequency range on radio systems and services. This concluded that to protect the RAS stations in Europe an exclusion zone should be implemented around RAS installations operating in the 76 - 79 GHz band, and 10 European RAS sites exist. The report did not conclude on the size of the exclusion zones and provided a procedure to determine at national level the size of the exclusion zone. However, the altitude of the rotorcraft has an essential impact on the separation distance; example calculations have shown at altitude 300 m a separation distance of 98 km, and at altitude 0 m a separation distance of 29 km. Our proposed maximum operating height is 120 m. In 2017 ECC Report 262 [i.1] was published following a co-existence study conducted with SE24. The study related to surveillance radar equipment operating in the 76 - 77 GHz range for fixed transport infrastructure. The fixed radars considered in this study have a mounting location of approximately 5 m above the road surface and 2 - 3 m laterally from the first running lane. The executive summary states that the incident power that may be received by an automotive radar from this fixed radar installation is of the same order of magnitude than can be received from a second automotive radar. However, the reports concluded that the scanning nature of the fixed installation radar contributes to the coexistence with automotive radars, as an interference mitigation method; this has led to a regulation for fixed infrastructure radars which requires fixed transportation infrastructure radars to be of a scanning nature in order to limit the illumination time and ensure a minimum silent time to achieve coexistence with automotive radar systems. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.2.3 Sharing and compatibility issues still to be considered | Based on the generic nature of ERC/REC 70-03 [i.2] Annex 1 band n1, airborne use is already permitted. ETSI assumes that the impact of drones on other users in the band 57 - 64 GHz need not be assessed. In the band 76 - 77 GHz the coexistence with existing and possible new applications should be considered. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 21 |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3 Transmitter parameters | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3.1 Transmitter output power / radiated power | Two cases are presented in the technical parameters clause above, dependent upon the operating band: • Medium range radar, 76 - 77 GHz: peak power to 55 dBm eirp and mean power to 50 dBm. • Short range radar, 57 - 64 GHz: mean power to 20 dBm eirp. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3.2 Antenna characteristics | • Medium range radar, 76 - 77 GHz: Antenna Gain typical 10 dBi to 25 dBi. • Short range radar, 60 - 64 GHz: Antenna Gain typical 10 dBi for 20 dBm eirp. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3.3 Operating frequency | 76 - 77 GHz for the SAA radar operating to medium range (that is 2 500 m). Having multiple functionalities from a single onboard UAS low SWaP-s sensor is advantageous as it reduces the overall sensor payload weight and increases UAS mission time. Radar Altimeters and Landing Radar have been implemented in the 60 - 64 GHz band, though could equally occupy 76 - 77 GHz. 57 - 64 GHz for short range SAA, or altimeters, or take-off/Landing flight support. The FCC rule making [i.6] has permitted radar devices deployed on UAS to operate within the frequency band 60 - 64 GHz, provided that the transmitter does not exceed 20 dBm peak e.i.r.p. The sum of continuous transmitter off-times, each of at least 2 ms, should equal at least 16,5 ms within any contiguous interval of 33 ms. Operation is limited to a maximum of 122 m (400 feet) AGL. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3.4 Bandwidth | The overall bandwidth is defined by the FM sweep pattern. This is typically in the range of 150 - 1 000 MHz for Medium or Short range SAA through 76 GHz radar, or up to 4 000 MHz from 60 - 64 GHz to support Short range SAA. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.3.5 Unwanted emissions | Unwanted emissions would be within the limits for out of band specified by ETSI EN 303 883-1 [i.16] and spurious emissions aligned with ERC/REC 74-01 [i.11]. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 7.4 Receiver parameters | Common radar models include a bi-static, dual antenna configuration. The radar receiver includes an active mixer that converts the Radio Frequency signal into an Intermediate Frequency range which covers to 15 MHz. For commonly available low SWaP-c SoC devices, the receiver Noise Figure is typically 12 dBm at 1 MHz. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 8 Radio spectrum request and justification | A low SWaP-c radar is required for UAS SAA in higher Air Risk Category operations, and is required by UAS regulators, under JARUS SORA [i.9]. Detection distances of up to 2 500 m for othership targets (targets other than the ego aircraft) including light aircraft will be required, as shown in Annex A. No suitable band currently exists. In 76 - 77 GHz a 2 500 m detection range is required (see Annex A) for 10 dBsm large objects (as defined in RTCA DO-366 [i.8]), to comply with JARUS SORA [i.9] in Air Risk Categories -c and -d. In these cases, a medium range radar is required, which leads to a transmit power requirement of 55 dBm peak. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 22 Shorter range radars are defined up to 150 m detection distances, where lower power less than 20 dBm is acceptable and operation within the 57 - 64 GHz or 60 - 64 GHz bands. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 9 Regulations | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 9.1 Current regulations | |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 9.1.1 ERC Recommendation 70-03 | The band 76 - 77 GHz is already used by many applications including ground-based vehicle and TTT infrastructure systems (ERC/REC 70-03 [i.2] Annex 5), obstruction/vehicle detection via radar sensor at railway level crossings (ERC/REC 70-03 [i.2] Annex 4), obstacle detection radars for rotorcraft use (ERC/REC 70-03 [i.2] Annex 5), HD-GBSAR (ERC/REC 70-03 [i.2] Annex 6) and LPR/TLPR (ERC/REC 70-03 [i.2] Annex 6). The technology that is discussed in the present document is that which is already in use in the 76 - 77 GHz band but limited to certain applications. There therefore exists a large body of experience in manufacturing and use of radars in this band. Dedicated semiconductor devices are available from several manufacturers. Design and manufacture of antenna systems has been perfected. In terms of technology, radars are typically FMCW with an RF power of the order of 10 - 100 mW. They all rely on digital processing, such as FFTs, to extract target information from the reflected radar signals and for post processing of this information. In terms of systems, the main applications are TTT infrastructure (ERC/REC 70-03 [i.2] Annex 5), HD-GBSAR (ERC/REC 70-03 [i.2] Annex 6) and radar sensors at railway level crossings (ERC/REC 70-03 [i.2] Annex 4), in addition to use on Manned Rotorcraft. The main application for mass market equipment is on ground-based vehicles (ERC/REC 70-03 [i.2] Annex 5). With the band listed in multiple Annexes in ERC/REC 70-03 [i.2], there is potential for uncertainty in exactly what applications and uses are permitted. The position with respect to uncrewed rotorcraft is also anomalous, when compared to ground vehicles. For the former, more restrictive peak and average e.i.r.p. are mandated for (manned) rotorcraft when compared with the numerous automotive radar that successfully manage mutual interference on a busy highway. Furthermore, it is, specifically, required that rotorcraft are 'manned', precluding useful deployment onto UAS systems, the uptake of which cannot have been foreseen at the time that ETSI TR 103 137 [i.5] was last updated in 2014. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 9.1.2 ITU study on UAS | In November 2010, ITU-R prepared a report [i.7] that identified bands that would be useful for UAS SAA, see also [i.9]. This report was possible because of work from RTCA special committee 203 (SC-203). The bands identified for airborne SAA applications on larger UAS were: • 4 200 - 4 400 MHz (see Table 3 of ITU Report [i.7]) • 5 350 - 5 470 MHz (see Table 3 of ITU Report [i.7]) • 8 750 - 8 850 MHz (see Table 3 of ITU Report [i.7]) • 9 300 - 9 500 MHz (see Table 3 of ITU Report [i.7]) • 13,25 - 13,40 GHz (see Table 3 of ITU Report [i.7]) • 15,40 - 15,70 GHz (see Table 5 of ITU Report [i.7]) ETSI ETSI TR 104 078 V1.1.1 (2025-06) 23 A further 2 bands are also offered for SAA by UAS by RTCA in standard RTCA DO-366 [i.8], being bands for general radio navigation, including aeronautical, maritime, and land navigation: • 24,24 - 24,65 GHz • 32,3 - 33,4 GHz A summary of the situation in these last 2 bands (as found in 2017) only follows, since the lower frequency bands are unlikely to be suitable for low SWaP-c SAA. Componentry and Antenna will be too large: • 24,24 - 24,65 GHz available for SAA applications in Americas and Asia, but not in Europe, Africa, of Middle East. SAA would have to share spectrum with land based maritime and land-based radio navigation systems as well as inter-satellite links in the US and fixed and mobile systems in other areas of the world. Available bandwidth is insufficient for the use cases presented. • 32,3 - 33,4 GHz available for SAA applications. SAA would have to share spectrum with land based maritime and land-based radio navigation systems as well as inter-satellite links. Available bandwidth is insufficient for the use cases presented. RTCA DO-366 [i.8] comments that there may be additional bands at higher frequencies (with radionavigation protection); however additional international and domestic rule making would be required to use such spectrum. |
a40bb51101196cec3be4d8fd3686e1c8 | 104 078 | 9.2 Proposed regulation and justification | 76 - 77 GHz: • For medium range radars (up to 2 500 m detection distance) it is proposed that the entry e1 of ERC/REC 70-03 [i.2] Annex 5 be extended to permit use onboard UAS. • The mean power of 50 dBm is defined over the signal repetition time. 57 - 64 GHz: • For shorter range radar (up to 150 m detection distance), ETSI assumes that the entry n1 of ERC/REC 70-03 [i.2] Annex 1 permits use onboard UAS. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 24 Annex A: ATAR - See and Avoid A.1 Concept of operations Micro-radar and sensing systems for the Uncrewed Aerial System (UAS) industry are currently in development. ATAR will provide a way to detect other aircraft and airspace users who may, or may not be, 'co-operative' i.e. detection occurs independently of any requirement upon the 'intruder's' equipage or response. It addresses only the 'detect' aspect of See And Avoid (SAA). It provides multiple simultaneous target acquisition and tracking data using a small solid state primary radar. The intention is that the detection data is subsequently processed either by a human operator, or an automated SAA system, which then performs the appropriate decision making and any avoidance functions. The mass, size and power consumption of the radar makes it possible to fit this equipment to 'small' UAS for use inflight. NOTE: 'Small' in this context refers to sub 25 kg UAs. Examples are shown in clause A.3. The objective of our current engagement with the Civil Aviation Authority (CAA) is to establish appropriate and proportionate criteria for: • Minimum Operational Performance Standards (MOPS); • Software (SW) standards; and • Airborne Electronic Hardware (AEH) standards. A.2 Operating environment The intended operating environment is: • below 500 feet Above Ground Level (AGL); and • outside approach and departure paths for licenced airports and heliports. Encounter aircraft, the activities they are anticipated to be performing, and their characteristic performance within the operating environment, are shown below in Table A.1. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 25 Table A.1: Intruder characteristics Activity Example aircraft type Example aircraft model RTCA DO-366 [i.8] aircraft classification (section 2.2.7) Performance characteristics Maximum speed considered for operations below 500 feet AGL Helicopter Emergency Medical Services (HEMSs) at low level due weather or on approach /departure from a non-licenced heliport Helicopter Airbus EC145 Large Approach, initial climb: 65 kts Normal cruise: 128 kts 95 kts (175 km/h) National Police Air Service (NPAS) at low level due weather, conducting search activity or on approach /departure from a non- licenced heliport Helicopter Airbus EC145 Large Approach, initial climb: 65 kts Normal cruise: 128 kts 95 kts (175 km/h) Infrastructure inspection e.g. pipeline or on approach /departure from a non-licenced heliport Helicopter Airbus EC135 Large Approach, initial climb: 65 kts Normal cruise: 122 kts 95 kts (175 km/h) GA aircraft on approach /departure from a non-licenced airstrip, or conducting practiced forced landings Fixed wing Cessna 172 Medium Approach (with flaps): 70 kts Initial climb: 74 kts Emergency landing: 70 kts Normal cruise: 120 kts 75 kts (139 km/h) GA aircraft on approach /departure from a non-licenced heliport, or conducting practiced auto-rotations Helicopter Robinson R44 Medium Take-off, initial climb and landing: 60 kts Autorotation: 70 kts Normal cruise: 110 kts 75 kts (139 km/h) Microlight aircraft on approach /departure from a non-licenced airstrip, or conducting practiced forced landings EuroFOX Small Approach and initial climb: 65 kts Cruise: 80 kts 65 kts (120 km/h) Microlight aircraft in low level cruise Powered hang glider Joker Trike Small Cruise: 48 kts 48 kts (89 km/h) Hot air balloons in low level cruise Balloon Small Maximum operating wind speed 10 kts 10 kts (19 km/h) A.3 Ownship characteristics The ownship to which the ATAR would be fitted are typically small UAS. These may be multi-rotor or lift-cruise UA with the ability to stop and or change direction in an agile way. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 26 A.4 Special condition - ATAR qualification per RTCA DO-366 RTO/DO-366 [i.8] was primarily written for a specific type of UAS (large, fast moving fixed wing UAS), engaged in operations above 500 feet AGL and considering fast moving intruder aircraft. Therefore, some of the performance specifications are inappropriate for small UAS operating at very low altitudes. A.5 Requirements for modification The affected requirements of RTCA DO-366 [i.8] include as follows: • 1.2.4: This requirement limits radar performance to above 500 feet. • 1.7.2: This requirement defines an alerting time of 85 seconds. • 2.2.3: This requirement defines the specific frequency bands for radar within the range 4 200 MHz to 33,4 GHz. • 2.2.6-2: This requirement defines the Field of Regard and the altitude at which it is effective (> 1 000 ft). • 2.2.7-11: This requirement defines the RDR for a small othership and an ownship UA turn rate of 3 degrees/s. This gives an RDR of 5,4 NM (10 km). • 2.2.7-12: This requirement defines the RDR for a medium othership and an ownship UA turn rate of 3 degrees/s. This gives an RDR of 6 NM (11,1 km). • 2.2.7-13: This requirement defines the RDR for a large othership and an ownship UA turn rate of 3 degrees/s. This gives an RDR of 6,7 NM (12,4 km). A.6 Proposed deviations The following are the proposed deviations to the requirements of RTCA DO-366 [i.8] listed above, along with the justification as to why the performance and safety of the system is unaffected: • 1.2.4: This requirement limits radar performance to above 500 feet. The operational environment of the UAS the radar is designed for is between ground level and 500 feet AGL, therefore the radar performance will be designed for up to 500 feet and intruders will only be considered in this operating altitude range. This provides an equivalent level of safety to the requirements of RTCA DO-366 [i.8]. • 1.72: This requirement defines an alerting time of 85 seconds, the radar has a much lower detect time which will be used. The 85 seconds takes into multiple factors and the operational requirements of a different vehicle and type of operation which is not relevant to the smaller and more responsive UAS which would have a response time of around 3 seconds time from the track being detected to the pilot being alerted. • 2.2.7-11: The Small Othership RDR is determined by UAS performance and the speeds at which a 'small' intruder may reasonably be anticipated to be operating at in the same low-level environment. Based on a worst-case scenario of a head-on encounter (i.e. minimum time) and no manoeuvring by the othership, a lower RDR of 1 000 m (0,54 NM) has been selected. • 2.2.7-12: The Medium Othership RDR is determined by UAS performance and the speeds at which a 'medium' intruder may reasonably be anticipated to be operating at in the same low-level environment. Based on a worst-case scenario of a head-on encounter (i.e. minimum time) and no manoeuvring by the othership, a lower RDR of 1 900 m (1 NM) has been selected. • 2.2.7-13: The Large Othership RDR is determined by UAS performance and the speeds at which a 'large' intruder may reasonably be anticipated to be operating at in the same low-level environment. Based on a worst-case scenario of a head-on encounter (i.e. minimum time) and no manoeuvring by the othership, a lower RDR of 2 500 m (1,3 NM) has been selected. ETSI ETSI TR 104 078 V1.1.1 (2025-06) 27 Annex B (informative): Bibliography • ETSI TR 103 148 (V 1.1.1): "Electromagnetic compatibility and Radio spectrum Matters (ERM); System Reference document (SRdoc); Technical characteristics of Radio equipment to be used in the 76 GHz to 77 GHz band; Short-Range Radar to be fitted on fixed transport infrastructure". ETSI ETSI TR 104 078 V1.1.1 (2025-06) 28 History Document history V1.1.1 June 2025 Publication |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 1 Scope | The present document identifies Resource Management (RM) at the Facilities Layer in relation to critical control requirements, capabilities, principles and parameters which could enable the definition of a mechanism supporting highly time and size dynamic data exchanging message services to operate robust, interoperable and backward compatible with existing ITS Release 1 and upcoming Release 2 message services in the 5,9 GHz ITS allocated band. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 2 References | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 2.1 Normative references | Normative references are not applicable in the present document. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 2.2 Informative references | References are either specific (identified by date of publication and/or edition number or version number) or non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the referenced document (including any amendments) applies. NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee their long-term validity. The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's understanding but are not required for conformance to the present document. [i.1] ETSI TR 103 439: "Intelligent Transport Systems (ITS); Multi-Channel Operation Study; Release 2". [i.2] ETSI TR 101 607: "Intelligent Transport Systems (ITS); Cooperative ITS (C-ITS); Release 1". [i.3] ISO 26262-9:2018: "Road vehicles — Functional safety, "Part 9: Automotive safety integrity level (ASIL)-oriented and safety-oriented analyses". [i.4] ETSI TR 103 903: "Intelligent Transport Systems (ITS); Framework; Basic principles; Release 2". [i.5] ETSI TS 103 141: "Intelligent Transport Systems (ITS); Facilities layer function; Multi-Channel Operation (MCO) for Cooperative ITS (C-ITS); Release 2". [i.6] ETSI TS 103 697: "Intelligent Transport Systems (ITS); Architecture; Multi-Channel Operation (MCO) for Cooperative ITS (C-ITS); Release 2". [i.7] Car2Car Communication Consortium: "Basic System Profile (BSP)" - C2C-CC Profile Release 1. [i.8] C-Roads profiles: "Release 2 of C-Roads Harmonised Communication Profile for C-ITS". [i.9] IEEE 802.11™-2025: "IEEE Standard for Information Technology -- Telecommunications and Information Exchange Between Systems Local and Metropolitan Area Networks -- Specific Requirements - Part 11: Wireless Local Area Network (LAN) Medium Access Control (MAC) and Physical Layer (PHY) Specifications", April 2025. [i.10] ETSI TS 103 836-4-1: "Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 4: Geographical addressing and forwarding for point-to-point and point-to- multipoint communications; Sub-part 1: Media-Independent Functionality; Release 2". [i.11] ETSI TS 103 695: "Intelligent Transport Systems (ITS); Access layer specification in the 5 GHz frequency band; Multi-Channel Operation (MCO) for Cooperative ITS (C-ITS); Release 2". [i.12] ETSI TR 101 612: "Intelligent Transport Systems (ITS); Cross Layer DCC Management Entity for operation in the ITS G5A and ITS G5B medium; Report on Cross layer DCC algorithms and performance Evaluation". ETSI ETSI TR 104 073 V2.2.1 (2026-02) 11 [i.13] ETSI TS 102 636-4-2: "Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 4: Geographical addressing and forwarding for point-to-point and point-to- multipoint communications; Sub-part 2: Media-dependent functionalities for ITS-G5". [i.14] ETSI TS 102 687 (V1.2.1): "Intelligent Transport Systems (ITS); Decentralized Congestion Control Mechanisms for Intelligent Transport Systems operating in the 5 GHz range; Access layer part". [i.15] ETSI TS 103 175: "Intelligent Transport Systems (ITS); Cross Layer DCC Management Entity for operation in the ITS G5A and ITS G5B medium". [i.16] ETSI EN 302 571: "Intelligent Transport Systems (ITS); Radiocommunications equipment operating in the 5 855 MHz to 5 925 MHz frequency band; Harmonised Standard covering the essential requirements of article 3.2 of Directive 2014/53/EU". [i.17] V. Todisco, S. Bartoletti, C. Campolo, A. Molinaro, A. O. Berthet and A. Bazzi: "WiLabV2Xsim: Performance Analysis of Sidelink 5G-V2X Mode 2 Through an Open-Source Simulator", IEEE™ Access, vol. 9, pp. 145648-145661, 2021, doi: 10.1109/ACCESS.2021.3121151. [i.18] Z. Wu, S. Bartoletti, V. Martinez, A. Bazzi: "A Methodology for Abstracting the Physical Layer of Direct V2X Communications Technologies", MDPI Sensors 2022, 22, 9330, doi: 10.3390/s22239330. [i.19] G. Bansal, B. Cheng, A. Rostami, K. Sjoberg, J. B. Kenney and M. Gruteser: "Comparing LIMERIC and DCC approaches for VANET channel congestion control", 2014 IEEE™ 6th International Symposium on Wireless Vehicular Communications (WiVeC 2014), Vancouver, BC, Canada, 2014, pp. 1-7, doi: 10.1109/WIVEC.2014.6953217. [i.20] M. Sepulcre, J. Mira, G. Thandavarayan, and J. Gozálvez: "Is Packet Dropping a Suitable Congestion Control Mechanism for Vehicular Networks?", 2020 IEEE™ 91st Vehicular Technology Conference (VTC2020-Spring), Antwerp, Belgium, 2020, pp. 1-5, doi: 10.1109/VTC2020-Spring48590.2020.9128822. [i.21] G. Thandavarayan, M. Sepulcre, and J. Gozálvez: "Cooperative Perception for Connected and Automated Vehicles: Evaluation and Impact of Congestion Control", IEEE™ Access, vol. 8, pp. 197665-197683, October 2020, doi: 10.1109/ACCESS.2020.3035119. [i.22] F. Marzouk, R. Zagrouba, A. Laouiti, and P. Mühlethaler: "An Empirical Study of Unfairness and Oscillation in ETSI DCC", 2015 International Conference on Performance Evaluation and Modeling in Wired and Wireless Networks (PEMWN'2015), November 2015. [i.23] A. Bazzi: "Congestion Control Mechanisms in IEEE 802.11p™ and Sidelink C-V2X", 53rd Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, USA, 2019, pp. 1125-1130, doi: 10.1109/IEEECONF44664.2019.9048738. [i.24] E. Xhoxhi and F. Schiegg: "Benefits of DCC Facilities in ITS-G5 Networks - First Simulated Results", IEEE™ 95th Vehicular Technology Conference: (VTC2022-Spring), Helsinki, Finland, 2022, pp. 1-7, doi: 10.1109/VTC2022-Spring54318.2022.9860904. [i.25] G. Thandavarayan, M. Sepulcre, J. Gozálvez and B. Coll-Perales: "Scalable Cooperative Perception for Connected and Automated Driving", Elsevier Journal of Network and Computer Applications, vol. 216, pp. 103-655, May 2023, doi: 10.1016/j.jnca.2023.103655. [i.26] A. H. Sakr: "Evaluation of Redundancy Mitigation Rules in V2X Networks for Enhanced Collective Perception Services", IEEE™ Access, vol. 12, pp. 137696-137711, 2024, doi: 10.1109/ACCESS.2024.3464514. [i.27] Q. Delooz. A. Willecke, K. Garlichs, A.-C. Hagau, L Wolf and A. Vinel: "Analysis and Evaluation of Information Redundancy Mitigation for V2X Collective Perception", in IEEE™ Access, vol. 10, pp. 47076-47093, 2022, doi: 10.1109/ACCESS.2022.3170029. [i.28] A. Figueiredo, P. Rito, M. Luís and S. Sargento: "Enhancing Vehicular Network Efficiency: The Impact of Object Data Inclusion in the Collective Perception Service", IEEE™ Open Journal of Intelligent Transportation Systems, vol. 5, pp. 454-468, 2024, doi: 10.1109/OJITS.2024.3437206. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 12 [i.29] F. Schiegg et al.: "Automated Vehicle Marshalling: The First Functionally Safe V2X Service for Connected Automated Driving", IEEE™ Open Journal of Vehicular Technology, vol. 6, pp. 927-947, 2025, doi: 10.1109/OJVT.2025.3538698. [i.30] G. Bansal and J. B. Kenney: "Achieving Weighted-Fairness in Message Rate-Based Congestion Control for DSRC Systems", 2013 IEEE™ 5th International Symposium on Wireless Vehicular Communications (WiVeC), Dresden, Germany, 2013, doi: 10.1109/wivec.2013.6698225. [i.31] Directive 2010/40/EU of the European Parliament and of the Council of 7 July 2010 on the framework for the deployment of Intelligent Transport Systems in the field of road transport and for interfaces with other modes of transport. [i.32] ETSI EN 303 797 (V2.1.1): "Intelligent Transport Systems (ITS); ITS-G5 Access layer in the 5 GHz frequency band; Release 2". [i.33] C. Campolo, A. Molinaro and R. Scopigno: "Vehicular ad hoc Networks, Standards, Solutions, and Research", Springer, 2015, ISBN: 978-3-319-15496-1 (Print), 978-3-319-15497-8 (Ebook), doi: 10.1007/978-3-319-15497-83. [i.34] Car2Car Communication Consortium: "Vehicle C-ITS Station Profile". [i.35] D. Smely, S. Rührup, R. K. Schmidt, J. Kenney and K. Sjöberg: "Decentralized Congestion Control Techniques for VANETs". In: Campolo, C., Molinaro, A., Scopigno, R. (eds) Vehicular ad hoc Networks. Springer, Cham, 2015, doi: 10.1007/978-3-319-15497-8_6. [i.36] R. K. Schmidt, T. Leinmüller and G. Schäfer: "Adapting the Wireless Carrier Sensing for VANETs", 6th International Workshop on Intelligent Transportation (WIT), Hamburg (2010). [i.37] D. Eckhoff, N. Sofra and R. German: "A Performance Study of Cooperative Awareness in ETSI ITS G5 and IEEE WAVE", 10th Annual Conference on Wireless On-demand Network Systems and Services (WONS), 2013, doi: 10.1109/WONS.2013.6578347. [i.38] Yongtae Park and Hyogon Kim: "Preventing congestion control oscillation in cellular vehicular communication", Electronic Letters, November 2021, Vol. 57, November 2024, doi: 10.1049/ell2.12319. [i.39] SAE J2945/1: "Surface vehicle standard, On-board System Requirements for V2V Safety Communications". [i.40] ETSI EN 302 636-4-1: "Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 4: Geographical addressing and forwarding for point-to-point and point-to- multipoint communications; Sub-part 1: Media-Independent Functionality". [i.41] ETSI EN 302 636-5-1: "Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 5: Transport Protocols; Sub-part 1: Basic Transport Protocol". [i.42] ETSI EN 302 636-6-1: "Intelligent Transport Systems (ITS); Vehicular Communications; GeoNetworking; Part 6: Internet Integration; Sub-part 1: Transmission of IPv6 Packets over GeoNetworking Protocols". [i.43] T. Tielert, D. Jiang, Q. Chen, L. Delgrossi and H. Hartenstein: "Design Methodology and Evaluation of Rate Adaptation Based Congestion Control for Vehicle Safety Communications", 2011 IEEE™ Vehicular Networking Conference (VNC), Amsterdam, Netherlands, 2011, pp. 116-123, doi: 10.1109/VNC.2011.6117132. [i.44] G. Bansal, J. B. Kenney and C. E. Rohrs: "LIMERIC: A Linear Adaptive Message Rate Algorithm for DSRC Congestion Control", in IEEE™ Transactions on Vehicular Technology, vol. 62, no. 9, pp. 4182-4197, November 2013, doi: 10.1109/TVT.2013.2275014. [i.45] CAMP Vehicle Safety Communications 3: "Interoperability Issues of Vehicle-to-Vehicle Based Safety Systems Project", Phase 1 Final Report, April 2014. [i.46] CAMP Vehicle Safety Communications 3: "Phase 2 Final Report Volume 1- Communications Scalability for V2V Safety Development" and "Phase 2 Final Report Volume 2- Communications Scalability for V2V Safety Analysis", February 2016. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 13 [i.47] ETSI TR 103 970 (V2.1.1_0.11): "Intelligent Transport Systems (ITS); Feasibility study on the use of UWB; Release 2". [i.48] IETF RFC 1042: "Standard for the transmission of IP datagrams over IEEE 802 networks". |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 3 Definition of terms, symbols and abbreviations | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 3.1 Terms | For the purposes of the present document, the following terms apply: application: computer software program that performs a specific task directly for an user or one another application automated system: computer (ICT) system that collects information and can react and perform tasks based on the data NOTE: Automated systems are made using three entities: Sensors. Microprocessors. Actuators. communication flows: data path to be followed by the information through an ITS from the source ITS-S to the sinking ITS-Ss direct communication: message exchange between data sourcing ITS-Ss and data sinking ITS-Ss, without involvement of intermediary networking routing functions NOTE: From a functional perspective, messages transported via Indirect Communication can be considered as Direct communicated when the messages cannot be processed (carried over in a secured envelope) along the indirect communication flow. equivalent bandwidth: computed value coming from a computation of relevant statistical properties of the traffic source (e.g. Markov-modulated arrivals, moment generating functions, or large-deviation techniques), ensuring that the probability of exceeding resource limits remains below a defined threshold NOTE: Equivalent bandwidth originates from wired networks and refers to the bandwidth that a given Message service effectively requires to meet its QoS constraints. extra ITS: external ITS-S communication using a non-interoperable ITS, or interference from non-ITS sources, also inter ITS heterogeneous: diverse in character or content indirect communication: message exchange between data sourcing ITS-Ss and data sinking ITS-Ss, with the involvement of intermediary networking routing functions such as the Internet intra ITS: internal ITS-S communication using interoperable ITS ITS constellation: group of ITS-Ss which can exchange information ITS-G5: ITS communication protocol based on IEEE 802.11 access layer ITS service: service provided by an application to an end user or automated function message service: Facilities Layer service which generates or processes data, triggered by an application or other message service, or a Facilities Layer service which executes system control functions for the purpose of correct operation of other message services or applications ETSI ETSI TR 104 073 V2.2.1 (2026-02) 14 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 3.2 Symbols | For the purposes of the present document, the following symbols apply: substitution constant parameter of the channel busy ratio limit or substitution parameter filter function factor substitution constant parameter of the channel busy ratio limit or substitution parameter parameter of the adaptive rate control function parameter of the adaptive rate control function measured channel busy ratio input for the adaptive congestion control upper limit of the channel busy ratio parameter of the adaptive rate control function cfn control function cfn control function of adaptive congestion control channel load equilibrium (i.e. steady state) channel load equilibrium channel load of the adaptive congestion control − channel load of the adaptive congestion control for the gain saturated case channel load produced by the adaptive congestion control + channel load of the adaptive congestion control for the gain saturated case envelope of channel load oscillations for the adaptive congestion control channel load of the adaptive congestion control with input filtering channel load of the adaptive congestion control for the gain saturated case measured channel busy ratio upper channel load limit upper channel occupation limit ∆ duration of a time step ∆ time step duration for network node j δmax parameter of the adaptive congestion control δmin parameter of the adaptive congestion control ∆ variation of within one time step ffn filter function ffn filter function of adaptive congestion control gain saturation of adaptive congestion control gain saturation parameter of adaptive congestion control gain saturation parameter of adaptive congestion control j enumerator for network nodes k enumerator for network nodes and time steps offset coefficient number of network nodes number of steps number of useful bits carried by an Orthogonal Frequency Division Multiplexing (OFDM) symbol of an ITS-G5 transmission number of steps to reach convergence number of steps to reach convergence number of steps to reach convergence number of bytes of the payload of an ITS-G5 frame estimated number of network nodes number of ITS-Stations (I.e. network nodes) contributing to DCC channel resources target message rate of the adaptive congestion control transmission rate of network node j j equilibrium rate of network node j sj equilibrium transmission rate of network node j for the gain saturated case of the adaptive congestion control time ETSI ETSI TR 104 073 V2.2.1 (2026-02) 15 duration of a time step, i.e. scaling factor for a limit transition integration constant having a unit of time integration constant having a unit of time integration constant having a unit of time integration constant having a unit of time parameter for the control function of the reactive congestion control parameter for the control function of the reactive congestion control convergence time duration of the transmission of the PLCP preamble and header of an ITS-G5 frame time after n time steps time step size used for numeric solutions time after n time steps for network node j !"# duration of an OFDM symbol of an ITS-G5 transmission transmission idle time equilibrium (i.e. steady state) transmission idle time upper transmission idle time limit upper limit lower limit offset coefficient total channel idle time observed during a measurement period transmission duration upper transmission duration limit total channel active time observed during a measurement period $ duration of the transmission of an ITS-G5 frame channel utilization equilibrium channel utilization % time derivative of channel utilization load of the adaptive congestion control channel utilization load of the adaptive congestion control for network node j $ channel utilization of network node k upper channel utilization limit dynamic upper channel utilization limit |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 3.3 Abbreviations | For the purposes of the present document, the following abbreviations apply: 3GPP 3rd Generation Partnership Project 5G-NR 5th Generation New Radio AC Admission Control AI Artificial Intelligence AIFS Arbitration Inter-Frame AL Access Layer ALI Access Layer Instance ASIL Automation Safety Integrity Level AVM Automated Vehicle Marshalling BE Best Effort BME Bandwidth Management Entity BTP Basic Transport Protocol C2C-CC Car2Car Communication Consortium CA Cooperative Awareness CAM Cooperative Awareness Message CAS Cooperative Awareness Service CBR Channel Busy Ratio CC Congestion Control CDF Cumulative Distribution Function C-ITS Cooperative Intelligent Transportation Systems CL Channel Load CP Collective Perception ETSI ETSI TR 104 073 V2.2.1 (2026-02) 16 CPM Collective Perception Message CPS Collective Perception Service CSMA/CA Carrier Sense Multiple Access with Collision Avoidance CW Contention Window DCC Decentralized Congestion Control DEN Decentralized Environmental Notification DENM Decentralized Environmental Notification Message DENS Decentralized Environmental Notification Service DoS Denial of Service ECC Electronic Communications Committee EIRP Effective Ideal Radiated Power EU European Union FAC Facility FL Facilities Layer FL-SDU Facilities Layer-Service Data Unit FUSA Functional Safety GCRA Generic Cell Rate Algorithm GPC GNSS Positioning Correction I2V Infrastructure to Vehicle I2X Infrastructure to Everything ICT Internet Communication Technology ITS Intelligent Transportation Systems ITS-S Intelligent Transportation Systems - Station IVI Infrastructure to Vehicle Information IVIM In Vehicle Information Message LLC Logical Link Control LTE Long-Term Evolution MAC Medium Access Control MAP Road topology MAPEM MAP (topology) Extended Message MCM Manoeuvre Coordination Message MCO Multi-Channel Operation MCS Modulation and Coding Scheme MIM Marshalling Infrastructure Message MSDU MAC Service Data Unit MTU Maximum Transmission Unit MVM Marshalling Vehicle Message NR New Radio NTL Networking & Transport Layer OFDM Orthogonal Frequency Division Multiplexing PDCP Packet Data Convergence Protocol PER Packet Error Rate PHY Physical layer PLCP Physical Layer Convergence Protocol PRR Packet Reception Ratio QAM Quadrature Amplitude Modulation QM Quality Management QoS Quality of Service QPSK Quadrature Phase Shift Keying RLC Radio Link Control RM Resource Management RSSI Received Signal-Strength Indicator RSU Roadside Unit RTCM Radio Technical Commission for Maritime services RTCMEM RTCM Extended Message RWW Roads Works Warning SAM Service Announcement Message SDV Slow Driving Vehicle SHB Single Hop Broadcast SINR Signal to Noise Interference Ratio SNAP Subnetwork Access Protocol SP System Profile ETSI ETSI TR 104 073 V2.2.1 (2026-02) 17 SPATEM Signal Phase And Timing Extended Message SREM Signal Request Extended Message SRTI Safety Related Traffic Information SSEM Signal request Status Extended Message TB Transport Block TC Traffic Class TDRC Transmit Data-Rate Control TLC Traffic Light Control TPC Transmit Power Control TR Technical Report TRC Transmit Rate Control TS Technical Specification UWB Ultra Wide Band V2V Vehicle to Vehicle V2X Vehicle to Everything VAM Vulnerable road user Awareness Message VoI Value of Information VRU Vulnerable Road User X2N2X Anything to Network to Anything |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4 Considerations | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.1 Introduction | The ITS Release 1 basic cooperative use cases, supported by the standards listed in ETSI TR 101 607 [i.2], are awareness oriented. When considered from an Automotive Safety Integrity Level (ASIL) perspective, according to ISO 26262-9 [i.3], ITS Release 1 supports some use cases of the Quality Management (QM) ASIL. For this level, there are no liability or other quality requirements defined, as identified in the ITS Framework Release 2 ETSI TR 103 903 [i.4]. The ITS Framework Release 2 (addressing both Cooperative-ITS (C-ITS) and non-C-ITS use cases), detailed in ETSI TR 103 903 [i.4], identifies specific Release 2 ITS services that are expected to support ASIL C and potentially ASIL D. The support of higher ASIL levels results in additional and more stringent data quality and system performance requirements, including additional QoS requirements. The present document focusses on two QoS related system improvement aspects: 1) how the ITS Service information dissemination could be better scheduled statistically; and 2) at the ITS-Station (ITS-S) system level, how the information dissemination could be better managed dynamically. Both aspects are system aspects from which the first should be specified in System Profiles (SPs). For Release 1 the C2C-CC Profiles [i.7] and the C-Roads Profiles [i.8] have specified this to enable a predictable static message exchange behaviour. Both profiles refer to the ETSI standards listed in the ETSI TR 101 607 [i.2] for Release 1. With regards to the second aspect, aspects related to the influence of ITS Services information exchange are identified in clause 4.2. Aspects related to the information dissemination QoS are clarified in clause 4.3. Finally, in clause 4.4, the system context is illustrated. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.2 Message handling | Message handling is the static setting of C-ITS information exchange in a specific environmental context. As stated in clause 4.1, the C2C-CC and C-Roads profiles limit the Release 1 information dissemination to V2V, I2V message exchange in a single direct communication specific 10 MHz channel. C-Roads identifies other I2X/X2I direct communication into other channels for early C-Roads Release 2 within the same C-ITS Ecosystem (in EU regulated by the ITS directive [i.31]) to realize the required QoS. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 18 In this context the C-Roads hybrid profile [i.8] (Internet based) is not considered at present as it does not provide dissemination QoS mechanisms for higher levels of safety and automation as expected for Release 2. At present according to current Functional Safety Assessments (FUSAs), Internet information exchange does not provide sufficient QoS to be used for safety related information exchange to satisfy higher ASIL levels. As such related information exchange is not considered. At the system level QoS is realized by means of stakeholder agreements and technical profiles such as the one from C2C-CC and C-Roads. As such they cannot be considered. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.3 Quality of Service improvement | In Information and Communication Technology (ICT), QoS cannot be fully guaranteed. The main three areas, in which QoS mechanisms are realized: 1) From a functional perspective internet communication capabilities are business driven statistically estimated and adjusted ensuring a general level of QoS. Initially this was only statically realized and later extended to support exceptional (large) events such as a football game with over 50 000 spectators and based on related statistics. For instance, when there is a large sport event, additional equipment is temporarily installed. This can be seen as scenario management. 2) Internet protocol communication in general is handshaking based which allows the disseminating station to verify whether information is received. Direct communications are sensor networked (broadcast/unicast) based in which this verification cannot take place. QoS is reached by system agreements and resource assessment. 3) From a technical perspective information can for example be given different priorities or be offered different communication capabilities fitting to the information behaviour. This however always goes with a cost in which lower priorities will have to pay for the burden of the higher priorities. In an open system such as the internet the large number of users (millions of users) damp the dynamic behaviour in general which has a positive effect on the QoS, by which the above three mechanisms are sufficient to realize an acceptable QoS for most internet services. Information exchange in the domain of transport has additional requirements up on standard internet-oriented services. ITS related QoS, depending on the use case, has more stringent requirements related to latency, range, data rate, ranging accuracy, positioning accuracy, the number of participating devices, and to many other communication parameters. Depending on the transport use case, they often have a safety impact and therefore require higher levels of QoS. While some of the ITS Release 1 use cases can still be realized with an internet level of QoS, extended ITS Release 2 use case QoS requirements are more demanding. One aspect of satisfying automated mobility systems is to provide higher levels of QoS to increase the reception probability so that information can be disseminated and reaches the users of that information in time. This presumes the accessibility of the network and that there is enough bandwidth available. Enough bandwidth is also reducing the sensitiveness to Denial-of Service (DoS, a typical cyber-attacks) attacks which make use of the bandwidth limitations of a network. Hence, only when common agreements in a closed communication environment are guaranteed, a high level of QoS can be reached. In case of safety related services, specific levels of QoS are required as otherwise the safety related operation cannot be guaranteed. As QoS are coming with a price, discussions are started to what level of costs the safety should be guaranteed. This is a society cultural question and generally out of scope of standardization. However, C-ITS is of interest for a specific group of stakeholders and not to the society as a hole, which allowed the user to come to an agreement about the requirement levels for QoS. For advanced C-ITS services and automated systems several approaches are being developed to realize a controlled system environment for extended trustworthiness and QoS needed for Release 2. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 19 In the C-ITS ecosystem the message behaviour depends on mostly non-predictable events and circumstances. Within this ecosystem various parties are interested in different C-ITS services and make use of the communications capabilities in different ways. To realize a trustworthy communication behaviour, organizational agreements (regulations and/or requirements from consortia) for establishing a stable environment are required. As the C-ITS information dissemination is very dynamic in time and size, additional technical measures are needed to enable applications and message services to be aware of the communication capabilities and to allow them to take the appropriate measures in any traffic scenario. As a result, for C-ITS, but also for other message services, resource management methods will increase the dissemination trustworthiness. As this is a system related trustworthiness, each participating ITS-S needs to address the trustworthiness and QoS in a standardized manner as otherwise the predicted resulting behaviour of a group of ITS-S's cannot be trusted. A basic trustworthiness and QoS improvement method is to increase the media (radio) access probability by e.g. increasing the communication bandwidth or by realizing agreements among the stakeholders on how to use the available media. This is useful for many of the non-safety or limited safety relevant message services. As spectrum is a sparce resource, the available bandwidth for exchanging information is limited. For more critical safety requirements (ASILs C-D), this limitation can be an issue, and therefore more stringent measures are needed to realize a better QoS. The way to realize an acceptable QoS depends on the network resources, they vary depending on the network configuration e.g. direct ITS-S to ITS-S (V2X) ITS-S - Service Provider - ITS-S (X2N2X) configurations (see ETSI TR 103 903 [i.4]. It could be that all stakeholders agree on what information should be disseminated. This is currently realized based on agreements (and associated profiles) between Car2Car Communication Consortium (C2C-CC) and C-Roads in Europe. This agreement comprises which Release 1 messages are at least disseminated in the single 10 MHz channel based on the IEEE 802.11p access layer technology, which is included in IEEE 802.11 [i.9]. This does not exclude that these messages can also be distributed via X2N2X networks, in parallel. Practice has shown that for basic Safety Related Traffic Information (SRTI) information exchange both methods are useful. When moving to more safety critical and automation Release 2 use cases, where more stringent QoS requirements are necessary, X2N2X QoS capabilities are expected not to grow with these needs, and direct data exchange between data source and data sink (ITS-S to ITS-S) without involvement of networks may therefore be the only possible solution for use cases with higher QoS needs. The present document intends to introduce a method which further increases the QoS by making the applications and message services aware of the current dissemination capabilities of the access layer and whether they are influenced by other applications in the same ITS-S or in other ITS-S. The present document therefore focuses on aspects which are relevant for the definition of a Resource Management (RM) method which supports an QoS improvement. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4 The system context | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.1 Overview | Previous work related to resource management was realized by the Multi-Channel Operation (MCO) work resulting in the MCO architecture for C-ITS as specified in ETSI TS 103 697 [i.6]. Present MCO functionalities are specified in the ETSI Release 2 standards, ETSI TS 103 141 [i.5], ETSI TS 103 836-4-1 [i.10] and ETSI TS 103 695 [i.11]. Figure 1 represents the MCO architecture as specified in ETSI TS 103 697 [i.6]. The RM has effect on the aspects of various layers, as clarified in the following clauses. The MCO specifications focused on the realization of the multi-channel use. While the Resource Management (RM) extends and generalizes this concept also for use by single channel systems, focusing on increasing the QoS. The MCO architecture and related functionalities are therefore the basis for a wider resource management specification. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 20 Figure 1: MCO Architecture (Source: ETSI TS 103 697 [i.6]) RM at the Facilities Layer (FL) is a functionality maintaining dynamic knowledge about the lower layer capabilities. It realizes higher levels of QoS by advising and directing applications about their message dissemination possibilities. This enables the applications to take appropriate functional and technical measures required for the use case or use cases it realizes. To allow the RM to fulfil its tasks, it communicates to the applications as well as to relevant lower layer functionalities as required. While the RM itself is technology agnostic, it can interface with technology specific entities at lower layers, although it expects technology agnostic parameter values. For the realization of the RM processes, RM related technology agnostic and technology specific functions can be expected at all layers. From a lower layer perspective, the RM is dependent on information about the Channel Load (CL), which is an aspect related to the Access Layer (AL) to be able to realize QoS improvements. In the following clause, relevant RM specific functional and technical aspects at different layers are recognized, and relevant studies are identified. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.2 Channel load | The bandwidth of a radio channel is always limited, and it needs to be used efficiently. This means maximizing the data that can be exchanged in such channel. At the same time, overusing the channel may cause radio interference that is resulting in an increased reception error probability at the receivers. As a result, there is a maximum possible channel throughput and approaching it allows to optimally use the resources. How this is handled depends on the channel management principle that is used. The specific split of radio resources, which can be in the frequency, time, space or code domain or a combination of these splitting principles, is technology specific. In ITS-G5 based on IEEE 802.11 [i.9] technologies resources are split in the time domain with a listen-before-talk mechanism, whereas in LTE-V2X and 5G-NR-V2X sidelink there is a grid of time and frequency slots with synchronous access and scheduling procedures. In ITS Release 1, there are separate mechanisms identified for ITS-G5 and for LTE-V2X sidelink. For ITS-G5 the mechanism is specified in several ETSI specifications as identified in ETSI TR 101 607 [i.2] and for LTE-V2X sidelink by ETSI and 3GPP specifications. In Release 2, an access layer technology agnostic solution at the higher layers needs to be defined to facilitate the use of various access layer technologies in the ITS spectrum bands. This implies the need to abstract the definition of used and available resources from the access layer to the layers above. The measurement of the channel occupation and the evaluation of the available resources remain anyway technology dependent and is therefore performed in the access layer, defined separately per each technology. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 21 When looking at the single channel and access technology, it is important to note that besides the inter ITS interference coming from ITS frames using the reference ITS access technology (ITS access technology used by the measuring ITS-S), extra ITS interference needs to be taken into account. When this external interference is above a given threshold it might also contribute to the channel load. Interference generated by a competing ITS technology can be also part of this extra (inter) ITS interference since it cannot be recognized by the receiver as useful signal. In the future, a differentiation between intra and extra ITS signal and interference might be required having in mind that ITS is operating in a shared radio environment. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.3 Access layer operation | The AL is composed of the Medium Access Control (MAC) and Physical sub-layers. It handles the radio spectrum access for the transmission of frames. Higher layers can provide data packets to the access layer. The access layer decides whether to send frames, and when to transmit them by the mechanisms of the media access protocol. With the aim to identify the occupation of the channel, in many systems including ITS-G5, the Channel Busy Ratio (CBR) is a metric derived from measurements which gather information about the load on the channel. This information can be provided to higher layers to inform them about the state of the channel from the local perspective of the ITS-S specific transceiver. The measurement performed locally can also be shared through higher layers with the neighbouring stations to improve awareness of channel use, as discussed in clause 4.4.5 in more detail. In Release 1, the CBR was directly related to a system CBR limit to control the amount of data that could be sent. The validity of such limit and the solutions defined in Release 1 to guarantee the respect of such limits are further elaborated in clause 5. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.4 Adjacent channel | The MCO report ETSI TR 103 439 [i.1] presented the influence of operating information dissemination in adjacent channels and concluded that the traffic in one channel affects the resources available in the adjacent ones. The result also shows that the second adjacent channel can be neglected. This means that the maximum occupation allowed in one channel may affect the performance and thus the maximum occupation of the adjacent ones. These aspects are further investigated in clause 5.4. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.5 Networking & transport layer operation | In Release 1, the Networking & Transport Layer (NTL) is realized by the GeoNetworking (GN) protocol as specified in ETSI EN 302 636-4-1 [i.40], complemented by the Basic Transport Protocol (BTP) as specified in ETSI EN 302 636-5-1 [i.41] and by the IPv6 over GeoNetworking Adaptation Sub-Layer (GN6ASL) as specified in ETSI EN 302 636-6-1 [i.42]. Together, these protocols provide the packet transport functions between ITS-S. The GeoNetworking protocol provides geographical addressing, meaning that the destination of a packet is expressed in geographical terms such as a point or an area rather than by a conventional network address. When the addressed destination is within the direct radio coverage of the transmitting station, the packet is delivered by a single-hop transmission without any relaying. When the destination lies beyond direct radio range, intermediate stations may act as relays and apply the forwarding algorithms defined in ETSI EN 302 636-4-1 [i.40], resulting in multi-hop forwarding. The protocol maintains state information in the Location Table and manages packet handling according to the selected transmission mode. Its main tasks are: • geographical addressing of packets for unicast, broadcast and anycast communications; • single-hop transmission of packets when the destination is within direct radio coverage; • multi-hop relaying of packets when the destination is outside the direct radio range of the source, using forwarding algorithms such as greedy forwarding and contention-based forwarding; • maintenance of a Location Table (LocT) containing position vectors of neighbouring stations, extended by LocTEX-G5 fields when Decentralized Congestion Control (DCC) is active; • exchange and storage of congestion-related information, namely local CBR measurements and CBR values received from neighbouring stations, which are combined to derive a Global CBR; • provision of an interface to the transport protocols above and of link-layer addressing to the Access Layer below; • handling of traffic classes to map facility layer service or message priorities to access layer priorities. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 22 The Basic Transport Protocol (BTP) provides a lightweight, connection-less transport service between protocol entities of the FL. Its main purpose is the multiplexing and demultiplexing of messages from different facilities processes, such as CAM and DENM. BTP defines two header types, BTP-A for interactive packet transport and BTP-B for non-interactive packet transport. It resides above GeoNetworking in the protocol stack and operates in a similar way to UDP in the IP suite. BTP does not itself contribute to congestion estimation or resource allocation. IPv6 integration is achieved through the GN6 Adaptation Sub-Layer (GN6ASL), which allows the transmission of IPv6 packets over GeoNetworking without modifications to IPv6. GN6ASL introduces the concept of geographical and topological virtual links in order to provide IPv6 with sub-IP multi-hop delivery and to extend connectivity towards infrastructure networks such as the Internet. The use of GN6ASL enables interoperability with IPv6-compliant systems but, like BTP, it does not have a direct role in Release 1 resource management. In summary, the NTL in Release 1 ensures the geographical addressing of packets, supports single-hop transmission when possible, and provides multi-hop forwarding where needed. It also enables the dissemination of congestion- related information required for DCC. BTP and IPv6/GN6ASL complete the transport functions of the layer and are important for message delivery and Internet integration, but they are not directly involved in congestion control or resource management. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 4.4.6 Applications, facilities layer and resource management | At the Facilities Layer (FL), the timely varying spectral resource capabilities come together with the timely changing functional requirements of applications and message services. As identified in the Multi-Channel Operation (MCO) concept study (ETSI TR 103 439 [i.1]), applications do not have any notion of the existence of other applications and their communication needs, therefore a Resource Management (RM) functionality at the FL (MCO management in ETSI TR 103 439 [i.1]) intends to facilitate a more predictable use of the actual radio resources. As RM takes care of the radio resource management in general, it is proposed to rename the MCO FL functionality as defined in ETSI TR 103 439 [i.1], into a more general RM for all kind of configurations (i.e. single-channel and multi-channel configurations). The RM is a FL functionality which replaces what is currently specified as the MCO-Facilities (MCO-FAC) in ETSI TS 103 141 [i.5]. The RM operation depends on the capabilities of the lower layers and on the proper operation of the applications and the message generating and manipulating message services. To allow the RM to provide robust operation of applications and message services, it is necessary that all dataflow functionalities at all layers, from applications to AL functionalities, operate with similar trustworthiness and QoS. In clause 5, relevant lower layer robustness improvements are depicted; in clause 6, the application and message service parameters and possibilities are considered; and in clause 7, the functional layer mechanisms are investigated and considered. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5 Decentralized Congestion Control optimization | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.1 Introduction | In Release 1, congestion control is managed separately for different Access Layer Instance (ALI) groups. The extended concept as outlined in the present document is based on the analysis of the ITS-G5 AL technology, but it is not limited to this specific AL technology. However, to use the concept with other technologies, additional studies might be needed. Clause 5 focusses on Decentralized Congestion Control (DCC) in a single channel. It contains a review of the related standards of Release 1 and identifies the aspects where Release 2 can benefit from modifications to comply with the MCO framework. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 23 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2 Decentralized congestion control in Release 1 | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.1 Introduction | This clause provides a synthetic review of DCC as it is defined in Release 1. First, it recalls the definition of the main metrics used for congestion control. Then details the limits that apply and the constrains imposed to the congestion control algorithm. The congestion control algorithm to be used is not specified by the standards but needs to respect the given constraints. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.2 Overview of Release 1 DCC | The general description of Release 1 DCC is provided in ETSI TS 103 175 [i.15], which details the entities at the various layers of the C-ITS protocol stack. In Release 1, DCC has entities at the access layer, detailed in ETSI TS 102 687 [i.14], and at the networking & transport layer, detailed in ETSI TS 102 636-4-2 [i.13]. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.3 Metrics | DCC is based on the concept of Channel Busy Ratio (CBR), as defined in ETSI EN 302 571 [i.16], which determines the level of load of the channel. More specifically, the receiver determines every &'(, set to 100 ms, the time when the strength of the received signal exceeds -85 dBm, called ) . The CBR is calculated every &'( as: = (1) NOTE: It is common use to consider the own transmission time as part of ) as specified in IEEE 802.11 [i.9], even if this is not explicitly mentioned in the current specifications. The measured CBR can be elaborated at the networking & transport layer and exchanged with the neighbouring nodes up to the second hop. This allows, as detailed in ETSI TS 102 636-4-2 [i.13], to obtain average values of the CBR in a larger area than the one directly observed by the station. This aspect is further discussed in clause 5.2.6. Other metrics that are relevant for the DCC process are: • The duration of a single transmission, denoted as . • The time between the end of a transmission and the beginning of the following one, denoted as . • The portion of time within a reference interval (set to 1 second) when the station is transmitting in the given channel, called duty cycle. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.4 Limit imposed to the channel load | The limit for the channel load of each station is indicated in the ETSI EN 302 571 [i.16], which in turn is based on ETSI TS 103 175 [i.15] and ETSI TR 101 612 [i.12]. The limit provides a minimum time between two consecutive transmissions where the transceiver is not allowed to generate a signal (minimum , as defined in clause 5.2.3), based on the duration of the last transmission ( , as defined in clause 5.2.3) and the measured CBR. For any value of the CBR, needs to be larger than 25 ms. If the CBR is equal or above to 0,62, the needs also to comply with equation (2): ≥min 1 000, × 4 000 × &'( ,*+ &'( −1 (2) with expressed in milliseconds. The rationale behind the equation above is discussed in Annex A. In addition, ETSI EN 302 571 [i.16] indicates a maximum of 4 ms and a maximum duty cycle of 3 %. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 24 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.5 Release 1 DCC at the access layer and constraints to the algorithm | ETSI TS 102 687 [i.14] details the constraints for the algorithm to be implemented at the access layer. The specific algorithm is not defined but left to the implementer. The constraints are as follows: • The algorithm runs on each frequency channel specified in ETSI EN 302 571 [i.16] independently. • The algorithm runs in an infinite loop. • The algorithm is activated at least every 200 ms. • The algorithm does not exceed the limits discussed in clause 5.2.4. In the same document, two possible classes and an example for each of them are reported. The first class is called reactive and consists in the use of a number of states. The reactive DCC algorithm transits from one state to another based on the measured CBR; the state defines the value of the minimum . The minimum that follows from each state can be dependent on (for example, two different values can be defined if the packet is small or large). ETSI TS 102 687 [i.14] does not specify the number of states or the values to be used per each state, but gives an example in its Annex A. In the reactive approach, at each step the station measures the CBR, identifies the state, and sets the minimum accordingly. The minimum has thus a granularity that is defined by the number of states and the associated values. The second class is called adaptive and increases or reduces the minimum based on the measured CBR. In this case, ETSI TS 102 687 [i.14] specifies the calculations that need to be performed from the measured CBR to the minimum , also including the values of the several parameters used by the calculations. ETSI TS 102 687 [i.14] also describes, in its Annex B, a possible implementation of the "gate keeping" function at the interface to the networking & transport layer to cope with messages of variable size. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.6 Release 1 DCC at the networking & transport layer | ETSI TS 102 636-4-2 [i.13] describes the media-dependent functionalities for ITS-G5 at the networking & transport layer, also called DCC_NET. If GeoNetworking is implemented, DCC_NET is mandatory. If it is implemented: • It maintains a number of DCC state variables (CBR_L_0_Hop, CBR_L_1_Hop, CBR_L_2_Hop, CBR_R_0_Hop, CBR_R_1_Hop, CBR_G, and CBR_Target), also using the Location Table Entry Extension for ITS-G5 (LocTEX-G5); LocTEX-G5 is an extension of the location table of the GeoAdhoc router. • It periodically calculates the global Channel Busy Ratio, CBR_G. • It processes and provides DCC-related information from/to DCC_CROSS. • It transmits and receives DCC-related information to other GeoNetworking routers using the extensions for GeoNetworking packet handling. • It optionally sets the transmission power limits, as specified in ETSI TS 103 175 [i.15], clause 6.2. Regarding the state variables CBR_L_0_Hop, CBR_L_1_Hop, and CBR_L_2_Hop, they are the CBR measured by the station (0 to indicate no hops), the CBR received by its neighbours (1 hop), and the CBR that its neighbours have forwarded as they have in turn received from them their neighbours (2 hops), respectively. CBR_L_1_Hop and CBR_L_2_Hop are conservative values derived from the possibly several CBR values received from the neighbours, with the specific calculations described in ETSI TS 102 636-4-2 [i.13], clause 5.3. CBR_R_0_Hop and CBR_R_1_Hop are the measured CBR shared by the station and the received CBR shared by the station, calculated as an average of those received from the neighbours. Per each received packet, each station stores the CBR_R_0_Hop and CBR_R_1_Hop received from all the neighbours in the LocTEX-G5, together with information about transmit power and Received Signal-Strength Indicator RSSI. Each station then calculates the global CBR CBR_G based on CBR_L_0_Hop, CBR_L_1_Hop, and CBR_L_2_Hop, which are in turn evaluated as a function of all received CBR_R_0_Hop and CBR_R_1_Hop. CBR_G is conservatively calculated, as detailed in ETSI TS 102 636-4-2 [i.13], clause 5.3. These calculations are performed every 100 ms. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 25 CBR_Target is finally the value used by DCC as target, which needs to be the same as the one used at the access layer; currently, it is set to 0,62. The value of CBR_Target is used during the calculation of CBR_G. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.2.7 Release 1 DCC at the facilities layer | The overall DCC architecture detailed in ETSI TS 103 175 [i.15], includes at the facilities layer a DCC_FAC function. However, DCC_FAC was not eventually defined in Release 1. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3 Critical analysis of Release 1 DCC | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.1 Introduction | The main aspects that could be revised in Release 2 are analysed in this clause. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.2 Analysis of global CBR sharing | Release 1 extends local congestion sensing by introducing the concept of the global Channel Busy Ratio (CBR_G) (see clause 5.2.6). CBR_G is formed by combining an ITS-S's own CBR measurement with CBR values received from neighbouring stations via 1‑hop and 2‑hop piggybacking. The intended effect is to extend awareness beyond the immediate carrier‑sense range, allowing nodes that do not sense local congestion to still contribute to congestion control efforts. This mechanism addresses known challenges in vehicular environments, such as hidden nodes, asymmetric sensing ranges, and local fading, which can make purely local decisions inadequate for achieving both local and global fairness e.g. as presented in the publication "Decentralized Congestion Control Techniques for VANETs" [i.35]. The concept was investigated in the CAMP V2V-Interoperability project [i.45], [i.46], where the LIMERIC algorithm [i.44] was tested with local, one-hop, and two-hop CBR sharing based on PULSAR [i.43]. LIMERIC remains stable in all cases, but the degree of spatial fairness improves with broader information sharing. Without sharing, nodes react only to their own CBR, achieving fairness locally. Sharing one-hop or two-hop values allows nodes to account for the interference they cause elsewhere, improving fairness - particularly with two-hop sharing, which approximates the interference range. Experimental results from the CAMP project [i.45], [i.46] showed that using the maximum CBR observed within two hops prevented spatial oscillations and unfair channel use seen in dense deployments. However, sharing also introduces delay, as received CBR values may refer to previous intervals. To preserve stability, the adaptive gains in LIMERIC were reduced, with a minor impact on convergence speed. In Release 1, CBR_G is computed by selecting either the maximum or the second-highest value from local, 1‑hop, and 2‑hop CBR reports within a validity window (T_Cbr) [i.13]. This approach is designed to be conservative, emphasizing the highest occupancy observed across the various reports. Consequently, the final CBR_G value may be influenced by a single, anomalous high report from the 1- or 2-hop data, even when the local channel does not exhibit signs of congestion. While sharing of global CBR is specified in in the ITS-G5 extensions of GeoNetworking [i.13] including corresponding header fields and location table entries, to date, the global CBR has not been adopted in the system profiles from C2C-CC [i.7] and C-Roads [i.8]. As a consequence, practical implementation of the global CBR has been limited, both in research prototypes and commercial implementations. Existing implementations either do not populate these fields or ignore them. Specifically, both the C2C-CC Basic System Profile [i.7] and the C-Roads Harmonized Profile [i.8] rely on local CBR measurement for congestion control. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.3 Analysis of the limit | When looking at the single generic channel, the limit imposed in Release 1 implies restrictions when the CBR exceeds 0,62. This number, which was chosen having a few message services and relatively small messages in mind, appears lower than the maximum channel load tolerated in ITS-G5. It appears therefore reasonable to verify if such value is still a good choice or it should be revised once moving to Release 2. It can be noted that an increase of the limit would mean that the already enrolled devices still comply, while a reduction of the limit would need to include exceptions for the devices implemented following the specifications in Release 1. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 26 With the aim to investigate this aspect, a simulation campaign was performed using the open-source WiLabV2Xsim [i.17]. Selected results are shown hereafter assuming a variable number of cars distributed on a highway segment of 8 km with 3 lanes per direction. The simulations are intended to obtain general considerations without focusing on a specific message service; for this reason, each vehicle is assumed to transmit 10 packets per second, and all packets are of the same size; different values for the packet size are assumed in different simulations. In particular, 350, 550, and 1 000 bytes are considered, where the first value is a reasonable reference for Release 1 messages and the last one for larger Release 2 messages. These three values correspond to a duty cycle of approximately 0,58 %, 0,84 %, or 1,44 %. ITS-G5 nodes are assumed to transmit at 23 dBm Effective Isotropic Radiated Power (EIRP), with 3 dBi antenna gain at the receiver, and 6 dB receiver noise figure. The access category Best Effort (BE) is adopted, corresponding to Arbitration Inter-Frame Spacing (AIFS) equal to 110 µs, Contention Window (CW) equal to 15. The sensing threshold, when a preamble is detected, is set to -85 dBm, whereas it is set to -65 dBm otherwise. The modulation and coding scheme corresponding to Quadrature Phase Shift Keying (QPSK) and 1/2 coding rate is assumed. The modified Electronic Communications Committee (ECC) rural model from ETSI TR 103 439 [i.1] is used for the propagation, with correlated shadowing with standard deviation 3 dB. The model implies a loss exponent of 2 up to 128 m, then 2,8 up to 512 m, and 3,3 m above 512 m. To evaluate the correctness of the decoding of each received signal, the average Signal to Noise and Interference Ratio (SINR) is calculated, where both noise and interference are assumed Gaussian and White. The reception is identified as correct when the SINR is above a given threshold as elaborated in "Methodology for Abstracting the Physical Layer of Direct V2X Communications Technologies" [i.18] Following the approach detailed in the same reference, the threshold is set to 1,2, 1,9, and 2,4, for 350, 550, and 1 000 bytes, respectively. Results are provided in terms of measured CBR, Packet Reception Ratio (PRR) and Packet Error Ratio (PER). The PRR is obtained dividing the number of correctly decoded packets to the number of packets attempted to be decoded. The PER is the complementary value, i.e. the number of packets not correctly decoded divided by the number of packets attempted to be decoded. Figure 2 represents the CBR varying the density and with different packet size. More specifically, it shows the median (i.e. the 50th percentile) of the distribution of the CBR measured in each interval by each station. As expected, the CBR increases almost linearly for low densities where collisions are limited and then slower when collisions become frequent. In addition, as expected, the increase is faster with larger packets. What is instead less obvious and therefore relevant to remark is that slower increase in CBR occurs at larger CBR values when the packets are larger; this effect is due to the lower relative impact of the inter-frame spaces when the packets are larger. Figure 2: Median of the CBR varying the density In Figure 3, the median CBR is shown in the x-axis and the PRR is plotted in the y-axis, when implicitly varying the density. Specifically, the PRR is calculated as an average at a distance of 400 m between transmitter and receiver. These results further show that the maximum CBR that can be tolerated changes when different packet size is assumed, with a higher CBR corresponding to a larger packet if the same PRR is assumed. This means that the larger packets expected in Release 2 may allow to increase the limit for congestion control. However, the increase appears limited. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 27 Figure 3: Average PRR at 400 m varying the median CBR In Figure 4, the median CBR and a measure of the average number of messages that are correctly decoded at 400 m per km and per second are shown varying the average messages sent per km and per second. Also in this case, the results are obtained by implicitly varying the vehicle density. Looking at the blue curves, corresponding to the median CBR, it can be noted that trends are consistent to those shown with the density in the x-axis in Figure 2; indeed, since 10 packets per second are sent by each vehicle, the average sent messages per km and second is proportional to the density. Looking at the red curves, which more specifically shows the product between the average number of messages sent per km per second and the PRR at 400 m, it can be noted that they increase for small values of the sent packets (i.e. for low vehicle density) and then decrease after a maximum is reached. In each of the three red curves, a star is added when the median CBR corresponds to 0,62; as observable, the marked values are close to the maximum of the curves, meaning that a load on the channel higher than the limit imply an excessive increase in the collision probability. Figure 4: Median CBR and measure of the average number of messages correctly received at 400 m per km per second Finally, in Figure 5, the PER at 400 m is shown together with the CBR varying the average messages sent per km and per second. Looking at the PER, it increases with larger packets, both due to a higher occupation of the channel and a higher SINR required to correctly decode the packet. Also in this case, the values corresponding to a median CBR equal to 0,62 are marked with a star, showing that the PER at 400 m is between 0,2 and 0,4 when the channel occupation is close to the limit imposed by current specifications. By looking jointly at Figure 4 and Figure 5, it can be noted that the maximum average number of correctly decoded messages at 400 m is achieved close to the channel occupation limit, despite the PER is larger than 0,2. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 28 Figure 5: Median CBR and frame error probability at 400 m, varying the messages sent per km per second The overall conclusion is that, when assuming that the packets are all of the same size, a different packet size could lead to a different maximum congestion threshold. Similar considerations apply also to other cases, such as if the packets are all with the same priority, but the priority is varied, because of different durations of the inter-frame spaces; or if the reference distance is changed. This means that it is not possible to determine a specific number for the optimum limit, instead this optimum depends on a large number of parameters which may change in time and space and are additionally depending on the use case. These results are consistent with what was discussed in the reference literature and with following main deriving observations: • A specific optimum value does not appear to be possible to derive, as it depends on a very large number of factors, which can also depend on the application and vary in time and space. • The threshold 0,62 as given currently in the standards appears reasonable also for Release 2. • A mechanism allowing a centralized authority to update the limit when deemed necessary could be introduced. The frequency of this kind of updates may be in the order of months or years and should introduce small variations based on considerations derived from the real implementations. The mechanism should be defined considering fairness between vehicles that may have temporarily different values during the update; this may be granted by only allowing small variations but needs further analysis. As an example, the parameter updates, when necessary, could be distributed along with the certificates. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.4 Analysis of algorithms | The current limits imposed to the algorithm that an implementation can use may not be sufficient. The current limits imposed by the standard ETSI TS 102 687 [i.14], in fact, do not guarantee fairness and may cause different implementations to have different levels of access to the channel in congested situations. With the objective to elaborate on this statement, hereafter some results are shown using the same simulator and settings of clause 5.3.2, adding the reactive and adaptive approaches described in ETSI TS 102 687 [i.14], without GeoNetworking. These results are not meant to comprehensively compare specific solutions, for which the reader can refer to the following references [i.19], [i.20], [i.21] and [i.22], but only to remark the different impact that different algorithms can have. In particular, hereafter either 200 vehicles per km are assumed to transmit messages of 350 bytes (scenario A), or 100 vehicles per km are assumed to transmit messages of 1 000 bytes (scenario B). These settings allow to investigate a scenario where the channel is slightly overloaded if no congestion control is implemented. Each vehicle generates one packet every + , where is set by the congestion control algorithm, based on the measured CBR. A new packet is generated at least every 1 second and at most every 100 ms. Each vehicle generates one packet every 100 ms if DCC is not active. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 29 When the reactive approach is used, the profile defined by the C2C-CC in the Vehicle C-ITS station profile, Release 1.6.7 is used. The profile requires that the CBR is updated making an average of the last measured CBR and the one calculated in the previous measurement interval. The profile also sets the values of the minimum for CAM messages to those of Table A.2 in ETSI TS 102 687 [i.14], which are therefore used for the 350 bytes packets; consistently, for the 1 000 bytes packets those of Table A.1 in ETSI TS 102 687 [i.14] are used. These values are also reported in Table 1. Table 1: Settings of reactive DCC State CBR Minimum with packets of 350 bytes Minimum with packets of 1 000 bytes Relaxed < 30 % 50 ms 100 ms Active 1 30 % to 39,99 % 100 ms 200 ms Active 2 40 % to 49,99 % 200 ms 400 ms Active 3 50 % to 65 % 250 ms 500 ms Restrictive > 65 % 1 000 ms 1 000 ms When the adaptive approach is used, the settings defined in ETSI TS 102 687 [i.14] are used. Figure 6 and Figure 7 refer to scenarios where all the vehicles implement the same algorithm, reactive or adaptive, or do not implement any algorithm. Figure 6, in particular, shows the Cumulative Distribution Function (CDF) of the CBR measured by every vehicle in every measurement interval &'(. The two subfigures refer to the two scenarios. The yellow curve, referring to no DCC implemented, provides a reference in the case all vehicles transmit exactly 10 packets per second. What can be observed is that both the reactive and the adaptive approaches reduce the measured CBR, which remains in most of the cases below or slightly above 0,6 as required by the limit imposed. The reactive approach implies a lower CBR than the adaptive approach, meaning that it reduces the average packet generation frequency stronger than the adaptive approach. A last observation that appears relevant is that in most of the cases the CDF tends to remain around a small range of the CBR; this is consistent with the stationarity of the scenarios and means that the algorithm tends to reach an equilibrium at a certain + . However, in Scenario B, where the channel is more congested, the CDF related to the reactive approach spans over a larger range of values of the CBR, which suggests that the algorithm does not converge to an equilibrium. (a) 200 vehicles per km, 350 bytes (b) 100 vehicles per km, 1 000 bytes Figure 6: Cumulative distribution function of the measured CBR The impact on the quality of the communication is shown in Figure 7, where the PRR multiplied by the average messages sent per vehicle per second is shown varying the transmitter-receiver distance. The metric observed corresponds to the average number of packets that can be correctly decoded by a receiver at the given distance and makes the performance of the various cases comparable, even if the number of generated packets is not the same. As observable, in both scenarios, when no DCC is implemented, the communication becomes unreliable at a shorter distance compared to the other cases. For example, looking at scenario a (Figure 7(a)), only one average packet can be correctly decoded at 600 m if there is no DCC, whereas more than two can be correctly decoded if a DCC algorithm is implemented. Comparing the two algorithms, the adaptive approach appears to provide a slightly higher number of correctly decoded packets, at least up to a certain distance. Overall, a higher transmission rate privileges the reception at shorter distance, whereas a lower transmission rate does the opposite. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 30 (a) 200 vehicles per km, 350 bytes (b) 100 vehicles per km, 1 000 bytes Figure 7: Average number of messages per second correctly received from a station at a given distance The results shown above suggest that the adaptive approach is able to maintain the channel occupation closer to the limit imposed, with a more stable value of the selected + and with slightly higher number of correctly decoded packets at short-to-medium distances. However, the difference between adaptive and reactive may not appear significant. In order to verify what happens if not all the vehicles implement the same algorithm, Figure 8 refers to a case where 50 % of the vehicles implement the reactive approach and the remaining 50 % implement the adaptive approach. In Figure 8, in particular, the minimum inter-packet time (i.e. + ) of two stations are shown varying the simulation time. The two stations are randomly selected, one among those implementing the reactive approach and the other among those implementing the adaptive approach. What can be observed is that the two stations have significantly different values of the minimum inter-packet time; the reactive approach, in particular, tends to use a larger gap, which means a lower frequency of packets sent. Having stations that generate packets at different frequencies because they are using different algorithms appears as a possible issue. (a) 200 vehicles per km, 350 bytes (b) 100 vehicles per km, 1 000 bytes Figure 8: Variation of the minimum inter-packet generation interval for randomly selected stations varying time The results shown are consistent with what has been discussed in the literature. The following observations can be made: • The specific algorithm that is adopted impacts on the system performance. The algorithm acts on the trade-off between the number of packets sent and the probability of collisions; reducing the packets sent may reduce the information available at shorter distance but increase that available at longer distance. This means that the distance from the limit at which the congestion control algorithm works has a significant impact on the system. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 31 • To comply with the limit described in clause 5.2.4 may not be sufficient to guarantee a certain performance and fairness among stations. Different algorithms may imply, in fact, that different amount of information is available at the receivers, which means that the principle of fair access to the channel may not be guaranteed. A solution may be to have a single algorithm or at least stricter rules for the definition of the algorithm. • The congestion control algorithm should (i) imply small variations of available resources when there are small variations of CBR, and (ii) allow the channel use to remain close to the maximum in congested conditions. For these reasons and given the detailed analysis in this clause and in the annexes, the use of the reactive algorithm as described in ETSI TS 102 687 [i.14] is not recommended. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.5 Analysis of variables that impact congestion control | Different options to cope with congested situations are suggested in Release 1 specifications, including Transmit Power Control (TPC), Transmit Rate Control (TRC), and Transmit Data-Rate Control (TDRC). For more details see the references [i.20], [i.23], and [i.24]. Variations of the power appear complex when looking at control at the facilities layer and also questionably effective: power variations, indeed, affect communication range but do not alter the CBR measurement close to the transmitting vehicle; this causes in turn variable power level and range in the spatial domain, which affects fairness and causes lower overall system performance due to interference coming from hidden nodes. Additionally, if all stations uniformly reduce the power consistently, the effect is only an average reduction of the ratio between the useful signal and the noise power levels, without an improvement of the ratio between the useful signal and the interference power levels [i.23]. Power control may still remain an option when dealing with unicast communications; in such case, in fact, the transmitter may control the power level in order to guarantee a sufficient reliability to the communication while minimizing the interference generated to the others. Data-rate control appears easier to control at the facilities layer and possibly more effective to control the trade-off between reliability and channel occupation. This means at the access layer to select a different MCS, which corresponds at the facilities layer to the selection of a different ALI. However, the choice of the MCS (or ALI) has also an impact on the range, since higher MCS implies lower protection to noise and interference. Since the header information is always sent with a robust MCS, a data-rate control does not impose additional hidden nodes, even the range for the payload is reduced. The use of data-rate control therefore requires careful considerations about the use cases that need to be supported. Still discussing the data-rate control, it is also worth observing that the impact may be limited. This is further elaborated hereafter, by assuming the eight MCSs of the basic version of ITS-G5 and calculating the duration of packet transmissions. Based on the specifications of ITS-G5, the duration $ of a packet transmission with payload of bytes can be approximated as follows: $ ≈ + ,-×. / !"# (3) where = 40 µs is the duration of the Physical Layer Convergence Protocol (PLCP) preamble and header, is the number of useful bits carried by an Orthogonal Frequency Division Multiplexing (OFDM) symbol, and !"# = 8 μs is the duration of an OFDM symbol. The value of depends on the specific MCS and ranges between a minimum of 24 and a maximum of 216, with 52 used as default (normally known as MCS 2 and corresponding to 4-QAM, coding rate 1/2). The duration deriving from this equation is shown in Figure 9 for packets of 350 or 1 000 bytes. As observable, compared to the use of MCS 2 (which is the default value in ITS-G5), a data-rate control can at most reduce the channel occupation by a factor close to 4; this however requires moving to the less reliable MCS, which strongly impacts on the range. Small variations of the MCS have limited impact on the channel occupation. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 32 (a) Duration of a transmission (b) Channel occupation compared to MCS 2 Figure 9: Impact on channel occupation of data-rate control in ITS-G5 The last option discussed is the transmit rate control, which means controlling the average number of packets sent and can be implemented in two ways. One way is discarding a portion of the packets before they are transmitted, and the other way is to control the generation rate. Looking at the former option, it can in turn be implemented at the facilities or at the access layer. If it is performed at the access layer, there is no possibility to differentiate among packets based on their content and the withdrawal is unavoidably performed in a random way. Among the various options, the control of the generation rate appears clearly preferable and the interaction between the facilities layer and the running applications and services can further help to optimally identify the information to be sent. When referring to Release 1, a withdrawal at the access layer appears as the only viable solution. Overall, the transmit rate control is the one that can impact more significantly on the channel occupation without affecting communication range. Its main drawback is a reduction of the information update, which is however balanced by a reduction of the packet error rate if the congestion control algorithm works properly. NOTE: It is observed that transmit rate control is the most effective approach to control congestion; it is also noted that it would be better implemented with a control made by the application rather than by discarding part of the packets. Data-rate control can also help in some cases, if the communication range can be traded-off. Transmit power control seems helpful only in specific cases, such as for unicast transmissions. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.3.6 Analysis of multi-hop forwarding | What discussed in the previous clauses assumes single hop broadcast communications. In principle, also multi-hop forwarding is however possible, and therefore it also needs to be considered. Multi-hop forwarding in Release 1 is based on the GeoNetworking protocol (see clause 4.4.5) and follows a packet-centric forwarding approach. In this model, packets are relayed by intermediate ITS-Ss according to geographical addressing information and forwarding rules such as greedy forwarding or contention-based forwarding. This mechanism is performed in the NTL, and it is transparent to the applications and the facilities layer, although it is controlled by them through the choice of packet type and addressing parameters. Packet-centric forwarding ensures that information is delivered to the addressed area without requiring facilities-layer interaction and is particularly suitable for message services with stringent latency or authentication requirements, such as Decentralized Environmental Notification (DENM), where messages need to be disseminated rapidly and without additional processing. An alternative forwarding concept is information-centric forwarding. In this case, the application or facilities layer interprets the received information, determines its relevance, and generates new messages that may aggregate, filter, or invalidate previously received content. Information-centric forwarding reduces redundancy and can adapt dissemination to the semantics of the information. For example, Cooperative Perception Messages (CPM) may be aggregated before being re-broadcast, and safety-related message services that are not strictly time-critical, such as Signal Phase and Timing (SPAT) and MAP, can benefit from controlled re-dissemination rather than pure packet relaying. 0 1 2 3 4 5 6 7 MCS 0 0.5 1 1.5 2 2.5 3 350 bytes 1000 bytes 0 1 2 3 4 5 6 7 MCS 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 350 bytes 1000 bytes ETSI ETSI TR 104 073 V2.2.1 (2026-02) 33 In Release 1, only packet-centric forwarding is supported by GeoNetworking, and there is no integration of information-centric mechanisms at the networking layer. This ensures simplicity and timeliness for critical message services but limits flexibility for other message services that could benefit from aggregation or content-based dissemination. For this reason, the use of GeoNetworking multi-hop forwarding should be restricted to safety-critical messages where low latency is essential or to long range messages that need authentication and that are sent seldom. message services without such requirements should rely on dissemination mechanisms at the facilities layer that are closer to information-centric forwarding. From a resource management perspective, Release 1 implements congestion control at the access layer through a gatekeeping mechanism that regulates transmission rates based on channel load. This mechanism manages forwarded packets in the same manner as those originating from upper layers, leading to a lack of distinction in how congestion is handled for forwarded traffic. As a result, forwarded packets utilize radio resources without explicit differentiation, which can lead to unpredictable effects during periods of congestion. Additionally, forwarding algorithms may introduce latency due to contention timers, impose an uneven load on stations that engage in more forwarding, and increase overhead in densely populated traffic scenarios. Buffer management is also limited, which may result in packet losses under high load. These limitations suggest that enhancing forwarding efficiency would be advantageous and should be considered for future releases. An additional aspect for packet centric forwarding is related to the use of traffic classes: In Release 1, the traffic class is a parameter that is passed from the FL through the NTL to the AL. The parameter is used to express the priority of a FL message and map it AL priorities, such as the Access Category (AC) in ITS-G5. In addition, the traffic class represents a data element in the GN header and is transmitted over the air. However, in Release 1, the traffic class is not further to control the forwarding of data packets. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.4 Interference from adjacent channels | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.4.1 Introduction | The congestion status of a given channel is not only dependent on the actual usage of the channel itself but also on the usage of the spectrum that is adjacent to it. This spectrum might be occupied by interoperable message services using the same protocol or by systems which are not interoperable or compatible. Interference management operations are basically different for the two cases. In the first case, it can be handled by internal RM control operations and functions that are part of the RM specifications. These operations can be called cross-channel RM functions, which include the cross-channel CBR evaluation and the cross-channel load control. In the non-interoperable case, the cross-channel RM control is not possible due to the lack of interoperability. These effects can only be considered as general interference from adjacent channels similar to co-channel interference from non-interoperable sources. All non-interoperable interference effects are included in the RM monitoring/evaluation process based on the CBR measurement at the access layer and thus contribute to the channel load in general. It has to be noted that these non- interoperable interference effects can lead to a significant increase of the channel load and to a reduced transmission probability/capability of the ITS systems due to the clear channel assessment performed by CSMA/CA before transmitting. In the RM and the MAC procedure a differentiation between the two different kinds of channel loading effect would be beneficial and could contribute to the proper operation of the ITS. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.4.2 Impact of interference from interoperable C-ITS in adjacent channels | The impact of interference caused by stations using the same protocols in adjacent channels is extensively studied in ETSI TR 103 439 [i.1]. In Figure 10 the basics of the interfering effects in an MCO operation are depicted with the focus onto the direct adjacent channel as considered in ETSI TR 103 439 [i.1]. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 34 Figure 10: Impact of an interfering transmitter and a victim receiver on the reception of wanted signals The study in ETSI TR 103 439 [i.1] was performed assuming a highway scenario with multiple lanes per direction and different levels of road traffic. Furthermore, different settings were assumed in terms of which channels were used by the vehicles moving in the scenario. Both cases where similar data traffic was generated on average in the various channels (balanced load) and cases where the load was imbalanced, were considered. The main conclusions of the study can be summarized as follows: • When transmissions are performed in channels that are not directly adjacent to each other, the interference that they cause reciprocally only slightly reduces the probability to correctly receive the packets. This also holds even when the channels are congested. This implies that there is no need to take the congestion status of the channels beyond the first adjacent channels into account in the internal RM operations. • Focusing on transmissions in the first adjacent channel, they do not impact relevantly on the clear channel assessment performed by CSMA/CA before transmitting, but impact on the calculation of the CBR. As a consequence, rarely transmissions are deferred due to interference from adjacent channels, but the interference may increase the CBR and therefore reduce the number of transmissions overall performed in the given channel. When two channels that are directly adjacent to each other are highly loaded, the effect is a reduction of the average number of transmissions due to congestion control, thus a reduction of the collision probability in each of the two channels. This in part counterbalances the increase of errors at the receivers. • A highly loaded first adjacent channel can cause a reduction of the probability of correct reception of the packets which leads to an estimated loss of range that can reach 25 % to 30 % (where the range is defined as the maximum distance at which a certain minimum reception probability is obtained). • Distributing the data traffic over two channels, even directly adjacent to each other, is always preferable to having all the data traffic in a single channel. This means that the reciprocal interference caused by transmissions in adjacent channels has an impact lower than the advantage obtained by halving the data load in a single channel. • Due to the current constraints imposed to the transmission mask, reducing the transmission power in one channel may not significantly reduce the interference it causes to the adjacent channels; even if this also depends on the specific implementation, the observation is that using power control to reduce the interference between adjacent channels may not be effective. • Reducing the congestion control limit in one channel may reduce the maximum interference that the transmissions in that channel causes to those in its first adjacent channels; reducing the limit in one channel may therefore be used to protect its first adjacent channels. Interfering TX spectrum RX selectivity Combinedinterference TX spectrum and RX selectivity 0 dB reference RX selectivity TX spectrum Interfering transmissionleaking into RX filteredchannel Frequency Receiver picking up signal in adjacent band Wantedsignal assignedbandwidth Most significant power components ETSI ETSI TR 104 073 V2.2.1 (2026-02) 35 Given these considerations, interference from stations using the same protocols potentially reduces communication performance in an adjacent channel, but in practice performance degradation does not appear to be critical. It may be helpful to reduce the congestion control limit in some ALI Groups to prevent the communication performance degradation in their first adjacent channels, including the available capacity and communication range. This would give priority to channels with less stringent requirements. In all the cases, distributing the traffic over multiple channels appears beneficial. Based on the finding presented in the present document the focus of any RM operation should be on the single operational channel. Adjacent channel effects are mainly relevant when all operation channels are in a high load situation close to a congestion. A single channel RM will keep this effect at a minimum in the first place. Longer term envisaged high load conditions due to a high C-ITS penetration might require some adjacent channel RM optimizations. Nevertheless, the actual specification of single channel RM work should permit the inclusion of adjacent channel RM in the long-term development. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.4.3 Adjacent channel monitoring | Different approaches for the monitoring operations of adjacent channel interference can be taken for single channel and multichannel ITS-S. A single channel device cannot differentiate between interoperable or non-interoperable deployments in the adjacent channels without switching to these channels, thus a single channel ITS-S can only passively monitor the interference increase as part of the channel load measurement, independently from the nature. Multichannel devices can perform this differentiation and can take the required control decisions. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.4.4 Cross-channel resource management control | In a single channel device, the adjacent channel interference will be treated as normal unspecified channel load element together with the co-channel load elements. The resulting levels of the load estimation in the access layer will be considered in the management algorithms. The co-channel load originating from an interoperable ITS system could be treated differently from the non-interoperable and adjacent channel interference. In a multichannel device a more detailed view of the interference characteristics in the adjacent channel can be evaluated. This information can be used for an optimization of the RM control algorithm. In order to differentiate between load originating from interoperable systems and load originating from non-interoperable systems, the channel load information from the co-channel and adjacent channel would need to be reported in a structured manner. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5 Modifications towards Release 2 | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.1 Introduction | This clause focuses on the modifications that are proposed to comply with the RM approach in Release 2. Modifications in channels that are already used today should not affect the performance of devices that are already on the road. To make the proper decisions, the RM should be aware of the radio resources that are available and the radio resources that are necessary to deliver each message. The following clauses therefore discuss the definition of radio resources, the approaches proposed for the determination of the available radio resources, and define the resources consumed by each message. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.2 Radio resources at the facilities layer | A first issue that needs to be solved is how to abstract the radio resources from the access layer to the facilities layer. The main difficulty is that the resources available and the resources used do not only depend on the average number of bits generated but also: i) on the number of packets, since each packet has its own headers adding overhead; and ii) on the specific settings at the access layer, including for example the MCS. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 36 The facilities layer should have, as much as possible, a precise and complete understanding of the options available. It should have for example all the information to decide, under congested situations, if for a given message it is better to use one or the other channel, one or the other MCS, or even discard it. The knowledge of the available channels and MCSs is already abstracted using the concepts of ALI and ALI group. What still needs work is the definition of the resources, the calculation of the available resources, and the calculation of the resource used to transmit a message in each ALI, possibly with some knowledge about the expected performance corresponding to the use of that specific ALI. The proposed definition of the technology-agnostic resources is as follows: • Portion of time and bandwidth. This definition is unitless and is valid for any bandwidth or time period. As an example, the resources corresponding to 0,005 are equivalent to the transmission of either: a) messages that occupy the whole bandwidth during 0,5 % of the time; or b) messages that occupy the half of the bandwidth during 1 % of the time; or c) messages that occupy one fifth of the bandwidth during 2,5 % of the time. This definition is valid for radio access technologies both with and without resource grids. As an example, ITS-G5 (IEEE 802.11 [i.9]) (Figure 11a) makes use of the whole bandwidth in all transmissions, providing a high granularity in time. On the other hand, LTE-V2X and 5G NR-V2X (Figure 11b) allow transmitting using part of the bandwidth but the time duration is fixed. (a) ITS-G5 (b) LTE-V2X or 5G NR-V2X Figure 11: Resources in time and frequency |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.3 Congestion control in Release 2 | In Release 1, the congestion control mechanism is fundamentally performed at the access layer, in which case it is technology-dependent and cannot affect the message generation. In Release 2, the RM functionality needs to be aware of the available resources and to manage them. The following three approaches appear possible: 1) congestion control is performed at the access layer, independently by each ALI group, and the available resources are abstracted to the facilities layer; the access layer provides in this case the available resources, through the networking & transport layer, to the RM; or 2) the access layer provides the information about the channel load to the facilities layer, with the information abstracted so that the facilities layer considers all the ALI groups equivalent; congestion control is then performed at the facilities layer in the same way for all ALI groups; the access layer provides in this case the CBR, through the networking & transport layer, to the RM; or 3) the access layer provides the information about the channel load to the facilities layer, which is configured considering the specific ALI groups; congestion control is then performed at the facilities layer, with ALI group-specific congestion control functions; also in this case, the access layer provides the CBR, through the networking & transport layer, to the RM. The first solution is closer to what implemented in Release 1, since the congestion control operations are placed at the access layer, but limits the control at the RM; in that case, in fact, the resources that are available per each ALI group are fixed by the access layer. Differently, the other two solutions allow the RM to have more control on the congestion control, for example considering which specific message services are active and with which priority, as explored in clause 7. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 37 The second solution allows to solve the limitations of the first approach maintaining independence from the access technology or technologies implemented; however, it requires that the information provided to the facilities layer and the congestion control algorithm that is implemented at the facility layer are defined technology agnostic, which appears difficult to realize. For the reasons above, the third solution appears as the preferable solution. It guarantees to the RM the possibility to impact on the congestion control operations, without the need to realize a technology agnostic congestion control function. The main implication of this option is that it requires that different congestion control functions are implemented for the different active ALI groups, or that the implemented function is designed considering the specific ALI groups that are active. In all the cases, the access layer, in addition to providing the available resources or CBR to the facilities layer, needs to inform it of relevant events, such as queue overflow, drop of packets, and time out of packets; their identification and notification by the access layer need therefore to be specified in Release 2. 5.5.4 Information required by the resource management for the ALI selection The proposed definition of resources can be also used for the resources needed to transmit a single message. This requires however to map the message size with the resource use. Given that the resources used for a packet transmission depend on the message size and on parameters known at the access layer, one proposed solution is to have at the facilities layer one table per each ALI indicating the resources required to transmit a message of a given size. The table should have at least: i) one column indicating the message size; and ii) one column indicating the resources required for the transmission of a message of that size. Using this table, the RM can calculate the resources it is using and apply procedures that require further work. As a simple example, if the facilities layer has a budget of 0,005 per second for an ALI group, and messages are generated every 100 ms on average, consuming in the selected ALI of that ALI group 0,0001, the RM knows that this flow consumes overall 10 × 0,0001 = 0,001, which is 1/5 of the available resources of the ALI group. An additional aspect to be considered is that the RM also needs indications on the performance expected when choosing among ALIs that consume different resources. For example, an ALI corresponding to a lower MCS will use more resources than another ALI corresponding to a higher MCS, but the advantage is that the transmission is more robust to noise and interference, which means that it is expected to provide more range. For this scope, one possibility is to store per each ALI a reference one-hop range, which may be for example calculated as the maximum distance at which the error rate is lower than 10 %, assuming a Gaussian channel, absence of interference, and predefined settings for the transmission power and receiver characteristics. As an example, it is assumed that there is one ALI group with two ALIs, ALI A and ALI B; the range of ALI A is 1 km and it requires 0,001 resources (defined as above discussed) to transmit a message of 800 bytes; the range of ALI B is 500 m and it requires 0,0005 resources to transmit the same message; given this information, the RM can decide if it is better to have a longer range but using more resources or the opposite. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.5 Overhead from protocol headers and security | One aspect that needs further clarification is how to account for the overhead added by headers of the protocol stack and security when calculating the total amount of data that will be transmitted over the air. Headers from NTL and AL protocols add to the overhead: • At the NTL, BTP and GeoNetworking headers have a fixed size. The GeoNetworking header size, however, depends on the GN transport type (e.g. single-hop broadcast SHB or geographically-scope broadcast). • The overhead at the AL is technology-specific: - For ITS-G5, it includes the LLC header - along with the SNAP sub-header - as well as MAC and PHY header. - For LTE-V2X and 5G NR V2X, the overhead is given by the header sizes of the PDCP, RLC, MAC and PHY protocols. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 38 A key point is that the overhead from protocol headers is known in advance of sending a message. Differently, for the security overhead the size of signatures and certificates is variable and not a-priori known. Consequently, instead of incorporating exact values, it is necessary to use either the maximum estimated overhead or an average estimated value in the RM calculations. In the context of overhead, it is also worth noting that the maximum size of the payload that can be transmitted in a single FL message is limited and access-technology specific. In Release 1, the size of a GeoNetworking packet sent over ITS-G5 is conventionally limited to 1,500 bytes (see Annex F). Beyond such protocol-specific constraints, the maximum FL message size may also be reduced due to radio-performance considerations: in general, smaller PHY frame sizes are preferred to mitigate synchronization and equalization challenges, particularly under high-mobility conditions. For Release 2, however, a larger maximum message size may be acceptable, taking into account the improved radio performance of the evolution of ITS-G5 based on IEEE 802.11 [i.9]. It should also be noted that for LTE-V2X and 5G NR-V2X other limitations for the maximum data unit apply and depend on the chosen access layer configuration. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.6 Introducing a mechanism for the update of parameters | One feature that is not available in Release 1 and might be worth introducing in Release 2 is the possibility to update the specific value for a limited number of parameters, such as the congestion control limit in each channel. Such updates would require the control from a central entity and a distribution to all ITS-S in a relatively short time, which in turn means that it would require the use of I2V links. EXAMPLE: One option to be explored is to exploit the distribution of the certificates to also include possible parameter updates. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 5.5.7 Networking & transport layer in Release 2 | Clause 5.3.6 has shown that, in Release 1, multi-hop forwarding is managed entirely by the GeoNetworking protocol in a packet-centric manner. Forwarding is subject only to the access-layer DCC, which limits transmissions through the gatekeeping mechanism based on channel load. As a consequence, forwarded packets are treated in the same way as packets generated by upper layers, with no explicit coordination or resource differentiation. This can lead to inefficiencies or uneven use of radio resources under congestion. In Release 2, the FL is expected to participate in the overall RM process, enabling a more comprehensive allocation of resources across message services. With this architectural evolution, different options can be considered regarding how multi-hop forwarding should interact with the emerging RM framework to ensure consistent and efficient use of network resources. Three approaches are identified: 1) Implicit forwarding without RM interaction: - GeoNetworking continues to perform forwarding autonomously. The consumed resources are considered negligible and therefore remain outside the RM budget. Optionally, a maximum forwarding rate (in bits/s or packets/s) may be defined locally to prevent excessive forwarding. When this limit is reached, GeoNetworking decides which packets to forward, buffer, or discard. This approach preserves the simplicity and real-time behaviour of Release 1. However, it provides no visibility to RM, which means that the effects of forwarding cannot be considered in admission control or congestion handling (see clause 7.2). 2) Periodic reporting of forwarding resource usage: - GeoNetworking estimates the amount of resources used for forwarding and periodically reports this information to RM, for example together with congestion-related metrics such as CBR. RM can then adjust the total available resources for other message services by subtracting the forwarding usage. This allows RM to incorporate forwarding into its scheduling and resource budgeting processes (see clauses 7.3 and 7.4). The approach improves coordination between layers but introduces additional measurement, signalling, and computational overhead within GeoNetworking. In addition, the introduction of priority flags for forwarded packets could enable the network layer to selectively discard lower-priority messages under congestion, ensuring that critical traffic is transmitted first. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 39 3) Forwarding message service abstraction at the FL: - A dedicated Forwarding message Service (FS) may be defined at the FL, acting as a consumer of resources in the same way as other message services. GeoNetworking would invoke this message service when forwarding is required, and RM would allocate resources through its normal admission control and scheduling mechanisms (see clause 7.5). This ensures homogeneous treatment of forwarding and other message services within the RM framework, offering maximum transparency and control. However, it increases architectural complexity and may introduce additional latency if frequent resource requests are needed for each forwarding action. This works for messages which are signed at the FL but when forwarding is realized according to the Release 1 concept the originating security is lost and the question is how to realize a working mechanism. Further research is needed. Each approach represents a trade-off between simplicity, control, and implementation complexity (Table 2): Approach 1 maintains architectural simplicity but offers no RM integration. Approach 2 provides moderate integration through lightweight coordination. Approach 3 achieves full integration into the RM framework but requires significant architectural changes and careful design to avoid degrading the responsiveness of safety-critical forwarding. Table 2: Comparison of approaches for integrating GeoNetworking forwarding into RM Aspect 1) Implicit forwarding 2) Periodic reporting 3) FS abstraction Integration with RM None - forwarding excluded from RM budget Partial - RM informed of forwarding resource use Full - forwarding handled as a message service managed by RM Implementation complexity Low Medium High Impact on latency None (same as Release 1) Minimal increase due to reporting overhead Possible increase due to RM coordination Fairness / resource accounting Not supported Basic resource accounting Full resource accounting and scheduling control Compatibility with Release 1 Full High Limited (requires new interfaces) Suitable for Safety-critical, low-latency message services Mixed or adaptive message services Non-safety-critical or managed message services Security headers Send's security headers preserved Send's security headers preserved Loss of send's security headers Implicit forwarding could be sufficient as long as system specifications ensure that forwarding is realized only at the lowest traffic class and that it is only used for functional reasons. Otherwise, the use of GeoNetworking multi-hop forwarding by message services should be limited to safety-critical messages where very low latency and direct dissemination is essential (e.g. DENM). This means that existing message service specifications should be extended to support forwarding as part of its service. To control the forwarding of data packets at the NTL, the concept of traffic classes should be extended. NOTE: Forwarding should not be used to avoid deployment of additional ITS-Ss. In addition to the proposed changes in forwarding behaviour, another point to be addressed in the NTL for Release 2 is the handling of the CBR_G. Clause 5.3.2 described how, in Release 1, CBR_G was derived from local, 1-hop, and 2-hop measurements to extend congestion awareness. For Release 2, it is proposed to maintain the inclusion of both 1-hop and 2-hop values to preserve the awareness range, while investigating potential modifications to the aggregation logic. Instead of relying solely on the maximum value, alternative formulations - such as weighted or averaged combinations of received CBRs - could provide a better balance between fairness and efficiency. Further research is required to assess the performance trade-offs and to validate whether these refinements yield measurable improvements under realistic traffic and network conditions. Finally, given the limited adoption of global CBR sharing in the system profiles defined by C2C-CC [i.7] and C-Roads [i.8], ETSI TC ITS Release 2 should take an active role in promoting the motivation, concept and specification of global CBR sharing, thereby supporting its practical implementation in research prototypes and commercial deployments. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 40 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6 Application requirements | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.1 Introduction | To ensure that active applications in an ITS-S are able to make appropriate decisions about when and how to exchange information with other ITS-Ss, they need to be aware of the available communication resources at any given time. In general, more than one application can be active in an ITS-S, and each application does not know the static and dynamic communication requirements from the other applications. The number of active applications in Release 1 is limited but for Release 2 they are significantly extended and therefore some resource management should be considered. The message behaviour is strongly dependent on the way the dissemination of messages is triggered and the amount of information it needs to disseminate. Furthermore, dissemination of a specific message type could be triggered by many applications, while other messages may be triggered by a single application or related to a specific use case. The DENM is typically a message type for which the message dissemination can be triggered by multiple applications, while the CAM is a typical message type which is related to a single use case, including message generating rules. The CAM is not triggered by applications but is a stand-alone functionality which has no interface to an application. Any resource management functionality most probably will only have to gather relevant management information for CAM dissemination directly from the CAS but as the DENS only knows that it has to disseminate a message when triggered, any resource management should in the first place get information from the triggering applications and possibly secondary from the DENS. In the following, for relevant dissemination information, reference is only made to applications while for several cases this could be also an message service. Since in Release 2 implementations the number of active applications is expected to significantly increase, it could be not sufficient to just have the static knowledge about the available resources. Having a generalized knowledge of the needs of all applications could allow a resource management functionality to distribute the available resources between all applications. This could be detailed with different granularity of the resource distribution, which could range from a binary switch-on/off of an application to a very fine allocation of resources to the applications active at any given time. Different methods exist in the initial Release 2 ETSI MCO concept. For the management of the resource allocation to the active applications in the ITS-S several aspects have to be considered: • the individual application or message service communication requirements and thus resource needs; • the application priority; • available resource management mechanisms and granularity of the resource allocation; • external resource limitation based on legal and regulatory restrictions. The first three aspects are ITS-S internal (intra ITS) whereas the last one is external to the ITS-S (inter ITS). The inter ITS aspect is an ITS intercommunity aspect which might require legal agreements or technical specifications agreed between stakeholders (profiles) to ensure which applications are active when and where in the available spectrum. It will need a secure control mechanism to allow for an external input of the required control information. C-ITS is defined by the European Union (EU) in the Directive 2010/40/EU [i.31] and its amendments, this EU Directive and related specifications should aim for a robust operation of all C-ITS applications in the assigned C-ITS spectrum. In the following clauses, the intra ITS aspects related to the way the dissemination of messages is triggered and the amount of information it needs to disseminate them are considered for the known ITS applications and message services. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 41 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2 Applications and message services | |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2.1 Introduction | There are applications which trigger event message generation and dissemination at the FL by message services (a specific message service) such as the Decentralized Environmental Notification Service (DENS). There are also Message Services (MSs) like Cooperative Awareness Service (CAS) which include message generation rules themselves, by which they trigger the generation and dissemination of messages. However, in future it cannot be excluded that such message services could be triggered as well. In general, message dissemination can be initiated by any application or message service with included message generation rules. As message services disseminate messages in the same channel as other message service, the dissemination of one message service can be influenced by the message dissemination of one other message service. This dependency does not occur when the channel occupation is not near to the set congestion limits, however, already ahead of the possibility of entering such state, it could be of interest to applications and message services to be aware of this and make different choices when this would apply. In principle, the understanding of the possible application message triggering and dissemination intensions are required for the realization of resource management. Considering that applications communicate with the ITS-Services and that therefore the message service knows the real disseminations requirements, currently it is considered sufficient that the dynamic behaviour expectations of the message service is communicated to the RM. The RM may also communicate with the applications for more general information, but this is not considered for the time being. In general, ITS-S-MSs dissemination behaviour can be categorized in three types as follows: • Broadcasted Event type, such as Decentralized Environmental Notification Message (DENM) triggered by Applications (such as the Triggering Conditions as specified by the C2C-CC [i.7]). • Broadcasted Awareness types such as: - Fixed Repetitive type, with fixed message size, such as MAPEM, IVI and SAM. - Adaptable Repetitive type, with predictable but no predefined message size, such as CAM, CPM, SPATEM, MIM/MVM, TLC (SREM, SSEM) and GPC (RTCMEM). • Broadcasted Streaming type, with fixed rate, fixed size and continuous, such as video streaming. In the following clauses the message services identified are further detailed with regards to their communication behaviour. NOTE: The mentioned ITS-S-MS are specified by ETSI and ISO but their references are not provided here as their reference number can differ for difference releases. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2.2 Decentralized Environmental Notification Message | The DENM is an event message (as identified in clause 6.2.1). Especially the dissemination of DENMs is triggered by one or more applications which can support the realization of different use cases and environmental scenarios. It can be expected that the number of use cases and scenarios will increase in the future. Depending on the safety impact of the specific use case, DENM dissemination via direct communication requires a specific priority over other message disseminations. In general, direct communications is direct safety impact oriented and Internet communications indirect safety impact oriented (more details in clause Ecosystems ETSI TR 103 903 [i.4]). While there are ITS use cases best serviced by either one of these communication solutions, there are also ITS use cases which can be serviced by both (see the Hybrid Solution by C-ROADs [i.8]). One example is the "End of Queue" use case. For this use case the information disseminated via direct communications can be used for a direct stop initiated by a driver of a vehicle or by the automated vehicle itself, while disseminated information via Internet communication will generally be used for rerouting of a trajectory and not predictably for a direct stop, thus they are the same use case but seen as different scenarios. Information disseminated via direct communication (as there are more stringent requirements) can also be used for those use cases which require the same information but can also be serviced by Internet communications. The communication requirements for the indirect scenario are out of scope of the present document. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 42 In Release 1, it was identified that high priority should be provided to these types of messages. It was recognized that events, as they have immediate impact, should be handled first compared to any other kind of other information dissemination. While for Release 1 only two levels were identified, it could happen that for later releases (at present it has not been defined for Release 2) the number of priorities should be extended. NOTE 1: Extension of the priority levels is not only something of relevance for DEN but also for other message services and therefore a generic extension should be considered. NOTE 2: Release 1 operational equipment is based on IEEE 802.11 [i.9] which includes four QoS traffic classes (TCs). These TCs are directly coupled to the four priorities used at the FL in C-ITS. At present two event types can be recognized: • An event which was not planned, mostly identified by an automated functionality in the vehicle, such as a Slow Driving Vehicle (SDV). This is a type of event which may just popup. • An event which was planned, typically managed by authorities and switched on and off by humans, such as a Roads Works Warning (RWW). While with the first type the dissemination of DENMs could take only a very short time, with the second type the disseminations of DENMs could take place for days or even month. This time related aspect is the only difference. Since at present both of the event types can be detected, activated and deactivated, the DENMs will be disseminated with a repetition rate fitting the environment and the use case as required. From an ITS-S communication perspective, DENM dissemination is not application but use case dependent. In general, an application does not statically know in advance what its communication needs are. Only dynamically, when a use case determines that it needs to disseminate DENMs, the application can notify its needs in terms of resources. This is valid for both the managed and not managed type of events. Possible parameter consideration for DENM: • Application statistically (at the time of application activation (application registration)): - Application indicator (internal for ITS-S). - Number of use cases supported (internal for ITS-S). - Expected priority level(s) to be supported. - Expected access technology and spectrum requirements (possibly including primary and secondary options). - Max message generation rate when activated. - Max message size when activated. • Application dynamically (based on activation of DENM dissemination cycle (application registration)): - Application indicator. - Use case indicator. - Dissemination initiation (Request): Actual priority of the cycle of DENMs to be disseminated. Actual technology and spectrum requirements (possibly including primary and secondary options). Real message generation rate. Real message size. Expected start time of dissemination. Expected termination time. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 43 - Dissemination termination: Termination indicator. NOTE 3: A dissemination cycle is considered to be the period in which a sequence of DENM with a specific rate are actively disseminated. NOTE 4: At present, a constant rate and constant message size is considered for DENM. Further it is considered that applications realize use cases with a similar message dissemination behaviour. NOTE 5: To allow any management of the DENM message as provided by several applications and their use cases, it should be clear to the resource management which application/use case initiate DENM dissemination and therefore it is required to provide related DENMs with indicators about which application/use case initiated the dissemination. As a result, each DENM from a cycle needs to include all application dynamic information. |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2.3 Cooperative Awareness (CA) Message | The CAM is a broadcast awareness message (as identified in clause 6.2.1) disseminated by an ITS-S which represents a road user e.g. vehicle, truck, motorbike, bicycle or pedestrian. The CAM provides information about the dynamic state of the represented road user. This information includes parameters such as location, time and dynamic parameters as speed and direction. Since a CAM only provides information about its own identified road user, it is considered as a single use case only. As specified in the CA ITS-S-MS specifications, the dissemination has two specific characteristics: • The message dissemination rate depends on the dynamic behaviour of the station. In principle, it depends on the speed, acceleration and movement. The dynamic dissemination behaviour is expected to depend on the type of represented road user and can differ for a vehicle compared to a bike. • The information to be shared as part of the CAM includes static as well as very dynamic parameters. As it is not that relevant to exchange the static parameters too often, related information is shared not as often as for the dynamic parameters, with the result that the disseminated message has a regular but not constant message size. CAM transmission can therefore be predictably estimated but depend on the environment (for vehicles, it differs for urban, sub-urban and highway). CAMs predictability depends on the intelligence of the system which disseminates these messages. This can be statically performed or done in various dynamic ways, including the use of Artificial Intelligence (AI). Possible parameter consideration for CAM: • Application statistically (at the time of application activation (application registration)): - Application indicator (internal for ITS-S). - Number of use cases supported (fixed = 1 for the time being) (internal for ITS-S). - Expected Priority level(s) to be supported. - Expected Technology and spectrum requirements (possibly including primary and secondary options). - Maximum message generation rate. - Minimum message generation rate. - Maximum message size. - Minimum message size. - Distribution type. • Application dynamically (based on activation of CAM dissemination (application registration)): - Application indicator. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 44 - Use case indicator (fixed = 1 for the time being). - Dissemination initiation (repeatedly based on environmental changes) can be repeated as long as needed: Actual Priority of the cycle of CAMs to be disseminated. Actual Technology and spectrum requirements (possibly including primary and secondary options). Expected average message generation rate. Expected maximum message generation rate. Expected minimum message generation rate. Expected average message size. Expected maximum message size. Expected minimum message size. Required operation limit message rate. Expected start time of dissemination. Expected termination time. Distribution type. - Dissemination termination (for instance when the vehicle is parked). |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2.4 Collective Perception (CP) Message | The CPM is a broadcasted awareness message (as identified in clause 6.2.1) that is continuously generated with variable interval and message size. The variable interval comes from the fact that the CPM is only generated when certain rules are satisfied and not based on a predefined interval. The CPM includes information recognized by the equipment from which the ITS-S is part off. The disseminate CPM can include (a subset of) the objects perceived by the station, information about the sensing capabilities, and information about the perception regions. The present Collected Perception Service (CPS) specification specifies a set of perceived object inclusion rules that significantly and dynamically influence the CPM size and generation rate (or interval). These rules are mainly based on the dynamics (e.g. position and speed) and type (e.g. VRUs vs vehicles) of the perceived objects. These object inclusion rules were extensively studied in related ETSI studies and in the scientific literature (see [i.25] and [i.28]). The CPM generation rules also include the possibility to dynamically include in each CPM a variable number of objects taking into account their value or utility, which is referred to as Value of Information (VoI) which refers to as redundancy in the scientific literature [i.26] and [i.27]. In addition, the object inclusion rules defined in the CPS standard allow the adaptation of the CPM size (or number of perceived objects) and interval based on inputs from RM. This ensures that the resources are used efficiently, and the amount of information sent by the collective perception service fits into the radio channel. As specified, the sender can design its own rules about the inclusion of perceived objects (with ObjectInclusionConfig flag set to "0"), which affect the CPM size and rate. The CPM is considered to support multiple use cases as the perceived objects represent different traffic participant types and other types such as empty road slots. As different traffic participant types could represent different message sizes and generation rules, one CPM can include objects representing of traffic participant types of one kind while other CPMs include objects representing of traffic participant types of one or more other kind. Also, for other reasons linking specific use cases to specific CPMs is advisable. This means not to disseminate object information of all objects recognizable by the sensor but select those object information relevant for the use cases to be supported. For system flexibility and robustness, it is advised not to include all the perceived objects in a single CPM. One of the main reasons is to avoid reaching the maximum message size and requiring the allocation of more than 10 MHz bandwidth. In general, from a system perspective it is better to keep awareness messages small so there is flexibility of making system choices. For the rest for CPM the same applies as for CAM. ETSI ETSI TR 104 073 V2.2.1 (2026-02) 45 |
fbcda0cf1f08aabcdc0778bd9c7a2a39 | 104 073 | 6.2.5 MIM and MVM | The Marshalling Infrastructure Message (MIM) and the Marshalling Vehicle Message (MVM) are used by the Automated Vehicle Marshalling (AVM) service or low-speed remote controlled automated driving (e.g. in parking areas or factories). The MIMs are disseminated by the infrastructure and the MVM are disseminated by vehicles. Each MIM sent by the infrastructure can target up to 32 vehicles and therefore its size can significantly change over time depending on the number of vehicles being remotely controlled. The MVM sent by vehicles have optional elements that also generate messages with variable size, but its variation is significantly lower than the MIM. A sequence of MIMs and MVMs are exchanged during the initialization of the AVM service. During the driving mode, the infrastructure and the vehicle have to periodically exchange MIMs and MVMs for the correct operation of the message service. To this aim, the AVM service introduces a message generation based on a mix of periodic and event- driven messages. By default, AVM messages are generated continuously at a recommended rate of 10 Hz. Additionally, the generation of MIMs and MVMs may be triggered by events, such as emergency stops. These events may cause the generation of one or more new messages and thus disrupt the periodic pattern. With regards to message dissemination behaviour, it can be expected that it has a more static behaviour as CAM but at high rates. It can be expected that the parameter set is quite similar to CAM/CPM. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.