hash stringlengths 32 32 | doc_id stringlengths 7 13 | section stringlengths 3 121 | content stringlengths 0 2.2M |
|---|---|---|---|
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.5.3 DownloadDataBlock | The DownloadData Message is defined in [4]. The use of the fields is defined in the DSM-CC specification (ISO/IEC 13818-6 [4]). |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.6 MPEG-2 Sections | ISO/IEC 13818-1 [2] defines a private_section structure which DSM-CC uses to provide re-assembly of Transport Stream Packets into DSM-CC messages. DSM-CC defines additional semantics on private_sections to support additional DSM-CC requirements. Called DSMCC_section, the structure is compatible with the private_section syntax so that compliant MPEG-2 Systems decoders may be used. The DSM-CC_section syntax is defined in ISO/IEC 13818-6 [4]. DVB Guideline: The encoding of the table_id_extension, version_number, section_number, and last_section_number are defined in table 4.16. Table 4.16: Encoding of DSMCC_section fields Message table_ id table_id_extension version_ number section_ number last_section_ number Download- ServerInitiate (DSI) 0x3B two LSB bytes of transaction_id of DSI 0x00 0x00 0x00 Download- InfoIndication (DII) 0x3B two LSB bytes of transaction_id of DII 0x00 0x00 0x00 Download- DataBlock (DDB) 0x3C moduleId module Version % 32 blockNumber % 256 Max(section_ number) DVB Guideline: For DownloadServerInitiate messages the 2 least significant bytes of the transaction_id shall be in the range 0x0000 - 0x0001. DVB Guideline: DownloadInfoIndication messages the 2 least significant bytes of the transaction_id shall be in the range 0x0002 - 0xFFFF. DVB Guideline: DVB has put some limitations to the basic DSM-CC specification regarding the transaction_id field to allow for easy filtering options to customer decoders. In particular, DSI messages have a value of 0x0000 or 0x0001 for the two LSB bytes. This enables receivers to bootstrap the carousel by setting up the section filters for table_id=0x3B (DownloadControlMessages) and table_id_extension= 0x0000 or 0x0001. Once the DSI message has been acquired the receiver can set up the section filter to listen to the other value of the two LSB bytes of the transaction_id. This shall trigger the receiver immediately once the carousel content is being updated. |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.7 Use of PSI descriptors | The object carousel specification in ISO/IEC 13818-6 [4] is network independent and is applicable for any type of Broadcast Network. Network independence is achieved by using the Tap concept. A Tap facilitates a reference to a particular network connection by means of an association tag. In the course of resolving an object, Clients have to associate the Taps to broadcast connections of the network. Clients need therefore an association table that makes the associations between the Taps and the connections of the broadcast network. For the implementation of U-U object carousels on top of Broadcast Networks that are based on MPEG-2 Transport Streams, the PSI mechanisms facilitate: 1) the association of a MPEG-2 Program (i.e. PMT) with an object carousel; 2) the association of a Tap with a PID or a MPEG-2 Program; 3) the localization of the PID on which the IOR of the Service Gateway is broadcast; and ETSI ETSI TR 101 202 V1.2.1 (2003-01) 45 4) the distributed implementation of an object carousel on top of multiple MPEG-2 Programs. This clause explains the use of three MPEG-2 descriptors that provide this functionality (see also ISO/IEC 13818-6 [4]). |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.7.1 Carousel identifier descriptor | The carousel identifier descriptor facilitates the association between a MPEG-2 Program and an object carousel. The syntax and semantics of the carousel_identifier_descriptor() are described in table 4.17 (see ISO/IEC 13818-6 [4]). This optional mechanism allows to acquire the ServiceGateWay of a ServiceDomain without first loading the Download Server Initiate and Download Indication Information messages. Table 4.17: carousel_identifier_descriptor Syntax bits Type Value Comment carousel_identifier_descriptor () { descriptor_tag 8 uimsbf 0x13 descriptor_length 8 uimsbf * carousel_id 32 uimsbf + FormatId 8 uimsbf Registered Identifier of the FormatSpecifier FormatSpecifier(){ FormatSpecifier_byte 8 uimsbf (see table 4.17a) N2 bytes } for ( i=0;i<N1;i++){ private_data_byte 8 uimsbf } } DVB Guideline: The carousel_identifier_descriptor() shall be inserted in the second descriptor loop of the PMT (ES_info) corresponding to the elementary stream carrying the DSI of the object carousel. This allows more than one object carousel per MPEG-program and implicitly identifies the PID on which each carousel should be booted from. The insertion of a carousel_identifier_descriptor() is also necessary to support the use of the DVBcarouselNSAPaddress, such as in the resolution of a LiteOptionsProfileBody reference. The FormatId identifies the format of a FormatSpecifier carried in the private data field of the descriptor. The syntax and semantics of this structure are defined in table 4.17a. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 46 Table 4.17a: FormatSpecifier in the carousel_identifier_descriptor FormatId Value Format Specifier Definition length [bits] Comment 0x00 no FormatSpecifier A value of 0x00 indicates the absence of a formatSpecifier. Thus the location of the ServiceGateway is only possible through the "standard" way interpreting the DSI and DII messages. 0x01 FormatSpecifier{ ModuleVersion ModuleId BlockSize ModuleSize CompressionMethod OriginalSize TimeOut ObjectKeyLength for (i=0;i<N1;i++{ ObjectKeyData } } 8 16 16 32 8 32 8 8 8 This FormatSpecifier is an aggregation of the fields necessary to locate the ServiceGateway, also found in the DSI and DII messages. NOTE: All field types are "uimsbf". timeout in seconds Object key of the service gateway object 0x02...0x7F reserved for future use The format Id values from 0x02 to 0x7F are reserved for future use of DVB 0x80...0xFF reserved for private use The format Id values from 0x80 to 0xFF are reserved for private use FormatId 0x01 identifies that the FormatSpecifier contains information (also found in the DSI and DII messages) that can be used to locate the ServiceGateway of the object carousel. Supporting this FormatID may have consequences for the broadcast server since this information must be kept consistent with changes to the ServiceGateway object and the module in which it is delivered. DVB Guideline: The presence of the FormatSpecifier with FormatId 0x01 implies that the DSI message and the module containing the ServiceGateway are carried on the same PID. |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.7.2 Association tag descriptor | The association_tag_descriptor (ISO/IEC 13818-6 [4]) facilitates the association between an association_tag and a PID and is therefore similar as the stream_identifier descriptor of DVB SI (EN 300 468 [6]). The assocation_tag descriptor uses however 16-bit assocation_tag (as opposed to the 8-bit component_tag of the stream_identifer_descriptor) and facilitates the identification of the PID on which the ServiceGateway is broadcast. The latter function allows receivers to bootstrap the object carousel efficiently from a PMT with a large number of PIDs. To label a PID with a particular association_tag value, the Server shall insert the association_tag descriptor in the descriptor loop of that PID. The syntax and semantics of the association_tag_descriptor are described in table 4.18. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 47 Table 4.18: association_tag_descriptor Syntax bits Type Value Comment association_tag_descriptor () { descriptor_tag 8 uimsbf 0x14 descriptor_length 8 uimsbf * association_tag 16 uimsbf + use 16 uimsbf 0x0000 0x0100-0x1FFF 0x2000-0xFFFF DSI with IOR of SGW DVB reserved user private if (use == 0x0000) { selector_length 8 uimsbf 0x08 transaction_id 32 uimsbf + transaction_id of DSI timeout 32 uimsbf + timeout for DSI } else if (use == 0x0001) { selector_length 8 uimsbf 0x00 } else { selector_length 8 uimsbf N1 for (i=0; i<N1; i++) { selector_byte 8 uimsbf } } for (i=0; i<N2; i++) { private_data_byte 8 uimsbf private data } } The use field may indicate the usage of the PID and shall specify the syntax and semantics of the selector field. If the use value equals 0x0000 then the DownloadServerInitiate message that carries the IOR of the Service Gateway is broadcast on this PID. In this case the data in the selector_byte fields shall contain the transaction_id and a timeout value. The semantics of the transaction_id and timeout fields are as follows. The value of the transaction_id field shall correspond to the transaction_id of the DownloadServerInitiate() message that conveys the IOR of the Service Gateway of the U-U object carousel. Except when the transaction_id in the association_tag_descriptor has the value of 0xFFFFFFFF. This value indicates that the transaction_id of the DownloadServerInitiate() message is not known at this point, but all DownloadServerInitiate() messages broadcast on the identified PID are valid. A transaction_id value of 0xFFFFFFFF may be used when the content of the DownloadServerInitiate() message is allowed to change (and thus the transaction_id in the message changes), without the need to update the PMT that contains the association_tag_descriptor. The timeout field shall indicate the time-out period in microseconds that may be used to time out the acquisition of the DownloadServerInitiate() message. A special value of the timeout (0xFFFFFFFF) indicates that no timeout value is known at this point. Allowing a 'static' PMT as described above. DVB Guideline: The default value for the use field shall be 0x0100. This means that the associated PID may or may not broadcast a DSI message. DVB Guideline: DVB reserves the range of 0x0101 to 0x01FF for the use field for future use. |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.7.3 Stream identifier descriptor | The stream_identifier_descriptor [DVB-SI] facilitates the association between a component_tag and a PID in an efficient way and may be used instead of (or in combination with) the association_tag descriptors. However since the component_tag field of a stream_identifier_descriptor is only an 8-bit field a mapping is necessary between component_tags and assocation_tags. DVB Guideline: A stream_identifier_descriptor in the descriptor loop of a PID shall be equivalent with an association_tag_descriptor for that PID with an association_tag value of LSB = <component_tag> and a use value of 0x0100. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 48 NOTE: This matching provides the flexibility to distribute the object carousel over multiple elementary streams and still use the same component_tag value in the different PMTs to refer to this particular data broadcast service. |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.7.4 Deferred association tags descriptor | An object carousel may use multiple PIDs, Services, and Transport Streams to broadcast the objects and associated control information. To facilitate Clients with the localization of all association_tags that are used in the different MPEG-2 Programs for the object carousel, a descriptor is defined that may be inserted in the first descriptor loop of the PMTs of the MPEG-2 Programs that implement the object carousel. The deferred_assocation_tags_descriptor contains association_tags that are used within the object carousel but that are not associated with a PID in the PMT in which the descriptor resides. The deferred_association_tags_descriptor contains therefore a forward reference to an MPEG-2 Program that does contain the PID to which the association tag is linked. Multiple deferred association tags descriptors may be inserted in a PMT if necessary. In addition a deferred_association_tag_descriptor may be used to refer to another DVB service (MPEG-2 program) as a result of a BIOP_PROGRAM_USE Tap. NOTE: Deferred_association_tags must be used whenever an object carrousel is broadcast using multiple services. For every service that carries a part of the carrousel, the list of deferred association_tags must be complete to avoid failing or false mapping of association_tags. The syntax and semantics of the deferred_association_tags_descriptor() are described in table 4.19. Table 4.19: deferred_association_tags_descriptor Syntax bits Type Value Comment deferred_association_tags_descriptor () { descriptor_tag 8 uimsbf 0x15 descriptor_length 8 uimsbf * association_tags_loop_length 8 uimsbf 2xN1 length in bytes for (n=0; n<N1 ; n++) { association_tag 16 uimsbf + } transport_stream_id 16 uimsbf + program_number 16 uimsbf + org_network_id 16 uimsbf + for (n=0; n<N ; n++) { Private_data_byte 8 uimsbf + } } |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.8 Information in the SI and PSI | For signalling just the use of the DVB object carousel the data_broadcast id shall be set to 0x0007. NOTE: If the use of the object carousel forms part of a specification which has registered a data_broadcast id, this alternative value (with the appropriate syntax for the selector fields) may be used instead. |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.8.1 SI Descriptor | The data_broadcast_descriptor in the SI can be used with the above value to indicate the presence of an DVB object carousel within a Service. In this case the selector field of the data_broadcast_descriptor contains a loop of object names that allows the bootstrapping of applications within the object carousel. The loop contains an ISO_639_language_code field which can be used (for example) to start an application based on preferred language. DVB Guideline: The object names used in the data_broadcast_descriptor shall exist in the object carousel. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 49 |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.8.2 Descriptors in PSI | The data broadcast_id descriptor can be used in a similar way as for data carousels (see clause 4.6.7.1). |
8772ed7b648c8fdde57091f4305eefdf | 101 202 | 4.7.9 Assignment and use of transactionId values | The use of the transactionId in the object carousel is inherited from its use as defined by the DSM-CC specification, and as such it can appear somewhat complex. The transactionId has a dual role, providing both identification and versioning mechanisms for control messages, i.e. DownloadInfoIndication and DownloadServerInitiate messages. The transactionId should uniquely identify a download control message within a data carousel, however it should be "incremented" whenever any field of the message is modified. NOTE: The term "incremented" is used in the DSM-CC specification. Within the scope of the UK DTT object carousel this should be interpreted as "changed". An object carousel are carried on top the data carousels and may be distributed over multiple data carousels. By a data carousel used below the object carousel, we mean in this specification a set of DownloadInfoIndication message transmitted on a single PID and the DownloadDataBlock messages carrying the modules described in the DownloadInfoIndication messages. The DownloadDataBlock messages may be spread on other elementary streams than the DownloadInfoIndication messages. The DownloadServerInitiate message in the context of object carousels is considered to be part of the top level of the object carousel and not associated with any data carousel. When a module is changed, the version number of the module needs to be changed. This implies that the DownloadInfoIndication message that references the module needs to be also updated. Since the DownloadInfoIndication is updated, the transactionId needs to be also changed. However, the transactionId of the DownloadInfoIndication message is used in other messages also, but the need to change the other messages should specifically be avoided and the implications of updating a module should be limited to the module itself and the DownloadInfoIndication that references the module. Therefore, additional rules on the usage of the transactionId have been specified as follows. The transactionId has been split up into a number of sub-fields defined in table 4.20. This reflects the dual role of the transactionId (outlined above) and constraints imposed to reduce the effects of updating a module. However, to increase interoperability the assignment of the transactionId has been designed to be independent of the expected filtering in target receivers. Table 4.20: Sub-fields of the transactionId Bits Value Sub-field Description 0 User-defined Updated flag This must be toggled every time the control message is updated 1 to 15 User-defined Identification This must and can only be all zeros for the DownloadServerInitiate message. All other control messages must have one or more non-zero bit(s). 16 to 29 User-defined Version This must be incremented/changed every time the control message is updated. 30 to 31 Bit 30 - zero Bit 31 - non-zero Originator This is defined in the DSM-CC specification ISO/IEC 13818-6 [4] as 0x02 if the transactionId has been assigned by the network - in a broadcast scenario this is implicit. Due to the role of the transactionId as a versioning mechanism, any change to a control message will cause the transactionId of that control message to be incremented. Any change to a Module will necessitate incrementing its moduleVersion field. This change must be reflected in the corresponding field in the description of the Module in the DownloadInfoIndication message(s) that describes it. Since a field in the DownloadInfoIndication message is changed its transactionId must be incremented to indicate a new version of the message. Also, any change in the DownloadServerInitiate message implies that its transactionId must also be incremented. However, when the transactionId is divided into subfields as specified above, updating a message will change only the Version part of the transactionId while the Identification part remains the same. Since the transactionId is used also for identifying the messages when referencing the messages in other structures, it is very desirable that these referenced would not need to be updated every time the control message is update. Therefore the following rule shall be applied when locating the messages based on the references: ETSI ETSI TR 101 202 V1.2.1 (2003-01) 50 When locating a message based on the transactionId value used for referencing the message, only the Identification part (bits 1...15) shall be matched. Using this rule, the implications of updating a module can be limited to the module itself and the DownloadInfoIndication message describing the module. Also, this implies that if a receiver wants to find out if a particular module that it has retrieved earlier has changed, it needs to filter the DownloadInfoIndication message that described that module and check if it has been changed. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 51 Annex A: DSM-CC messages for data carousel This annex contains the syntax of the DSM-CC Download messages as defined per July 12 1996. The semantic description of each field indicates where possible the value to use when implementing a DVB Data Carousel using this protocol. A.1 dsmccMessageHeader Table A.1: MPEG-2 DSM-CC Message Header Format Syntax Number of Bytes dsmccMessageHeader() { protocolDiscriminator 1 dsmccType 1 messageId 2 transactionId 4 reserved 1 adaptationLength 1 messageLength 2 if(adaptationLength>0) { dsmccAdaptationHeader() } } The protocolDiscriminator field is used to indicate that the message is a MPEG-2 DSM-CC message. The value of this field shall be 0x11. NOTE: The use of protocolDiscriminator 0x11 is dependent upon the response of ITU-T SG11 and ISO/IEC JTC1 to a liaison letter requesting that this value be assigned to DSM-CC. The dsmccType field is used to indicate the type of MPEG-2 DSM-CC message. The value of this field shall be 0x03 to indicate that the message is a U-N Download message. The messageId field indicates the type of message which is being passed. The values of the messageId are defined within the scope of the dsmccType. The transactionId field is used for session integrity and error processing and shall remain unique for a period of time such that there will be little chance that command sequences collide. The transactionId is of local significance, i.e. the value should be chosen by the broadcast server. The reserved field is ISO/IEC 13818-6 [4] reserved. This field shall be set to 0xFF. The adaptationLength field indicates the total length in bytes of the adaptation header. The messageLength field is used to indicate the total length in bytes of the message following this field. This length includes any adaptation headers indicated in the adaptationLength and the message payload indicated by the messageId field. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 52 A.2 dsmccDownloadDataHeader Table A.2: DSM-CC Download Data Header Format Syntax Number of Bytes dsmccDownloadDataHeader() { ProtocolDiscriminator 1 DsmccType 1 MessageId 2 DownloadId 4 Reserved 1 AdaptationLength 1 MessageLength 2 for(adaptationLength>0) { dsmccAdaptationHeader() } } The protocolDiscriminator field is used to indicate that the message is a MPEG-2 DSM-CC message. The value of this field shall be 0x11. NOTE: The use of protocolDiscriminator 0x11 is dependent upon the response of ITU-T SG11 and ISO/IEC JTC1 to a liaison letter requesting that this value be assigned to DSM-CC. The dsmccType field is used to indicate the type of MPEG-2 DSM-CC message. The value of this field shall be 0x03 to indicate that the message is a U-N Download message. The messageId field indicates the type of message which is being passed. The values of the messageId are defined within the scope of the dsmccType. The downloadId field is used to associate the download data messages and the download control messages of a single instance of a download scenario The reserved field is ISO/IEC 13818-6 [4] reserved. This field shall be set to 0xFF. The adaptationLength indicates the total length in bytes of the adaptation header. The messageLength field is used to indicate the total length in bytes of the message following this field. This length includes any adaptation headers indicated in the adaptationLength and the message payload indicated by the messageId field. A.3 DownloadServerInitiate Table A.3: DownloadServerInitiate message Syntax Number of Bytes DownloadServerInitiate() { dsmccMessageHeader() serverId 20 compatibilityDescriptor() privateDataLength 2 for(i=0;i<privateDataLength;i++) { privateDataByte } 1 } The serverId field shall be set to 20 bytes with the value 0xFF (i.e. the field is not used). ETSI ETSI TR 101 202 V1.2.1 (2003-01) 53 The compatibilityDescriptor() structure shall only contain the compatibilityDescriptorLength field of the compatibilityDescriptor as defined in DSM-CC (ISO/IEC 13818-6 [4]). It shall be set to the value 0x0000 (i.e. the field is not used). The privateDataLength field defines the length in bytes of the following structure. The privateDataByte fields shall convey the GroupInfoIndication structure defined in the DVB Specification for Data Broadcasting (EN 301 192 [1]). A.4 DownloadInfoIndication Table A.4: DownloadInfoIndication message Syntax Number of Bytes DownloadInfoIndication() { dsmccMessageHeader() downloadId 4 blockSize 2 windowSize 1 ackPeriod 1 tCDownloadWindow 4 tCDownloadScenario 4 compatibilityDescriptor() numberOfModules 2 for(i=0;i< numberOfModules;i++) { moduleId 2 moduleSize 4 moduleVersion 1 moduleInfoLength 1 for(i=0;i< moduleInfoLength;i++) { moduleInfoByte } 1 } privateDataLength 2 for(i=0;i< privateDataLength;i++) { privateDataByte } 1 } The downloadId field is the identifier of the download scenario in progress. The downloadId shall be uniquely defined within the Network for data carousel scenario and unique within the connection for the flow-controlled and non-flow-controlled scenarios. This identifier shall be used in all of the subsequent DownloadDataBlock, DownloadDataRequest, and DownloadCancel messages used by the download scenario in progress. The blockSize field is the length in bytes of the data in every block carried in the DownloadDataBlock messages, except for the last block of each module which may be smaller than blockSize. The windowSize is unused for broadcast data carousel scenarios and shall be set to 0. The ackPeriod is unused for broadcast data carousel scenarios and shall be set to 0. The tCDownloadWindow is unused for broadcast data carousel scenarios and shall be set to 0. The tCDownloadScenario field indicates the time out period in microseconds for the entire download scenario in progress. The compatibilityDescriptor() structure shall only contain the compatibilityDescriptorLength field of the compatibilityDescriptor as defined in DSM-CC (ISO/IEC 13818-6 [4]). It shall be set to the value 0x0000 (i.e. the field is not used). ETSI ETSI TR 101 202 V1.2.1 (2003-01) 54 The numberOfModules field is the number of modules described in the loop following this field. For flow-controlled and non-flow controlled download scenarios, the loop describes all the modules that have to be downloaded by the Client. For the data carousel scenario, the loop describes a subset of all the modules associated with this data carousel, although it may describes all of them. The moduleId field is an identifier for the module that is described by the moduleSize, moduleVersion, and moduleInfoByte fields. The moduleId is unique within the scope of the downloadId. The moduleSize field is the length in bytes of the described module. The moduleVersion field is the version of the described module. The moduleInfoLength field defines the length in bytes of the moduleInfo field for the described module. The moduleInfoByte fields shall convey a list of descriptors. Each list will define one or more attributes of the associated module. Note that the interpretation of these fields is different when the moduleId is in the range 0xFFF0 to 0xFFFF. In this case, these fields carry the ModuleInfo structure as defined by DAVIC. The privateDataLength field defines the length in bytes of the following privateDataByte field. The privateDataByte field is user defined. A.5 DownloadDataBlock Table A.5: DownloadDataBlock Syntax Number of Bytes DownloadDataBlock() { dsmccDownloadDataHeader() moduleId 2 moduleVersion 1 reserved 1 blockNumber 2 for(i=0;i<N;i++) { blockDataByte } 1 } The moduleId field identifies to which module this block belongs. The moduleVersion field identifies the version of the module to which this block belongs. The reserved field is reserved by ISO/IEC 13818-6 [4] and shall be set to 0xFF. The blockNumber field identifies the position of the block within the module. Block number 0 shall be the first block of a module. The blockDataByte conveys the data of the block. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 55 A.6 DownloadCancel Table A.6: DownloadCancel message Syntax Number of Bytes DownloadCancel() { dsmccMessageHeader() downloadId 4 moduleId 2 blockNumber 2 downloadCancelReason 1 reserved 1 privateDataLength 2 for(i=0;i<privateDataLength;i++) { privateDataByte } 1 } The downloadId field is the identifier of the instance of the download scenario in progress. It shall be used this to associate the DownloadCancel message to a particular download scenario in progress or data carousel. The moduleId and blockNumber fields indicate the last processed DownloadDataBlock message at the time of the cancel. If no data blocks have been processed, these fields shall be set to 0. The downloadCancelReason field contains a reason code that explains the reason for the cancellation. The reserved field is reserved by ISO/IEC 13818-6 [4] and shall be set to 0xFF. The privateDataLength field defines the length in bytes of the following privateDataByte fields. The use of the privateDataByte field is not specified by the DVB Data Carousel and may be used for proprietary information. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 56 Annex B: Encapsulation of DSM-CC messages in MPEG-2 sections This annex illustrates how DSM-CC messages are encapsulated in MPEG-2 sections. Please refer to the DSM-CC specification for the precise semantics. When DSM-CC Download messages are encapsulated in MPEG-2 Transport Streams, the DSMCC_section syntax shall be used. This structure inherits all of the Private_section syntax as defined in ISO/IEC 13818-1 [2]. Special semantics apply to the encoding of particular fields in the DSMCC_section header. The mapping of the DSMCC_section into MPEG-2 Transport Stream Packets and the maximum length of a DSMCC_section are governed by the semantics for Private_sections defined in ISO/IEC 13818-1 [2]. In some implementations, it is desirable to use the CRC_32 available in Private_sections. Because some systems may have difficulty calculating a CRC_32, the DSMCC_section syntax defines an alternative to using CRC_32. To be consistent with ISO/IEC 13818-1 [2], if the section_syntax_indicator is set to '1', then the CRC_32 shall be present and correct. In the case where the section_syntax_indicator is '0', the syntax of the section is the same as when the section_syntax_indicator is '1', except that the CRC_32 field is replaced with the checksum field. The resultant syntax is still compliant to ISO/IEC 13818-1 [2], since the payload following the section_length field shall be treated as private data. Since the section_syntax_indicator bit itself may be subject to a bit error, the private_indicator field shall be set to the complement of the section_syntax_indicator value. If the section_syntax_indicator is '0', then the private_indicator shall be verified to be '1', and if it is not, the section has suffered an error. Similarly, if the section_syntax_indicator is '1' then private_indicator shall be '0'. When section_syntax_indicator is '0' (CRC is not used) and the checksum field has been set to 0, another form of error detection shall be provided at a different layer. This requirement is imposed to ensure the DSMCC_section maintains the minimal requirements this specification imposes on its transport protocol. For syntax and semantics related to the carriage of private_sections (and therefore DSMCC_sections) within the MPEG Transport Stream, see ISO/IEC 13818-1 [2], clause 2.4.4, Program specific information. This includes the setting of the payload_unit_start_indicator, the presence of the pointer_field in the Transport Stream packet payload, and the use of packet stuffing bytes. Unless otherwise restricted, DSM-CC tables (i.e., one or more DSMCC_sections with the same table_id) may be contained in Transport Stream packets with the same value PID as other private_section formatted tables (e.g. in ISO/IEC 13818-1 [2] stream_type 0x05), if table_id parsing is done. When DownloadDataBlock messages are carried in MPEG-2 Transport Streams, only DownloadDataBlock messages with the same value of downloadId shall be contained in Transport Stream packets with the same value PID. This means that each PID can only deliver download data messages from a single data carousel. There is no such restriction specified for downloadcontrol messages, allowing such messages from any number of data carousels to be transported in the same elementary stream. In these cases the transactionId of a particular top-level control message must be explicitly identified using the data_broadcast_descriptor in SI to achieve predictable behaviour. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 57 Table B.1: DSM-CC Section Format Syntax N° of bits Mnemonic DSMCC_section() { table_id 8 uimsbf section_syntax_indicator 1 bslbf private_indicator 1 bslbf reserved 2 bslbf dsmcc_section_length 12 uimsbf table_id_extension 16 uimsbf reserved 2 bslbf version_number 5 uimsbf current_next_indicator 1 bslbf section_number 8 uimsbf last_section_number 8 uimsbf if(table_id == 0x3A) { LLCSNAP() } else if (table_id == 0x3B) { userNetworkMessage() } else if (table_id == 0x3C) { downloadDataMessage() } else if (table_id == 0x3D) { DSMCC_descriptor_list() } else if (table_id == 0x3E) { for (i=0;i<dsmcc_section_length-9;i++) { private_data_byte } } if(section_syntax_indicator == '0') { checksum 32 uimsbf } else { CRC_32 32 rpchof } } NOTE 1: The DownloadServerInitiate message, the DownloadInfoIndication message, and the DownloadCancel message are in the userNetworkMessage class. NOTE 2: The DownloadDataBlock message is within the downloadMessage class. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 58 Annex C: Naming of objects in directories In DSM-CC, the Directory objects provide a hierarchical tree-like directory structure (actually, the directory structure can be even more general graph than a tree). Each Directory object may contain references to other Directories (i.e. subdirectories) and other objects. When a Object is bound to a Directory, a name string is assigned to it that uniquely identifies the object within that directory. The ServiceGateway object is the root directory of the directory hierarchy. The path that includes the names starting from the ServiceGateway via possible subdirectories to an object identifies that object uniquely within the object carousel. An object can possibly be bound to multiple directories and thus have many paths all pointing to the same object. The following conventions for delimiting names and directories are the following: a) The forward slash "/" shall be used as a delimiter between directory names and object names. b) The forward slash is not allowed as part of a name. Relative path names are not required therefore no other convention is necessary. For various reasons, DSM-CC and the object carousels use many slightly different data structures for storing the path in different contexts. Thus, in different contexts the data structures that are used for storing the path may be different while they are still referring to the same path, i.e. name strings in the data structures are the same. C.1 Data structures used for names in DSM-CC User-to-User API The User-to-User API uses two different data structures for the path in different contexts. The reason for this is that the Directory object of DSM-CC inherits from the CosNaming::NamingContext object of CORBA. DSM-CC however adds some functionality to that and for this additional functionality it has been necessary to define a separate data structure for passing the path information. The CosNaming::NameComponent is the basis for all name data structures. It represents one part of the whole path name to the object, i.e. the name within one subdirectory. The NameComponent structure contains two fields: id and kind. The id field contains the actual name string and the kind field contains the type of the object. The CosNaming::Name is a sequence of NameComponets and represents the whole path. However, this structure normally identifies the relative path starting from the directory where it is used in. When the Name is used in the ServiceGateway, it naturally represents the full absolute path. In some functions, the Name is carried inside a structure called CosNaming::Binding. In addition to the Name, the Binding contains a field that identifies the BindingType. The purpose of the BindingType is to identify a classification of the object that the Name points to. DSM-CC has defined another data structure for the path, the DSM::PathSpec. The PathSpec consists of a sequence of DSM::Step structures. The Step contains the same NameComponent as is used in the CosNaming::Name and also an additional process flag that is used in some functions to inform if the operation should be applied to this part of the path or not. When the PathSpec is used in DSM-CC, there is usually another parameter also: a PathType. The PathType identifies the way how the PathSpec should be interpreted. It differentiates between the two different ways how the PathSpec is used. When the PathType is DEPTH, the meaning of the PathSpec is equivalent to the Name, i.e. it is a relative path down the directory hierarchy starting from the current directory. However, when the PathType is BREADTH, the NameComponents in the PathSpec are used to identify multiple different objects within the same directory. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 59 C.2 Data structures used for names in object carousels For optimizing the transportation, the object carousels use slightly different data structures than the U-U API. These data structures are however intended to be equivalent with the ones that are used in the API. The BIOP::NameComponent is equivalent to the CosNaming::NameComponent, but the maximum lengths for the strings have been added to optimize the encoding. The BIOP::Name is equivalent to the CosNaming::Name, but it defines an upper bound for the number of NameComponents in it to optimize the encoding. The DirectoryMessage of the object carousels provides the necessary information for implementing the Directory object. The Directory message contains BIOP::Bindings that include the Name that identifies the path to the object starting from this directory and the Interoperable Object Reference that contains the necessary information to locate the actual object. The BIOP::Binding is different from the CosNaming::Binding so that the BIOP::Binding contains the object reference while the CosNaming::Binding does not. This is because in the object carousels, it is used to carry the location of the object, while in the API the location of the object is not visible to the application but internal to the Directory object. C.3 CORBA strings in object carousels In a number of places object carousel messages include text strings. These are all formatted in accordance with clause 12.3.2 of CORBA V2.0. I.e. the text is preceded by a length of 1, 2 or 4 bytes, depending of the coding, field and followed by a null terminator. In general this is can be seen clearly in the syntax tables that follow. However, for clarity CORBA format strings are used in the following places: Table C.1: Location of CORBA format strings String location objectKind_data BIOP::FileMessage syntax objectKind_data, id_data, kind_data BIOP::DirectoryMessage syntax objectKind_data BIOP::StreamMessage syntax objectKind_data, eventName_data BIOP::StreamEventMessage syntax type_id_byte BOP::IOR syntax id_data, kind_data Syntax of Options Profile Body with ServiceLocation component ETSI ETSI TR 101 202 V1.2.1 (2003-01) 60 Annex D: Example of an object carousel Figure D.1 illustrates an object carousel that is distributed over three elementary streams belonging to the same service. The DownloadServerInitiate (DSI) message is carried on the first elementary stream. It contains the object reference that points to the service gateway. The tap with the BIOP_ DELIVERY_PARA_USE points to a DownloadInfoIndication (DII) message that provides the information about the module and the location where the module is being broadcasted. In the example, the service gateway object is in the module number 1 that is carried on the second elementary stream (indicated by a BIOP_OBJECT_USE tap structure in the DII message). The Service Gateway object is a root directory that, in this example, references three subdirectories. Taps with BIOP_DELIVERY_PARA_USE are used in the object references of the subdirectories to provide links to the modules via the DownloadInfoIndication (DII) message. The two first subdirectories "dir1" and "dir2" are referenced in the DII message that is carried in the first elementary stream. The third subdirectory is referenced in the DII message carried in the third elementary stream. In this example, the two first elementary streams carry the messages of one logical data carousel while the third elementary stream carries the messages of another logical data carousel. All these belong to the same object carousel. In the example, the third elementary stream contains the objects in the "dir3" subdirectory and the objects in the "dir1" and "dir2" subdirectories are distributed over the first and second elementary stream. It is important to note that the third elementary stream may originate from a completely separate source than the first two elementary streams. The directory hierarchy and objects contained in the third elementary stream are "mounted" in the root directory by providing the "dir3" directory entry with the appropriate location information. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 61 Figure D.1: Example object carousel ETSI ETSI TR 101 202 V1.2.1 (2003-01) 62 This type of structure could be used, for example, in a national information service that contains some regional parts. The common national parts could be carried in this example case on the two first elementary streams that are distributed unmodified in the whole country. The regional parts are carried in the third elementary stream that is locally inserted at each region. From the application's point of view, the common national parts are in the "dir1" and "dir2" subdirectories while the regional parts are in the "dir3" subdirectory. Another example where this type of structure could be used is if the service contains multiple independent applications. In this case, each application could be placed in its own subdirectory and these subdirectories might be carried as separate data carousels on different elementary streams. ETSI ETSI TR 101 202 V1.2.1 (2003-01) 63 Annex E: Bibliography ISO/IEC 8802-1: "Information technology - Telecommunications and information exchange between systems - Local and metropolitan area networks - Specific requirements - Part 1: Overview of Local Area Network Standards". ISO/IEC 8802-2: "Information technology - Telecommunications and information exchange between systems - Local and metropolitan area networks - Specific requirements - Part 2: Logical link control". IETF RFC 1951: "DEFLATE Compressed Data Format Specification version 1.3". ISO/IEC 13818-2: "Information technology - Generic coding of moving pictures and associated audio information - Part 2: Video". ISO/IEC 13818-3: "Information technology - Generic coding of moving pictures and associated audio information - Part 3: Audio". ETSI ETSI TR 101 202 V1.2.1 (2003-01) 64 History Document history V1.1.1 February 1999 Publication V1.2.1 January 2003 Publication |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 1 Scope | The present document is intended as a companion document to the full specifications, EN 300 707 [1] and EN 300 708 [2], covering the data format and transmission via Teletext of an Electronic Programme Guide (EPG). It is primarily aimed at EPG/Teletext service providers and network operators with the intention that the specifications are interpreted in a consistent way while recognizing that there are many options available to all the parties involved in creating the total system. The present document outlines the basic EPG concepts and highlights the key parameters for a successful service. It makes recommendations as to how aspects of the specifications should be implemented and suggests strategies to maximize the transmission efficiency of both normal Teletext and EPG services when they coexist in the same TV channel. The present document has been revised in the light of the knowledge and experience gained from operating real services with a variety of decoders available. It is anticipated that the operation of enhanced EPG may require a further revision. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 2 References | For the purposes of this Technical Report (TR) the following references apply: [1] ETSI EN 300 707: "Electronic Programme Guide (EPG); Protocol for a TV Guide using electronic data transmission". [2] ETSI EN 300 708: "Television systems; Data transmission within Teletext". [3] ETSI EN 300 706: "Enhanced Teletext specification". [4] ETSI EN 300 231: "Television systems; Specification of the domestic video Programme Delivery Control system (PDC)". [5] ETSI TS 101 231: "Television systems; Register of Country and Network Identification (CNI), Video Programming System (VPS) codes and Application codes for Teletext based systems". [6] ETSI TR 100 287: "Television systems; Code of practice for enhanced Teletext". |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 3 Definitions and abbreviations | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 3.1 Definitions | For the purposes of the present document, the following terms and definitions apply: application information: Data block providing the name of the EPG service provider and a list of the networks supported. The total number of programmes and number of days covered for each network is indicated. attributes: additional "machine-readable" information on a programme event, e.g. "live" or "subtitled" NOTE: Can be used by a decoder as a filter when searching the database. Also known as "Feature Flags". bundle information: Data block applicable to all data broadcasting applications within a given stream. It enables the number of applications and their type to be identified. category: content of a programme event; e.g. "News", "Sport", "Drama" composite EPG decoder: decoder which compiles a multiple channel display by scanning several EPG services on different networks conditional access: method by which network operators/EPG service providers can restrict access to all or part of their service to a particular group of viewers ETSI ETSI TR 101 288 V1.3.1 (2002-12) 8 database: The EPG service provider's store of all programme-related data. In a decoder context, the sub-set of the EPG transmission which the decoder has stored. data stream: continuous sequence of EPG-related data NOTE: In order to maximize efficient use of the VBI capacity and to guarantee a maximum performance for an EPG service, the total EPG data stream can be split into two separate streams (Stream 1 and Stream 2) for transmission purposes. decoder: collects and decodes the transmitted EPG data. It processes and stores the data and under user control selects the information for display NOTE: Decoders can differ in their storage capacity and display capability. EPG service provider: generic term for the different parties involved in compiling an EPG database and formatting it ready for transmission event area: part of the EPG display screen where programmes are listed or menu items are displayed far information: programmes which are not scheduled for transmission today or tomorrow but for the third day onwards NOTE: Compare with Near information. feature flags: See "Attributes". filler packets: dummy packets inserted onto otherwise unused VBI lines which exist as a result of obeying the 20 ms rule full EPG: multiple channel EPG service which includes navigation and sorting information header area: The top-most part of the EPG display screen. Its contents are defined by the EPG service provider. housekeeping data: elements within an EPG transmission that are essential to its operation but which do not form part of the programme database Level 1, 1.5, 2.5, 3.5: Teletext presentation levels message area: part of the EPG display screen where text messages defined by the EPG service provider are displayed NOTE: Normally the text will be linked to a highlighted event in the Event Area. multiple channel EPG: EPG service transmitted on a particular network which comprises information on programmes from more than one network or TV channel navigation: method by which the viewer interacts with the decoding system via menus, leading him to the desired programme information navigation area: the bottom-most part of the display screen where the decoder displays locally generated user-interface prompts and messages to enable the viewer to access the EPG navigation information: Data block used to create a menu structure for navigation purposes within a Full EPG. It defines the text to be displayed and the links to the next level of menu or programme information. near information: programmes scheduled for transmission today or tomorrow network operator: generic term for the different parties responsible for the delivery of the EPG data now and next: details about the current TV programme (or programmes in the case of a multiple channel service), plus the programme(s) that follow on immediately OSD information: data block used to define display data for areas of the display screen that are under the control of the EPG service provider programme information: data block containing information about one programme event. It includes channel, times, ratings, themes, etc. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 9 refresh procedure: constant transmission of the complete EPG database NOTE: Different parts of the database can be transmitted at different rates according to the priority of the data. Stream 1: The Teletext pages carrying Near Information. Their transmission obeys the 20 ms page clearing rule. The pages are distinguishable from those in Stream 2 through the allocation of a value of "0" to the S3 component of the page sub-code. Stream 2: The Teletext pages carrying the remaining EPG data that is not included in Stream 1. Their transmission does not have to obey the 20 ms page clearing rule. The pages are distinguishable from those in Stream 1 through the allocation of a value of "1" to the S3 component of the page sub-code. transparent strings: Sequences of characters and attributes defined by the EPG service provider as part of the EPG database. Spacing attributes (a sub-set of those available with Level 1 Teletext) can be used within each string. Accented characters and symbols found in Level 1.5 Teletext transmissions are also accessible. update procedure: transmission of information which enables a decoder to update quickly a section of its database when changes occur in the programme schedule 20 ms page clearing rule: This rule defines the minimum interval between the transmission of the page header (row 0) of a Teletext page and the transmission of the remaining packets. It is essential for some existing Teletext decoders to give them time to erase the old page from memory. Level 2.5 (and above) decoders can operate without such a delay being necessary. This is referred to as the 20 ms rule in EN 300 707 [1]. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 3.2 Abbreviations | For the purposes of the present document, the following abbreviations apply: AI Application Information BI Bundle Information CNI Country and Network Identification code EACEM European Association of Consumer Electronic Manufacturers EPG Electronic Programme Guide HI Helper Information LI Language Information MI Message Information MIP Magazine Inventory Page NI Navigation Information OI OSD Information OSD On Screen Display PDC Programme Delivery Control PI Programme Information TI (sub-)Title Information TV TeleVision UI Update Information VBI Vertical Blanking Interval VCR Video Cassette Recorder VPS Video Programming System |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 4 Introduction to Electronic Programme Guides (EPG) | EPGs are to be seen as a major improvement of broadcasting information about TeleVision (TV) programmes to the viewer. The technology used to present or display the information may be different from that used to transmit the data. The subject of the present document is the EPG standard conceived by the European Association of Consumer Electronic Manufacturers (EACEM). The full coding details are specified in EN 300 707 [1]. The present document also covers the transmission of the data as one form of data broadcasting via page-format Teletext. These aspects are dealt with in EN 300 708 [2]. A commercial name for EPG services based on these specifications is given in annex A. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 10 Electronic programme guides go far beyond the possibilities of programme listings provided by normal Teletext services. It is to be seen as a major improvement resulting from the transmission of additional programme information and a new generation of decoders especially designed for the purpose of handling an EPG service. The vertical link between hardware and software manufacturers on the one side and the Teletext broadcasters on the other will provide the viewer with a fast, attractive and easy to use information service about TV programmes. As a consequence, it may improve the attractiveness of a broadcaster's programmes as well as the products of a TV or VCR manufacturer who chooses to implement an EPG decoder. In addition, it can be implemented in a multimedia PC equipped with a video/Teletext capture card. An EPG will offer easy, attractive and fast access to a listing of programmes in the "near" future (for example, today and tomorrow) for one or more channels. Depending on the editorial policy of the broadcaster concerning the scope of his EPG service, the guide may also cover programmes further ahead across several channels. Going beyond the chronological listings of TV programmes, the more elaborate implementations of an EPG will enable the viewer to select programmes by personal criteria, e.g. programme theme. Another important part of an EPG service will be the navigation elements which will help the user to view and select the various categories of programme information displayed by the EPG decoder. Navigational elements can also make the application attractive by their design and presentation and, therefore, they should focus on the graphical possibilities of Level 2.5 Teletext as defined in EN 300 706 [3]. Compatibility with lower Teletext levels shall still be maintained as some EPG displays will be very simple. Depending on the display hardware used in the decoder, the appearance of the display may well exceed that available even with Level 3.5 Teletext, for example in a multimedia PC. An EPG does not only include various ways of listing programme information but also offers an easy method of programming VCRs through its common link to the established PDC and VPS protocols as defined in EN 300 231 [4]. In addition, it is possible to create a special service for programme information, which can be a separate service from the traditional Teletext service. The EPG data is carried by the network operator's video signals as part of the Teletext stream. It is additional data to the "normal" data transmitted in a Teletext service. It is transmitted to the TV receiver or VCR where it is stored in a database. (From an editorial point of view, it does not matter whether the decoder resides in a TV receiver or a VCR.) The "computing power" of the decoder processes the database and under the user's command extracts the programme data and formats it for display. There may be many EPG services available to the viewer, either single or multiple channel and transmitted on one or more TV channels or networks. The viewer will have to select the EPG services he wishes to decode, store and use. In a simplified form the system can be presented as shown in figure 1. This code of practice aims to: - provide the essential background information about an EPG service; - highlight the key parameters and concepts for successful EPG operation; - make suggestions on how an EPG service may best be exploited; - give recommendations and examples of how an EPG service may be implemented; - suggest strategies to maximize the efficiency of both the Teletext and EPG services. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 11 TV set / VCR EPG Data Normal Teletext Service EPG Database xxxxxxxxxxx xxxxxxxxxxx EPG Decoder TV Info Teletext Database xxxxxxxxxxx xxxxxxxxxxx Teletext Decoder Teletext Figure 1: Basic system concept |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 5 Fundamentals of an EPG | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 5.1 A non-proprietary and non-discriminatory system | The ETSI standard EN 300 707 [1] is an open system which is non-proprietary and non-discriminatory. It is to be considered as an enabling specification which any broadcaster can adopt providing that they respect the fundamental agreements. NOTE: There are no licensing costs for a network operator or EPG service provider. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 5.2 Key concepts | When thinking about an EPG one has to be clear as to what one is actually considering. Often very similar terms are used for the various types of transmission and capabilities of decoders. Similarly, there are very many possibilities so there are often paradoxes where the capabilities of the specification exceed the Teletext transmission capacity. Fundamental to an EPG is the transmission of a large database of programme information. The decoder's database may be of a different size in different products. Only the minimum size is stated in the specification. Through the use of input filters only a subset of the total database will usually be stored. The information is stored in the TV receiver or VCR and when the information is retransmitted, the opportunity may be taken to modify it. This refresh operation is different from the explicit update of information to the decoder's database. The programme information can be categorized by the following terms: This channel: data relating to the network/channel carrying the EPG; Other channels: data relating to other networks/channels; Near information: data for programmes scheduled for transmission within two days; Far information: data for programmes further in the future. Near and Far Information may be transmitted in slightly different ways but this will be transparent to the viewer. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 12 If one assumes that a decoder will be tuned to a channel for a few minutes each day, it will be able to store the frequently transmitted Near Information. When the viewer turns on the TV tomorrow, he can be presented with the information for the current day from the stored database. Because a VCR is likely to be powered continuously, there is ample opportunity for its stored database to be refreshed and updated. However, a TV receiver can only acquire new information if it is turned on and tuned to the channel carrying the EPG service. NOTE: Some high-end receivers will contain a second tuner to support Picture-in-Picture operation. In some circumstances this will allow the EPG data to be refreshed and updated while the viewer is watching a different channel. Because of this, the transmission of the entire database should be completed within 20 to 30 minutes (the duration of a typical TV programme), the This Channel/Near Information within 30 s, This Channel Now and Next Four Programmes within 10 s. An EPG decoder can start to acquire data at any point in the refresh cycle and, accordingly, a fairly constant transmission is preferred over infrequent bursts of data. EPG and the existing Teletext services can share the same VBI lines. A complex EPG service can be sent in the same way as display enhancement data for Level 2.5 or 3.5 Teletext, occupying some or all of the spare capacity that cannot be used by the normal Teletext service due to decoder constraints. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 5.3 Basic editorial decisions | The attractiveness of an EPG service depends upon the performance of the decoder implemented in the TV receiver or VCR and on the quality of the programme information. The main editorial decisions to be considered are as follows: - the kind of programme data to be transmitted: - the number of days the information will cover; - the number of programmes and channels; - the depth of information per programme. - the refresh and update procedure of the data; - the arrangement of navigational elements; - Conditional Access (CA); - copyright; - the organization of the data transmission within the Teletext stream. Recommendation: Decisions on the amount of programme information to be transmitted via Teletext to the EPG decoder can have several consequences for the Teletext service. The aim is to find a "balance" between the EPG data and the "normal" Teletext pages of the service. The transmission of EPG data should not adversely affect the capacity of the Teletext service. There are several different technical measures that may be taken to minimize the effect of the EPG on the Teletext service. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 5.4 EPG and Teletext | Even if the EPG and Teletext represent different services there are more common links besides the fact that the source data for EPG is transmitted via Teletext. Both services not only share VBI lines they also obey the same display specifications (at least Level 1.5 Teletext with the corresponding character sets and serial attributes). Both systems are compatible with PDC/VPS and use page-based Teletext transport. To identify a broadcaster, both systems make use of the country and network identification data (CNl) which is transmitted in packet 8/30 format 2 of the normal Teletext service or via VPS. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 13 The EPG protocol includes the possibility of the reuse of text from existing Teletext pages, but the reading of an information block out of a Teletext page is quite a complicated process where many problems may occur. For example, the text block has to be referenced not only by its page number but also by its exact position (row and column). In addition, it requires more decoding than when handling explicit EPG text data. It is dependent on the Teletext service and so the text will take longer to acquire. If the reuse of Teletext information is required in order to reduce the amount of transmitted data for the EPG, it should be kept in mind that the perceived advantages are off-set by a number of disadvantages. The reading of Teletext pages is only allowed for the "long information" block (see clause 8.3.1). The only data that is imported from a Teletext page is the foreground character information as would be displayed after the addition of any Level 1.5 accented and supplementary characters carried in packets 26, and replacing any colour control characters or other attributes with "space". This text is then processed, stored and displayed as if it were explicitly transmitted in the EPG data stream. The colours of the text and background of the message box are defined by the decoder manufacturer. Recommendation: For consistency, the same source of basic programme data should be used for both EPG and Teletext services. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 6 Characteristic elements of EPG services | The provider of an EPG has to consider the four main points which determine the functionality of an EPG service: - single channel; - many channels; - navigation; - Near and Far programmes. The EPG service as broadcast is independent of the complexity of the decoder used by the TV set or VCR. Therefore it is necessary to consider both the various types of broadcast (this clause) and the types of decoder (clause 7). Recommendation: The range and depth of the information carried by an EPG shall be at least as comprehensive as that on the existing Teletext service; e.g. a broadcaster whose Teletext service already offers multiple channel programme listings should also offer a multiple channel EPG. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 6.1 This Channel EPG | As shown in figure 2, a This Channel EPG service only contains information on the programmes of the network or TV channel on which the EPG service is broadcast. "This channel" information Network operator / EPG service provider Near Far This channel EPG with Near and Far information Far database Near database Figure 2: This channel EPG ETSI ETSI TR 101 288 V1.3.1 (2002-12) 14 Recommendation: Ideally, all broadcasters should transmit at least This channel/Near Information on all of their channels every 10 s, and the remainder of This Channel/Near at a reasonable transmission rate. However, the minimum service should contain Now information and the next four programmes for the channel although this is not likely to be regarded as a true EPG service. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 6.2 Multiple channel EPG | A Multiple Channel EPG service is shown in figure 3 and comprises information on the programmes of more than one network. The EPG service provider has to maintain a database of all the channels in his service. The theme of a programme is defined only by the categories included in the specification. These should be used by all broadcasters. Multiple channel EPG with Near & Far information Network operator / EPG service provider "This channel" information Channel 2 Channel 3 Channel 4 Channel 5 Channel 6 Far database Near database Figure 3: Multiple channel EPG |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 6.3 Full EPG | A Full EPG service, as shown in figure 4, is a Multiple Channel EPG service with the addition of navigational elements which are under the control of the service provider. These elements should enable the viewer to identify easily programmes that meet their personal criteria. A Full EPG service allows an extensive sorting of programmes by themes defined by the service provider. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 15 Full EPG with Near & Far information plus navigation Network operator / EPG service provider "This channel" information Channel 2 Channel 3 Channel 4 Channel 5 Channel 6 Far database Near database Navigational elements Figure 4: Full EPG with navigation |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 6.4 The "Near" and "Far" distinction | The other characteristic of an EPG is the editorial/technical division of the data into Near or Far Information. Near Information will be refreshed more frequently and may contain more details about each programme than Far Information. The Near Information for at least the first channel shall be transmitted in Stream 1, see clause 8.6. This restriction may cause an editor to limit the number of programmes included for a particular channel. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7 Types of decoder | A range of decoders with different storage capacities, functionality and display features is envisaged. Editors may offer a level of service which simple decoders are not capable of handling in full. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7.1 Simple decoder | These decoders will have a very limited amount of memory but they will support at least the minimum service level (Now and the next four programmes for This Channel). |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7.2 Single channel decoder | These decoders will have sufficient memory to handle the This Channel information (Near and Far) extracted from the selected EPG service. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7.3 Multiple channel decoder | These decoders will have sufficient memory to handle the Near and Far Information from a multiple channel service. Any navigation features are defined by the decoder. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 16 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7.4 Full EPG decoder | These decoders will have sufficient memory to handle the Near and Far Information from a multiple channel service. They implement the navigation features defined in a Full EPG service. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 7.5 Composite EPG decoder | A Composite EPG decoder, figure 5, scans any available EPG service and extracts the This Channel information. It then composes a multi-channel EPG. As the information comes from many different sources, the display and any navigation features are defined by the decoder. Network operator / EPG service provider A "This channel" information Network operator / EPG service provider B "This channel" information Network operator / EPG service provider C "This channel" information Network operator / EPG Service provider D "This channel" information Composite EPG Far database Near database Figure 5: Composite EPG decoder ETSI ETSI TR 101 288 V1.3.1 (2002-12) 17 7.6 Comparison of composite and full EPG systems and decoders Composite Feature Full As many as can be received by the decoder. Number of channels As many as the service carries. In theory, as many as each EPG service carries. In practice, limited by the size of the decoder's memory. Number of days As many as the service carries; limited by the size of the decoder's memory. Scans all the channels at some time each day: approximately 1 minute per channel for Near Information. Acquisition method Tuned to one channel for, say, 20 minutes each day: about 30 s for Near Information. Decoder manufacturer. Screen layout EPG provider. Very limited; decoder defined. Navigation Flexible; defined by the EPG service provider. Themes defined in EN 300 707 [1], if supported by the decoder. Programme categories Themes defined in EN 300 707 [1] plus those defined by the EPG service provider. Decoder manufacturers may offer additional features. The This Channel information from the channels received (includes Multiple Channel EPG transmissions). Source of information The selected Full EPG service. The channel is omitted from the channel list. What happens if a channel does not have an EPG? Irrelevant; the service provides this information. Preferred order is the order the TV channels are stored in the decoder. Order of channels displayed Full EPG service provider defined; CNI list order. Has to be obtained by the individual EPG service providers. Copyright of information Has to be obtained by the Full EPG service provider. Under the control of individual service providers for parts of the EPG. Conditional Access (CA) Under the control for all or part by the EPG service provider. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8 The structure of an EPG service | An EPG database can be divided into five main parts: - Bundle Information block; - Application Information block; - programme database comprising a number of Programme Information blocks; - database for display structures; - database for navigational structures (Full EPG only). There are a number of other minor elements but the majority of the data consists of the above. At some point in the transmission chain the data blocks will be encoded for transmission and then packed into Teletext pages in order to be broadcast. There are instances where the number of bits for a particular field is different in different structures. At all times the value of the field shall be the same. The Block numbers in each structure should be contiguous to aid the memory management in the decoder. The broadcaster should note that if a decoder has insufficient memory to store all the blocks of a structure, the latest PIs in time and date may be disregarded. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 18 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.1 The Bundle Information block | EPGs are but one form of data broadcasting and the Teletext pages used to transmit an EPG can also be used to convey data for other applications, including other EPGs from different service providers. The other applications will be coded in a similar manner to the EPG data and a Bundle Information block is transmitted frequently to inform decoders of the number and type of the applications within the data stream. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.2 The Application Information block | The Application Information (AI) block is a single entity containing data concerning the complete EPG database. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.2.1 Block contents | The principal contents of the AI block is: - the name of the EPG service provider; - the number of networks supported. For each network: - its name (as a text string) and CNI code; - number of days covered in the listings; - an indication of the first and last programmes in the listings (the first programme will be assumed to be the Now programme by a decoder); - the Teletext page containing conventional Teletext-style listings; - a network identification number that then appears in each Programme Information block belonging to that network. - the network providing This Channel information (see clause 8.2.3); - the version number of the database; - indications of the number of each type of data block in the total guide. Some of the information is included twice when there is a need to inform the decoder as to how the data is split between Streams 1 and 2. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.2.2 Transmission aspects | Since the Application Information block defines basic reference data for each network it has to be received by a decoder before any Programme Information blocks can be interpreted correctly. Consequently, it has to be transmitted frequently, and only in Stream 1 so that it is accessible to all types of decoder. It shall be updated and re-transmitted on a change of programme on any network since the data relating to at least the first (i.e. Now) programme for that network will have changed. Ideally, this should be linked in real time to the programme change but this may not be practical for other than the This Channel network. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.2.3 Identification of the broadcaster | Where the decoder acquires a multiple channel EPG service there may be problems with the identification of the broadcaster. Thus there is a component within in the Application Information block (the this_network_operator_no value) which identifies the source of This Channel information. The order of the remaining channels is at the EPG service provider's discretion. To allow a TV or VCR to tune automatically to a channel listed by the EPG, the EPG service should only include data for channels that are broadcasting a CNI code via a packet 8/30 format 2 or VPS. This includes the channel carrying the EPG. The allocation of NI codes to networks is defined in TS 101 231 [5]. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 19 If a broadcaster wishes to supply only a Default EPG (see clause 13) a Magazine Inventory Page (MIP) shall be transmitted. Some EPG decoders use the last 8 characters in packets X/0 to determine the time. These characters do not necessarily carry the time so it is desirable to the transmitted Teletext stream contains packet 8/30 format 1 with the correct UTC and Local time offset so that the date and local time can be decoded. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.3 Programme Information blocks | A Programme Information (PI) block is transmitted for each individual programme event in the EPG. It contains text and "machine-readable" data. PI blocks will form the bulk of any EPG database. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.3.1 Text content | The three text fields in a PI block are: - Title: the name of the programme, maximum length of 40 characters; NOTE: Early decoders may truncate titles to around 30 characters. - The Short Information can provide details about the programme. Alternatively it may be used for other purposes such as advertising. On the display it will appear in the Message Area and shall always be displayed when the programme is selected or otherwise highlighted by the user. Subsequent user actions may result in other information appearing in the Message Area. The maximum length of Short Information is 255 characters; - The Long Information will be used for longer critiques of films or for information such as recipes or contact addresses associated with the programme. Not every programme will have a Long Information within a typical EPG service. This information will be displayed in the Message Area or full screen, and the maximum length is 1 000 characters. Recommendations: For the Title it is recommended to be as efficient as possible remembering that spacing attributes count as a character. If a Title is greater than 40 characters it should be repeated in full in the Short Information. A carriage return command should not be used within a Title to prevent complications within a decoder. The recommended length of the Short Information is typically 140 characters as it has to fit within the Message Area (the size of which is defined by the EPG Service Provider via the OSD Information block). The Short Information should explain what the Title does not. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.3.2 Attributes, categories and ratings | A PI block includes "machine-readable" information about a programme such as network, transmission data and time, PDC/VPS code, sorting categories and other attributes. Attributes (or feature flags) provide the viewer with extra information about the programme. Attributes can be used to filter and sort programmes in the stored database according to the user's preferences. A list of attributes and other parameters that can be defined for each programme event is given in annex B. To allow sorting of programmes by category (i.e. theme or genre) each PI block carries thematic information. In other than a Full EPG service, the categories are those defined in the EPG specification, see annex C. A Full EPG service can define its own categories as well as using the pre-defined set. The pre-defined table of EPG themes is identical to those defined for PDC in EN 300 231 [4]. Recommendations: Attributes are very important. They are the means by which the viewer can make an individual selection from the programme information available. Even when operating a Full EPG service, it is recommended to build upon the pre-defined theme categories. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 20 Different categories and attributes can be combined within the decoder to produce new filter criteria. For example, the viewer can choose to select "Football/Live" or "Movie/Widescreen". Ratings can also be used to help the viewer find a suitable programme. Obviously ratings cannot be considered as objective criteria. Two kinds of ratings are distinguished: - parental rating (indicating a recommended minimum age for a group of viewers); - editorial rating (offering a global recommendation concerning the quality of a single programme event). In both cases a field in the PI block defines values for the event. Editors should set the values appropriately, see in EN 300 707 [1] annex F. Other ratings to advise on specific content issues, e.g. sex, violence, bad language, can be added in the future in such a way that compatibility is maintained with the present system. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.4 Display related blocks | In addition to the text within AI and PI blocks, text information can be conveyed via other blocks: - OSD Information blocks: used to define the contents of the Header Area and the text for menus; - Navigation Information blocks: conveys the text associated with navigation menus in a Full EPG; - Message Information blocks: used for text messages that are not related to a particular event. For the Enhanced EPG there are three additional structures: - Object definition structures that support the display of a string contained in an MI NI, PI, or OI structure; - DRCS definition structures that allow the definition of graphic characters either within the EPG or from the Teletext text service; - CLUT definition structure structures that allow the definition of Colours either within the EPG or from the Teletext text service. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.5 Navigation Information blocks | In other than Full EPG decoders, the navigation aspects are determined by the decoder manufacturer. In a Full EPG service the EPG service provider is not only responsible for the layout of the screen displays but also for the linking and interaction of the navigational structure. This should be designed to aid the viewer to find the information simply, easily and logically. Navigation Information blocks convey the data to achieve this. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 8.6 The transmitted data stream | The transmission of data broadcasting information via Teletext is covered by EN 300 708 [2]. The EPG data is additional data that has to be transmitted along with the normal Teletext service. It is the network operator's/EPG service provider's responsibility to ensure that the EPG data is organized correctly and multiplexed with the normal Teletext data in a way that is compatible with all types of decoders which comply with the specifications. In order to make effective use of the VBI and ensure that the most important programme data is available quickly for the viewer, EPG transmissions are split into two streams. Within the Stream 1 the normal Teletext 20 ms page clearing rule applies, within Stream 2 it does not. Stream 1 carries the Near Information for at least This Channel supported by the EPG service. The two streams are distinguished by a decoder through the use of different subcodes. Stream 1 contains the Application Information, the Programme Information for at least the Near Information of This Channel, the OSD Information and the Bundle Information. Stream 2 carries the remaining information blocks. The splitting of the data stream in this way will accelerate the transmission of the more important data. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 21 Recommendations: Network operators/EPG service providers should aim for a minimum repetition rate of 10 s for at least the Now and Next of the This channel component of the Near Information within Stream 1. Apart from allocating the Near component of at least one channel to Stream 1, the network operator/EPG service provider is free to split the information between the two streams in whatever way he wishes. The hexadecimal Teletext page number used for the EPG data is redefinable in the MIP. At least one hexadecimal digit is used to make the page "invisible" to normal Teletext decoders. Recommendations: A page number which includes at least one hexadecimal digit should be used to transport the EPG service (see EN 300 708 [2]). The default Teletext page for EPG data is 1DF but is redefinable in the MIP if necessary. The transmission of a MIP is mandatory. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 9 Display aspects | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 9.1 Screen layout | The display standard of the EPG is based on Teletext. Generally, the display screen is based on Level 1.5 Teletext (24 rows, 40 columns), with a screen aspect ratio of 4:3. The screen is divided into four areas, as shown in figure 6. - Header Area: rows 1 to 3. Content and size are defined by the network operator/EPG service provider to identify his service. Starting from the first column position in row 1, contiguous locations will be filled row by row, left to right, with groups of 40 characters (including serial attributes) provided by the appropriate transparent string in an OSD Information block. - Event Area: row 4 until the start of the Message Area. Visual layout and navigational elements are presented by the decoder and the displayed information is supplied by the EPG service. The programme title and other parameters such as channel, time and date of transmission will be displayed. In a Full EPG decoder, the results from the viewer's selection by theme or other criteria will appear here. - Message Area: Number of rows determined by the size of the data to be displayed. Minimum of 1 row, maximum of 8. If a Message Area is defined, it always finishes on row 23. This area can contain the Short or Long Information, corresponding to the selected event. Alternatively, it can be used for messages, promotion of programmes or advertisements, etc. Starting from the first column position in the first row, contiguous locations will be filled row by row, left to right, with groups of 40 characters (including spacing attributes) provided by the appropriate transparent string in the Programme or OSD Information blocks. The same appears if the text information is referenced from a Teletext page although colour and other attributes will be added by the decoder, if required. If the text information is smaller than the message area, the manufacturer is free to position it vertically anywhere within the Message Area. - Navigation Area: row 24. This area is used by the decoder to display navigational prompts. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 22 Header Area Event Area Message Area Navigation Area Row 24 Rows 23-16 Event list between 12 & 21 rows long Rows 1-3 Between 0 and 8 events. Content comes from the EPG publisher. Navigation and layout is defined by the decoder. The user selects one event. Decoder EPG Provider Defined by Figure 6: Screen layout NOTE: A Composite EPG decoder may not necessarily display the information in this way. Recommendations: The display seen by the viewer will play a large part in determining the impact and success of an EPG service. It is essential that the service provider is aware of the possible screen layouts to ensure that the information is displayed in a pleasing and correct way. The image SHALL be checked by the service provider via an EPG decoder prior to transmission. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 9.2 The definition of text | Text can be transmitted explicitly as part of the EPG transmission. Transparent text strings are sequences of characters and spacing attributes. A limited selection of Level 1 spacing attributes are available plus a Carriage Return command for formatting text strings into rows. Accented characters and other symbols available in Level 1.5 Teletext transmissions can be inserted as required. To ensure that some early decoders display text strings in a consistent manner, the escape sequences should be used to overwrite characters only and not attributes. A more technical discussion of Transparent strings can be found in clause 13. Alternatively, text from pages in the normal Teletext service can be "cut-and-pasted" into the EPG and a complete Teletext page can be used as "Long Information". However, the use of these techniques is not recommended for the reasons outlined in clause 5.4. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 23 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 9.3 Use of the Carriage Return attribute | A Carriage Return command can be transmitted as a spacing attribute within any transparent string. If it is known that only decoders with Teletext type displays are receiving the EPG transmission, such a command should not be used to break a row from column 35 onwards. Explicit spaces should be transmitted instead. This is because one Carriage Return command occupies 6 transmitted bytes. The use of a Carriage Return command in a string should only be used where all the attributes are set to their default values. NOTE: Present developments centre around decoders which use Teletext display techniques and, generally, the displays will be character based, with 40 characters per row. Future decoders using more advanced display techniques may be capable of displaying more than 40 characters per row. Such decoders should ignore Carriage Return commands which are inserted for the sole benefit of normal Teletext decoders. However, a new Carriage Return command that is interpreted only by more advanced decoders will be required in the future in addition to the existing one. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 9.4 Enhanced EPG | The enhanced EPG provides the display attributes of Enhanced Teletext to the EPG. For an editorial based overview of these features see clause 4 of TR 100 287 [6]. In adapting these enhancements to the EPG it shall be remembered that the screen is divided into Areas and any particular enhancement shall be contained totally within that area. In addition there are features such as a rotating messages. There are a number of restrictions within the specification in terms of numbers of objects; DRCS and Colours available; these are unlikely to cause any limitations to normal editorial operation. Enhancements require more memory in the decoder and more data volume in the transmission. The Enhanced EPG decoder may be able to share its memory between text content and the enhancement data but there is likely to be a limit of 128 kbyte on the enhancement data. The transmission requirements may be only once per complete transmissions of the database, but there may be some elements that need to be transmitted more frequently. The provision of these enhancements makes the EPG far more attractive and gives the broadcaster more control over the look of the EPG. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 10 Copyright and access control | Any information used in an EPG service may be subject to copyright protection for legal reasons. NOTE: A distinction has to be made between programme information and programme listings. Some broadcasters will not want their listings copied and in some countries legislation may exist to prevent this. Each EPG data block contains a copyright protection flag which indicates if the block can be used outside of the EPG service in which it was transmitted. By this means the network operator/EPG service provider can prevent the use of information blocks by, for example, Composite EPG decoders. For various reasons all or part of an EPG service may be placed under conditional access control. The network operator/EPG service provider can restrict access to the data for instance if an extended EPG service is not free of charge. Each data block contains a set of flags allowing three modes of conditional access plus free access. More information can be found in EN 300 707 [1]. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 24 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11 Scope and depth of an EPG | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.1 General considerations | Technical staff may well be asked to replicate the calculations shown in clause 12.5 regarding the volume of data and the effect on the normal Teletext service to match the needs of a particular network. There are three main decisions: - How much information per channel, per day? - What depth of information should any particular day/channel have? - When will more information be added to the EPG? The information per channel, per day is best obtained by manually coding the printed version of what is required. A very careful check should be made to ensure the material is typical. The inclusion of Long Information can make a difference to the attractiveness of the service at the expense of the total size of the database. For example, it may be required to provide full details of the films to be shown at the weekend during the previous week. Of course, even within a set of title-only listings, there can be a major event(s) which should have Short and/or Long Information, etc. The depth of information has a more complex idea to grasp if there is sufficient information already stored in an EPG decoder with non-volatile storage, it need not be sent so rapidly. The concept of Near being today and tomorrow was chosen to ensure that a suitable decoder need only be powered-up for a short time the day before to acquire and store sufficiently up-to-date information for the current day. Previously, editors thought of just providing the current days information but this would have had to be transmitted very frequently so that a decoder would have today's information for all channels within a very short time of turning on. Thus it may be useful to transmit a day or two with a listing of the title only; at least this will ensure that the programme title and type/genre are known to the decoder if acquisition has been halted for a day or two. There is also an important concept regarding the time of day at which new data is added to the EPG. Broadcasters think of large sections of programming, perhaps morning, afternoon, early evening, late evening and overnight. It is thus more likely that the EPG will be updated with sections of programming rather than holding a constant number of programmes. Also, the EPG may grow in data volume at certain times. For example, if the EPG usually covers only the Near period (two days) it is likely that by early Friday evening the whole of the EPG for Saturday and Sunday will be need to be sent so that viewers can plan their weekend viewing. This will mean that there would be, say, two and a half days worth of information needing to be transmitted and stored in the EPG decoder which has obvious implications. Likewise, the Christmas/New Year period is likely to place a great strain on programme listings. It is not easy to make EPG calculations with absolute accuracy. Clauses 12.3 and 12.5 give typical figures as a starting point. A particular service will deviate from these figures and the only proof is to generate the database. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.2 Prioritization | Generally prioritization should be used with caution. The more frequently information is sent, the greater the data rate. One important concept is that once an EPG service is running, the majority of the decoders will have the majority of the information stored, and thus need minimal refreshing. The EPG can operate with the programme information blocks being transmitted in any order and so there is no technical reason why prioritization cannot be applied. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.2.1 The whole EPG | As reception of the EPG is dependent on the decoder being tuned to the channel there may be occasions, for instance during news bulletins, where the majority of the viewers will be watching this channel. Although the transmission frequency of the EPG is, say, 20 minutes, which is well within a half hour programme, it may be that to send the EPG more rapidly will ensure absolutely that the decoder is updated. Thus it may be a good time to add the next day's guide to the EPG. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 25 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.2.2 Near information | This Channel Near is transmitted very frequently for the benefit of low-end decoders. For other channels, the Near information is likely to be stored in the decoder, but perhaps not in the depth that is desired from an editorial point of view. A practical solution needs to be found for each channel. For example, a refresh frequency of, say, 5 to 10 minutes for popular channels and closer to 20 minutes for the less popular ones. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.2.3 Far information | In general, Far Information is for events sufficiently remote from Near that it may be deemed to be less important and thus can be refreshed at the most infrequent rates. However, This Channel Far may be treated with a similar rate to Other Channels Near. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3 Editorial guidance | There are many issues that the editor of an EPG should take into account in the way in which the service is delivered. This clause links some of the key parameters that need consideration following on from the general considerations detailed in clause 11.1. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3.1 Data volume, data rates and prioritizing | The amount of Teletext capacity in the transmission is likely to be limited. Thus the editor should decide on the best way of using the capacity, which could be a small database transmitted frequently or a larger database transmitted less frequently but with prioritizing so that some information is more rapidly acquired and stored e.g. popular channels, and updates being sent rapidly so that the EPG seems to be accurate at all times. The frequent repetition of information will often slow down the remaining information very considerably. As information may tend to be compiled a day or day segment at a time, the data volume will decrease during the day as the programmes are transmitted and then increase when the next segment of information is added to the EPG. This can be a very technical matter, and advice should be sought. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3.2 Time to acquire | The updating of a database starts within a few seconds of the EPG signal being recognized by the decoder. It is suggested that the complete database should be able to be down loaded within the duration of the typical programme, say twenty minutes. The prioritization of near information and the now and next can make the EPG provide some useful information to the viewer within a minute or so of starting to acquire the EPG. Many decoders will store the EPG and so will have information available at the time the EPG is used by the viewer. It should be noted that at installation a decoder will take some time to scan all the channels to provide the viewer with a list of EPGs that can be selected, and then take say 20 minutes to load the complete database. If the EPG is not on page 1DF the scanning time can be reduced by the frequent transmission of the part of the MIP containing the pointer to the EPG page. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3.3 Version control | A decoder may totally reconstruct the database when there is a new version signalled, or it may modify the existing information stored. Thus consideration should be given to when the version of the database is changed so that the viewer has the information available when they next access it. For instance if a decoder clears its memory when a New version is transmitted, it would be best to do a version change just after that start of a popular programme as it is unlikely that the viewer will need to access the EPG. If a major editorial change, such as the deletion of a network operator from the EPG, is being considered, it is highly desirable to delete the higher structures when there is no information in the lower structures. For example if you are deleting the network operator at the end of the month do not send any information for the following month, so that at the end of the month there is only a structure with no information to delete. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 26 Similarly the point in the day when the information for the next days is included in the transmission, which lengthens the time to acquire may need to be considered. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3.4 Viewers preferences | The viewer has two basic decisions: what type of decoder and then what level of EPG service to receive. The editor should be aware of the limitations of the lower end decoders which will tend to store less and not retain the information in non volatile RAM and thus need more frequent updating and retransmission of the database. But assuming that he majority of decoders are multiple channel the editor should be aware of the mix between a full EPG and a composite EPG amongst the viewership. There are many features of each and it is likely that early adopters of EPG transmissions will favour the Full EPG as a way of trying to exclude other channels from being within a (composite) EPG. The navigation in a full EPG is totally under broadcaster control but tends to work in terms of "today" or "tomorrow" with the time of midnight being the divider. A Composite EPG with the navigation being provided by the manufacturer may operate with the next few hours thus providing continuity across midnight for what may be a programme days listings. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 11.3.5 Enhanced EPG | The enhanced EPG enables a more attractive service to be designed. Because of the inherent templating in the way that the EPG is set up and transmitted, the editorial overhead of doing some of the enhancements is very low. Careful use of objects and DRCS - which could be thought of a small icons- can improve the look of the EPG tremendously. As these features requires the memory and capabilities of a high level EPG decoder, the overall functionality of the EPG will be improved for the viewer. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12 Technical background | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.1 Outline | The purpose of this clause is to give some indication of the effects of choosing to transmit a certain size of EPG and the resulting effects on other Teletext services. An EPG is a complex and interrelated system and editorial aspirations may have to be altered to fit the technical limits. It is not easy to discuss one aspect without making reference to many others. This clause has four main topics: 1) Transmission related aspects, e.g. where does the capacity for Streams 1 and 2 come from? 2) The size of the constituent parts of the total EPG, and the minimum repetition rates required. 3) Examples of possible service, showing the effects of decisions on the scope of the service and the cycle time. 4) A number of other operational technical issues. When planning an EPG the basic parameters that have to be considered are: - What has to be transmitted (e.g. number of channels supported, number of days, depth of information, etc.)? - How much data is there to be transmitted? - How frequently does a particular type of data need to be transmitted? - How much transmission space is there? Overall these parameters define the shape and style of an EPG. With careful thought, the technical features can be applied to enable a practical and distinctive service to be created. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 27 This clause tries to give some feeling for the size and dynamics of an EPG. There are, of necessity, a number of technical approximations made and these are clearly indicated. Also, the amount of editorial information has been worked out for typical services, other services will need to modify them accordingly. An EPG is a data broadcast operation transferring information from the EPG service provider's database to the decoder's database where it is stored and then accessed by the viewer. Thus the speed and frequency of transmission are not major issues in themselves. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2 Transmission aspects | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.1 Page format | Full details can be found in EN 300 708 [2] clause 4. An EPG transport page is of the data broadcasting type Page Format - Clear. A standard Page Format - Clear data broadcasting Teletext page comprises a page header and up to 23 normal packets. The EPG data appears in the 23 packets, each with a capacity of carrying 39 bytes of data. (The first byte is used to indicate where a new block of data starts within the packet). Thus each full EPG transport page can convey a maximum of 897 bytes of data. The actual quantity of EPG data carried will be less than this as the control data elements within the database are Hamming 8/4 coded and thus require two bits to transmit one bit of data. The text components are parity protected. The S2 and S4 parts of the page sub-code in the header are used to inform a decoder of the number of packets that will be broadcast within this page. The page can be transmitted in fragments, i.e. a header followed by some of the packets. This is repeated until the final packet (as indicated by S2 and S4) has been transmitted. This technique is more likely to be used for Stream 2 as only a few VBI lines become available at any one time. Fragmented transmission is described in EN 300 706 [3] clause B.6. The S1 component of the sub-code is used as a continuity index to ensure a decoder processes the pages in the correct order. There will be separate indices for Streams 1 and 2. S3 is used to distinguish between Stream 1 and Stream 2 pages. A packet 28/0 may be appended to the page to define its function as a data broadcasting page of type Page Format - Clear and to prevent it from being erroneously decoded by equipment designed to receive the original type of data broadcasting page known as Page Format - CA (as defined in EN 300 708 [2] clause 5). EPG reception may be corrupted on some first generation decoders when streams other than 1 and two are transmitted (for other applications) on the same page number as the EPG. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.2 Stream 1 | The EPG database is split into two streams for transmission as page-format Teletext. The specification defines that the pages in Stream 1 shall contain This Channel Near Information (i.e. the programme information for the next two days on this channel) and at least the Bundle Information, Application Information and the OSD Information block defining the contents of the Header Area. A Stream 1 page is identified by a value of 0 for the S3 part of the page sub-code. Stream 1 pages are broadcast obeying the 20 ms rule like normal Teletext pages and have to be captured by all types of decoder. Thus Stream 1 pages will either take capacity from the existing Teletext service, or result in a slower cycle time, regardless of whether the transmission mode is serial or parallel. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.3 Stream 2 | Stream 2 pages contains the rest of the EPG database, the Near Information for other channels, Far Information for all channels and, where applicable, further navigation and menu data. It is not constrained by the 20 ms rule and thus can be transmitted in the filler packet space (see clause 12.2.4) in a similar manner to Level 2.5 Teletext enhancement data (see EN 300 706 [3] clause B.6). Accordingly, the data will not be accessible by the simpler types of decoder. A Stream 2 page is identified by a value of 1 for the S3 part of the page sub-code. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 28 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.4 Filler packet space | The situation where a VBI line allocated for Teletext is not actually taken up happens quite frequently in many transmissions, especially those operating in serial mode. Unused lines occur because in normal Teletext transmissions it is not permitted to send the page header and another packet for the same page in the same VBI period; the 20 ms rule. Let us assume that a page consists of 24 packets and it is being transmitted on 10 VBI lines. NOTE: A normal displayable Teletext page consists of a header (row zero) and a number of other packets. A full page takes 24 packets, but if a row has no information it need not be transmitted. This is referred to as Row Adaptive transmission. Fastext, local enhancement data and PDC use more packets per page. Because of the 20 ms rule only the header packet is transmitted in the first VBI. 10 packets are sent in the next VBI, 10 packets in the following VBI, and the final 3 packets in the VBI after that but there will be 6 VBI lines that cannot be used for the normal Teletext service, assuming the header of the next page is transmitted on the final line. This "space" is often filled with duplicates of the next page header or a packet 8/25. It is in this filler packet space that Stream 2 can be transmitted. Table 1 shows how many filler packets result from the transmission of given size of page using a given number of VBI lines per field. Stream 2 is intended to recover the otherwise lost capacity and the normal text service resumes with the page header for the next page on the last of the VBI lines available. An EPG page can be a minimum of two packets, a page header and one data packet. Therefore, there has to be more than two filler packets available in any VBI to allow Stream 2 data to be carried. Table 2 shows the effective data rate that can be achieved using the filler packets alone, assuming 40 data bytes per packet and all the pages in the transmission have the same number of rows per page. It will become apparent later that for the majority of services there is ample filler packet space for both EPG and enhanced Teletext services. Level 2.5 Teletext has a maximum enhancement data rate requirement of 500 packets in 20 s = 1 kbyte/s. Many systems use row adaptive transmission and do not have pages of constant length. Either calculation from page lengths or the use of Teletext analysers will determine the filler packet space that is available. Further, there may be other users of the space, e.g. packets 31, see EN 300 708 [2] clauses 6 and 7. However, they are unlikely to exceed 1 kbyte/s. Table 1: Filler packets created per page in normal transmissions Number of rows in text page VBI lines 30 29 28 27 26 25 24 23 22 21 20 19 18 1 2 1 1 1 1 1 1 3 1 2 1 2 1 2 1 2 4 2 3 1 2 3 1 2 3 1 2 5 1 2 3 4 1 2 3 4 1 2 6 1 2 3 4 5 1 2 3 4 5 7 5 6 1 2 3 4 5 6 1 2 3 8 2 3 4 5 6 7 1 2 3 4 5 6 9 6 7 8 1 2 3 4 5 6 7 8 10 1 2 3 4 5 6 7 8 9 1 2 11 3 4 5 6 7 8 9 10 1 2 3 4 12 6 7 8 9 10 11 1 2 3 4 5 6 13 9 10 11 12 1 2 3 4 5 6 7 8 14 12 13 1 2 3 4 5 6 7 8 9 10 15 1 2 3 4 5 6 7 8 9 10 11 12 16 2 3 4 5 6 7 8 9 10 11 12 13 14 ETSI ETSI TR 101 288 V1.3.1 (2002-12) 29 Table 2: Data rates achievable through the use of filler packets (kbyte/s) Number of rows per text page VBI lines 30 29 28 27 26 25 24 23 22 21 20 19 18 1 2 3 0,2 0,2 0,2 0,3 4 0,2 0,5 0,3 0,6 0,3 0,7 0,4 5 0,3 0,6 1,0 0,4 0,8 1,2 0,4 6 0,4 0,7 1,1 1,6 0,4 0,9 1,4 2,0 7 1,5 2,0 0,4 0,9 1,4 1,9 2,4 0,5 1,1 8 0,4 0,9 1,3 1,8 2,4 2,9 0,5 1,1 1,7 2,4 3,1 9 2,3 2,9 3,4 0,5 1,1 1,7 2,3 3,0 3,8 4,6 10 0,5 1,1 1,7 2,3 3,0 3,7 4,4 5,2 0,7 11 1,1 1,7 2,3 2,9 3,6 4,3 5,1 5,9 0,7 1,5 2,3 12 2,9 3,5 4,2 4,9 5,7 6,5 0,7 1,5 2,3 3,1 4,0 13 4,8 5,6 6,3 7,2 0,7 1,5 2,2, 3,1 4,0 4,9 5,9 14 7,0 7,8 0,7 1,4 2,2 3,0 3,9 4,8 5,8 6,8 7,9 15 0,7 1,4 2,2 3,0 3,9 4,8 5,7 6,7 7,8 8,9 10,1 16 0,7 1,4 2,2 3,0 3,8 4,7 5,6 6,6 7,6 8,7 9,8 11,0 12,3 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.5 Transmission relationship between Streams 1 and 2 | Stream 1 and Stream 2 are independent and asynchronous feeds of data. However, there is a technical requirement to simplify decoder design that the transmission of a page header at the start of a page shall be separated by at least 200 ms (10 fields) from the transmission of another EPG page. Other than this there are no constraints relating the two streams. The 200 ms rule governs only the true start of pages, that is when the page header will be followed by packet 1. It does not apply to later fragments. If a Stream 2 page is sent in fragments in the filler packet space, it is possible for it to overlap a Stream 1 page. Fragmented transmission is covered in EN 300 706 [3] clause B.6. Figure 7 shows an example of a transmission using 9 lines per VBI. A Stream 2 page (EPG 2) starts in field 0 and uses the filler packet space as it occurs while row adaptive pages (prefix "M") are being transmitted in the normal service. (The page headers of the later fragments of the Stream 2 page are marked "epg 2".) A Stream 1 page (EPG 1) cannot commence until at least field 10, and in this example field 12 has the earliest opportunity. In effect the Stream 2 page is interrupted during fields 13-15 while the Stream 1 page is transmitted. Field 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 M19 M2 M12 epg 2 M1 M10 M19 M1 M10 M20 M1 M10 M19 E1 E10 E19 M20 M3 M13 e3 M2 M11 M20 M2 M11 M23 M2 M11 M20 E2 E11 E20 M21 M4 M14 e4 M3 M12 M21 M3 M12 epg 2 M3 M12 M21 E3 E12 E21 M22 M5 M16 e5 M4 M13 M22 M4 M13 e12 M4 M13 M22 E4 E13 E22 M23 M6 M17 e6 M5 M14 M23 M5 M15 e13 M5 M14 epg 2 E5 E14 E23 EPG 2 M7 M18 e7 M6 M15 epg 2 M6 M16 e14 M6 M15 e17 E6 E15 epg 2 e1 M8 M19 e8 M7 M16 e10 M7 M17 e15 M7 M16 e18 E7 E16 e20 e2 M9 M21 e9 M8 M17 e11 M8 M18 e16 M8 M17 e19 E8 E17 e21 M0 M10 M23 M0 M9 M18 M0 M9 M19 M0 M9 M18 EPG 1 E9 E18 M0 Figure 7: Transmission sequence example |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.2.6 Serial versus parallel transmissions | With reference to table 1 it can be seen that for a typical serial transmission using, say, 9 lines per field, there will be occasions when 8 filler packets occur. However, for a 9 line parallel transmission organized as 3 magazines, each on 3 VBI lines, the maximum number is only 2. While parallel transmissions are more efficient for normal text services, there is much less inefficiency that can be exploited for EPG Stream 2 use. Also, the transmission equipment may be capable of allocating every available VBI line, perhaps by inserting packets 30 or 31. Consequently, a more significant allocation of transmission capacity to the EPG service may be required when parallel mode is employed and it may be simpler to assume that BOTH streams obey the 20 ms rule. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 30 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.3 Database components | This clause states the required transmission rate and size of each block assumed in the later calculations of complete services. Block sizes are quoted in terms of the number of bytes that will be transmitted after the data has been encoded for transmission (i.e. Hamming and parity protection bits have been added) and assembled into Teletext packets. A decoder is likely to remove the protection data prior to storing the database and the volume of stored data will be less than the transmitted volume. Sample lengths have been chosen for the text strings that form part of each block under consideration. To a first approximation, block sizes with different text string lengths can be calculated by adding or subtracting one byte per character as required. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.3.1 Bundle Information | The Bundle Information amounts to 14 bytes. It will be absorbed into the overall transmission without any noticeable effect despite having to be transmitted frequently. Consequently, it is omitted from the later calculations although it may well be considered to be part of the AI block as it should be transmitted at the same rate. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.3.2 Application Information | The size of the AI block depends upon the number of channels supported by the guide. The following examples assume an EPG service name of 20 characters, and, for each channel, a network operator's name of 10 characters. Channels AI size (bytes) 1 103 2 140 4 212 10 427 20 786 When any type of decoder is turned on it needs to find the top level EPG data fairly quickly so that it can start collecting and processing the EPG specific data. This requires the AI and BI blocks to be transmitted every 3 to 4 s. Further, to ensure that decoders looking for information on the current programme display the correct information, the AI block has to be updated and re-transmitted when one programme finishes and the next one starts on ANY of the networks covered by the EPG. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.3.3 Programme Information | The size of PI blocks is governed largely by the amount of text data they contain. In the following examples a PI block is assumed to have a title length of 32 characters, two themes and one sort criteria: PI contents PI size (bytes) Abbreviation Title only 69 Title PI Title + Short Info of 80 characters (2 rows) 177 Mini PI Title + recommended Short Info of 140 characters 243 Short PI Title + maximum Short Info (256 characters) 361 Title + Long Info of 960 characters (24 rows) 1 140 Long PI The above abbreviations are used in clause 12.4 when discussing examples of services. The transmission frequency of a PI block depends upon the channel it belongs to and how long before the associated programme will be broadcast. To enable a low-end decoder with limited memory to provide a This Channel Now And The Next Four Programmes service, the PI blocks for these items have to be transmitted every 10 s or faster and appear in Stream 1. The user can then be presented with the information soon after a channel change. However, this is also a requirement for more sophisticated decoders if they do not include non-volatile storage. The remaining Near Information for This Channel has to be transmitted at least every 30 s maximum. The transmission rate for the remaining programme information is at the service provider's discretion but a maximum cycle time of 20 minutes is recommended. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 31 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.3.4 OSD Information, Navigation Information and other blocks | The likely quantity of OSD, Navigation and Message Information is difficult to gauge as it depends on the editorially defined "look and feel". A figure of 15 to 25 kbytes is used later in the multi-channel examples. However, if the service provider wants his name to appear in the Header Area on all decoders he has to transmit an OSD Information block (with a block number of 0) in kbytes Stream 1. By way of example, an OI block containing a name of 30 characters will amount to 62 bytes. The OSD Information in Stream 1 should be transmitted no slower than the minimum PI rate, i.e. every 10 s. OSD and Navigation Information in Stream 2 can be much slower, perhaps every 4 to 5 minutes. The sizes of the other Housekeeping blocks that will be present in the EPG are not evaluated here as they are very small compared to the total PI component. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.4 Typical transmission decisions | It is recommended that the whole EPG should be transmitted once every 20 minutes. The time between the transmission of the same information is important because: - it determines the time that it will take a decoder starting from nothing to assemble an EPG; - for the decoder that has a full EPG already in memory, it determines the maximum time (if no other updating techniques are employed) by which the decoder's database is out of step with the service provider's database; - it is the minimum time that a decoder has to be tuned into that channel to obtain a refresh of the database. A decoder that can make use of all the information in a Full EPG service is likely to be constantly updating the information and thus there is no need to send the information very frequently. Assuming This Channel Near is carried in Stream 1 then Stream 2 will have three main constituents, Other Channels Near, All Channels Far and Housekeeping data such as OSD and Menu Information, messages, etc. It may be necessary to transmit these at different frequencies and this in turn leads to an increase in the data rate. The prioritization of this information can be very complex and so a fairly simple view is taken in the service examples in clause 12.5. It can also lead to a waste of transmission capacity and so should be used with caution. The Housekeeping information depends upon the editorial style and "look-and-feel" adopted. Its overall size is not likely to be significant compared to the PI component and a transmission frequency of around 4 minutes should be acceptable as it is unlikely to change very often. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5 Service scenarios | How much information to include about each programme and how many programmes should be covered are likely to be the editorial matters of most concern and the greatest variability. This clause presents a number of different scenarios and calculates the volume of data for each. They are illustrations and show the method that can be used to work out the approximate amount of data within an EPG. Refresh rates are set and the transmission implications for each database, using two streams if necessary, are calculated accordingly. The size of the data volumes for each category of information in a particular EPG should be calculated at the planning stage. These figures will give an indication of the total volume required for a particular service. Although the examples are generalizations they should be sufficiently accurate to give implementors a feel for a service and the size of the database required. The real test is to set up an EPG with the selected parameters, valid programme listings and other information, and then to check the volume it occupies. Quite minor changes in the amount of data for each item, when multiplied by the large number of items in an EPG, can make noticeable changes to the data volume. In particular, including several Long Information fields can inflate the data volume considerably. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 32 An EPG database is not static. It can go up and down in size, both in data volume for the same number of programmes, and also in the number of days covered. For example, a third day of listings in detail may be useful for the weekend (Friday, Saturday and Sunday). Thus on a Friday the database would contain an extra day's worth of detailed programme information compared to the rest of the week. The following assumptions have been made and constraints imposed on the analysis of the example services: - there are between 32 and 40 programme events per day (36 is assumed in the calculations); - the total database for a Multiple Channel EPG or Full EPG should not exceed 256 kbytes once it has been encoded into Teletext packets. The PI component should amount to around 230 kbytes, leaving space for navigation and menu data. This also allows some extra space when programme listings for an extra day are required to cover a weekend or public holiday; - for the multi-channel scenarios, each channel is placed in one of four categories on a day-by-day basis and a standardized volume of data is assumed: the Major Channel Near category is quite detailed and 12 kbytes per day (24 kbytes in total) is allocated. This may comprise 32 events per day, each with a PI containing a Short Info of the recommended 140 characters, and 4 events where the PI has a Long Info of 24 rows; the Minor Channel Near and Channel Far categories contain some programme detail and each is allocated 9 kbytes per day per channel. (Thus Minor Channel Near will total 18 kbytes.) This may be achieved with 33 events per day, each with a PI containing a Short Info of about two rows of text, and 3 events where the PI has a Long Info of 24 rows; programmes in the Titles Only category do not have any additional information. 36 events per day amounts to 2 kbytes. - AI and BI blocks are transmitted every 3 s; - This Channel Now and The Next Four Programmes should be transmitted every 10 s (the minimum stated in the specification); - the remaining This Channel Near information should be transmitted every 30 s (the minimum stated in the specification); - the total EPG should have a maximum cycle time of 20 minutes; - Stream 1 should be limited to one page per second, if practical; - the required EPG data and page rates are calculated individually for Streams 1 and 2 and then combined into an overall figure as this may be more representative of the true rate in a parallel transmission system; - the data broadcasting pages transmitted are used to carry EPG data only. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5.1 Minimum EPG service; This Channel Now and Next | The calculation assumes that each of the 5 PI blocks required in the minimum This Channel Now and the Next Four Programmes service has a Short Info of 140 characters. All this data has to be transmitted in Stream 1. Component Size (bytes) Tx rate Bytes/s Pages/s Short PI (243 × 5) 1 215 10 s 122 0,13 AI (single channel) 103 3 s 34 0,04 OI (service title only) 62 10 s 6 0 Total 1 380 162 0,18 Thus a minimum EPG service can be supplied using 0,18 pages per second in Stream 1. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 33 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5.2 This Channel Today | Programme information for one day for This Channel; 32 Short PI + 4 Long PI. This split of Short and Long PIs is chosen to achieve the PI size constraint of 12 kbytes per day for This Channel Near. All this data has to be transmitted in Stream 1. Component Size (kbytes) Tx rate kbyte/s Pages/s Short PI (32 × 243 bytes) 7,59 10 s 0,76 0,85 Long PI (4 × 1 140 bytes) 4,45 10 s 0,44 0,49 AI (single channel; 103 bytes) 0,10 3 s 0,03 0,04 OI (service title only) 0,06 10 s 0 0 Total 12,20 1,23 1,38 This has exceeded the one page per second target for Stream 1 and so some of the data has to be transmitted at a slower rate. Assuming there is one Long PI within the first five programmes, the target rate is achieved if the PI data outside of the minimum EPG is transmitted over 15 s. Component Size (kbytes) Tx rate kbyte/s Pages/s Minimum EPG, Short PI (4 × 243 bytes) 0,95 10s 0,10 0,11 Minimum EPG, Long PI (1 × 1 140 bytes) 1,11 10s 0,11 0,12 Rest of today, Short PI (28 × 243 bytes) 6,64 15s 0,44 0,49 Rest of today, Long PI (3 × 1 140 bytes) 3,34 15s 0,22 0,24 AI (single channel; 103 bytes) 0,10 3s 0,03 0,04 OI (service title only) 0,06 10s 0 0 Total 12,20 0,90 1,00 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5.3 This Channel Near | Two days worth of programme information for This Channel, each day with 32 Short PI and 4 Long PI. All this data has to be transmitted in Stream 1. The target of one page per second is achieved if the PI data outside of the minimum EPG is transmitted over 34 s. Component Size (kbytes) Tx rate kbyte/s Pages/s Minimum EPG, Short PI (4 × 243 bytes) 0,95 10s 0,10 0,11 Minimum EPG, Long PI (1 × 1 140 bytes) 1,11 10s 0,11 0,12 Rest of Near, Short PI (60 × 243 bytes) 14,24 34s 0,42 0,47 Rest of Near, Long PI (7 × 1 140 bytes) 7,79 34s 0,23 0,26 AI (single channel; 103 bytes) 0,10 3s 0,03 0,04 OI (service title only) 0,06 10 s 0 0 Total 24,25 0,89 0,99 Unfortunately, the figure of 34 s is just outside the specification maximum of 30 s. To achieve this the page rate would have to be 1,08. This scenario defines the Major Channel Near category in the following multi-channel, multi-day examples. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5.4 Service A: This Channel Only for 14 days | A guide to one channel over 14 days, with less detailed information after the first two days. The two Near days each comprise 32 Short PI and 4 Long PI, as in the previous example. The 12 days of Far Information comprises 33 Mini PI and 3 Long PI per day to achieve the PI size constraint for This Channel Far per day: (33 × 177) + (3 × 1 140) = 9 kbytes. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 34 Leaving aside the transmission rate aspects, the data volume calculation based on the PI component alone amounts to: Channels Days Component Size Near Far 1 2 Major Channel Near 1 × 2 × 12 K 24 K 1 12 Channel Far 1 × 12 × 9 K 108 K Sub-totals 24 K 108 K PI Total 132 K The service can be represented graphically in the following manner, with the number of days horizontally and the number of channels vertically: Near Far Today/ Tomorrow Day 3 4 5 6 7 8 9 10 11 12 13 14 12 kbytes/day, full details 9 kbytes/day, some detail As all of Near Information belongs to This Channel, it has to be transmitted in Stream 1. A cycle time of 10 minutes is chosen for the Far Information and the size of the navigation and menu components is estimated as 25 K, with a repetition rate of 4 minutes. Component Size Tx Rate Stream 1 Stream 2 Combined AI 103 3 s 34 This Channel Now and Next 2 K 10 s 205 Rest of This Channel Near 22 K 30 s 751 PI Far 108 K 10 m 184 OI, etc, 25 K 4 m 107 Total 157 K 990 byte/s 294 byte/s 1 284 byte/s Page rate 1,08 pages/s 0,32 pages/s 1,40 pages/s 12.5.5 Service B: 4 channels for 7 days in some depth, plus 16 channels for 3 days, titles only During the Near period, two channels are presented in detail and two others are covered to a lesser extent. On the third day, all 4 channels are treated equally and for the rest of the week only titles are included. For the first three days, the programmes titles on 16 other channels are included. Channels Days Component Size Near Far 2 2 Major Channel Near 2 × 2 × 12 K 48 K 2 2 Minor Channel Near 2 × 2 × 9 K 36 K 16 2 Titles Only 16 × 2 × 2 K 64 K 4 1 Channel Far 4 × 1 × 9 K 36 K 16 1 Titles Only 16 × 1 × 2 K 32 K 4 4 Titles Only 4 × 4 × 2 K 32 K Sub-total 148 K 100 K Total 248 K ETSI ETSI TR 101 288 V1.3.1 (2002-12) 35 The original intention was to present four channels in some detail and the titles only for 16 other channels for a week. However, this accumulates to almost 500 kbytes. The compromise adopted would allow the listings for a weekend to be transmitted from late on a Thursday night. The service can be represented graphically in the following manner: Near Far O t h e r C h a n n e l s This Channel Today Tomorrow Day 3 4 5 6 7 12 kbytes/day, full details 9 kbytes/day, some details 2 kbytes/day, titles only The menu and navigation components are estimated at 20 kbytes. A 5 minute rate is chosen for the Near Information from other channels, and a 20 minute rate for all Far Information. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 36 Component Size Tx Rate Stream 1 Stream 2 Combined AI (20 channels) 786 4 s 197 This Channel Now and Next 2 K 10 s 205 Rest of This Channel Near 22 K 30 s 751 Other PI Near 124 K 5 m 423 PI Far 100 K 20 m 85 OI, etc. 20 K 4 m 85 Total 268 K 1 153 byte/s 593 byte/s 1 746 byte/s Page rate 1,25 pages/s 0,64 pages/s 1,89 pages/s Note how the 20 channel AI component in Stream 1 has become comparable with the This Channel Now and Next component. 12.5.6 Service C: 2 channels in detail plus 9 other channels, titles only, for 7 days Nine channels are covered for one week. Two channels are presented in some depth and only titles are provided for the remainder. Channels Days Component Size Near Far 2 2 Major Channel Near 2 × 2 × 12 K 48 K 7 2 Titles Only 7 × 2 × 2 K 28 K 2 5 Channel Far 2 × 5 × 9 K 90 K 7 5 Titles Only 7 × 5 × 2 K 70 K Sub-total 76 K 160 K Total 236 K The service can be represented graphically in the following manner: Near Far O C t h h a e n r n e l s This Channel Today Tomorrow Day 3 4 5 6 7 12 kbytes/day, full details 9 kbytes/day, some details 2 kbytes/day, titles only The menu and navigation components are estimated at 25 kbytes. A 3 minute rate is chosen for the Near Information from other channels, and a 20 minute rate for all Far Information. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 37 Component Size Tx Rate Stream 1 Stream 2 Combined AI (9 channels) 393 4 s 98 This Channel Now and Next 2 K 10 s 205 Rest of This Channel Near 22 K 30 s 751 Other PI Near 52 K 3 m 296 PI Far 160 K 20 m 137 OI, etc. 25 K 4 m 107 Total 261 K 1 054 byte/s 540 byte/s 1 594 byte/s Page rate 1,15 pages/s 0,59 pages/s 1,74 pages/s 12.5.7 Service D: 1 channel in some depth plus 20 other channels, titles only, for 5 days Twenty-one channels are covered over 5 days. The This Channel coverage gets progressively less detailed. Only titles are presented for the remaining channels. Channels Days Component Size Near Far 1 2 Major Channel Near 1 × 2 × 12 K 24 K 20 2 Titles Only 20 × 2 × 2 K 80 K 1 1 Channel Far 1 × 1 × 9 K 9 K 1 2 Titles Only 1 × 2 × 2 K 4 K 20 3 Titles Only 20 × 3 × 2 K 120 K Sub-total 104 K 133 K Total 237 K It should be noted that if one week's worth of information was required, the number of other channels would have to be reduced from 20 to about 12. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 38 The service can be represented graphically in the following manner: Near Far O t h e r C h a n n l s This Channel Today Tomorrow Day 3 4 5 12 kbytes/day, full details 9 kbytes/day, some details 2 kbytes/day, titles only The menu and navigation components are estimated at 15 kbytes. A 5 minute rate is chosen for the remaining Near Information, and a 20 minute rate for all Far Information. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 39 Component Size Tx Rate Stream 1 Stream 2 Combined AI (21 channels) 822 4 s 206 This Channel Now and Next 2 K 10 s 205 Rest of This Channel Near 22 K 30 s 751 Other PI Near 80 K 5 m 273 PI Far 133 K 20 m 113 OI, etc. 15 K 4 m 64 Total 252 K 1 162 byte/s 450 byte/s 1 612 byte/s Page rate 1,26 pages/s 0,49 pages/s 1,75 pages/s |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.5.8 Conclusions and impact on the normal Teletext service | Based on the multi-channel, multi-day examples (services A, B, C and D), the following conclusions can be drawn: - Stream 1 is going to require between 1 and 1,25 pages per second for any type of service other than the bare minimum; - on the basis of the figures shown here, it is possible to transmit Stream 2 at less than 600 byte/s and thus majority of serial mode services will have enough filler space for both EPG and enhanced Teletext; - a database of around 256 kbytes, with the repetition rates chosen here, is going to require at least 1,75 pages per second when transmitted in parallel mode, assuming that no filler packet space is available; - even if Stream 1 was allowed to be transmitted without obeying the 20 ms rule, it can NOT be guaranteed that sufficient filler packet space would be available over 30 s to ensure it was transmitted at the required rate. Thus its transmission would need to be scheduled and the capacity would still be taken from the normal Teletext service. To put the minimum one page per second requirement for Stream 1 into some kind of perspective, table 3 shows the approximate number of pages transmitted per second in normal Teletext services by mapping the average number of rows per page against the number of VBI lines in use. Table 4 shows the percentage reduction as a result of allocating one page per second to the EPG service. Table 3: Pages per second in normal Teletext services Average number of rows per text page VBI lines 30 29 28 27 26 25 24 23 22 21 20 19 18 1 1,6 1,7 1,7 1,8 1,9 1,9 2,0 2,1 2,2 2,3 2,4 2,5 2,6 2 3,1 3,2 3,3 3,4 3,6 3,7 3,8 4,0 4,2 4,3 4,5 4,8 5,0 3 4,5 4,7 4,8 5,0 5,2 5,4 5,6 5,8 6,0 6,3 6,5 6,8 7,1 4 5,9 6,1 6,3 6,5 6,7 6,9 7,1 7,4 7,7 8,0 8,3 8,7 9,1 5 7,1 7,4 7,6 7,8 8,1 8,3 8,6 8,9 9,3 9,6 10,0 10,4 10,9 6 8,3 8,6 8,8 9,1 9,4 9,7 10,0 10,3 10,7 11,1 11,5 12,0 12,5 7 9,5 9,7 10,0 10,3 10,6 10,9 11,3 11,7 12,1 12,5 13,0 13,5 14,0 8 10,5 10,8 11,1 11,4 11,8 12,1 12,5 12,9 13,3 13,8 14,3 14,8 15,4 9 11,5 11,8 12,2 12,5 12,9 13,2 13,6 14,1 14,5 15,0 15,5 16,1 16,7 10 12,5 12,8 13,2 13,5 13,9 14,3 14,7 15,2 15,6 16,1 16,7 17,2 17,9 11 13,4 13,8 14,1 14,5 14,9 15,3 15,7 16,2 16,7 17,2 17,7 18,3 19,0 12 14,3 14,6 15,0 15,4 15,8 16,2 16,7 17,1 17,6 18,2 18,8 19,4 20,0 13 15,1 17,1 19,0 20,6 22,1 23,5 24,7 25,9 27,0 28,1 29,2 30,3 31,4 14 15,9 17,7 19,4 20,9 22,3 23,6 24,8 25,9 27,0 28,1 29,2 30,3 31,4 15 16,7 18,2 19,7 21,1 22,4 23,6 24,8 25,9 27,1 28,2 29,2 30,3 31,4 16 17,4 18,7 20,0 21,3 22,5 23,7 24,8 26,0 27,1 28,2 29,2 30,3 31,4 ETSI ETSI TR 101 288 V1.3.1 (2002-12) 40 Table 4: Percentage reduction in the normal page transmission rate through allocating one page per second to the EPG service Average number of rows per text page VBI lines 30 29 28 27 26 25 24 23 22 21 20 19 18 1 62,0 60,0 58,0 56,0 54,0 52,0 50,0 48,0 46,0 44,0 42,0 40,0 38,0 2 32,0 31,0 30,0 29,0 28,0 27,0 26,0 25,0 24,0 23,0 22,0 21,0 20,0 3 22,0 21,3 20,7 20,0 19,3 18,7 18,0 17,3 16,7 16,0 15,3 14,7 14,0 4 17,0 16,5 16,0 15,5 15,0 14,5 14,0 13,5 13,0 12,5 12,0 11,5 11,0 5 14,0 13,6 13,2 12,8 12,4 12,0 11,6 11,2 10,8 10,4 10,0 9,6 9,2 6 12,0 11,7 11,3 11,0 10,7 10,3 10,0 9,7 9,3 9,0 8,7 8,3 8,0 7 10,6 10,3 10,0 9,7 9,4 9,1 8,9 8,6 8,3 8,0 7,7 7,4 7,1 8 9,5 9,3 9,0 8,8 8,5 8,3 8,0 7,8 7,5 7,3 7,0 6,8 6,5 9 8,7 8,4 8,2 8,0 7,8 7,6 7,3 7,1 6,9 6,7 6,4 6,2 6,0 10 8,0 7,8 7,6 7,4 7,2 7,0 6,8 6,6 6,4 6,2 6,0 5,8 5,6 11 7,5 7,3 7,1 6,9 6,7 6,5 6,4 6,2 6,0 5,8 5,6 5,5 5,3 12 7,0 6,8 6,7 6,5 6,3 6,2 6,0 5,8 5,7 5,5 5,3 5,2 5,0 13 6,6 6,5 6,3 6,2 6,0 5,8 5,7 5,5 5,4 5,2 5,1 4,9 4,8 14 6,3 6,1 6,0 5,9 5,7 5,6 5,4 5,3 5,1 5,0 4,9 4,7 4,6 15 6,0 5,9 5,7 5,6 5,5 5,3 5,2 5,1 4,9 4,8 4,7 4,5 4,4 16 5,8 5,6 5,5 5,4 5,3 5,1 5,0 4,9 4,8 4,6 4,5 4,4 4,3 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.6 Technical tailoring | This clause is intended for technical staff who wish to minimize the effect of an EPG service on existing Teletext services. The major concern of Teletext service providers will be the loss of the page transmission space taken by the Stream 1 pages. The use of such pages will enable low-end decoders to acquire a very simple EPG, typically This Channel only, or, by scanning a number of channels, a composite EPG. As these low-end decoders will have little memory, the information will be displayed when it is received; rather like an existing Teletext service. Even when a decoder has the 24 kbytes of memory required to store This Channel Near it is unlikely to be non-volatile and so all the data will have to be acquired from the point at which the decoder is tuned to the channel. Considering the various parameters, the following actions are possible: - reduce the amount of PI data; but this is unlikely to be acceptable as it is very limiting editorially. It is recommended to have full information for at least two days; - increase the time between sending AI, etc. (and decrease the amount of data in the AI, etc.). This means that it will take longer for a decoder to start to acquire the EPG (and the size of AI is determined by the number of channels); - increase the time between sending Now and Next Four Channels information. Again, an editorial point, and similar information on the text service is often transmitted as frequently (but is only a few hundred bytes). So it can be seen that there is little that can be done with the data in Stream 1 to reduce the amount of capacity consumed by the EPG. The operation of a composite EPG decoder should also be considered. It tunes to the first channel, waits, say, a maximum of 4 s to acquire the AI, etc., to identify that an EPG is present, then waits about 45 s to acquire This Channel Near. It then scans the next channel. Overall, it takes almost a minute per channel which is probably too slow. However, a small saving can be made at the Teletext transport layer by preventing the insertion of EPG pages from creating additional filler packet space, as shown by the following example. Assuming the Teletext service is operating on 10 VBI lines per field, a Stream 1 page comprising the full 24 packets will require 3 VBIs to transmit packets 1 to 23. However, it is occupies only 23 of the 30 lines available. If the page was only 20 packets long, it would take only 2 VBI to transmit, thus saving the capacity of one VBI per second. This very efficient transmission reduces the overall data rate at one page a second from 920 bytes to 760 bytes. As a consequence, the page rate has to increase to 1,2 per second. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 41 Over 5 s we have either: 5 × 1 × 920 bytes/page = 4 500 bytes, taking 5 × 3 = 15 VBI periods; or 6 × 1 × 760 bytes/page = 4 560 bytes, taking 6 × 2 = 12 VBI periods. Thus in 5 s there is the saving of 3 VBI periods, the equivalent of ONE normal Teletext page. Thus the EPG is displacing 4 pages rather than 5 every 5 s. Table 5 shows the optimum number of rows for an EPG page (including the page header) and the saving in filler packet space that results. When the number of rows is less than 24 it is necessary to send two separate pages in order to transmit the 920 data bytes (23 rows) and meet the required data rate. Thus an extra page header is required and one packet is deducted from the number of filler packets shown for a 24 row page in table 1 when calculating the saving in filler packets. Implementation of this concept should be easy to achieve in the origination equipment. Table 5: Optimum number of rows for an EPG page VBI Lines Optimum number of rows for EPG page Saving in filler packet space 1 24 0 2 24 0 3 24 0 4 24 0 5 20 0 6 24 0 7 21 3 8 24 0 9 18 2 10 20 5 11 22 8 12 24 0 13 13 1 14 14 3 15 15 5 16 16 7 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7 Other operational issues | 12.7.1 Numbering, scheduling and transmission of Programme Information blocks It is informative to consider the rules for the numbering, scheduling and transmission of Programme Information blocks. However, some detailed knowledge of the coding of AI and PI blocks is required, see EN 300 707 [1] clauses 11.2 and 11.3. Each PI block contains a block_number that is unique to a given network. PI block_numbers within one network are continuous over the whole range of transmitted PIs and over the Stream 1 and Stream 2 boundary as shown in figure 8. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 42 stream 1 stream 2 first PI = “Now” for network A n = prog_start_no in AI block last PI for network A n+x+y = prog_stop_no_swo in AI block PI n+x PI n+x-1 PI n+3 PI n+2 PI n+1 PI n PI n+x+3 PI n+x+2 PI n+x+1 PI n+x+y PI n+x+y-1 n+x = prog_stop_no in AI block Figure 8: PI numbering in Streams 1 and 2 PIs are transmitted sequentially in ascending order, i.e. PI1, PI2, PI3,... PIn, PIn + 1. Updates of individual PIs or PIs with a higher repetition rate, e.g. Near PIs or PIs from Stream 1, may interrupt any other PI sequence with a lower repetition rate, e.g. PIs from Stream 2. In operating a real service the following situations are likely to occur which will necessitate changes to the AI block and a number of PI blocks: a) A current programme finishes: All PI block_numbers on the network concerned remain unchanged. In the AI block the prog_start_no value used to indicate the current (i.e. Now) programme on the network concerned is incremented. b) One programme becomes longer or shorter than originally scheduled: Some or all of the following programmes on the network concerned will be shifted in time. Changes to the contents of PI blocks will occur (i.e. start/stop times) but PI block_numbers will remain the same. It is necessary to set a new version number in the AI block in order to force decoders to reload the entire database. c) A new programme is inserted: With reference to figure 8, assume that a new programme is inserted between PI(n) and PI(n + 1). The new PI becomes PI(n + 1) and all the following PI block_numbers are incremented by one. Start/stop times within the following PI blocks are modified accordingly. In the AI block the progr_stop_no and/or progr_stop_no_swo values (for Streams 1 and 2 respectively) are incremented and a new version number is entered in order to force decoders to reload the entire database. d) A programme is cancelled: With reference to figure 8, assume that the programme PI(n + 2) is cancelled. PI(n + 3) becomes PI(n + 2) and all the following PI block_numbers are decremented by one. Start/stop times within the following PI blocks are modified accordingly. In the AI block the progr_stop_no and/or progr_stop_no_swo values (for Streams 1 and 2 respectively) are decremented and a new version number is necessary in order to force the decoder to reload the entire database. e) Changing attributes or feature flags in an existing PI block: It is assumed here that network, time and date of transmission are NOT being altered and that other PI blocks are not affected by the change. Either a new version number is set in the AI block in order to force the decoder to reload the whole database or an Update block is transmitted to identify the single PI block that has been changed. The latter approach is not recommended, see clause 12.7.4. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 43 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7.2 Operations at the end of a programme | At the end of a programme the AI block is updated to indicate the new current programme. Assuming that the service provider does this at the end of the programme, the decoder would have responded within perhaps 10 s. This channel and the Next Four Programmes feed will have changed as well within, say, 20 s and the main body of the EPG within, say, a minute. For other channels the information for that channel will be updated every 20 minutes at the outside. However, the EPG decoder can identify the current programme from the AI block because it is transmitted frequently. Consequently, the EPG decoder can provide very rapidly a "What's on now" display across all channels. Thus the database changes required at the end of programmes should be handled by the usual refresh cycle. In addition, the "current programme" data in the packet 8/30 format 2 will be changed by the PDC system. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7.3 Major event rescheduling | Rescheduling at short notice has to be taken into account when providing an EPG service. For instance, the live football match has gone to extra time and so the schedule for the rest of the evening is different. For This Channel the revised schedule can be output immediately as part of the normal Stream 1 refresh cycle. (Note this is still slower than an immediate page transmission on the normal Teletext service.) For other channels, either their information is prioritized in the Stream 2 transmission to reflect the changes or the update mechanism is used, see clause 12.7.4. Then, the new schedule should be transmitted via the usual refresh cycle, or as soon as possible. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7.4 Update mechanism | There is an update mechanism allowed for within the specification whereby a single PI can be addressed and altered. It is felt that this will be rarely invoked as in practice a section of the EPG for that channel will be transmitted, either as part of the normal refresh cycle or with some prioritization. Even if an update is transmitted the decoder has to be tuned to the EPG channel for it to be received, thus it is possible, or likely, that the decoder will not receive an isolated Update data block. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7.5 Diagrammatic representation of refreshing | The refreshing of PI and all the Housekeeping information is shown diagrammatically in figure 9. It is extremely simplified as there can be many more circles of information being refreshed, each with different cycle times. This shows each component of the Housekeeping data in Stream 2 being updated at the same time; this may not be the case in practice. Likewise, the editorial divisions of Stream 2 are not shown in detail. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 12.7.6 Bits in each field | There are instances where the number of bits for a particular field is different in different structures. At all times the value of the field shall be the same. The Block numbers in each structure should be contiguous to aid the memory management in the decoder. The broadcaster should note that if a decoder has insufficient memory to store all the blocks of a structure, the highest block numbers will be disregarded. Thus high block numbers should be used for the least used structures. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 44 Stream 1 This Channel Near PInow + Next 4 PIs AI, OI, BI AI, OI, BI PInow + Next 4 PIs Stream 2 Near Stream 2 MI, UI, NI LI, TI, HI Stream 2 MI, UI, NI LI, TI, HI Stream 2 Far Stream 2 MI, UI, NI LI, TI, HI Stream 2 Near Stream 2 MI, UI, NI LI, TI, HI Stream 1 This Channel Near PInow + Next 4 PIs AI, OI, BI AI, OI, BI PInow + Next 4 PIs Figure 9: Representation of refreshing 13 Default EPG operation using the TV-related pages of a Teletext service |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.1 General | Teletext is a well-established medium for the distribution of TV programme information. The TV-related pages in existing services are very popular and can be regarded as a simple form of Electronic Programme Guide (EPG). Channels broadcasting true EPG data may be few and far between, at least in the early days, and not all decoders will be capable of storing data across a channel change. Not unreasonably, the viewer might still expect some level of programme information on his EPG equipped TV regardless of the channel he is watching. If a broadcaster does not wish to provide a proper EPG service for whatever reason, he may be sympathetic to the idea of making his Teletext service more "usable" for very little overhead. This clause indicates the additional data that would need to be transmitted as part of the Teletext service to allow an EPG decoder designed to EN 300 707 [1] to display programme information from the Teletext service in the absence of any valid EPG data. The extra data is necessary to provide the confidence that the decoder has selected the correct page(s) and to eliminate the need for any user action to configure the system. Through the combination of true EPG transmissions and a small amount of additional information within a Teletext service, it should be possible to provide some form of EPG functionality on the majority of TV channels. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 45 |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.2 TV-related Teletext pages | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.2.1 Content | The TV-related pages in normal Teletext services generally carry the following information: - single and multi-channel programme listings for at least "today", some including VCR programming data (PDC or VPS); - "Now and Next" programme information; - previews and synopses; - programme back-up information, for example contact addresses, recipes and URLs; - subtitles. The information can be updated very rapidly to reflect schedule changes. At the editor's discretion, a listing can be limited to the programme title and time of transmission, or include a short description. The more comprehensive services allocate a complete Teletext page per programme in order to provide further details. The "Now and Next" page defines the current and pending programmes for one or more channels. It is often presented in "subtitle" format, with the text inset into the TV picture. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.2.2 Access issues | Page identification and access is not always that easy for the viewer. The relevant pages are likely to have different page numbers on each channel, making them difficult to remember. However, it is not reasonable to expect broadcasters to agree on common page numbers for Europe as some Teletext service providers are allocated a limited number of pages or restricted to using certain magazines. The viewer may find it necessary to consult an index page on route to the TV-related pages, increasing the time the receiver is in Teletext mode and during which the viewer is unable to watch the TV picture. Some decoders allow the viewer to enter their favourite page numbers for each channel and priority is given to their capture. Even if the page numbers are those of the TV-related pages, the viewer may have to press several buttons on his remote control handset to make, for example, the "Now and Next" page appear on the screen. Some TV receivers already have a Subtitle key on the handset for immediate access to the subtitle page. This is relatively easy to implement as there is unique data within the page header packet to enable such pages to be identified. Many EPG receivers will have a dedicated key on the remote control handset to select the EPG feature. As a default mode of operation in the absence of any true EPG data, pressing the key could cause an appropriate Teletext page to be displayed. Unless any extra knowledge could be derived from the transmission, it would be necessary for the viewer to define a relevant page number for each channel. These would be stored in non-volatile memory. However, the design of a simple to use programming interface could be difficult, deterring viewers from using the feature. Further, if a broadcaster rearranges his database, the page number may no longer be valid and would require updating. These problems can be overcome if the Teletext broadcast includes additional information as described in the following clause. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.3 Magazine Inventory Page (MIP) | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.3.1 General principles | EN 300 706 [3] defines a special page which conveys the function or content of each page in a magazine. This is known as a Magazine Inventory Page (MIP). The page number is fixed at MFD, where M defines the magazine to which the contents apply. Each page in the Teletext magazine is defined by a two byte code at a fixed location within the MIP. A number of TV-specific categories are defined. Clause 8.6 asks that a Teletext service carrying EPG data should include a MIP to ensure a decoder can identify the EPG data transport page. ETSI ETSI TR 101 288 V1.3.1 (2002-12) 46 Ideally, if a given page is being broadcast the relevant packet of the MIP would be transmitted to define the function of that page. However, this is not mandatory and EN 300 706 [3] states that the absence of a packet within the MIP does not necessarily imply that the pages covered by that packet are not being broadcast. This approach was adopted to minimize transmission capacity, for example if the sole reason for transmitting a MIP is to define the EPG data transport page. Under these circumstances only two packets (the page header and the packet containing the definition of the relevant page) need to be transmitted. To implement the default EPG scheme proposed here it is also necessary to transmit the packets covering the TV-related pages. If these page are not in the same magazine as the EPG data transport page they will be in a different MIP, requiring an additional page header. It is unlikely that more than three packets will be required. Because a MIP is potentially useful to all Teletext decoders, it has to be transmitted obeying the 20 ms rule. The MIP can be updated at any time, for example if the Teletext database is re-organized. |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.3.2 TV-related categories | The TV-related categories listed below have explicit entries in the coding scheme used by the MIP. The hexadecimal code value is given in brackets. For some types it is possible to indicate that there is more than one sub-page and, for some, the number of sub-pages in a multi-page set. TV Index page, single page (7F); TV Index page, multi-page set (7E) TV schedule page, single page (81) TV schedule page, multi-page set, number of sub-pages in the range 2 to 79 (82 - CF) TV schedule page, multi-page set, number of sub-pages in the range 80 to 212 -1 (D0) TV schedule page, multi-page set, number of sub-pages in the range 212 to 213 -2 (D1) Current TV programme information page, single page (7C) Current TV programme information page, multi-page set (7B) Current TV programme warning page (7A) "Now and Next" TV programme (7D) Subtitle page (70 - 77) |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.4 Operational aspects | |
5a4d9711ae9929bfa143700e1f8d2372 | 101 288 | 13.4.1 Default operation of decoders | The transmission of a MIP in which the TV-related Teletext pages were indicated would enable manufacturers to offer some EPG functionality in the absence of valid EPG data. The actual interpretation is at the discretion of the decoder manufacturer. For example, pressing the EPG key could result in the "Now and Next" page from the Teletext service being displayed automatically. The decoder will have interpreted the MIP data and requested the relevant page with a high degree of confidence in the validity of the page number and the contents of the page. The viewer does not have to remember specific page numbers, nor has he had to pre-programme the system. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.