Datasets:

---

ExponentialScience

/

DLT-Scientific-Literature

like 0

Follow

Exponential Science 4

ERROR: type should be string, got "https://doi.org/10.1007/s10796 021 10150 8\n\n# A System to Access Online Services with Minimal Personal Information Disclosure\n\n**Antonia Russo[1]** **· Gianluca Lax[1]** **· Baptiste Dromard[2]** **· Menad Mezred[2]**\n\n\nAccepted: 21 May 2021\n© The Author(s) 2021\n\n\n/ Published online: 19 July 2021\n\n\n**Abstract**\nThe General Data Protection Regulation highlights the principle of data minimization, which means that only data required\nto successfully accomplish a given task should be processed. In this paper, we propose a Blockchain-based scheme that\nallows users to have control over the personal data revealed when accessing a service. The proposed solution does not rely\non sophisticated cryptographic primitives, provides mechanisms for revoking the authorization to access a service and for\nguessing the identity of a user only in cases of need, and is compliant with the recent eIDAS Regulation. We prove that the\nproposed scheme is secure and reaches the expected goal, and we present an Ethereum-based implementation to show the\neffectiveness of the proposed solution.\n\n**Keywords Blockchain · Smart contracts · eIDAS Regulation · Access control · Privacy**\n\n\n## 1 Introduction\n\nIn the digital era, information is a valuable asset: for\ninstance, think about social networks, which collect\ninformation of hundreds of millions of people such as\npersonal data, friends, visited places, listened to music,\nwatched TV, preferences, and interests. Such data are\nmonetized in several ways, such as to produce contextaware information that influences users’ preferences to\nrecommend specific items (custom advertising). For this\nreason, often accesses to services require that a user\nauthenticates.\n\n� Gianluca Lax\n[lax@unirc.it](mailto: lax@unirc.it)\n\nAntonia Russo\n[antonia.russo@unirc.it](mailto: antonia.russo@unirc.it)\n\nBaptiste Dromard\n[dromard@ecole.ensicaen.fr](mailto: dromard@ecole.ensicaen.fr)\n\nMenad Mezred\n[mmezred@ecole.ensicaen.fr](mailto: mmezred@ecole.ensicaen.fr)\n\n1 University of Reggio Calabria, Reggio Calabria, Italy\n\n2 ENSICAEN, Caen, France\n\n\nHowever, in many situations, there is not a real need to\nbe aware of personal information, which is done only to\ncollect rich data. Consider the case of a merchant selling\nproducts for adults (e.g., liquor or cigarettes), which only\nneeds to check that acquirers are of a certain age: the request\nof the identity card to show the birth date has the side effect\nof disclosing personal information, such as name, surname,\nnationality. Although this situation is not very worrying in\nreal life, the problem is relevant in the digital world because\ndisclosed data can be stored and processed automatically.\nAs collected data may contain private information that could\nbe transferred to unauthorized parties, privacy-preserving\nproposals in this context are gaining attention (Shin, 2010;\nKim et al., 2018). Indeed, recently many service providers\nrequire the disclosure of less sensitive information.\nThis privacy problem is well-known and has been\nrecently remarked also by the issuance of the General\nData Protection Regulation (European Parliament, 2016),\nthe new European Union privacy law that puts guidelines\nand regulations on how data have to be processed, used,\nstored, or exchanged to protect and ensure individuals data\nprivacy (Lee et al., 2020; Shin et al., 2019). In this context,\nthe problem we address is how to build a system able to\nreach four research goals:\n\nRG1. when accessing a service, a user should be\nallowed to provide the minimal amount of personal\ninformation needed to access the service;\n\n\n-----\n\nRG2. only authorized users should access a service;\nRG3. there should be the possibility to revoke the\npermission given to a user to access a service;\nRG4. the user’s identity should be revealed in case of\nneed.\n\nFor this purpose, we exploit the power of Blockchain\n(Nakamoto, 2008), a recent technology used in many\nfields, such as finance, smart cities, society progress\ndriving (Shin & Ibahrine, 2020). Blockchain is a fully\ndistributed repository that stores transactions. Several nodes\ndistributed in a peer-to-peer fashion have the control over\nstored information and run programmable rules in the\nform of smart contracts (Karamitsos et al., 2018). By\nreplacing a single centralized party with a distributed ledger\nof replicated, shared, and synchronized data, Blockchain\nguarantees transparency, traceability, and immutability of\nregistered information.\nIn this paper, we propose a Blockchain-based system\nthat allows users to prove the possession of some attributes\nwithout disclosing their identity. Moreover, our proposal\nprovides suitable mechanisms to allow revocation and\naccountability. Revocation concerns the possibility to make\ninvalid a credential when a user loses possession of an\nattribute (for example, a driver’s license) or the credential\nis stolen or expired. Accountability is a feature that\nallows a party in cooperation with other trusted parties to\nguess the identity of a user in cases of need. Differently\nfrom the state of the art, our proposal does not rely on\nsophisticated cryptographic primitives, which reduce the\nefficiency of a solution. An important aspect is related to\nthe Regulation (EU) N 910/2014 (European Commission,\n2016), which regards electronic identification and trust\nservices for electronic transactions in the EU internal\nmarket. This regulation provides a normative basis to enable\nsecure electronic interactions between businesses, citizens,\nand public authorities. Among others, eIDAS introduces\nthe role of the Attribute Provider, an entity responsible\nfor providing information about electronic identities. The\nissuance of eIDAS opens the possibility to design new\nsolutions for attribute certification that can be very effective\nbecause it is expected that in the next years eIDAS will\ninvolve most of EU people. We exploited this opportunity\nso that the proposed solution is compliant with the\neIDAS infrastructure, which increases its effectiveness. We\ninstantiated the general solution to a real-life scenario and\ndescribed the detailed data workflow to show how our\napproach can be implemented by Ethereum, the most used\nBlockchain enabling smart contracts.\nThe paper is structured as follows. In the next section,\nwe survey the most significant proposals of the state of\nthe art related to the addressed problem. We introduce the\nbackground needed to understand the proposal in Section 3.\nIn Section 4, we describe the faced problem. In Section 5,\n\n\nwe define the approach used to allow a user to disclose some\nattributes without revealing personal data. In Section 6, we\ninstantiate the proposed scheme to a specific scenario, and\nwe provide the technical details about how our solution\nworks. The security analysis is presented in Section 7.\nFinally, in Section 8, we draw our conclusions.\n\n## 2 Related Work\n\nOne of the problems addressed in this paper concerns access\ncontrol, which ensures that users can access a piece of data\nonly if they are authorized (Goyal et al., 2006). AttributeBased Access Control (ABAC) is defined as logical access\ncontrol methodology where authorization to perform a\nset of operations is determined by evaluating attributes\nassociated with the subject, object, requested operations,\nand, in some cases, environment conditions against policy,\nrules, or relationships that describe the allowable operations\nfor a given set of attributes. Hu et al. (2013) define ABAC\nto understand the real applications of this mechanism.\nAttribute-Based Access Control is analyzed in real use\ncases to improve scalability, feasibility, and performances\nof applications in which the information sharing within and\nbetween organizations is expected.\nSahai and Waters (2005) propose the concept of\nAttribute-Based Encryption. Authors provide an original\ntype of Identity-based Encryption in which the identity\nconsists of an attribute set. Users, with an identity and their\nattributes, can decrypt a ciphertext encrypted with the same\nattributes.\nGoyal et al. (2006) develop a new cryptographic system\nfor fine-grained access control of shared encrypted data,\ncalled Key-Policy Attribute-Based Encryption (KPABE). In\nthis system, ciphertexts are matched with sets of attributes,\nand private keys are associated with access structures.\nThe concept of Ciphertext-Policy Attribute-based\nEncryption is formalized by Bethencourt et al. (2007). In\nthis solution, the policy is associated with the ciphertext and\nthe attributes with the key. A user can decrypt a ciphertext\nif the user’s attributes pass through the ciphertext’s access\nstructure.\nAttrapadung and Imai (2009) propose a solution combining CP-ABE with KP-ABE. The authors consider a scheme\nin which both policy and attributes are associated with the\nciphertext and key. The attributes are related to the ciphertext, and the policy designs the users who can decrypt. The\npolicy states the kind of ciphertext the user can decrypt.\nThe Attribute-based proxy re-encryption scheme\n(ABPRE) (Liang et al., 2009) extends the ABE scheme\nempowering users with delegating capability in the access\ncontrol environment. A proxy can be chosen by users to\nre-encrypt a ciphertext related to a specific access policy.\n\n\n-----\n\nConcerning the revocation of attributes, Hur and Noh\n(2010) propose a solution exploiting ciphertext-policy\nattribute-based encryption for an access control mechanism.\nThe solution is related to an efficient implementation\nprovided with an attribute and user revocation capability.\nAttribute-based encryption has always been considered\nas a technology for solving the problem of data privacy\nand fine-grained access control in traditional cloud storage\nsystems based on a centralized storage architecture.\nThe development of Blockchain technology allows the\nbuilding of a decentralized storage mode that could\novercome the problem of a single point of failure in\ntraditional cloud storage. Wang et al. (2018) propose a\nframework that combines the Ethereum Blockchain and\nABE technology to implement data storage and sharing\nscheme for decentralized storage systems.\nMaesa et al. (2017) deal with a new approach to access\ncontrol based on Blockchain technology. The policies\nthat express the right to access a resource are published\ninside Blockchain. That way, every user can check if\npolicies and resources match. Considering a Blockchain,\nits capabilities of transparency and immutability allow a\ndistributed consensus and auditability preventing a party\nfrom denying the rights granted by the policy.\nBlockchain technology can provide patients with\nimmutable records regarding their medical data, said Electronic Health Records (EHRs). Guo et al. (2018) present\nan attribute-based signature scheme with various authorities to enforce the validity of EHRs stored in a Blockchain.\nThis system allows patients to possess the control of generating, managing, and sharing EHRs with other authorized\ndata consumers in a secure environment.\nIn a cloud computing environment, service providers\ncan be allowed to take care of confidential data, and this\npermission may raise potential security and privacy issues\n(Wang et al., 2010). The cloud service provided, by adopting\nan encryption system, has to support fine-grained access\ncontrol and also provide high performance and scalability.\nThe authors propose a scheme to help companies use cloud\nservers to share confidential data efficiently.\nPinno et al. (2017) provide a new architecture for access\ncontrol in the IoT. This framework, based on Blockchain\ntechnology, overcomes the FairAccess (Ouaddah et al.,\n2016) and other issues derived from the architecture,\nevaluating a new decentralized and authorization process\nfor authorization in IoT environments. FairAccess is\nan authorization management framework that is fully\ndecentralized and privacy preserving and uses Blockchain\nas a decentralized access control manager. With the help of\nBlockchain, this solution introduces various and new types\nof transactions to delegate, revoke, and grant access.\nFrom the review of the state of the art here reported, it\nemerges the importance of the attribute-based access control\n\n\nproblem, which is faced by various sophisticated cryptographic approaches. The first observation is that most of the\nproposals in the field of selective disclosure (for example,\nHern´andez-Ramos et al. (2018) and Lor¨unser et al. (2016))\nrely on sophisticated cryptographic primitives. In contrast,\nour proposal relies only on efficient and straightforward\ncryptographic operations, which are considered secure and\nefficient. Moreover, our proposal is designed for taking\nadvantage of the attribute providers defined in eIDAS. To\nthe best of our knowledge, our proposal is the first one compliant with the eIDAS Regulation, which will be a de-facto\nstandard in the next years (at least, in the European Union).\n\n## 3 Blockchain and eIDAS\n\nIn this section, we recall some concepts used in this paper,\nwhich are Blockchain technology and eIDAS ecosystem.\nBlockchain is a technology that could have the capacity\nand potency for enhancing and changing various aspects\nof economy and society, and for this reason, it can be\nconsidered as a disruptive technology (Swan, 2015). It\nwas proposed by Nakamoto (2008), and it is defined as a\ndistributed ledger that stores, in a transparent and immutable\nway, transactions executed among users. Information is\nstored inside blocks, whose number and dimensions are\ncontinuously growing. Every block is linked to the chain\nby its header. The header contains the hash of the previous\nblock and a timestamp. The transactions that take place in\nBlockchain are stored inside blocks and contain information\non the recipient’s public address, the characteristics of\nthe transaction, and the cryptographic signature, which\nguarantees the integrity and authenticity of the transaction.\nEvery operation has to be confirmed and validated by\nBlockchain participants, and this concept is summed up by\ndistributed consensus. This way, users can trust the system\nof the public ledger, without trusting a central authority or\na third-party intermediary. It has been proved that perceived\nprivacy in using blockchain positively could affect and\ninfluence users’ trust and attitudes towards blockchain\ntechnology (Shin, 2019; Shin & Hwang, 2020). Over time,\nBlockchain assumed different meanings and definitions: the\nfirst one is called Blockchain 1.0, and it is referred to as\nthe Bitcoin paradigm. This kind of system represents a\nplatform in which it is possible running and deploying all\nthe operations carried out with cryptocurrency in digital\npayment systems.\nEthereum (2020) is a Blockchain 2.0 that enables the\npossibility to create and run smart contracts, programs\nexecuted over the Ethereum computing infrastructure. A\nsmart contract is defined as a digital protocol to verify\nor digitally enforce the negotiation of a contract, with no\nneed for an intermediary. The code of a smart contract\n\n\n-----\n\nautomatically verifies the occurrence of specific conditions\nand automatically carries out actions when the conditions\ndetermined between the parties are reached. Smart contracts\nare written in Solidity (2020), an object-oriented and\nhigh-level language, and a real application relies on the\ncreation and the deployment of distributed applications\n(Ethereum dApps, 2020). It exists a practical and conceptual\nissue about the external data used to verify and perform\ndecision inside smart contracts, that is the “oracle” presence\nin Ethereum. An oracle has the purpose of connecting\ndecentralized applications with third-party services in a\ntrust and secure way, in order to get data from outside\nBlockchain and execute any API call preventing data\nintegrity and authenticity. Provable blockchain (2020) is\nthe most famous oracle service for smart contracts and\nBlockchain applications. It consists of three main entities:\ndata-source, query, and oracle. When a smart contract\nrequires data from a data source outside Blockchain, it sends\na query to Provable and calls a function passing the result\nof the query as an input. Provable aims to demonstrate that\nthe data taken from the original data source is authentic, by\nlinking returned data with an authenticity proof document.\nThe Regulation (EU) N 910/2014 (European Commission, 2016) on electronic identification and trust services\nfor electronic transactions in the internal market provides\na normative basis to enable secure electronic interactions\nbetween businesses, citizens, and public authorities. The\neIDAS principles are based on security, trust, and interoperability of electronic services carried out by citizens all\nover EU countries. All Member States have to notify their\neID schemes (national electronic identification schemes)\nto the European Commission, which are published in the\nOfficial Journal of the European Union. Both people and\ncompanies can access public services provided by an EU\nMember State using the eID of another EU Member State:\nthis concept aims at promoting cooperation between states.\nInteroperability between different eID-schemes is reached\nby defining the interfaces between eIDAS-Nodes. However,\nthe eID ecosystem consists of various actors that should\nbe available in all EU countries: the most important is the\nnode operator, which controls that an eID node behaves correctly and implements the function of the connection point\nbetween the attribute provider, the identity provider and the\nservice provider.\nIn our proposal, we will focus on the role of attribute\nproviders and, in particular, those compliant with eIDAS\nRegulation. An attribute provider is in charge of providing\ninformation about electronic/digital identities: specifically,\nthis entity has to verify the real possession of different\nattributes from a digital identity. A second entity, the\nidentity provider, creates and manages the public digital\nidentity of a physic or legal individual. Instead, a\nservice provider is a public or private entity that makes\n\n\navailable online services using eID for cross-border\nauthentication. Inside the eIDAS framework, eIDASConnector and eIDAS-Service need to exchange messages\nregarding personal and technical attributes to support crossborder identification and authentication processes. To do\nthis, they use SAML (2020).\nSAML (Security Assertion Markup Language) is\nan XML-based open-source framework created for the\nexchange of authentication and authorization information\nbetween the service provider and the identity or attribute\nprovider in a secure way. The SAML 2.0 specification\ndefines a series of request/response protocols and assertions\nthat allow an application to request or query an assertion or to ask a user for authentication. Regarding the\nproof of attribute possession during an authentication process, the complete list of attributes included in the eIDAS\nminimum data sets is declared in the technical document\n(CEF Digital, 2019) concerning eIDAS SAML Attribute\nProfile.\n\n## 4 Scenario and Problem Formulation\n\nIn this section, we introduce a scenario to present the\naddressed problem.\nThe scenario is composed of the following actors:\n\n– _Users, physical or legal people with an eIDAS-_\ncompliant digital identity.\n– _Identity Providers, eIDAS-compliant entities that create_\nand manage digital identities.\n– _Attribute Providers, eIDAS-compliant entities that are_\nin charge of verifying and validating the possession of\nattributes.\n– _Service Providers, which supply users with (online)_\nservices.\n– a Blockchain, a Distributed Ledger in which smart\ncontracts can be deployed.\n\nWe consider the case of a user who needs to access an\nonline service supplied by a service provider, but this access\nis granted, provided that the user has the permissions and\nattributes to access it. The considered service is a car rental\nbooking, in which users need to demonstrate the possession\nof two attributes: a driver’s license and the age (because\nthe rental price depends on the user’s age). Thus, a service\nprovider that offers a rental car service needs to be sure\nabout the possession of the driver’s license and to know\nthe age of any user who wants to rent a car. Generally,\nin such cases, a service provider asks users to provide all\npersonal data, which characterize them as legal and natural\npeople. In some cases, the identification can be performed\nwith the support of a third party (such as a social network\nauthentication procedure).\n\n\n-----\n\nThe problem is that users have to reveal many personal\nand (possible) sensitive data that are not useful to gain the\nservice requested, and this could damage their privacy and\nexpose them to various attacks or future vulnerabilities.\nThe disclosure of specific attributes can help to reduce this\neffect: users are responsible for the information they want\nto share with third parties, that is, they can decide to show\ndata in a granular way.\nIn summary, the goals to reach are:\n\n1. the service provider should be aware of the minimal\namount of the user’s information needed to access the\nservice;\n2. only authorized users should access a service;\n3. there should be the possibility to revoke the permission\ngiven to a user to access a service;\n4. the user’s identity should be revealed in case of need.\n\nConcerning the last item, it is worth noting that the\nrecent definition of the eIDAS environment can help to face\nthis problem with new tools that did not exist until a few\nyears ago (Priesnitz Filho et al., 2019). Indeed, the identity\nproviders defined by eIDAS keep a series of information\nrelated to the digital identity of users, called unqualified or\nelementary attributes. Furthermore, the attribute providers\ndefined by eIDAS are authorized to certify a qualification\nand can add other attributes to the digital identity of the user.\nAs a consequence, we propose a solution based on\nthe disclosure of the attributes selected by users who\nwant to access the service. Besides eIDAS, our proposal\nalso relies on the recent technology of Blockchain, which\nallows us to design a decentralized approach to maintain a hidden link between users and their attributes.\nBlockchain (see Section 3) is in charge of guaranteeing transparency and immutability of actions (or transactions) and allowing stakeholders to perform and verify a\nsecure access control relying on attributes through smart\ncontracts.\n\n## 5 Conceptual Model\n\nStarting from the scenario described in the previous\nsection, here we describe the solution designed to solve\nthe considered privacy issue. The entities in our scenario\ncooperate by performing the tasks described in the\nfollowing.\n\n1. _Digital Identity issuing. Any user running our solution_\nneeds to have an eIDAS digital identity. To do this, the\nuser needs to register to one of the available identity\nproviders, which is responsible for the verification of\nthe user identity before issuing the credential of the\ndigital identity.\n\n\nThe identity provider of a user knows a list of\nelementary attributes (e.g., date of birth) of the user.\nOn the other side, there are also attribute providers that\nmanage other and not elementary attributes related to\nthis identity (for example, a Motor Vehicle Office plays\nthe role of attribute provider for a driver’s license).\n2. _Blockchain registration. In this proposal, users will be_\nreferred by their Ethereum address, so that any user\nhas to create an external owned account, characterized\nby a public address and controlled by a private\nkey. The public key derives from the private one\nand is computed by a cryptographic function of\ntype elliptic curve point multiplication. Typically, the\nBlockchain address is derived from the public key by\na cryptographic hash function. For example, in Bitcoin,\na user with public key K has an address computed\nas RIPEMD160(SHA256(K)), where SHA-2 (2020) is\na cryptographic hash developed by National Security\nAgency (NSA) and returns a 256-bit digest, whereas\n(RIPEMD160, 2020) is a cryptographic hash designed\nin the open academic community and returns a 160-bit\ndigest (i.e., the address).\n3. _Credential issuing. This operation is carried out by the_\nuser when a credential for one or more attributes is\nneeded and involves an attribute provider (say AP [1]).\nAccording to the eIDAS protocol, AP acts as a service\nprovider and needs to identify the user by an eIDAScompliant scheme, as illustrated in Fig. 1.\nFirst of all, the user sends a request for a credential to\n_AP (Step 1). Then, AP replies with an authentication_\nrequest to be forwarded to the Identity Provider (Step\n2). The Identity Provider performs a challenge-response\nauthentication with the user (Steps 3 and 4). In case\nof successful user authentication, the Identity Provider\nprepares the assertion of user authentication, which is\nforwarded to AP (Step 6). This way, AP is aware of the\nuser’s digital identity and can verify if the user owns the\nrequired attributes. In this case, AP performs a second\ntask to know the user’s Blockchain address (Step 7):\n_AP sends a random string to the user and waits for_\nreceiving a Blockchain transaction with this string from\nthe address of the user, say A.\nIf such a transaction is received, then AP replies\nto the user with the requested credential: it consists\nof an assertion reporting the Blockchain address A\nof the user, the verified attributes, and a (suitable)\nURL belonging to the Web domain of the attribute\nprovider (e.g., [http://www.attributeprovider.com/](http://www.attributeprovider.com/8f7b19f38f4c4b10b52de9727e9f0538)\n[8f7b19f38f4c4b10b52de9727e9f0538).](http://www.attributeprovider.com/8f7b19f38f4c4b10b52de9727e9f0538) Finally, the\n\n1For the sake of simplicity, let us assume only one attribute provider\nhandles the needed attributes: in case more attribute providers are\ninvolved, this operation is repeated for each of them.\n\n\n-----\n\n**Fig. 1 Data flow in assertion**\nissuing process\n\nattribute provider publishes at this URL the digest of\nthe credential as a proof of authenticity and integrity of\nthe credential (this replaces a cryptographic signature).\n4. _Credential using. When a user needs to access a_\nservice granted only to people with some attributes\n(e.g., a driver’s license or being of age), the user\ncan exploit a credential for such attributes obtained\nwith the procedure described in the previous step. The\nuser sends the credential assessing the possession of\nthe requested attributes to the service provider (say\n_SP ) supplying this service. The verification of the_\ncredential validity is carried out from SP by the call\nto a function of the smart contract SC, which receives\nthis credential and executes the following steps: (1)\nextracts from the credential the value of the URL field,\n(2) verifies that this URL belongs to the domain of the\nattribute provider, (3) downloads the file at this URL,\n(4) calculates the digest of this file, say D, and (5)\ncalculates the digest of the received credential, say D[′].\nIf and only if D = D[′], then the function returns that\nthe credential is valid. Only in this case, the user will\nbe able to access the service; otherwise, the access is\ndenied.\n5. _Credential Revocation. Revocation concerns the possi-_\nbility to make invalid a credential when a user loses possession of an attribute (for example, a driver’s license)\n\n\nor the credential is stolen or expired. Revocation is carried out by the attribute provider: for each credential\nto be revoked, the attribute provider extracts the specified URL where the credential is published and removes\nthe document on this URL, making the file unreachable. This way, the user cannot provide the proof of\nthe requested attribute to any service provider, which\ncannot find the credential.\n\nThe proposal here presented is defined at a conceptual\nlevel and does not consider several (orthogonal) aspects,\nsuch as the smart contract and the exchanged data. These\naspects are the subject of the next section.\n\n## 6 Implementation and Proof of Concept\n\nIn this section, we instantiate the general approach\npresented above to a real-life scenario to help the reader to\nunderstand better how our solution can work. We describe\nthe detailed data workflow to show how our approach can\nbe implemented in a real Blockchain: we used Ethereum,\nwhich is the most used one enabling smart contracts.\nFor the sake of presentation, we refer to a simplified\nscenario in which the service to access requires the user\nto be of age (for example, in the case of age-restricted\n\n\n-----\n\n**Fig. 2 User authentication**\n\nvideos, which are not visible to users who are under\n18 years of age). Our implementation is based on the\nEthereum Blockchain, and the environment in which we\ntested our solution is the Ropsten testnet Blockchain,\na free Blockchain based on Ethereum using proof of\nwork (Ropsten Testnet Explorer, 2020). The smart contract\nis build using (Solidity, 2020) and exploits (Provable\nblockchain, 2020) to import data from external sources (see\nSection 3). We implemented a JAVA decentralized web\napplication (DAPP) by the Web3j SDK (2020) library and\nused Infura (Ethereum and IPFS APIs, 2020) as Blockchain\ninfrastructure to access the Ethereum Blockchain.\nNow, we describe how the operations defined in our\nproposal are implemented in this scenario.\n\n1. _Digital Identity issuing. The public digital identity_\nof a user U can be issued in order to perform a\nsecure authentication based on attributes. The Identity\nProvider IP stores a list of the elementary attributes\nof the user. The Attribute Provider AP manages not\nelementary attributes and checks the identity of the\nuser.\n2. _Blockchain registration. We created an Ethereum_\naddress for any entity involved (user, identity provider,\nand attribute provider). First, we generated a couple of\nasymmetric keys for each of them. Then, the Ethereum\naddress is computed from the public key by applying\nKeccak-256 (Bertoni et al., 2009) and taking the last\n20 bytes of that hash. In practice, we installed the\n_MetaMask extension in Google Chrome, created the_\nnew accounts, and saved the seed words for restoring\nthe MetaMask accounts.\n3. _Assertion issuing. This operation is carried out by the_\nuser when a proof of attribute is needed.\nTo obtain the assertion, the user connects to the\n_AP ’s site by a browser and sends the request for an_\nassertion. Now, the user is authenticated by the chosen\n\n\neID (this part is skipped in our implementation because\nit is a standard procedure). Then, AP replies with a\nchallenge-response authentication with the user (see\nFig. 2). In case of successful user authentication, the\nattribute provider shows a page with all the user’s\nattributes so that the user can select the ones to be\ncertified, as shown in Fig. 3.\nNow, AP prepares a response that includes the\nassertion of the attributes selected by the user: this\nmessage is based on the SAML format and according\nto eIDAS eID Technical Subgroup (2019) the most\nrelevant fields are:\n\n_•_ an attribute Version with the version of SAML\nused in the message;\n\n_•_ an element StatusCode with the outcome of the\nrequest;\n\n_•_ an attribute ID containing the assertion identifier;\n\n_•_ an attribute IssueInstant, which specifies the\ninstant at which the assertion is issued;\n\n_•_ an element Issuer, which refers to the issuer of\nthe message;\n\n_•_ an element AuthnInstant, which specifies\nthe instant at which the authentication has been\nperformed.\n\n_•_ an element AuthnContextClassRef, which\nspecifies the used authentication method based on\na particular class reference.\n\nAn example of a response message is shown in\nFig. 4. The status code contains the URL used by\n_AP to publish the assertion. The assertion contains the_\nversion, the ID, the issue instant, the reference to the\nissuer, the Blockchain address of the subject, the end of\nvalidity, and the attribute name. Moreover, the assertion\nhas temporal data regarding the start of validity instant,\nthe timestamp of the authentication, and the type of\ncertified attributes (proof of age).\n\n\n-----\n\n**Fig. 3 Attribute selection**\n\nMoreover, AP publishes at the URL reported in the\nassertion a file containing the assertion and calculates\nthe digest of this file. Finally, AP calls the function\n_indexDigest of the smart contract (see Fig. 6, Lines_\n54-59), which generates a token. This token is the\ndigest of a nonce generated from the block timestamp\nand incremented by one every time a new token is\ngenerated. This ensures its uniqueness. Observe that it\nis not a secret value (all data in the smart contract are\nnot secret). This token is used as an index to find such\nan assertion digest in a mapping (Line 57).\nFinally, the attribute provider sends to the user a\nJSON file containing the URL where the assertion is\navailable, and the token t. An example of this file is\nreported in Fig. 5.\n4. _Assertion using. Suppose the user needs a service_\nsupplied by the service provider SP and U is required\nto possess an attribute, for example, a proof of age. To\n\n**Fig. 4 Example of response**\nmessage\n\n\nprove to be of age, U sends the JSON file received by\nthe attribute provider in the previous step to SP .\nTo verify this credential, SP calls the function\n_verifyAssertion of the smart contract SC (Fig. 6, Lines_\n38-49), giving as input the token and the location\nextracted from the JSON file. This function is payable\n(i.e., it can receive Ether) and exploits the Provable\noracle to download the assertion located at the given\nURL (Line 44). Then, it retrieves the digest of the\nassertion previously stored in digestSet (Line 57) and\nadds it to verifiedDigest, a mapping between digests and\nquery identification (i.e., content in Line 45). Moreover,\nthe query is included in a set of pending query (line 46).\nOnce Provable returns the query result, the callback\nfunction is called automatically. This function has\nthe oracle query identification Id and the query\nresult as parameters. First, the digest of the assertion\ndownloaded by Provable is calculated (Line 29)\n\n\n-----\n\n**Fig. 5 Example of JSON**\ncredential\n\nand compared to the digest previously stored in\n_verifiedDigest (Line 30). The result of this comparison_\nis stored in resultSet.\nThen, the Service provider SP calls the function\n_checkResult, which takes the query identification Id as_\na parameter. This function returns the content of the\n_resultSet at the index Id, which represents the result_\nof the previous function (Lines 50-53). Specifically, the\nvalue 1 denotes a valid assertion, the value 2 denotes\nan incorrect assertion, whereas the value 0 denotes that\nthe query result is not available (for example, in case of\nwrong id of the query).\n\nConcerning the smart contract, we showed a more extensive implementation in which also events are used (see Lines\n11-17 and the calls to function emit). Logged events can\nprovide support in case of need (for example, in the event\nof a complaint). The implementation of this solution by a\n[Java prototype is available on Github at the address https://](https://github.com/DroBaptiste/SelectiveDisclosureOfAttribute)\n[github.com/DroBaptiste/SelectiveDisclosureOfAttribute.](https://github.com/DroBaptiste/SelectiveDisclosureOfAttribute)\nConcerning the smart contract, it has been deployed on Rop[sten and is reported at https://ropsten.etherscan.io/address/](https://ropsten.etherscan.io/address/0x2cF05A44F23A92581088c17e7C8c7D88B2F8d0f2#code)\n[0x2cF05A44F23A92581088c17e7C8c7D88B2F8d0f2#](https://ropsten.etherscan.io/address/0x2cF05A44F23A92581088c17e7C8c7D88B2F8d0f2#code)\n[code,](https://ropsten.etherscan.io/address/0x2cF05A44F23A92581088c17e7C8c7D88B2F8d0f2#code) where the provable.sol library has been\nincluded.\n\n## 7 Security Analysis\n\nIn this section, we discuss how our proposal reaches the\nexpected goals, which are:\n\n1. the service provider should be aware of the minimal\namount of the user’s information needed to access the\nservice;\n2. only authorized users should access a service;\n3. there should be the possibility to revoke the permission\ngiven to a user to access a service;\n4. the user’s identity should be revealed in case of need.\n\nLet us start with the first property. The service provider\nis aware of (i) the JSON file and (ii) the assertion. The\nformer does not contain any information about the user; the\nlatter contains only data that the user selected. Provided that\n(1) there is no collusion among identity provider, attribute\nprovider, and service provider, and (2) the user does not\nselect more information than the needed one to access the\nservice, the first goal is reached.\n\n\nConcerning the second item, it is clear that authorized\nusers can access the service by following the protocol.\nThus, we have to prove that unauthorized users cannot\ncreate a valid assertion. In our analysis, we assume that the\nprotocol is correctly run by identity and attribute providers\nbecause they are trusted entities. Consequently, an attacker\ncannot tamper with his/her identity or attributes, which are\nguaranteed by identity and attribute providers. When the\nattacker is able to create an assertion that passes all the\nsecurity checks, we have to consider the following two\npossibilities. The attacker has created a false assertion and\nis able to create the signature done by the attribute provider.\nThis means to break the cryptographic primitives or to\nguess the private key. The probability of any of these cases\nis negligible. Another possibility is that the attacker has\nviolated the smart contract. If we assume no error exists\nin the smart contract code, then the adversary can only try\na 51% attack (Sayeed & Marco-Gisbert, 2019). Again, the\nprobability of this attack is negligible.\nConcerning the attribute revocation, it can occur for\nseveral reasons: 1) most of the attributes have an expiring\ndate from which their validity ends, 2) a user can choose\nto change the status of a single attribute (in our scenario\nattributes could be separate and disjointed) or 3) the\nmost significant cause is when the attribute provider\nought to revoke the expired attribute. How the revocation\noccurs is evident in the protocol. Thus, we focus on a\npossible misconduct of attribute providers, that is the unfair\nrevocation of an attribute. Consider the case of an attribute\nprovider AP that has issued an adult proof of age credential\nto a user U . Now assume that AP revokes this credential\nunfairly so that the credential validation fails. However, if U\nhas stored this credential, then by computing the digest d of\nthe credential and by invoking the function indexDigest of\nthe smart contract (see Fig. 6, Lines 54-59), U can prove that\na token is associated with d, and thus, that AP had issued\nthis credential. This possible malicious behavior of AP is\ndetected and, thus, contrasted. Clearly, in some cases, other\ninformation could be necessary to close the complaint.\nAs for the last goal, we reach this result by exploiting\nan eIDAS-compliant identification scheme. The robustness\nof an identification scheme depends on the degree to\nwhich it adheres to the technical specifications and best\npractices. Even if the standards used for identification\nsystems can vary by country, the compliance with eIDAS\nensures acceptable robustness. Our scheme allows a party in\ncooperation with other trusted parties to guess the identity\n\n\n-----\n\n**Fig. 6 Code of smart contract**\n\nof a user in cases of need. This possibility is given because\neach credential has an identifier, and each attribute provider\nstores the mapping between each generated credential and\nthe user identity. Recall that a service provider is not aware\n\n\nof the identity of the user accessing the service, and the only\ninformation known is about the user’s attributes certified\nby the credential. However, in case of valid reasons (for\ninstance, a terrorist who rented a car), the cooperation\n\n\n-----\n\nbetween the service provider (which knows the credential\nidentifier) and the attribute provider that issued such a\ncredential (which knows the user’s identity associated with\nthis credential) allows us to uncover the user’s identity.\nConcerning this aspect, we observe that there is the\npossibility of running off-line the scheme by relaxing some\nof the security requirements. Consider the case in which\nthe URL is not available: the protocol should deny the\nrequest of access request because the credential cannot\nbe retrieved. However, as it is done for micro-payments\nwhen the economic value of the service is limited, the\nservice provider could accept to receive the credential from\nthe user instead of downloading it from the (unavailable)\nURL. Observe that in this case, the credential verification\nis done again by the function indexDigest, thus proving that\na token is associated with this credential. Clearly, in this\nway, the check of credential revocation has not been carried\nout: however, if the cost of making unavailable the web\nsite of the attribute provider (for example, by a Denial-ofService attack) is higher than the price of the service, it\nis not advantageous for a malicious user to run this attack\nfor obtaining the service. The choice about providing the\nservice at this risk is left to the service provider.\n\n## 8 Conclusion\n\nIn this paper, we proposed a system that allows users to\nprove the possession of some attributes without disclosing\ntheir whole identity. Our solution relies on Blockchain,\nwhich is a very recent technology but considered mature\nand already widely adopted in many application contexts.\nThe use of a Blockchain platform such as Ethereum\nallowed us to make transparent cryptographic operations to\nuser and system, thus making the solution more effective,\nrobust, and secure. Using certain features of Blockchain,\nthe proposed system provides suitable mechanisms for\nrevocation and accountability. The proposed solution has\nbeen implemented by a JAVA decentralized Web application\nthat exploits Ethereum as Blockchain. As a real scenario\nfor validation, we considered an infrastructure compliant\nwith the eIDAS Regulation, in which the attribute providers\ndefined by eIDAS are authorized to certify a qualification or\nsome attributes of a digital identity.\nMost of the proposals of the state of the art rely\non sophisticated cryptographic primitives. In contrast,\nour proposal relies only on efficient and straightforward\ncryptographic operations, which are based on elliptic curves\nand considered secure and efficient.\nThe proposed solution has practical implications, primarily related to the General Data Protection Regulation\n\n\n(European Parliament, 2016), which remarks the importance of seven basic principles of data protection: 1) lawfulness, fairness, and transparency, 2) purpose limitation,\n3) data minimization, 4) accuracy, 5) storage limitation, 6)\nsecurity, 7) accountability. Although how to apply these\nprinciples is not stated, they represent the spirit of the\nregulatory framework. Thus, compliance with these principles is fundamental to build any data-processing framework\nin practice. Indeed, this regulation emphasizes the importance of applying these principles to any company, and\nit is not possible to be GDPR-compliant without implementing these rules in the data life cycle of the company.\nThe third principle regards data minimization (Article 5.1.c)\nand requires entities to process only adequate, relevant,\n_and limited personal data that is necessary. This regu-_\nlation does not define what the terms adequate, relevant,\nand limited means but states that data processing should\nonly use as much data as is required to successfully accomplish a given task, and data collected for one purpose cannot\nbe used for a different purpose without obtaining a new consent. This means that companies must limit personal data\ncollection to data that are absolutely necessary for carrying\nout the purpose for which data are processed.\nIn this paper, we described an example in which a\ncompany needs to know if a user is of age but collects\nthe user’s date of birth. This data processing violates the\nminimization principle and, according to Article 83 of\nEuropean Parliament (2016), this infringement is subject\nto administrative fines up to 20M EUR or, in the case of\nan undertaking, up to 4% of the total worldwide annual\nturnover of the preceding financial year.\nThis aspect highlights the practical importance of our\nsolution: the scheme we proposed allows a company\nto comply with the data minimization principle stated\nby GDPR, yet ensuring that access-control policies are\nrespected. Although the use of our solution does not allow a\ncompany to know the identity of users accessing a service,\nin case of valid reasons (for instance, a terrorist who rented\na car), it is possible to uncover user’s identity with the\nsupport of a trusted party, which is the attribute provider\nthat certified user’s age. This is an important added value\nof our proposal, especially in this period in which the\nbalance between privacy right and security right is difficult\nto determine.\n\n**Funding Open access funding provided by Universit`a degli Studi**\nMediterranea di Reggio Calabria within the CRUI-CARE Agreement.\n\n### Declarations\n\n**Conflict of Interests The authors declare that they have no conflict of**\ninterest.\n\n\n-----\n\n**Open Access This article is licensed under a Creative Commons**\nAttribution 4.0 International License, which permits use, sharing,\nadaptation, distribution and reproduction in any medium or format, as\nlong as you give appropriate credit to the original author(s) and the\nsource, provide a link to the Creative Commons licence, and indicate\nif changes were made. The images or other third party material in this\narticle are included in the article’s Creative Commons licence, unless\nindicated otherwise in a credit line to the material. If material is not\nincluded in the article’s Creative Commons licence and your intended\nuse is not permitted by statutory regulation or exceeds the permitted\nuse, you will need to obtain permission directly from the copyright\n[holder. To view a copy of this licence, visit http://creativecommons.](http://creativecommons.org/licenses/by/4.0/)\n[org/licenses/by/4.0/.](http://creativecommons.org/licenses/by/4.0/)\n\n## References\n\nAttrapadung, N., & Imai, H. (2009). Dual-policy attribute based\nencryption. In M. Abdalla, D. Pointcheval, P. A. Fouque, &\nD. Vergnaud (Eds.) Applied cryptography and network security\n(pp. 168-185). Berlin: Springer.\nBertoni, G., Daemen, J., Peeters, M., & Van Assche, G (2009). Keccak\nspecifications. Submission to nist (round 2), pp 320–337.\nBethencourt, J., Sahai, A., & Waters, B (2007). Ciphertextpolicy attribute-based encryption. In _2007_ _IEEE_ _Sympo-_\n_sium_ _on_ _Security_ _and_ _Privacy_ _(SP_ _’07)_ (pp. 321–334).\n\n[https://doi.org/10.1109/SP.2007.11.](https://doi.org/10.1109/SP.2007.11)\n[CEF Digital (2019). eID Profile. https://ec.europa.eu/cefdigital/wiki/](https://ec.europa.eu/cefdigital/wiki/display/CEFDIGITAL/eIDAS+eID+Profile)\n[display/CEFDIGITAL/eIDAS+eID+Profile, Accessed 13 January](https://ec.europa.eu/cefdigital/wiki/display/CEFDIGITAL/eIDAS+eID+Profile)\n2021.\neIDAS eID Technical Subgroup (2019). eIDAS SAML Message Format. [https://ec.europa.eu/cefdigital/wiki/download/attachments/](https://ec.europa.eu/cefdigital/wiki/download/attachments/82773108/eIDAS%20SAML%20Attribute%20Profile%20v1.2%20Final.pdf)\n[82773108/eIDAS%20SAML%20Attribute%20Profile%20v1.2](https://ec.europa.eu/cefdigital/wiki/download/attachments/82773108/eIDAS%20SAML%20Attribute%20Profile%20v1.2%20Final.pdf)\n[%20Final.pdf, Accessed 13 January 2021.](https://ec.europa.eu/cefdigital/wiki/download/attachments/82773108/eIDAS%20SAML%20Attribute%20Profile%20v1.2%20Final.pdf)\n[Ethereum (2020). https://www.ethereum.org, Accessed 13 January](https://www.ethereum.org)\n2021.\n[Ethereum and IPFS APIs (2020). https://infura.io/, Accessed 13](https://infura.io/)\nJanuary 2021.\n[Ethereum dApps (2020). Explore Decentralized Applications. https://](https://www.stateofthedapps.com)\n[www.stateofthedapps.com, Accessed 13 January 2021.](https://www.stateofthedapps.com)\nEuropean Commission (2016). eIDAS Regulation (Regulation\n(EU) N 910/2014). [https://ec.europa.eu/futurium/en/content/](https://ec.europa.eu/futurium/en/content/eidas-regulation-regulation-eu-ndeg9102014)\n[eidas-regulation-regulation-eu-ndeg9102014,](https://ec.europa.eu/futurium/en/content/eidas-regulation-regulation-eu-ndeg9102014) Accessed 13\nJanuary 2021.\nEuropean Parliament (2016). General data protection regulation.\n\n[https://eur-lex.europa.eu/eli/reg/2016/679, Accessed 13 January](https://eur-lex.europa.eu/eli/reg/2016/679)\n2021.\nGoyal, V., Pandey, O., Sahai, A., & Waters, B (2006). Attributebased encryption for fine-grained access control of encrypted data.\nIn Proceedings of the 13th ACM conference on Computer and\n_communications security (pp. 89–98). ACM._\nGuo, R., Shi, H., Zhao, Q., & Zheng, D (2018). Secure attributebased signature scheme with multiple authorities for blockchain in\nelectronic health records systems. IEEE Access, 6, 11676–11686.\nHern´andez-Ramos, J.L., P´erez, S., Hennebert, C., Bernab´e, J.B.,\nDenis, B., Macabies, A., & Skarmeta, A F (2018). Protecting\npersonal data in IoT platform scenarios through encryption-based\nselective disclosure. Computer Communications, 130, 20–37.\nHu, V.C., Ferraiolo, D., Kuhn, R., Friedman, A.R., Lang, A.J., Cogdell,\nM.M., Schnitzer, A., Sandlin, K., Miller, R., Scarfone, K., & et al.\n(2013). Guide to attribute based access control (abac) definition\nand considerations (draft). NIST special publication, 800(162).\nHur, J., & Noh, D.K. (2010). Attribute-based access control\nwith efficient revocation in data outsourcing systems. IEEE\n\n\n_Transactions on Parallel and Distributed Systems, 22(7), 1214–_\n1221.\nKaramitsos, I., Papadaki, M., & Al Barghuthi, NB (2018). Design of\nthe blockchain smart contract: a use case for real estate. Journal of\n_Information Security, 9(3), 177–190._\nKim, J., Baskerville, R.L., & Ding, Y (2018). Breaking the privacy kill\nchain: Protecting individual and group privacy online. Information\n_Systems Frontiers, 1–15._\nLee, J.K., Chang, Y., Kwon, H.Y., & Kim, B (2020). Reconciliation\nof privacy with preventive cybersecurity: the bright internet\napproach. Information Systems Frontiers, 1–13.\nLiang, X., Cao, Z., Lin, H., & Shao, J (2009). Attribute\nbased proxy re-encryption with delegating capabilities. In Pro_ceedings of the 4th international symposium on informa-_\n_tion, computer, and communications security (pp. 276–286 ),_\n[https://doi.org/10.1145/1533057.1533094.](https://doi.org/10.1145/1533057.1533094)\nLor¨unser, T., Slamanig, D., L¨anger, T., & P¨ohls, HC (2016).\nPrismacloud tools: a cryptographic toolbox for increasing security\nin cloud services. In 2016 11Th international conference on\n_availability, reliability and security (ARES) (pp. 733–741). IEEE._\nMaesa, D.DF., Mori, P., & Ricci, L (2017). Blockchain based\naccess control. In IFIP international conference on distributed\n_applications and interoperable systems (pp. 206–220). Springer._\nNakamoto, S. (2008). Bitcoin: A peer-to-peer electronic cash system.\n\n[https://bitcoin.org/bitcoin.pdf.](https://bitcoin.org/bitcoin.pdf)\nOuaddah, A., Abou Elkalam, A., & Ait Ouahman, A (2016).\nFairaccess: a new blockchain-based access control framework for\nthe internet of things. Security and Communication Networks,\n_9(18), 5943–5964._\nPinno, O.J.A., Gregio, A.R.A., & De Bona, L C (2017). Controlchain:\nBlockchain as a central enabler for access control authorizations\nin the IoT. In GLOBECOM 2017–2017 IEEE Global Communica_tions Conference (pp. 1–6). IEEE._\nPriesnitz Filho, W., Ribeiro, C., & Zefferer, T (2019). Privacypreserving attribute aggregation in eid federations. Future\n_Generation Computer Systems, 92, 1–16._\nProvable blockchain (2020). The Provable blockchain oracle for\n[modern DApps. http://provable.xyz/, Accessed 13 January 2021.](http://provable.xyz/)\nRIPEMD160 (2020). [https://en.wikipedia.org/wiki/RIPEMD,](https://en.wikipedia.org/wiki/RIPEMD)\nAccessed 13 January 2021.\nRopsten Testnet Explorer (2020). [https://ropsten.etherscan.io,](https://ropsten.etherscan.io)\nAccessed 13 January 2021.\nSahai, A., & Waters, B. (2005). Fuzzy identity-based encryption. In\n_Annual international conference on the theory and applications of_\n_cryptographic techniques (pp. 457–473). Springer._\n[SAML (2020). https://en.wikipedia.org/wiki/SAML 2.0, Accessed 13](https://en.wikipedia.org/wiki/SAML_2.0)\nJanuary 2021.\nSayeed, S., & Marco-Gisbert, H. (2019). Assessing blockchain\nconsensus and security mechanisms against the 51% attack.\n_Applied Sciences, 9(9), 1788._\n[SHA-2 (2020). https://en.wikipedia.org/wiki/SHA-2, Accessed 13](https://en.wikipedia.org/wiki/SHA-2)\nJanuary 2021.\nShin, D., & Hwang, Y. (2020). The effects of security and traceability\nof blockchain on digital affordance. Online Information Review.\nShin, D., & Ibahrine, M. (2020). The socio-technical assemblages\nof blockchain system: How blockchains are framed and how the\nframing reflects societal contexts. Digital Policy, Regulation and\nGovernance.\nShin, D.D. (2019). Blockchain: The emerging technology of digital\ntrust. Telematics and Informatics, 45, 101278.\nShin, D.D., Fotiadis, A., & Yu, H (2019). Prospectus and limitations\nof algorithmic governance: an ecological evaluation of algorithmic\ntrends. Digital Policy, Regulation and Governance.\nShin, D.H. (2010). The effects of trust, security and privacy in social\nnetworking: a security-based approach to understand the pattern\nof adoption. Interacting with computers, 22(5), 428–438.\n\n\n-----\n\n[Solidity (2020). https://solidity.readthedocs.io/en/v0.5.8, Accessed 13](https://solidity.readthedocs.io/en/v0.5.8)\nJanuary 2021.\nSwan, M. (2015). Blockchain: Blueprint for a new economy. Boston:\nO’Reilly Media Inc.\nWang, G., Liu, Q., & Wu, J (2010). Hierarchical attribute-based\nencryption for fine-grained access control in cloud storage\nservices. In Proceedings of the 17th ACM conference on Computer\n_and communications security (pp. 735–737). ACM._\nWang, S., Zhang, Y., & Zhang, Y (2018). A blockchain-based\nframework for data sharing with fine-grained access control in\ndecentralized storage systems. IEEE Access, 6, 38437–38450.\n[Web3j SDK (2020). https://web3j.io/, Accessed 13 January 2021.](https://web3j.io/)\n\n**Publisher’s Note Springer Nature remains neutral with regard to**\njurisdictional claims in published maps and institutional affiliations.\n\n**Antonia Russo is PhD student in computer science at the University**\nMediterranea of Reggio Calabria. She received her MsC Degree in\nTelecommunication Engineering from the University Mediterranea of\nReggio Calabria in July 2018. Her research interests include security,\nprivacy, and social network analysis.\n\n\n**Gianluca Lax is Associate Professor of Computer Science at the**\nUniversity Mediterranea of Reggio Calabria, Italy. In 2005, he received\nhis PhD in computer science from the University of Calabria. In\n2018, he got the habilitation as a Full Professor of Computer Science.\nHis research interests include information security and social network\nanalysis. He is an author of more than 120 papers published in\nleading international journals and conference proceedings. He serves\nas a referee for many international journals and is in the program\ncommittee of many conferences. He is also included in the editorial\nboard of several international journals and participates in many funded\nprojects.\n\n**Baptiste Dromard is a MsC Degree student at Ensicaen, France. His**\nresearch interest is focused on Blockchain and Bitcoin.\n\n**Menad Mezred is a MsC Degree student at Ensicaen, France. His**\nresearch interest is focused on Blockchain and Ethereum.\n\n\n-----\n\n"