Elsevier

Computer Networks

Volume 111, 24 December 2016, Pages 129-140
Computer Networks

SNVC: Social networks for vehicular certification

https://doi.org/10.1016/j.comnet.2016.08.030Get rights and content

Highlights

  • The proposal uses relationship and prior knowledge of users to establish a trust degree in social networks.

  • SNVC mechanism uses certificates in social networks to define the reliability of a message.

  • The attribution of reputation can promote the recognition of users who favored their social networks.

  • The results encourage the increase of friends as a way to improve the exchange of reliable messages.

  • SNVC represents greater security for nodes regarding the trust and authentication on the network.

Abstract

Vehicular Disruption Tolerant Networks appear due to the search of information by drivers, which are organized into mobile ad hoc networks that may suffer from long interruptions. In such networks, both humans as well as machines contribute with information about the traffic and road conditions. As humans can be biased or may be compelled to disseminate false information for personal gains, the network should incorporate mechanisms to assert the trust and reputation of the information sources. This article proposes an inter-vehicle certification mechanism based on social networks to enable reliable exchange of messages called SNVC - Social Networks for Vehicular Certification. It enables mobile devices to share keys through direct contacts between two acquaintances that warrant their identity, so they mutually sign their certificates, establishing trust in the cyber-physical system. Certificates signed by the friend of a user can be validated if the public key of this friend is validated by a friend in common in the social network. Further, a reputation system rates users or devices based on the reliability of the information that they produce. The evaluation of the mechanism shows the behavior of vehicular networks using real mobility traces. The results show the feasibility of the trust in friends as a way to improve the exchange of reliable messages in a Mobile Opportunistic Network. Also, the attribution of reputation for reliable information promotes the recognition of trustworthy users.

Introduction

Vehicular networks are formed by mobile devices and fixed equipment situated along the margins of streets and highways. Vehicular Disruption Tolerant Networks (vDTN) [1] aim to support new applications, which can increase the safety in the roads by means of communication among users of the transportation system. In urban areas or on roads, the network infrastructure is usually not available or does not cover contiguous areas. Vehicular networks can take advantage of the existent cities’ infrastructure and the availability of the users’ mobile devices. Those devices can be connected to the mobile Internet, whose coverage is expanding in urban areas, to provide information about traffic.

In vehicular networks [2], communication occurs between vehicles (V2V) and between vehicles and access points (V2I) as shown in Fig. 1. Networks formed by these elements present various limitations concerning security and availability, since certain requirements are not always satisfied. End-to-end communication is rarely available among nodes and centralized authorities. Humans provide information about the traffic and road conditions, although they can be biased or disseminate false information. Information exchanged among devices can lead to redefinition of vehicular routes. In this context, a user, for instance a gas station, may attempt to influence the routes in order to increase the flow of vehicles near it. Thus, it is necessary to guarantee the trust and integrity of the messages, even without an instantaneous connection among the nodes exchanging messages.

In the communication among vehicles, end-to-end connectivity is highly susceptible to interruptions. The concepts and technology of DTN (Disruption Tolerant Networks) [3] are the most adequate to networks with intermittent connectivity or long delays. Communication in DTN is asynchronous and does not require end-to-end paths. Nodes follow the store-carry-forward communication paradigm, in which messages are stored temporarily in a buffer in case of lack of a direct connection from source to destination. In this work, the type of vehicular network considered is an Opportunistic DTN, in which encounters or contacts occur unexpectedly, when after it communication can happen; this is different from the Probabilistic or Scheduled types of DTN, in which information about encounters is known as a priori [3].

There are many open issues in the security domain concerning the interaction among users to exploit and generate information in a cyber-physical system [4]. Security is more challenging in vDTN due to the specific characteristics of this type of network, such as unpredictable mobility and variable latency [3]. Because of the sporadic connectivity and possibility of high delays in message transmission, it is necessary to eliminate expired messages and to avoid information leaks. These characteristics have to be considered in vDTNs, once they aim to encourage the participation of all users in a social network, without the need of Internet connections. Social networks are graphs modeling relationships among people, in general called friends.

The problem addressed in this work is the key management in vDTN [1], which enables authentication and reliable message exchange among drivers. At any instant, reliable end-to-end communication is unlikely to be available between DTN nodes and a key management center [4]. Key management solutions or mechanisms that tolerate disconnections are still an open matter [6]. To date, no key management scheme is deemed suitable for DTN [5], [6]. So, knowledge and strategies for message authentication are demanded. Our most important contributions here presented are (a) a certification mechanism based on the interaction of users in a social network; and (b) a trust model, based on trust degrees derived from the user's social network, which is applied to a Mobile Opportunistic Network.

We propose a social network-based mechanism for certification in vDTN (SNVC - Social Networks for Vehicular Certification, through which cryptographic material is exchanged in daily relationships, like meetings with friends or helping another vehicle. SNVC establishes trust degrees among users in the social network. Furthermore, a reputation mechanism allows benefited users to assign a reputation for the user who helped them. The reputation mechanism can identify users that collaborate in the generation of reliable information at a cyber-physical Mobile Opportunistic Network, hence increasing the amount of information sources deemed reliable.

The remainder of this work is organized as it follows. Section 2 presents the related work. Section 3 describes the network model adopted along this work. Section 4 describes the proposed SNVC mechanism and its relation to social networks. Section 5 presents the evaluation and Section 6 concludes the work.

Section snippets

Related work

Vehicles receiving information from other vehicles or network entities should be aware of the trustworthiness of the originator of information. Providing security is a crucial challenge in vehicular networks, where trust plays a key role [2], thanks to the large number of independent nodes involved and the presence of human factors, which can increase the occurrence of bad behavior. Also, trustworthiness is related to privacy and anonymity, which is still an open issue in those networks [7].

Network model

Heterogeneous vehicular networks rely on several technologies for communication [2]. One such technology is the IEEE 802.11p protocol, an approved amendment to the IEEE 802.11 standard, also known as WAVE (Wireless Access for Vehicular Environments). WAVE is required to support the Intelligent Transportation Systems (ITS) applications in the short-range communications. The speed of participating vehicles is limited to exchange data also in opposite ways, using the IEEE 802.11p protocol as

SNVC mechanism

This article proposes certification by means of a social network (SNVC) that uses reputation and cryptographic material received in daily relationships, such as friends meeting or helping other users. A user is a node of the social network that can have friendship relations with others. A friend is a reliable person, previously known, who can sign the certificate of others. A friend of a friend is a user with an indirect relationship with other users by means of a friend in common. A benefited

Evaluation

In order to validate the proposed mechanism, simulations were realized to verify the efficacy of the certification using social networks. The purpose of the simulations is to show the behavior of the vehicular network as function of the number of friends of each user and the average number of reliable users.

SNVC was evaluated using three traces of node mobility in vDTN networks: DieselNet [21], Chicago [22] and Seattle [23]. To perform this analysis, initially the traces were converted to the

Conclusions

This article has proposed social networks as a form of certification between vehicles, which is the basis of the novel mechanism SNVC (Social Networks for Vehicular Certification). Social network incorporates mechanisms to assess the trust and reputation of the information sources, allowing the cyber-physical system to exchange messages in a reliable way. SNVC uses a social network to establish trust degrees among users, considering relationships and valuations of the user's past behavior.

Acknowledgments

The authors would like to thank CNPq, CAPES and Fapemig for their financial support.

Thiago Rodrigues de Oliveira is an Assistant Professor at Universidade Federal de São João Del Rei (UFSJ) in the area of Applied Computing to Engineering at Campus Alto Paraopeba. PhD in progress at Computer Science from the Federal University of Minas Gerais, completed a Masters in Computer Science at the institution and obtained his Bachelor's degree in Computer Science also from UFMG.

References (27)

  • ZhouJ. et al.

    Autonomic group key management in deep space DTN

    Wireless Pers. Commun.

    (2014)
  • S.F. Symington, S. Farrell, H. Weiss, P. Lovell, Bundle security protocol specification Available at...
  • M.N.M. Bhutta et al.

    An efficient, scalable key transport scheme (ESKTS) for delay/disruption tolerant networks

    Wireless Networks

    (2014)
  • Cited by (9)

    View all citing articles on Scopus

    Thiago Rodrigues de Oliveira is an Assistant Professor at Universidade Federal de São João Del Rei (UFSJ) in the area of Applied Computing to Engineering at Campus Alto Paraopeba. PhD in progress at Computer Science from the Federal University of Minas Gerais, completed a Masters in Computer Science at the institution and obtained his Bachelor's degree in Computer Science also from UFMG.

    Cristiano Maciel Silva obtained his BS in Computer Science (2000), MS in Computer Science (2004), MBA (2008), and PhD in Computer Science (2014). He published over 30 papers in top venues along the last three years. He is an Assistant Professor and the Associate Head of the Department of Technology at Universidade Federal de São João Del-Rei, Brazil, and he also serves the Committees of several IEEE conferences. His research interests are focused on the development of technologies and applications for Intelligent Transportation Systems.

    Daniel Fernandes Macedo is an Assistant Professor in the Computing Department (DCC) in Federal University of Minas Gerais (UFMG), Brazil. He has productivity in research scholarship (Bolsa PQ – CNPq) level 2. He was a post-doc researcher in UFMG, Brazil. He holds a PhD in computer science from Université Pierre et Marie Curie-ParisVI (2009). He also holds a M.Sc. and a B.Sc. in Computer Science from Federal University of Minas Gerais (2006). His research interests are autonomic computing, wireless networks and network management. He has worked on national projects, as well as international collaboration projects. Daniel frequently serves as TPC member on Brazilian and international conferences such as WGRS, WTF and SBCUP, as well as international conferences such as IEEE WCNC, IEEE ISCC and IFIP Wireless Days. He served as track co-chair in IFIP Wireless Days 2013 and CCNC 2016.

    José Marcos Silva Nogueira is a Titular Professor in the Computing Department (DCC) in Federal University of Minas Gerais (UFMG), Brazil. Graduate in Electrical Engineering from the Universidade Federal de Minas Gerais (1975), Master's in Computer Science from the Universidade Federal de Minas Gerais (1979) and Ph.D. in Electrical Engineering from the Universidade Estadual de Campinas (1985). Has experience in Computer Science, acting on the following subjects: computer networks, wireless sensor networks, network management, dtn. He has productivity in research scholarship CNPq - level 1C - Computer Science.

    View full text