Content delivery enhancement in Vehicular Social Network with better routing and caching mechanism

https://doi.org/10.1016/j.jnca.2020.102952Get rights and content

Abstract

Vehicular Social Network (VSN) emerges as one meaningful scenario of Mobile Social Network (MSN) and aims at addressing the traffic congestion and security problems in smart cities. However, the current IP based network connection cannot efficiently solve the high mobility and dynamically changing topology issues caused by VSN, which results in long routing latency and low cache hit ratio. In this regard, this work intends to optimize the routing latency and the cache hit ratio by solving the content delivery problem in VSN, during which the concepts of fog/edge computing and Information-Centric Networking (ICN) are introduced, since they can provide a new perspective of network connection in VSN by concentrating on contents instead of IP addresses. Firstly, the ICN routing table structure is expanded for the purpose of providing key routing basis for both interest and data packets. Secondly, the vehicle social information is fully explored from the points of relationship strength and interest degree to offer reliable data foundation for community detection and packet forwarding. Thirdly, the dynamic vehicle communities are constructed based on an improved community overlap propagation algorithm. By monitoring the content popularity and cache distribution in the dynamic communities, this paper proposes the vehicle-adaptive caching scheme which guarantees the cache diversity and improves the packet delivery rate. After that, a vehicle-adaptive interest packet routing scheme is proposed based on relationship strength and interest degree for both inter/intra communities. The experiment results indicate that the proposed mechanism can improve the network performance in terms of the packet delivery rate, the average hop, the average delay and the overhead.

Introduction

The explosive development of information and communication technologies promotes the fact that more and more people having the same interests or backgrounds prefer using the Online Social Network (OSN) for communication and information sharing. Currently, the social networking services are mostly visited by mobile devices, which results in the booming of Mobile Social Network (MSN) (Shi et al., 2018). MSN has been defined as a kind of delay tolerant network which is composed of many mobile nodes with social attributes. Different from OSN, MSN focuses more on the user mobility and relies on exploring the social relationship of these mobile users to judge whether they will meet or not. Once two mobile users meet each other, the corresponding devices are in the communication range that they can fulfill the message delivery by the opportunistic communication technology and hop-by-hop forwarding method (Ko et al., 2017).

Unlike MSN that focuses on the interactions among human, the Vehicular Ad-hoc Network (VANET) (Bhatia et al., 2018) studies the interactions among vehicles. Despite this, they have the same social structure with the mobility attribute, such that their integration contributes to a new mobile network communication mode, that is, Vehicular Social Network (VSN), which supports the interconnection among vehicles, human and base stations (Moriyama et al., 2019). In particular, VANET provides the underlying network infrastructure for VSN, while MSN enables the social attributes of VSN. Hence, VSN includes two parts. On one hand, it supports both the traditional Vehicle-to-Vehicle (V2V) and the Vehicle-to-Infrastructure (V2I) communication modes. On the other hand, it enables the human attributes such as mobility, selfishness and interest preferences (Hail, 2019). The basis framework of VSN is shown in Fig. 1, where a virtual network graph is abstracted from the underlying VSN.

VSN is regarded as one specific and promising scenario of MSN, which can actually improve the experience of both vehicle drivers and passengers. For example, a lot of vehicles are now equipped with multimedia devices which offers entertainment to passengers. These multimedia contents can be shared by passengers in different vehicles. Besides, the number of vehicles and the corresponding vehicular services both grow rapidly nowadays. These trends all lead to the close relationship between human and vehicles. However,the VANET mainly focuses on studying the relationship among vehicles, while VSN intends to connect human, machine and vehicles. Apart from this, the high mobility characteristic of vehicle causes the corresponding topology to change dynamically, which may lead to high delay and low content delivery rate using the IP-based communication model. On the contrary, VSN offers a new communication model which can well accommodate these features. Moreover, it is known that the vehicles are typically driven by human, which means that the mobility of vehicle is also affected by the human social activity. For instance, the drivers having same hobbies may encounter with each other in some specific places related to their interests. Hence, we can conclude that the vehicles also have social attributes. In VSN, the social relationship among different vehicles (or vehicle drivers) can be better leveraged to improve the content delivery rate, the response speed to content requests, and etc. Due to the above motivations, VSN has already attracted a lot of attentions from both industry and academic.

Despite the benefits of VSN, it still faces many challenges, among which one most promising research topic is the content delivery. However, efficient content delivery becomes extremely difficult in VSN. On one hand, the vehicles, passengers and drivers can be regarded as different types of nodes in VSN, which results in massive content requests that may not be efficiently satisfied due to the limitation of storage and wireless spectrum resources. In particular, these different types of nodes are usually heterogeneous, which makes the content delivery even more challenging. To address this awkward situation, determining where to cache the popular content and how much should be cached become vital important. On the other hand, most of the nodes in VSN have the mobility characteristics, which means that the VSN topology would change frequently (Vegni and Loscr, 2015). To solve this kind of situation, a corresponding routing method should be explored to help adapting the dynamically changing topology and finding the demanded content quickly. Now, jointly taking the above two aspects into consideration, we need to address both the caching and routing issues for the purpose of achieving an efficient content delivery in VSN.

Most of the related research work (e.g., (Fang et al., 2018a; Luo and An, 2019; Liao and Zhang, 2018)) rely on using a centralized structure to solve the content caching and routing issues in VSN. Despite the centralized control structure may offer optimum caching and routing strategies in a global perspective, it also makes the central controller becoming a bottleneck, because massive amount of new content requests will be sent to the controller for caching and routing processing. To alleviate or even avoid such situation, many other researches (e.g., (Fang et al., 2015; Hajimirsadeghi et al., 2017; Tran et al., 2019)) begin to introduce the concepts of fog computing and edge computing into their work. Specifically, they distribute a certain part of the decision tasks and contents from the central cloud to edge servers, thus to ease the burden of the centralized controller. Despite this, these work still rely on using the traditional IP network for communication, which cannot actually address the high mobility and dynamically changing topology issues efficiently. In this regard, this paper proposes to address the content delivery problem in VSN by integrating the technologies of fog computing and Information-Centric Networking (ICN). Specifically, fog computing supports to sink the computing and storage capacity from cloud to edge, which can help addressing the caching issue of content delivery in VSN. ICN supports content mobility naturally and focuses on contents instead of IP addresses, which can help addressing the routing issue of content delivery in VSN. In this case, we summarize the main contributions of this work as follows:

  • A fog computing and ICN based VSN framework is built to support efficient content delivery, during which the caching and routing issues are addressed. Briefly speaking, the technology of fog computing brings contents to the place near users, while ICN enables a novel network connection which supports mobility naturally.

  • An improved community overlap propagation algorithm is designed for community detection and construction based on users' common interests or preferences. The nodes within the same community have similar social interests or preferences, and they would meet each other frequently. Thus, the content sharing and exchange among them can improve the content hit rate.

  • The content delivery involves both the caching and routing. Hence, we propose an in-network caching mechanism and a mobility supported routing mechanism respectively based on the community detection and construction. The caching mechanism includes content popularity statistics, caching policy and cache replacement strategy, while the routing mechanism includes interest packet routing and data packet routing.

The rest of this work is organized as follows. Section 2 summarizes the related work and Section 3 presents the system mode and framework. Section 4 explains the proposed mechanisms in detail. The experiment results are shown in Section 5. Section 6 concludes this work.

Section snippets

Related work

VSN is still in the infancy. Despite this, many widely known technologies (e.g., fog computing and edge computing) can actually enable and promote the development of VSN. In this way, a lot of academic researchers begin to explore the relationship between VSN and these technologies, and try to make an integration among them for the purpose of improving the VSN performance. Since the main target of this work is to address the content delivery problem in VSN based on ICN and fog computing, the

System framework

VSN is the integration of social network and vehicle network. Hence, the system framework of this work should take the characteristics of both social network and vehicle network into consideration. Firstly, we begin designing the system framework (as shown in Fig. 2) by considering the basic entities in vehicle network, which include the vehicle nodes, Road Side Unit (RSU) and Mobile Edge Server (MES). Vehicle nodes include the public and private transportation such as car and bus. RSU plays as

VSN caching and routing mechanisms

VSN is a kind of social-oriented network. Hence, how to accurately describe the social relationship among vehicles becomes very important for VSN caching and routing. In this section, the social measurements are first defined and then the caching and routing mechanisms for VSN are proposed.

Environment

The proposed mechanisms are implemented using the Java language and tested on the Opportunistic Network Environment (ONE) (Opportunistic network environment) platform. As shown in Fig. 7(a), the simulation structure is mainly composed of the movement model, the event generator, the simulation engine, the virtualization and the content delivery. In particular, the movement model simulates the moving tracks of vehicle nodes by using many track protocols such as the map-based movement, the random

Conclusion

This work intends to address the content delivery problem in VSN via integrating the social network and the vehicle network. In particular, the content delivery problem in VSN includes two parts which are the content caching and routing. To address these issues, we first propose a system framework, in which four tables are designed to improve the forwarding efficiency and cache hit ratio. After that, we propose different mechanisms to solve the caching and routing issues respectively.

Credit author statement

Bo Yi: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation. Xingwei Wang: Supervision, Writing – review & editing, Funding acquisition. Min Huang: Resources, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This work is supported by the National Key Research and Development Program of China under Grant No. 2019YFB1802800; the National Natural Science Foundation of China under Grant No. 61872073 and 62002055; the Major International (Regional)Joint Research Project of NSFC under Grant No. 71620107003; LiaoNing Revitalization Talents Program under Grant No. XLYC1902010; the Postdoctoral Science Foundation of China under Grant No. 2020M680972; the Fundamental Research Funds for the Central

Bo Yi received the B.S. and M.S. degrees in computer science from South-Central University for Nationalities, Wuhan, China in 2012 and 2015 respectively, and Ph.D degree in computer science from Northeastern University, Shenyang, China in 2019. He is currently a lecture at the College of Computer Science and Engineering, Northeastern University, Shenyang, China. His research interests include routing and service function chain in SDN, NFV, deterministic networking, cloud computing, etc.

References (24)

  • C. Fang et al.

    An energy efficient distributed in network caching scheme for green content centric networks

    Int. J. Comput. Telecommun. Netw.

    (2015)
  • X. Liu et al.

    Information centric mobile ad hoc networks and content routing: a survey

    Ad Hoc Netw.

    (2017)
  • S. Arshad et al.

    Recent advances in information-centric networking-based Internet of things (ICN-IoT)

    IEEE Internet Things J.

    (2019)
  • A. Bhatia et al.

    Realization of flexible and scalable VANETs through SDN and virtualization

  • S. Choi et al.

    A networking framework for multiple-heterogeneous unmanned vehicles in FANETs

  • F.D. Cunha et al.

    Socially inspired data dissemination for vehicular ad hoc networks. Proceedings of the 17th ACM international conference on Modeling, analysis and simulation of wireless and m obile systems

    ACM

    (2014)
  • C. Fang et al.

    Edge cache based intelligent content delivery in information centric wireless networks

  • C. Fang et al.

    A survey of mobile information centric networking: research issues and challenges

    IEEE Commun. Surv. Tutorials

    (2018)
  • R.A. Ghalehtaki et al.

    A bee colony-based algorithm for micro-cache placement close to end users in fog-based content delivery networks

  • M.A. Hail

    IoT-NDN: An IoT architecture via named data netwoking (NDN)

  • M. Hajimirsadeghi et al.

    Joint caching and pricing strategies for popular content in information centric networks

    IEEE J. Sel. Area. Commun.

    (2017)
  • H. Ko et al.

    An opportunistic push scheme for online social networking services in heterogeneous wireless networks

  • Cited by (0)

    Bo Yi received the B.S. and M.S. degrees in computer science from South-Central University for Nationalities, Wuhan, China in 2012 and 2015 respectively, and Ph.D degree in computer science from Northeastern University, Shenyang, China in 2019. He is currently a lecture at the College of Computer Science and Engineering, Northeastern University, Shenyang, China. His research interests include routing and service function chain in SDN, NFV, deterministic networking, cloud computing, etc.

    Xingwei Wang received the B.S., M.S., and Ph.D. degrees in computer science from the Northeastern University, Shenyang, China in 1989, 1992, and 1998 respectively. He is currently a Professor at the College of Computer Science and Engineering, Northeastern University, Shenyang, China. His research interests include cloud computing and future Internet, etc. He has published more than 100 journal articles, books and book chapters, and refereed conference papers. He has received several best paper awards.

    Min Huang received the B.S. degree in automatic instrument, the M.S. degree in systems engineering, and Ph.D. degree in control theory from the Northeastern University, Shenyang, China in 1990, 1993, and 1999 respectively. She is currently a Professor at the College of Information Science and Engineering, Northeastern University, Shenyang, China. Her research interests include modeling and optimization for logistics and supply chain system, etc. She has published more than 100 journal articles, books, and refereed conference papers.

    View full text