Elsevier

Computer Networks

Volume 107, Part 2, 9 October 2016, Pages 220-232
Computer Networks

A joint active time and flow selection model for cellular content retrieval through ITS

https://doi.org/10.1016/j.comnet.2016.03.015Get rights and content

Abstract

Operators need to address increased data demands to meet subscribers’ growing requirements by offloading a portion of cellular traffic onto other types of networks. In this paper, we investigate the possibility of using vehicular ad hoc networks (VANETs) for this purpose. We study joint data flow selection and contention resolution in a hybrid VANET-cellular system. We formulate the problem as an optimization problem called FOSAA, which considers vehicle-to-vehicle and vehicle-to-infrastructure link quality, channel access, inter-nodal interference and node active time. The problem is solved through an iterative approach. FOSAA is compared to other proposed schemes. The performance results show that offloading fraction is significantly affected by the data volume, vehicle density and number of hops from the infrastructure to the downloader vehicles.

Introduction

Recent studies have confirmed that, in the near future, cellular networks will most likely become overloaded and congested, especially in hot zones and urban areas. Moreover, tablet subscriptions are expected to grow from 250 million in 2012 to approximately 850 million in 2018, exceeding the number of fixed broadband subscriptions [1]. The average smartphone usage increased by 81% in 2012 [2]; in addition, the total number of smartphone subscriptions will be tripled and is expected to increase to 3.3 billion by the end of 2018 [1]. The growing number of such mobile devices results in an increase in data demand and Internet queries, which become the dominating traffic request of wireless cellular network users. By the end of 2018, it is estimated that a smartphone will generate approximately 2GB of data per month, and a mobile PC will generate over 10GB of data [1]; moreover, by 2017, two-thirds of the world’s mobile data traffic will be video data [2]. Data traffic has doubled in only one year, between 2011 and 2012, in contrast to mobile voice traffic, which continues to grow at a steady rate [1]. According to [3], this increase in mobile data traffic demand is expected to continue to increase at a rate of more than 100% annually. Operators need to address this increase in data traffic demand and ensure sufficient RAN (Radio Access Network) capacities to meet this growing demand.

Numerous works have studied systems for offloading cellular traffic via FemtoCells or Wi-Fi hotspots [4]. Accordingly, it will also be interesting to consider and investigate possible offloading through mobile ad hoc networks formed by opportunistic communications between intelligent vehicles, wherein vehicles are wirelessly connected and form a Vehicular Ad hoc Network (VANET). In this context, this work provides analysis and a comparison of analytical models proposed for cellular traffic offloading. We discuss and evaluate the capacity and ability of VANETs to offload a portion of cellular traffic while considering the constraints related to the intermittent vehicular nodes’ connectivity in I2V and V2V links and to the data volume. We propose a cooperative traffic transmission problem formulation in a joint 4G LTE Advanced cellular infrastructure and VANET network where VANET nodes cooperate with the LTE infrastructure by offloading a portion of the cellular traffic.

This solution is very promising and motivated by many reasons. A dedicated frequency band, i.e., 5.86–5.92 GHz, has been allocated for Intelligent Transportation System (ITS) communications. In addition, according to the ETSI 102 638 technical report, by 2027, almost 100% of vehicles will be equipped with On Board Units (OBUs), therein providing V2V (vehicle-to-vehicle) and I2V (infrastructure-to-vehicle)/V2I (vehicle-to-infrastructure) communications, thereby forming ITS [5]. Thus, with the increasing number of vehicles equipped with OBU devices, several related applications are emerging such as safety-related applications (e.g., collision warning, emergency information, assistance for safe driving, and remote vehicle diagnostics), automobile high-speed Internet access, and multimedia content sharing. Moreover, most subscribers currently use their smartphones and tablets during transit. Vehicles are where citizens spend a substantial portion of their time during a day, with more time being spent only in homes and offices. According to [6], commuters spend 500 million hours per week in a car. Thus, opportunistic contacts between vehicles offer high capacity for data transmission.

In summary, traffic offloading using VANETs is a promising solution for partially supporting the exponential growth of mobile data traffic, which otherwise could not be supported even by 4G cellular networks [2] for two main reasons. First, the dedicated frequency band allocated for VANET communications represents an attractive capacity potential compared to the freely shared band used for Wi-Fi. Second, subscribers use their mobile devices during most of the time that they are using modes of transportation. According to the Alcatel-Lucent’s study performed in 2009 [6], 22 % of interviewed consumers would be willing to pay from 30$ to 65$ per month for services proposed while traveling on the road. Therefore, using VANETs to offload a portion of cellular traffic seems to be an attractive solution to investigate.

The remainder of this paper is structured as follows. In Section 2, we present related works. In Section 3, we describe the system model. In Sections 4 and 5, we present the FOSAA model and its resolution procedure. In Section 6, benchmarking schemes are detailed. Finally, a performance evaluation and conclusions are presented in Sections 7 and 8.

Section snippets

Related works

The most popular proposed offloading solutions include femtoCells for indoor offloading and Wi-Fi for outdoor offloading. Numerous analytical models and studies have been conducted in this context [4], [7], [8]. However, very few works have focused on cooperation between cellular infrastructure and opportunistic communications between vehicles whereby the cellular network is offloaded through VANETs. Most of these works study the I2V link for the purpose of improving the performance of VANETs

System modeling

The system model is based on a hybrid network architecture composed of two systems: an LTE cellular infrastructure and a VANET network. The cellular infrastructure is composed of eNodeBs connected via an S1 interface to the LTE Evolved Packet Core (EPC) [25]. The ITS system is composed of fixed Road Side Units (RSUs) deployed over a road (e.g., highway) and vehicular nodes traveling along this road. Only intelligent vehicles that have wireless communication capabilities are considered in this

Problem formulation

We present a multi-constraint optimization problem to evaluate the maximum data content that can be downloaded via the VANET network through either direct I2V downloading or V2V relayed links. The proposed problem formulation is called the Flow Offloading Selection and Active time Assignment problem (FOSAA).

FOSAA resolution procedure

The FOSAA problem is difficult to solve directly because it contains nonlinear constraints and integer variables. Thus, we propose to solve it approximately via an iterative approach. In each iteration, one algorithm, namely, FOSAA-φ, determines flow assignments, and another algorithm, namely, FOSAA-a, determines contention resolution.

The goal of FOSAA-φ-n, where n is the iteration index, is to determine the φi, f that maximizes the objective function (1). Initially, bi is fixed and set to bi=1

Greedy scheme

With the Greedy offloading and active time assignment scheme, both the node active time computation and the selection of offloaded traffic flows are performed step by step in an iterative process until reaching network capacity saturation.

Let us define Hk to be the set of nodes that are at k hops from the RSU and Ai to be the set of active nodes that are exposed to vehicle i. For traffic flow selection, the Greedy approach is applied by first checking the data flows of first-hop nodes ( ∈ H1).

Performance evaluation

In this section, we evaluate the effectiveness of the FOSAA scheme to determine the capacity of a vehicular network to support additional traffic originating from the cellular infrastructure. We compare the FOSAA scheme to the Greedy and Cont-A schemes. Very few works have studied the potential of using vehicular nodes for offloading LTE base stations. The originality of this paper is that it goes beyond existing works by including several additional major parameters such as channel contention,

Conclusions

In this work, we presented FOSAA, an analytical model that determines the potential of using vehicular nodes for cellular traffic offloading. We studied joint data flow selection and contention resolution in a hybrid VANET and cellular infrastructure system, where the objective is to maximize the set of data flows offloaded through the vehicular nodes. The evaluation results show that FOSAA prioritizes the offloading of 1- and 2-hop flows and that the adequate node time activity assignment

Ghayet El Mouna Zhioua obtained her Ph.D. degree jointly from Telecom ParisTech (Paris-FRANCE) and Sup’Com (the Higher School of Communication of Tunis- TUNISIA) in 2014. She received her Master’s degree in Telecommunication from Sup’Com in 2010. She had publications in refereed International Journals and Conferences, such as IEEE Transactions, IEEE WCNC, IEEE VTC, ACM MSWiM. Her research interests are: LTE network, 4G Heterogeneous Wireless Networks, Vehicular Communications, IEEE 802.11p,

References (35)

  • J. Whitbeck et al.

    Push-and-track: Saving infrastructure bandwidth through opportunistic forwarding

    Pervasive Mob. Comput.

    (2012)
  • Ericssson, Ericsson Mobility Report, http://www.ericsson.com/res/docs/2012/ericsson-mobility-report-november-2012.pdf...
  • Cisco, Cisco visual networking index: Global mobile data traffic forecast update, 2012–2017, http://www.cisco.com (6...
  • K. Samdanis et al.

    Traffic offload enhancements for eUTRAN

    IEEE Commun. Surv. Tutor.

    (2012)
  • K. Lee et al.

    Mobile data offloading: How much can WiFi deliver?

    Proceedings of the Sixth International Conference

    (2010)
  • D. Camara et al.

    Propagation of public safety warning messages: A delay tolerant network approach

    Proceedings of the 2014 IEEE Wireless Communications and Networking Conference (WCNC)

    (2010)
  • G. Araniti et al.

    LTE for vehicular networking: A survey

    IEEE Commun. Mag.

    (2013)
  • S. Wietholter et al.

    Performance evaluation of selection schemes for offloading traffic to ieee 802.11 hotspots

    Proceedings of the IEEE International Conference on Communications (ICC)

    (2012)
  • L. Al-Kanj et al.

    Offloading wireless cellular networks via energy-constrained local ad hoc networks

    Proceedings of the IEEE Global Telecommunications Conference (GLOBECOM)

    (2011)
  • Y. Li et al.

    Coding or not: Optimal mobile data offloading in opportunistic vehicular networks

    IEEE Trans. Intell. Transp. Syst.

    (2014)
  • F. Malandrino et al.

    Offloading cellular networks through its content download

    Proceedings of the Ninth Annual Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON)

    (2012)
  • F. Malandrino et al.

    Content download in vehicular networks in presence of noisy mobility prediction

    IEEE Trans. Mob. Comput.

    (2014)
  • C. Makaya et al.

    Seamless data offload and flow mobility in vehicular communications networks

    Proceedings of the IEEE Wireless Communications and Networking Conference Workshop (WCNCW)

    (2012)
  • A. Balasubramanian et al.

    Enhancing interactive web applications in hybrid networks

    Proceedings of the Fourteenth ACM International Conference on Mobile Computing and Networking (MobiCom)

    (2008)
  • B. Han et al.

    Mobile data offloading through opportunistic communications and social participation

    IEEE Trans. Mob. Comput.

    (2012)
  • S. Yoon et al.

    MoPADS: A mobility profile aided file downloading service in vehicular networks

    IEEE Trans. Veh. Technol.

    (2009)
  • N. Lu et al.

    Capacity and delay analysis for social-proximity urban vehicular

    Proceedings of the 2012 IEEE INFOCOM

    (2012)
  • Cited by (1)

    Ghayet El Mouna Zhioua obtained her Ph.D. degree jointly from Telecom ParisTech (Paris-FRANCE) and Sup’Com (the Higher School of Communication of Tunis- TUNISIA) in 2014. She received her Master’s degree in Telecommunication from Sup’Com in 2010. She had publications in refereed International Journals and Conferences, such as IEEE Transactions, IEEE WCNC, IEEE VTC, ACM MSWiM. Her research interests are: LTE network, 4G Heterogeneous Wireless Networks, Vehicular Communications, IEEE 802.11p, IEEE 802.11s standard and wireless mesh communications.

    Jun Zhang received his B.Eng. (2002) in computer science and technology from Shanghai Jiaotong University, and Ph.D. (2007) in computer science from Hong Kong University of Science and Technology. He was a research fellow in the Hong Kong City University (2007–2009), the Hong Kong Polytechnic University(2009–2011), a postdoctoral fellow in the Hong Kong University of Science and Technology(2011–2013), and currently a postdoctoral researcher in Telecom Paristech. His research interests include wireless multi-hop networks (ad hoc/sensor/mesh networks), wireless LAN and vehicular networks.

    Houda Labiod received a Ph.D. degree in Computer Science from the University of Versailles Saint-Quentin-en-Yvelines (France) in 1998. From 1999 to 2000, she worked as an assistant researcher at Eurecom Institute in Sophia-Antipolis (France) in the Mobile Communications Unit. Since, September 2000, she is an associate professor at Telecom ParisTech in the INFRES (Computer Science and Network) department. Her main research interests include optimization of mobile and wireless networking and mobile ad hoc networks (unicast routing, multicast routing, security, QoS routing).

    Nabil Tabbane is currently an Associate Professor in the Computer Networks Department at Higher School of Communications, Tunis: SUP’COM. He received his Ph.D. degree in SUP’COM and in the University of Versailles Saint-Quentin-en-Yvelines (France) in joint guardianship. His Ph.D. research was on stochastic models for forecasting the quality of service in multi-media ad hoc networks. His research interests are in ad hoc and sensor networks.

    Sami Tabbane is currently a Professor in the Computer Networks Department at Sup’Com. He is a specialist of mobile communication systems. He has graduated from Ecole des Mines de Paris in 1998 and got a Ph.D. of Telecom ParisTech in 1991. He started his carreer at France Télécom in 1992–1994 and was recruted by Sup’Com in 1994. He has achieved many missions for ITU in the domain of mobile networks planning, management and training as well as spectrum management in regulatory agencies. He is the co-author of “Réseaux GSM” (Hermes, 1995) and “Ingénierie des Services de Télécommunications” (Hermes-Lavoisier, 2005). He is the author of “Réseaux Mobiles” (Hermes, 1997), “Handbook of Mobile Networks” (Artech House, 2000), “Ingénierie des Réseaux Cellulaires” (Hermes-Lavoisier, 2002).

    View full text