Computer Networks 83(2015)184-198 Contents lists available at ScienceDirect Computer Computer Networks Networks ELSEVIER journal homepage:www.elsevier.com/locate/comnet MobiCache:Cellular traffic offloading leveraging cooperative CrossMark caching in mobile social networks Sheng Zhang3*,Jie Wu b,Zhuzhong Qian3,Sanglu Lu State Key Laboratory for Novel Software Technology.Nanjing University.China PDepartment of Computer and Information Sciences,Temple University.USA ARTICLE INFO ABSTRACT Article history: Offloading cellular traffic through mobile social networks has arisen as a promising way for Received 3 November 2014 relieving cellular networks.Prior studies mainly focused on caching data in a number of Received in revised form 10 February 2015 pre-selected helpers.However,such a strategy would fail when mobile users enter and Accepted 2 March 2015 leave the target area over time.In this paper,we examine the research decisions and design Available online 20 March 2015 tradeoffs that arise when offloading cellular traffic in such a dynamic area of interest, referred to as a MobiArea,and we design an offloading framework,MobiCache,for maximiz- Keywords: Cellular traffic offloading ing cellular operators'revenues and minimizing the overhead imposed on mobile devices. Content floating On the user side,we propose a content floating-based cooperative caching strategy that Cooperative caching caches data in geographical floating circles.instead of selected helpers in previous studies. Mobile social networks to cope with the dynamics.A geographical routing scheme is designed for delivering data and queries towards floating circles.We also develop a cache replacement scheme to improve caching cost-effectiveness inside floating circles.On the operator side.query his- tory and feedback are maintained for cellular operators to optimize framework parameters that maximize their revenues.Extensive trace-driven simulations show that,compared with a state-of-the-art scheme,MobiCache offloads up to 52%more traffic with 15%shorter delay and 6%less forwarding cost. 2015 Elsevier B.V.All rights reserved. 1.Introduction already made several measures for throttling its customers [2].One straightforward solution to tackling this problem The past few years have witnessed the explosive popu- would be deploying more base stations to expand cellular larity of smart-phones and tablets.According to Cisco network capacity [3].which is limited,however,since it Visual Networking Index (VNI)[1,the mobile data traffic requires high financial input but gets low and diminishing generated in 2014 was nearly 30 times the size of the returns.Addressing this problem needs some paradigm-al- entire global Internet in 2000,and global mobile traffic will tering approaches. grow at a compound annual growth rate of 57 percent Due to the inherent proximity-based sharing ability of from 2014 to 2019;in other words,it will increase 10-fold mobile devices and the delay-tolerant nature of many cel- and reach 24.3 exabytes per month by 2019.This huge lular contents,offloading cellular traffic through mobile amount of data traffic,as a consequence,has degraded ser- social networks (MSNs)[4]has arisen recently as a promis- vice quality and created immense pressure on the limited ing method for relieving cellular networks [5-11].When spectrum of cellular networks.For example,AT&T has users are rewarded with proper incentives [9.10],it is very likely that they will be willing to wait for a predetermined Corresponding author.TeL:+8625 83681369. delay to obtain some cellular contents,and to help store- E-mail addresses:sheng@nju.edu.cn (S.Zhang).jiewu@temple.edu carry-forward them.Traffic offloading then can be realized (J.Wu).qzz@nju.edu.cn (Z.Qian).sanglu@nju.edu.cn (S.Lu). http://dx.doi.org/10.1016/j.comnet.2015.03.011 1389-1286/2015 Elsevier B.V.All rights reserved.MobiCache: Cellular traffic offloading leveraging cooperative caching in mobile social networks Sheng Zhang a,⇑ , Jie Wu b , Zhuzhong Qian a , Sanglu Lu a a State Key Laboratory for Novel Software Technology, Nanjing University, China bDepartment of Computer and Information Sciences, Temple University, USA article info Article history: Received 3 November 2014 Received in revised form 10 February 2015 Accepted 2 March 2015 Available online 20 March 2015 Keywords: Cellular traffic offloading Content floating Cooperative caching Mobile social networks abstract Offloading cellular traffic through mobile social networks has arisen as a promising way for relieving cellular networks. Prior studies mainly focused on caching data in a number of pre-selected helpers. However, such a strategy would fail when mobile users enter and leave the target area over time. In this paper, we examine the research decisions and design tradeoffs that arise when offloading cellular traffic in such a dynamic area of interest, referred to as a MobiArea, and we design an offloading framework, MobiCache, for maximiz￾ing cellular operators’ revenues and minimizing the overhead imposed on mobile devices. On the user side, we propose a content floating-based cooperative caching strategy that caches data in geographical floating circles, instead of selected helpers in previous studies, to cope with the dynamics. A geographical routing scheme is designed for delivering data and queries towards floating circles. We also develop a cache replacement scheme to improve caching cost-effectiveness inside floating circles. On the operator side, query his￾tory and feedback are maintained for cellular operators to optimize framework parameters that maximize their revenues. Extensive trace-driven simulations show that, compared with a state-of-the-art scheme, MobiCache offloads up to 52% more traffic with 15% shorter delay and 6% less forwarding cost. 2015 Elsevier B.V. All rights reserved. 1. Introduction The past few years have witnessed the explosive popu￾larity of smart-phones and tablets. According to Cisco Visual Networking Index (VNI) [1], the mobile data traffic generated in 2014 was nearly 30 times the size of the entire global Internet in 2000, and global mobile traffic will grow at a compound annual growth rate of 57 percent from 2014 to 2019; in other words, it will increase 10-fold and reach 24.3 exabytes per month by 2019. This huge amount of data traffic, as a consequence, has degraded ser￾vice quality and created immense pressure on the limited spectrum of cellular networks. For example, AT&T has already made several measures for throttling its customers [2]. One straightforward solution to tackling this problem would be deploying more base stations to expand cellular network capacity [3], which is limited, however, since it requires high financial input but gets low and diminishing returns. Addressing this problem needs some paradigm-al￾tering approaches. Due to the inherent proximity-based sharing ability of mobile devices and the delay-tolerant nature of many cel￾lular contents, offloading cellular traffic through mobile social networks (MSNs) [4] has arisen recently as a promis￾ing method for relieving cellular networks [5–11]. When users are rewarded with proper incentives [9,10], it is very likely that they will be willing to wait for a predetermined delay to obtain some cellular contents, and to help store￾carry-forward them. Traffic offloading then can be realized http://dx.doi.org/10.1016/j.comnet.2015.03.011 1389-1286/ 2015 Elsevier B.V. All rights reserved. ⇑ Corresponding author. Tel.: +86 25 83681369. E-mail addresses: sheng@nju.edu.cn (S. Zhang), jiewu@temple.edu (J. Wu), qzz@nju.edu.cn (Z. Qian), sanglu@nju.edu.cn (S. Lu). Computer Networks 83 (2015) 184–198 Contents lists available at ScienceDirect Computer Networks journal homepage: www.elsevier.com/locate/comnet