PERFORMANCE ANALYSIS OF MOBILE DATA OFFLOADING IN HETEROGENEOUS NETWORKS
ABSTRACT
An unprecedented increase in the mobile data traffic volume has been recently reported due to the extensive use ofsmartphones, tablets and laptops. This is a major concern for mobile network operators, who are forced to often operate veryclose to their capacity limits. Recently, different solutions have been proposed to overcome this problem. The deployment ofadditional infrastructure, the use of more advanced technologies (LTE), or offloading some traffic through Femtocells and WiFiare some of the solutions. Out of these, WiFi presents some key advantages such as its already widespread deployment and lowcost. While benefits to operators have already been documented, it is less clear how much and under what conditions the usergains as well. Additionally, the increasingly heterogeneous deployment of cellular networks (partial 4G coverage, small cells,etc.) further complicates the picture regarding both operator- and user-related performance of data offloading. To this end, in thispaper we propose a queueing analytic model that can be used to understand the performance improvements achievable by WiFibaseddata offloading, as a function of WiFi availability and performance, user mobility and traffic load, and the coverage ratioand respective rates of different cellular technologies available. We validate our theory against simulations for realistic scenariosand parameters, and provide some initial insights as to the offloading gains expected in practice.
EXISTING SYSTEM:
Authors in propose to exploit opportunistic communicationsfor information spreading in social networks.Their study is based on determining the minimum numberof users that are able to reduce maximally the amounttransmitted through the cellular network. A theoreticalanalysis with some optimization problems of the offloadingfor opportunistic and vehicular communication are givenin. The LTE offloading into WiFi directis subject of study in. The work in is mainlyconcerned with studying the conditions under which ratecoverage is maximized, for random deployment of APsbelonging to different networks. Contrary to most of theother works, authors in consider the situation in whichcellular operators pay for using the AP from third parties.They use game theory to consider different issues, such asthe amount of data and money a cellular operator shouldpay for utilizing the APs. In, a solution for mobile dataoffloading between 3GPP and non-3GPP access networks ispresented. A WiFi based mobile data offloading architecturethat targets the energy efficiency for smartphones waspresented in. An interesting work on determining thenumber of WiFi AP that need to be deployed in order toachieve a QoS is presented in.As more related to mobile data offloading are the paperswith measurements, [9]. Authors in have trackedthe behavior of pedestrian users and their measurementssuggest heavy-tailed periods of WiFi availability. The sameholds for the time when there is no WiFi connectivityin the proximity of the mobile user. Similar conclusionsfor the availability periods are given in, where authorsconduct measurements for vehicular users. These users areon metropolitan area buses.
PROPOSED SYSTEM:
To this end, in this paper we propose a queueing analyticmodel for performance analysis of on-the-spot mobiledata offloading. Our contributions can be summarized asfollows:• We consider a simple scenario where the user canchoose between WiFi and a single cellular technology,and derive general formulas, as well as simplerapproximations, for the expected delay and offloadingefficiency (Section 2).• We generalize our analysis to the case where multiplecellular technologies (and respective rates) are availableto a user, e.g., depending on her location, and/ordifferent rates are offered by the same technology (e.g.,rate adaptation, indoor/outdoor, etc.) (Section 3).• We validate our model in scenarios where most parametersof interest are taken from real measured data, andwhich might diverge from our assumptions (Section 4).• We use our model to provide some preliminary answersto the questions of offloading efficiency anddelay improvements through WiFi-based offloading
CONCLUSION
In this paper, we have proposed a queueing analytic modelfor the performance of on-the-spot mobile data offloadingfor generic number of access technologies, and we validatedit against realistic WiFi network availability statistics.We have provided approximations for different utilizationregions and have validated their accuracy comparedto simulations and the exact theoretical results. We alsoshowed that our model can be applied to a broader class ofdistributions for the durations of the periods between andwith WiFi availability. Our model can provide insight on theoffloading gains by using on-the-spot mobile data offloadingin terms of both the offloading efficiency and delay. Wehave shown that the availability ratio of WiFi connectivity,in conjunction with the arrival rate plays a crucial role forthe performance of offloading, as experienced by the user.
REFERENCES
[1] “Cisco visual networking index: Global mobile data traffic forecastupdate, 2014-2019,” Feb. 2015, http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/whitepaper c11-520862.pdf.
[2] “Mobile data offloading through WiFi,” 2010, Proxim Wireless.
[3] T. Kaneshige, “iPhone users irate at idea of usage-basedpricing,” Dec. 2009, http://www.pcworld.com/article/184589/ATTIPhoneUsersIrateatIdeaofUsageBasedPricing.html.
[4] “Growing data demands are trouble for Verizon, LTE capacitynearing limits,” http://www.talkandroid.com/97125-growing-datademands-are-trouble-for-verizon-lte-capacity-nearing-limits/,2012.
[5] http://www.3gpp1.eu/ftp/Specs/archive/23series/23.829/.
[6] M. Qutqut, F.M.Al-Turjman, and H. Hassanein, “MWM: Mobilefemtocells utilizing WiFi (a data offloading framework for cellularnetworks using mobile femtocells),” in Proc. of IEEE ICC, 2013.
[7] C. B. Sankaran, “Data offloading techniques in 3GPP rel-10 networks:A tutorial,” IEEE Communications Magazine, June 2012.
[8] K. Lee, I. Rhee, J. Lee, S. Chong, and Y. Yi, “Mobile data offloading:How much can WiFi deliver,” in Proc. of ACM CoNEXT, 2010.
[9] A. Balasubramanian, R. Mahajan, and A. Venkataramani, “Augmentingmobile 3G using WiFi,” in Proc. of ACM MobiSys, 2010.
[10] J. Eriksson, H. Balakrishnan, and S. Madden, “Cabernet: VehicularContent Delivery Using WiFi,” in Proc. ACM MOBICOM, 2008.
[11] S. M. Ross, Stochastic Processes, 2nd ed. John Wiley & So