Research On Seed Node Selection Algorithm Under Opportunity Diversion

With the emergence of affordable smart mobile devices(such as smart phones,tablets)running innovative applications have severely overloaded the cellular network.Increasing data traffic consumption makes cellular networks overwhelmed,causing serious traffic load and network congestion problems.In order to solve this problem effectively,data offloading emerges as a hot research direction.Mobile data offloading,also called mobile cellular data offloading,is designed to transfer the data that is originally planned to transmit through cellular networks to other complement networks,which is targeted to alleviate the overburdened cellular network traffic.Opportunity data offloading,as a kind of data offloading,essentially uses the contact opportunities created by mobile nodes to transmit data to assist data offloading.The cellular network disseminates the data directly to the seed nodes,and then sends the data to other nodes requesting data as the seed nodes moves.The quality of the selected seed node determines the amount of data to be offloading,therefore seed node selection algorithm has become the main research goal of mobile data offloading.This thesis mainly studies the seed nodes selection algorithm from two aspects.The main research contents are as follows:(1)A Selfishness and Centrality based mobile Data Offloading algorithm(SCDO)in Opportunistic Network.Firstly,the cellular network disseminates the data directly to the seed nodes,and then sends the data to other nodes requesting data as the seed nodes move.If a node does not receive data when the delay tolerance expires,it downloads the data directly from the cellular network.(2)The traffic offload problem is formulated as a link prediction problem in opportunistic D2 D network.Three link prediction algorithms(CN,Katz,LRW index)are applied to reconstruct a more realistic network through network reconfiguration,so as to select better seed nodes,reduce the weight of the cellular network,and reduce the time delay.Simulation results prove the feasibility and reasonability of the work in the thesis.The proposed two algorithms achieve good performance in data offloading ratio and time delay,thus significantly alleviating the overburdened cellular network traffic and reduce the delay time.