Peer-assisted Data Offloading and Distributed Channel Selection for Mobile Cloud Computing

Abstract

The rapid growth of intelligent applications makes it increasingly desirable to offload mobile terminals (MTs) traffic to data centers for smart computing. In this paper, we propose a distributed channel selection scheme with cooperative offloading, in which MTs help each other in each channel to offload the data for the purpose of saving energy. Particularly, MTs that are in the same channel and close to each other in a certain geographical area can form into cooperation cluster to perform local computing, then, offload the data to a cloud server in a certain wireless channel. In this way, the energy consumption for data offloading can be reduced. We adopt game theory to formulate the channel selection problem as a non-cooperation game. We then prove that the formulated game is a potential game, and, thus, guarantee the existence of Nash equilibrium. A Markov approximation approach is then applied to design a distributed channel selection algorithm so that each MT can self-organize into stability without information exchange across the whole network. Analytical and numerical results show that the distributed algorithm is efficient in comparison to the centralized optimization solutions.