Offloading computation to a mobile cloud is a promising solution to augment the calculation capabilities of the cell phone devices. In this project, we consider selfish mobile phones in a dense wireless network, in which individual mobile phones can offload calculations through multiple access point or through the base station to a mobile cloud to minimize their calculation costs. We give a game-theoretical investigation of the issue, prove the existence of pure strategy Nash equilibria, and give an effective decentralized algorithm to computing an equilibrium. For the case when the cloud computing resources scale with the number of mobile phones, we demonstrate that all improvement ways are finite. Besides, we give an upper bound on the price of anarchy of the game, which serves in as an upper bound on the approximation ratio of the proposed decentralized algorithms. We utilize simulations to evaluate the time complexity nature of computing Nash equilibria and to provide bits of knowledge into the price of the anarchy of the game under practical situations. Our results demonstrate that the equilibrium cost might be near ideal, and the convergence time is relatively linear in the number of mobile phone devices.