共享通信信道下多用户竞争的能效计算卸载机制

Energy Efficient Computation Offloading Algorithm for Competing Users on a Shared Communication Channel

移动云计算环境中,移动用户需要决定是否将应用卸载至云端处理,即计算卸载决策问题。在共享信道下,多竞争用户进行应用卸载时,必然会导致应用执行的延时。为了解决截止时间约束下执行能耗的优化问题,提出了一种非合作博弈算法。博弈过程中,每个用户以满足应用执行期限的同时最小化自身能耗为目标,将多竞争用户共享通信信道建立为非合作博弈过程,使单个用户可以独立作出计算卸载决策。证明了在同质和半同质环境下,算法总能产生Nash均衡解。并设计了一种基于高斯赛德尔模型的方法求解了Nash均衡解。仿真实验结果证明,算法可以在一定次数迭代后收敛于Nash均衡点处,在满足时间约束的同时,极大降低用户局部执行时的总体能耗。

In mobile cloud computing environment, a mobile user needs to decide whether offload the application to the cloud, i.e. computation offloading decision problem. In a shared channel, when multiple users choose to offload the application, the delay of executing application will be happened. For solving the energy consumption optimization problem of executing application under the deadline constraint, a non-cooperative game algorithm is presented. In the game process, each user is interested in minimizing its own energy consumption under satisfying the deadline of executing application. This sharing communication channel among multiple competing users is modelled as a non-cooperative game process, where each user can independently adjust its computation offloading decision. We prove that in homogeneous and semi-homogeneous environment, the algorithm can always generate Nash equilibrium solution. And, we design a Gauss-Seidel-like method to solve the Nash equilibrium. Simulation experimental results show that our algorithm can converge to Nash equilibrium in certain number of iterations, and greatly reduce the total energy consumption of local executing applications under satisfying the deadline.