Abstract: An offloading decision and resource optimization scheme for cache-assisted edge computing is proposed to further reduce the energy consumption of terminal devices in the mobile egde computing(MEC)system.Firstly,the optimization problem is established to minimize the worst-case energy consumption of user during the task execution,and the mixed integer programming problem is transformed into a non-convex quadratic constrained quadratic programming(QCQP)model.Semidefinite-relaxation and randomization probability mapping are used to obtain the pre-selected offloading set assisted by caching;Secondly, the Lagrangian dual decomposition method and the bisection method are utilized to acquire the optimal transmission power and edge computing resource under constraints.By comparing the energy consumption of the set of devices,an ideal set of offloading decision and resource allocation scheme are got.Experiment shows that the proposed scheme can effectively reduce the energy consumption and improve the service performance of the edge computing system.

Key words: mobile edge computing, caching, computational offloading, semidefinite-relaxation, resource allocation