Energy-Efficient Secure Computation Offloading in Wireless Powered Mobile Edge Computing Systems.

This paper investigates secure offloading for a wireless powered mobile edge computing (MEC) system. Specifically, an energy-constrained user is first charged by a power beacon (PB) via wireless power transfer (WPT) and then leverages the harvested energy to offload its computation tasks to a MEC server in the presence of multiple eavesdroppers. To enhance the security of computation offloading, we propose a novel physical layer security-assisted scheme, in which the PB can also play the role of a cooperative jammer by transmitting jamming signals to interfere with the eavesdroppers. The objective is to maximize secrecy energy efficiency (SEE) of the MEC system by jointly optimizing the transmit power for WPT, offloading, and jamming, the time allocated for WPT and offloading, and the computation task partition while satisfying the secrecy offloading rate constraint and the energy causality constraint. To deal with the highly non-convex problem, we develop an efficient two-layer algorithm by resorting to variable substitutions, successive convex approximation, and the Dinkelbach method. Simulation results demonstrate that our designed scheme yields a SEE performance enhancement as compared to benchmarks.