A Continuous Policy Learning Approach for Hybrid Offloading in Backscatter Communication

In wireless powered communication networks (WPCNs), wireless devices (WDs) can offload tasks to edge server by both passive backscatter and active transmission with low or even no energy consumption. Multi WDs equipped with one antenna each share the same channel. Work modes as well as time sharing for energy harvesting, backscatter and active RF transmission should be properly managed to optimize the system performance. This article proposes a deep deterministic policy gradient (DDPG) algorithm for hybrid data offloading, by which the system can search the best action in consecutive domain to minimize the overall offloading delay with the consideration of fairness among WDs. The complexity is analyzed and the numerical results show that the approach can achieve minimal offloading delay and enhance the energy harvesting efficiency.