In this paper, we propose a learning-based solution for resource allocation in a wireless powered communication network (WPCN). We provide a study and analysis of a deep neural network (DNN) which can reasonably effectively approximate the iterative optimization algorithm for resource allocation in the WPCN. In this scheme, the deep neural network provides an optimized solution for transmitting power with different channel coefficients. The proposed deep neural network accepts the channel coefficient as an input and outputs minimized power for this channel in the WPCN. The DNN learns the relationship between input and output and gives a fairly accurate approximation for the transmit power optimization iterative algorithm. We exploit the sequential parametric convex approximation (SPCA) iterative algorithm to solve the optimization problem for transmit power in the WPCN. The proposed approach ensures the quality of service (QoS) of the WPCN by managing user throughput and by keeping harvested energy levels above a defined threshold. Through numerical results and simulations, it is verified that the proposed scheme can best approximate the SPCA iterative algorithms with low computational time consumption.
Exploiting a Deep Neural Network for Efficient Transmit Power Minimization in a Wireless Powered Communication Network
