In this paper, a deep learning (DL)-based method is proposed to rapidly determine the optimal phase for maximizing power delivery to a receiver in distributed microwave power transmission (DMPT) systems. The proposed deep neural network (DNN)-based phase optimization method has a low computational time because it learns the optimized phase in advance, eliminating the redundant tasks of existing methods. The proposed method was compared with two existing methods in terms of the computation time and received power through Monte Carlo simulations. The analysis results were also experimentally verified using a 4 × 4 DMPT system. Compared to the two existing methods, it achieves the lowest computation time while maintaining a comparable received power level, with less than 1 dB difference from the best-performing baseline.
Lee et al. (Sun,) studied this question.