Wireless Powered D2D Communication Security Using LSTM in Emergency Communication System

This paper investigates the security of wireless-powered device-to-device (D2D) communication networks in an emergency communication system with multiple eavesdroppers, where an unmanned aerial vehicle (UAV) acts as a temporary base station providing wireless power for the D2D transmitters to guarantee power efficiency for the D2D network, and serving as cooperative jammer (CJ) to interfere with the eavesdroppers. Considering the privacy of D2D communication, we formulate the establishment of D2D pairs as a multi-class classification problem, and use the Long Short-Term Memory (LSTM) algorithm to classify potential D2D transmitters. Specifically, the LSTM algorithm is based on the remaining energy of the temporary base station, the mobile state of the potential D2D transmitter and the content request amount of the D2D receiver to establish D2D communication that achieves the optimal security performance. The simulation results show that the LSTM-based D2D establishment scheme can effectively improve the security capability of D2D communication, and prove that the amount of energy harvested by the D2D transmitter affects the security capacity of the D2D communication system.