Sparse Deep Neural Network Based Directional Modulation Design

When receivers change their locations, the transmitter array needs to change its weight coefficient to steer the mainbeam pointing to the corresponding direction. However, the traditional direction modulation (DM) design takes time to optimize the weight coefficient, which may not give a quick response. To solve the problem, a sparsely connected deep neural network (SDNN) based directional modulation method is proposed in this paper. The sparse deep neural network is built using the uniform linear array as an example, and the weighted vector of the array is obtained by training to actualize the necessary DM function. The outcomes demonstrate the viability of the suggested design.