User Identification and Channel Estimation by Iterative DNN-Based Decoder on Multiple-Access Fading Channel

In the user identification (UI) scheme for a multiple-access fading channel based on randomly generated (0, 1, -1)-signature code, previous studies used signature code over noisy adder channel, and only state information (USI) was decoded by decoder. However, considering communication model as compressed sensing process, it is possible to estimate coefficients while identifying users. this study, improve efficiency of decoding we propose an iterative deep neural network (DNN)-based Simulation results show that proposed DNN-based decoder requires less computing time than classical signal recovery algorithm in achieving higher UI estimation (CE) accuracies.