An Efficient Model Driven Deep Learning Based Approximate Message Passing Detector for MIMO Systems

This paper presents an improved signal detection method for multiple-input multiple-output (MIMO) systems. The approximate message passing (AMP) algorithm is one of the promising signal detection methods which can achieve near optimal error rate performance. The proposed method enhances the performance of an existing AMP method by applying a model-driven deep learning network. In the proposed method, a trainable parameter is selected and optimized using a neural network. Simulation results illustrate that the proposed method can improve the bit error rate performance with lower computational complexity, compared to the existing methods.