Unfolding for Symbol Detection in Task-Based Quantized MIMO Receivers

Abstract

This letter proposes a Task-based Alternating Direction Method of Multipliers unfolded Network (TAMNet) for symbol detection in multiple-input-multiple-output (MIMO) systems with quantized measurements at the receiver. Task-based systems apply pre-processing, to mitigate the effect of quantization nonlinearity, on the incoming signal before it is sent to the quantizer. In this letter, we provide a closed-form expression for a suitable pre-processing block and then modify its structure to let it be trained by the incoming data. The quantized symbol is then sent to an unfolded symbol detector that is based on the Alternating Direction Method of Multipliers (ADMM) iterations. We investigate the significance of the pre-processing block in TAMNet and conclude about its efficacy. We also provide numerical results to demonstrate that TAMNet yields an improvement of upto 3.5 dB over the existing unfolding-based detector DeepHOTML, in terms of the Signal-to-Noise Ratio $(\rm {SNR})$ , in achieving a symbol error rate (SER) of $\approx 3\times 10^{-4}$ .