Trainable Projected Gradient Detector for Sparsely Spread Code Division Multiple Access

TL;DR

This paper introduces a trainable deep unfolding-based detector for SCDMA systems that achieves near-BP performance with lower computational cost, and jointly learns signature sequences to enhance detection in low SNR conditions.

Contribution

The paper presents a novel trainable projected gradient detector for SCDMA, combining deep unfolding with joint signature and detector learning for improved performance.

Findings

Achieves detection performance comparable to belief propagation with lower complexity.

Joint learning of signatures and detector improves performance in low SNR regimes.

The proposed detector is scalable for massive SCDMA systems.

Abstract

Sparsely spread code division multiple access (SCDMA) is a promising non-orthogonal multiple access technique for future wireless communications. In this paper, we propose a novel trainable multiuser detector called sparse trainable projected gradient (STPG) detector, which is based on the notion of deep unfolding. In the STPG detector, trainable parameters are embedded to a projected gradient descent algorithm, which can be trained by standard deep learning techniques such as back propagation and stochastic gradient descent. Advantages of the detector are its low computational cost and small number of trainable parameters, which enables us to treat massive SCDMA systems. In particular, its computational cost is smaller than a conventional belief propagation (BP) detector while the STPG detector exhibits nearly same detection performance with a BP detector. We also propose a scalable joint learning of signature sequences and the STPG detector for signature design. Numerical results show that the joint learning improves multiuser detection performance particular in the low SNR regime.