Decentralized Massive Access Random Scheme in User-Centric Cell-Free Massive MIMO System

This paper considers a decentralized massive random access scheme applicable to a user-centric cell-free massive MIMO system. In this system, a large number of user equipment (UEs) select access points (APs) based on the quality of channel, and adjacent APs can achieve lossless data exchange with a finite data volume. Each AP is equipped with an Edge Distributed Unit (EDU) capable of independently performing Maximum Likelihood (ML) estimation of the large-scale fading coefficients (LSFC) for the UEs associated with. Subsequently, the APs collectively assess the activity of associated UEs based on the estimated LSFC vectors obtained from adjacent APs, and employ a heuristic method to identify primary interference sources. On this basis, each AP's EDU uses the approximate message passing - sparse Bayesian learning (AMP-SBL) algorithm for channel estimation (CE). The proposed scheme concentrates on the associated user set, reducing computational complexity and signaling overhead while maintaining accurate performance. Numerical simulations validate the effectiveness and superiority of the approach presented in this paper.