Joint parameter priority metric formulations for APs subset selection in user-centric cell-free massive MIMO

In this paper, we investigate various joint parameters priority metric formulations (PMFs) for access point (AP) subset selection in the uplink of user-centric Cell-Free Massive MIMO systems. Unlike conventional approaches that activate all APs, we propose a selective activation strategy based on priority metrics that capture large-scale gain, line-of-sight (LoS) power, distance, and interference. Three formulations are introduce —linear, exponential, and sigmoid—to score and rank APs for each user equipment (UE). The top-κ APs are then activated, and the closed-form uplink spectral efficiency (SE) expressions under phase-aware MMSE estimation and large-scale fading decoding (LSFD) are evaluated on the reduced system. Analytical modeling of computational complexity and energy efficiency also developed. Simulation results shows that the sigmoid based formulation achieves up to 40–45% higher SE and 6–8 times higher EE than the two-layer LSFD baseline, with approx. 65% lower runtime. The exponential metric provides comparable performance, while the linear metric offers a balanced trade-off between efficiency and complexity. Activating only 45 and 65% of APs attains near optimal SE with substantial energy and computational savings, demonstrating the proposed framework’s suitability for scalable and sustainable Cell-Free Massive MIMO deployments.