Comparative analysis of uplink power control schemes in user-centric cell-free massive MIMO

Abstract

User-centric cell-free massive MIMO is emerging as a key architecture for next-generation wireless networks, offering improved spectral efficiency, seamless connectivity, and enhanced fairness by eliminating cell boundaries and leveraging cooperative transmission from distributed access points (APs). However, to fully exploit its potential, especially in dense deployments, effective uplink power control (UPC) mechanisms are essential to manage interference while balancing throughput and user fairness. This paper compares three UPC strategies: Full power control, fractional power control (FPC) with exponent υ = 0.5, and a fixed-point algorithm (FPA) designed for max-min fairness. Simulation results under varying network scales reveal critical trade-offs. In per-user spectral efficiency (SE) distribution, Full yields the best median (3.089 bit/s/Hz) but poor fairness. FPA maximizes fairness with a minimum SE of 2.1 bit/s/Hz, outperforming Full (1.585) and FPC (0.47). In total SE, Full achieves 34.81 bit/s/Hz (cumulative distribution function=0.5), higher than FPC (28.566) and FPA (21.05). Scalability analysis shows Full and FPC improve total SE with more APs (up to 47.23 and 40.56 bit/s/Hz at 100 APs), while FPA leads in minimum SE (3.347). In larger user equipment scenarios, FPA maintains fairness (min SE drops mildly from 2.36 to 1.63), while Full suffers (1.9 to 0.98). Computation-wise, FPA runs up to 10× faster than FPC (0.017 ms vs. 0.153-0.493 ms). Overall, Full suits throughput-focused designs; FPA excels in fairness; and FPC offers a scalable, tunable balance between the two.