A Joint Optimization of Backhaul Traffic and Queue Length in Multi-CPU Cell-Free Networks

Cell-free massive multiple-input multiple-output (MIMO) systems are emerging as a key technology for 6th Generation (6 G) wireless communication, offering superior network management, high spectral efficiency, and optimal resource utilization. However, large-scale cell-free deployments pose significant challenges, particularly in efficiently handling user association and resource allocation across a vast network. While multi-Central Processing Unit (CPU) configurations are necessary for scaling, the resource allocation problem in such multi-CPU environments has been largely unexplored. In this work, we introduce a novel multi-CPU backhaul and queue model and propose an optimization algorithm that jointly minimizes backhaul traffic and queue size by optimizing access point (AP) association, uplink and downlink power, and bandwidth allocation. Our extensive simulations demonstrate that the proposed algorithm reduces backhaul traffic and queue size by up to twenty times compared to reinforcement learning and other benchmark methods. This model not only provides valuable insights into efficient resource management in large-scale cell-free networks but also paves the way for future research into multi-CPU cell-free systems.