Frequency Channel Selectivity in Vehicular CFmMIMO Systems: a Multi-Objective Optimization Approach

As 6G networks aim to meet the increasing demand for high data capacity and ultra-reliable, low-latency communication, cell-free massive Multiple-Input Multiple-Output (CFmMIMO) systems emerge as a key technology by eliminating traditional cell boundaries and ensuring seamless coverage. However, a critical challenge arises when considering the frequency dimension in the channel and system model, specifically the need to address frequency selectivity and bandwidth sharing. These factors complicate resource allocation, particularly in dynamic environments like vehicular networks, where maintaining Quality of Service (QoS) becomes increasingly difficult. This paper tackles these issues by proposing a novel multi-user allocation strategy within shared subbands, designed to optimize spectral efficiency (SE), ensure fairness, and minimize interference under real-time user mobility conditions. To ensure practical and realistic performance evaluations, we base our analysis on real-world mobility patterns and channel characteristics derived from empirical data. A Simulated Annealing (SA) algorithm is employed to solve the multi-objective optimization problem, with comparisons made to Genetic Algorithm (GA) and Ant Colony Optimization (ACO). Our results show that the SA-based approach significantly improves SE and achieves up to 40% savings in frequency resources, providing a scalable and robust solution for frequency management in CFmMIMO systems, particularly for dynamic vehicular communication scenarios.