Enhancing Spectral Efficiency in Cell-Free Massive MIMO Systems Using K-Means++ Clustering and AP-UE Association

Current cellular networks are based on autonomous cells, which often struggle to support large numbers of users due to uneven coverage. As the world becomes increasingly dependent on wireless communication, there is a growing need for cellular networks that offer higher spectral and energy efficiency through multiple wireless access points. Cell-free Massive Multiple-Input Multiple-Output (MIMO) networks successfully meet this demand. In addition to meeting modern wireless communication requirements, these networks can also mitigate many existing interference challenges. This study aims to lower the computational burden of cell-free Massive MIMO systems, enhancing their practicality for large-scale deployment and addressing one of their major operational challenges. We propose a novel access point selection algorithm that combines a machine learning approach for clustering, specifically the K-means++ algorithm and the AP-UE association. Based on simulation findings and evaluation metrics, the proposed algorithm consistently outperforms existing methods, demonstrating notable improvements in efficiency and performance.