USER CENTRIC BASED CLUSTERING FOR MITIGATING INTERCELL INTERFERENCE FOR 5G NETWORK USING MACHINE LEARNING APPROACH

Abstract

The introduction of fifth-generation (5G) mobile communication networks has brought about revolutionary improvements in user experience, latency reduction, and data speeds. However, issues like frequency reuse-induced inter-cell interference continue to exist and impede the best possible network performance. In order to reduce inter-cell interference in dense 5G networks, this study investigates a novel user-centric clustering strategy that makes use of machine learning techniques. The suggested approach optimizes scheduling and resource allocation by dynamically classifying users according to interference levels and proximity, guaranteeing increased spectrum efficiency and user satisfaction. Through the use of a variety of clustering algorithms, including KM-Means clustering, the model successfully handles network variability. The outcomes show notable increases in network efficiency, decreased interference, and improved quality of service, providing a scalable answer to the needs of contemporary communications. In particular, after KM-means clustering, the user's interference from the various base stations (BS1=1.9dBm, BS2=2.6dBm, BS3=2.55dBm, BS4=1.91dBm, BS5=1.4dBm) was decreased to (BS1=0.97dBm, BS2=1.25dBm, BS3=1.23dBm, BS4=0.98dBm, BS5=0.7dBm). The network can improve data rate and provide a more seamless user experience with fewer interference.

KEYWORDS  Clustering, Machine Learning, Inter-cell Interference, 5G Network.