Machine Learning-Enhanced eICIC Framework for Mitigating Inter-Cell Interference in Multi-Layer LTE-A Networks

Résumé fourni par la source

Inter-cell interference (ICI) has become much more acute and critical owing to many newly deployed co-channel multi-layer Long Term Evolution-Advanced (LTE-A) networks which affected the quality of service (QoS) and overall network performance. The present research paper is dealing with an Enhanced Inter-Cell Interference Coordination (eICIC) approach for co-channel multiple layer LTE-A networks. The proposed eICIC scheme combines the contemporary approaches involving dynamic resource control, power control, and scheduling to avoid ICI efficiently. Applying an innovative self-developed algorithm based on ML for real-time interference estimation and resource management, this framework improves the network's availability, throughput, and spectral efficiency. Computer simulation is also used to test the proposed eICIC scheme and indicates that the new scheme significantly relieves interference and enhances the user throughput in contrast to the traditional methods of operation. The study reveals that the proposed ML-based predictive model effectively solves the problem of the dynamic and heterogeneity of the network while optimising resource utilisation and increasing QoS. This study also pays attention to influence of different network parameters on the efficiency of the eICIC framework, thus serving as reference for the improvement of the next-generation LTE-A network.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.