A Machine Learning Solution for Video Delivery to Mitigate Co-Tier Interference in 5G HetNets

The exponential demand for multimedia services is one reason behind the substantial growth of mobile data traffic. Video traffic patterns have significantly changed in past two years due to coronavirus disease (COVID-19). worldwide pandemic has caused many individuals work from home and use various online video platforms (e.g., Zoom, Google Meet, Microsoft Teams). As a result, overloaded macrocells are unable ensure high Quality Experience (QoE) all users. Heterogeneous Networks (HetNets) consisting small cells (femtocells) promising solution mitigate this problem. A critical challenge with deployment femtocells HetNets interference management between Macro Base Stations (MBSs), Femto (FBSs), FBS FBS. Indeed, dynamic can lead co-tier interference. With rolling out 5G network, it becomes imperative operators maintain network capacity manage different types Machine Learning (ML) considered challenges HetNets. In paper, we propose Interference Classification Offloading Scheme (MLICOS) address problem delivery. Two versions MLICOS, namely, MLICOS1 MLICOS2, proposed. former uses conventional ML classifiers while latter employs advanced algorithms. Both MLICOS compared classic Proportional Fair (PF) scheduling algorithm, Variable Radius (VR+PF) Cognitive Approach (CA). models assessed based on prediction accuracy, precision, recall F-measure. Simulation results show that outperforms other schemes by providing highest throughput lowest delay packet loss ratio. statistical analysis was also carried depict degree faced users when employed.