Paper Title: A Survey- ML based Offloading in HetNets
Author's name: Duraimurugan J, Dr.Indra Gandhi K,
The deployment of ultra-dense heterogeneous networks (HetNets) composed up of macro, micro, pico, and femto cells is required due to the exponential expansion of mobile users and the critical necessity for excellent service quality. Each cell type of HetNets delivers a various level of cell range and a unique system capacity. As a result, there is an urgent requirement to balance the loads between small cells, especially given the users' erratic distribution over different mobility axes. The intelligent load balancing models, including those based on machine learning (ML) technology, that have been created in HetNets are surveyed in this work. The evaluation offers a framework and an approach to design load balancing models for future HetNets that are affordable, adaptable, and intelligent. Additionally, a general explanation of the load balancing issue is provided. The idea of load balancing is initially introduced, followed by explanations of its goal, functionality, and assessment standards. In addition, a basic Offloading frame and its operation are given. The next step is to undertake a thorough literature review that includes approaches and fixes for the load balancing issue. To demonstrate the historical evolution of load balancing models, a thorough literature study of ML-driven load balancing methods is completed. Finally, the existing difficulties in putting these models into practice are discussed, along with the potential operational implications of load balancing in the future.