Mobility management in HetNets consider on QOS and improve Throughput

Abstract

HetNet networks are one of the key features of LTE and the ones beyond. Despite the development created in HetNet, recent studies have shown that HOFs and PPs can cause serious problems given the small size of the cell. On these networks, given the difference in base station's coverage, the handover performance of the user equipment (UE) may significantly reduce; this is especially obvious in situations where UE passes through the small cells at a high-speed. In this paper, a context-aware mobility management (MM) approach is presented for small-cellular networks, where reinforcement learning. BSs collectively learn their long-term traffic loads and optimal extension of their cell range and schedule their UE according to the speed and rate of history data, along with the consideration of QOS, which is exchanged over rows. The proposed method not only provides better performance than MM in terms of power, but also provides better fairness conditions that reduce the likelihood of a Handover and Ping Pong failure. In the survey, macro and picocells learn how to optimize their longterm traffic, whereas in the short term, the UE communication process is based on history and timing based on speed and QOS. Thus, a long-term load balancing approach aims to improve overall system capacity while reducing the probability of Hand Over Failure (HoF) and Ping Pong (PP).