A Novel Unified Handover Algorithm for LTE-A

In today's mobile networks, handover (HO) decisions are performed based on static thresholds of e.g. the “Received Signal Radio Power” (RSRP) or the “Received Signal Radio Quality” (RSRQ). In LTE networks the HO Events A1-A6 have been defined. However, threshold based HO decisions often lead to poor network resource utilization, increased call blocking probabilities and provide low adaptivity w.r.t. user mobility pattern changes. In this paper, we propose a novel unified HO Algorithm based on Discrete Stochastic Dynamic Programming (DSDP), taking into account not only the radio conditions but also the overall resource utilization (i.e. the past, current and predicted future cell loads) and the impact of individual HO decisions on the serving and target cells. The HO algorithm operates in a decentralized manner - it is executed at each eNodeB (eNB) and attempts to achieve a balanced cell load. Our method can be readily integrated in legacy networks, as all required input parameters have been defined in the 3GPP rel. 11. To demonstrate the performance of the proposed algorithm, we implemented it in a NS3-GYM simulation environment and investigated an indoor LTE network scenario with X2-based HO capability.