Analysis and design of an efficient handoff management strategy via velocity estimation in HetNets

In this paper, we present an efficient handoff decision strategy and a mobility state detection (MSD) scheme for heterogeneous networks (HetNets). Due to base station (BS) densification, HetNets occasionally face unnecessary handovers and service failures. Since velocity plays a crucial role in handover process, its knowledge is imperative for effective mobility management. Furthermore, due to limited power in mobile devices, emphasis is in investigation of velocity estimators that are at service provider end. Thus, we propose two maximum likelihood–based velocity estimators exploiting handover count and sojourn time measurements, respectively. Our analysis illustrates that both the estimators are asymptotically unbiased and efficient. In addition, we also notice that variance of sojourn time–based estimator is smaller than that of handover count–based estimator. Considering superiority of sojourn time–based estimator, we propose a handover decision strategy to minimize unnecessary handovers. Here, we first predict the sojourn time of upcoming BSs on the basis of velocity estimated via sojourn time samples. Next, BSs whose predicted sojourn time are greater than a prespecified threshold are then considered to identify the next serving BS. Finally, we select next serving BS on the basis of minimum cost for satisfying quality‐of‐service parameters such as bandwidth, security, monetary cost, and available power at user device. This work also exploits estimated velocity to investigate MSD specified in LTE standards. The numerical and simulation results validate our approach by illustrating accuracy in velocity estimation, reduction in frequent handovers and service failures, and improvement in MSD. Due to base station (BS) densification and user mobility, Heterogeneous Networks (HetNets) occasionally face unnecessary handovers and service failures. Two maximum‐likelihood (ML) based techniques exploiting handover count and sojourn‐time measurements are proposed to estimate user velocity at the service provider end. Further, recognizing superiority of sojourn‐time based velocity estimate, an efficient mobility management strategy is proposed to minimize unnecessary handovers and call drops.