Optimal eNodeB Estimation for 5G Intra-Macrocell Handover Management

Abstract

In next generation 5G intra-macrocell deployment due to the high number of small cells existing in the network, one of the main concerns is the increased handover rate, followed by frequent, unnecessary and ping-pong handover challenges. That can also lead to high packet loss, dropped and blocked calls. Moreover, in 5G intra-macrocell deployments, due to the control and data channel separation handover operation must be executed in two tiers (both data and control channels). For these reasons, handover management in this specific 5G deployment becomes a challenging issue. We believe that,having an optimal and accurate eNodeB estimation, handover overhead in these deployments can be dramatically decreased. In this paper, we propose an optimal eNodeB selection mechanism for 5G intra-macrocell handovers based on spatio-temporal estimations. In this approach, Kriging Interpolator with Semivariogram Analysis is supported by the Autoregressive model for selecting the optimal eNodeB before the connection setup. The stochastic and statistical behaviors of Kriging Interpolation provide better modeling performance. These operations are performed by the proposed EnodeB Estimation Entity. Also, these estimations are applied to both the data and control channels independently. As a result of the proposed management scheme, unnecessary, frequent and ping-pong handover rates are decreased by %35, %37 and %17 respectively compared to the traditional handover method.