Individualized optimization of handover parameters in 5G (NR) indoor and outdoor co-frequency scenarios

In wireless communication system, due to the uniform setting of handover parameters in a cellular, it is liable to handover too late or too early in some scenarios for some terminals, which will affect the user's perception and system performance. In this paper, a novel approach to recognizing the character of scenes and Individualizing handover parameters for 5G communication system is presented. Based on the received RSRP (Reference Signal Receiving Power) values from multi-cell reported by users, and combined with t-SNE (t-distribution stochastic neighbor embedded) algorithm and DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method, it can realize the division of handover parameter configuration of multi scene in one cellular coverage. Furthermore, the handover parameter requirements of different spatial clusters are obtained by statistical analysis. The simulation results show that the average SINR is increased by 9.5dB in the worst coverage positions of handover process.