Intelligent Recommendation-Based User Plane Handover With Enhanced TCP Throughput in Ultra-Dense Cellular Networks
Yan Peng,Yiqing Zhou,Ling Liu,Jintao Li,Zhengang Pan,Gangcan Sun
DOI: https://doi.org/10.1109/tvt.2021.3129832
IF: 6.8
2022-01-01
IEEE Transactions on Vehicular Technology
Abstract:In ultra-dense cellular networks (UDNs) with user/control plane(U/C) splitting, frequent handovers in user planes are unavoidable, which seriously degrades the transmission control protocol (TCP) throughput of mobile station (MS). To enhance the TCP throughput in UDNs, this paper proposes an intelligent recommendation-based user plane handover scheme. Firstly, based on intelligent recommendation algorithms, a mobility prediction algorithm called content-based collaborative hybrid filters (CCHF) is proposed to predict the target small base station (SBS). When the MS moves into the cell-edge of the source SBS, it can pre-access the predicted target SBS and set up connections to the predicted target SBS and the source SBS simultaneously. This is the proposed CCHF-dual-handover scheme. With an accurate prediction and a simultaneous connection, CCHF-dual-handover can present enhanced signal to interference and noise ratio (SINR) at cell-edge, reduced handover interruption ratio (HIR), and improved MSs TCP throughput. Moreover, TCP throughput of CCHF-dual-handover is analyzed to show the impact of various key parameters (such as MSs velocity and pre-access threshold). Finally, simulations are carried out to evaluate the performance of the proposed CCHF-dual-handover. Given random trajectory, the prediction accuracy using CCHF is increased by more than 100 compared with existing prediction algorithms. Given accurate prediction, the CCHF-dual-handover can improve the TCP throughput by up to 150 compared with that of existing handover schemes.
telecommunications,engineering, electrical & electronic,transportation science & technology