A CONTEXTUAL AWARENESS-LEARNING APPROACH TO MULTI-OBJECTIVE MOBILITY MANAGEMENT IN 5G ULTRA-DENSE NETWORK

Qing Wang, Liping Teng, and Shuang Zhao

Keywords

5G, UDN, mobility management, handover, user satisfaction-ratio

Abstract

With the rapid development of mobile Internet and Internet of Things, the current mobile communication data traffic is facing ex- ponential growth. The 5G (fifth generation) mobile communication network has been put on the agenda. Ultra-dense network (UDN) deployment is a key technology, while the mobility management issues usher in new challenges. This paper aims to solve the FHO (frequent handover) problem and candidate base stations’ (BSs’) selection in UDN. We use a multi-objective non-stochastic learning method with known context to solve the problems. An improved non-stochastic learning model is proposed to collect the dynamic Contextual Competition Scores (CCS) of small base stations (SBSs) to determine the sequence of selected BSs. And the concept of handover overhead is introduced to strictly control the occurrence of handover. Finally, the performance of the BS selection scheme is evaluated by the user satisfaction-ratio parameter and handover times. The simulation shows the validity of the proposed algo- rithm in the case of ideal UDN and non-ideal UDN with delay or packet loss.

Important Links:



Go Back