A study on the impact of mobility on caching in non-standalone 5G vehicular networks

Abstract

Responding to numerous requests of mobile users and maintaining the quality of service is one of the important challenges in designing and implementing future vehicular networks. One of the solutions to reduce network traffic is caching on the radio access network and close to users. In this paper, we investigate the impact of the mobility of users on caching performance and propose novel mobility-aware schemes to turn the mobility challenge into an opportunity to reduce latency. We study vehicular networks with Non-Standalone (NSA) 5G deployment in which a mobile user (MU) can access the 5G Next Generation NodeB (gNB) stations with caching capability in addition to accessing the 4G Evolved NodeB (eNB) stations. We provide an analytical framework and present a closed-form expression of the average end-to-end data transfer time in cache-enabled vehicular networks with NSA 5G deployment, considering users' mobility and the stochastic arrival of user requests. We study various caching schemes and propose novel mobility-aware caching schemes. Using the map-based mobility patterns of various vehicles including cars and buses and also pedestrians, it is shown that the proposed mobility-aware schemes improve the performance of mobile networks in terms of latency.