Base Station Selection and Task Offloading of the Mobile Edge Computing System

Abstract

Based on the two decision variables, service location and base station selection, construct a computational model of the switching delay, communication delay, and queuing delay patterns of a mobile edge computing system in each time horizon; minimize the non-switching latency to obtain the service deployment and base station selection decision at the initial time horizon; compute the switching latency and non-switching latency of the current time horizon based on the decision of the previous time horizon, and determine the current time horizon based on the principle of larger non-switching latency tolerance. If the service is not reallocated, the decision for the current time slot remains the same as the decision for the previous time slot; if the service is reallocated, the non-reach time of the current time slot is minimized and the service allocation and base station selection decision for the current time slot is obtained; the service allocation and base station selection decisions for all time slots are iteratively computed. This paper provides low-latency processing and quality of service for users to satisfy the randomness of user movement.