Affiliations: [a] Engineering Trainning Center, Zhengzhou University of Light Industry, Zhengzhou, 450000, China | [b] School of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou, 450000, China
Abstract: In order to overcome the problems of high unloading time cost, long unloading task delay and poor load balance of traditional offloading methods, this paper studies the mobile edge computing task offloading method of wireless network based on improved genetic algorithm. Based on the wireless network mobile edge computing architecture, a wireless network mobile edge computing task scheduling scheme is constructed to lay the foundation for subsequent task offloading. Then, the improved genetic algorithm is used for initial operation allocation and offloading priority ranking, and the mobile edge computing task offloading is realized by dynamically adjusting the trade-off coefficient. The experimental results show that the offloading time cost of this method is between 0.16 min–0.31 min, the offloading task delay is between 1.05 s–1.47 s, and the load balance can reach 97.9%, indicating that it effectively realizes the design expectation.