Research on Multi-level Offloading Scheduling Algorithm for Mobile Edge Computing

With the rapid development of wireless sensor networks, existing sensor devices are difficult to meet the needs of low latency and high reliability applications. In order to improve the efficiency of the program, some tasks are migrated from the local device to the edge cloud or edge server through edge computing technology to reduce the running time of the application and the power consumption of the device. This paper proposes a multi-level offloading model for edge computing, and studies the offloading scheduling algorithm to minimize the power consumption of the sensor device under the specified delay. This paper proposes a fast algorithm based on greedy strategy, which optimizes the particle swarm optimization algorithm combined with greedy strategy. The experimental results show that the total cost of the HGA algorithm in multi-level offloading is reduced by 23.6% and 8.4%, respectively, compared to the traditional single-level offloading. The proposed HPSO algorithm reduces the total cost of the HGA algorithm by 12.1%.