Deep Reinforcement Learning Based Task Offloading Mechanism for Edge Computing in Wireless Body Area Network

Wireless body area network can be regarded as an unmanned network system to realize remote health care monitoring. The electronic mobile medical system based on wireless body area network is expected to bring huge changes to the medical industry in the future. Inter-network interference is prone to occur between multiple body area networks. Collecting and managing stable and reliable data effectively and quickly between multiple networks is an important direction in body area networks. The paper uses the multiple area network architecture based on edge computing to decompose the task offloading problem into task classification and resource allocation through combining the movement characteristics and data characteristics of the body area network. Oriented by the unmanned system, a resource allocation strategy based on A3C algorithm is proposed. Simulation and experimental results show that the strategy can achieve better performance compared with the existing offload strategy.