雾计算辅助的无线体域网任务卸载方法

基于无线体域网（Wireless Body Area Network，WBAN）的智慧医疗在取得广泛关注的同时也面临诸多挑战，传统方法引入云计算解决体域网网关设备（HUB）资源受限问题，但实时性差。采用低时延的雾计算辅助的方法对体域网生理数据进行处理，提出了一种雾计算目标节点优化选择及任务卸载方法。该方法充分考虑紧急任务的实时性需求，采用抢占式任务调度及有效的资源分配方式降低紧急任务的总时延。同时该方法根据任务紧急与否自适应调整目标节点的评价尺度，在紧急情况下，以时延及任务负载最小为优化目标，在非紧急情况下，以时延及可靠性为优化目标以确定目标任务卸载目标节点。仿真结果表明，该方法可以有效提升系统的实时性，保证系统的可靠性，尤其是可以在很大程度上降低紧急任务处理时延，满足紧急任务的低时延需求。

Fog-Computing-Assisted Task Offloading Method of Wireless Body Area Networks

WBAN（Wireless Body Area Network）-based smart healthcare has attracted wide attention but faces many challenges. The traditional method introduces cloud computing to solve the resource-constraint problem of WBAN gateway device（HUB）, but its real-time performance is poor. A low-latency fog-assisted approach is proposed to process the physiological data of the WBAN, and a target node optimizing and task offloading method is designed. This method takes full account of the real-time demand of the urgent tasks and uses preemptive task scheduling and effective resource allocation to reduce the total delay of the urgent tasks. At the same time, the evaluation metrics of the target node is adjusted adaptively according to whether the task is urgent or not. When in emergency the goal is to minimize the time delay and task load, and when in non-emergency the goal is to optimize the time-latency and reliability of the target fog node for task processing. Simulation results show that the proposed method can effectively improve the real-time performance of the system and ensure the reliability of the system. Particularly, it can dramatically improve the processing delay of urgent tasks to meet the low delay processing requirements.