无线体域网已成为5G智慧医疗系统的核心组成部分，其推广应用必须突破“低能效”带来的连接不稳定、续航时间短、传输不可靠等瓶颈问题。对此，本项目重点研究无线体域网的高能效技术，拟采用深度学习信道预测解决无线信道状态的不确定性问题，基于信道状态数据驱动解决资源管理的复杂性问题，并通过能量采集和中继协作解决无线传输的弱信号问题。具体研究内容包括：①提出一种深度无线体域网（Deep-WBAN）的架构，采用Shimmer传感器设计一套远程健康状态监测系统，进行信道测量获得复杂条件下的信道状态数据，研究基于深度时间卷积网络的信道预测方法；②研究基于信道预测的中继选择、调度和传输功率控制机制，提高能量效率；③研究时间和功率平衡分配下无线信息和能量协作同传的最佳传输机制。本项目的实施对于推动5G智慧医疗系统应用，缓解我国人口亚健康化、空巢化、老龄化以及重大公共卫生突发事件对医疗体系的压力，有重要现实意义。

Wireless body area network has become the core component of 5G intelligent medical system, its application must break through the bottleneck problems such as unstable connection, short endurance and unreliable transmission brought by "low energy efficiency". In this regard, this project focuses on the high energy efficiency technologies of wireless body area networks. It is proposed to use deep learning channel prediction to solve the uncertainty of wireless channel state, to solve the complexity of resource management based on channel state data drive, and to solve the weak signal problem of wireless transmission through energy harvesting and relay cooperation. The specific research contents of this project include: (1) a deep WBAN architecture is proposed, a remote health monitoring system is designed with Shimmer sensors, the channel state data under complex conditions is obtained by channel measurement, and the channel prediction method based on deep time convolution network is studied; (2) the relay selection, scheduling and transmission power control mechanism based on channel prediction is studied to improve the energy efficiency; (3) study the optimal transmission mechanism of wireless information and energy simultaneous interpreting under the time and power balance allocation. The implementation of this project is of great practical significance for promoting the application of 5G intelligent medical system, relieving the pressure of sub-health, empty nest, aging and major public health emergencies on the medical system.