无线供能体域网能量优化理论与方法

With the breakthrough in the technique of wireless power transfer (WPT), the problem of sustained energy provision in body area networks (BAN) can be resolved and BAN applications are further extended. However, it comes with the optimization problems of energy scavenging and scheduling. On the one hand, affected by the human behavior patterns, the amount of harvested energy by the wireless powered BAN (WPBAN) nodes has uncertainty but also follows some rules. This brings difficulty to the optimization of energy scavenging as well as the possibility of huge improvement. On the other hand, there exists diverse types of traffic in BAN, but the dynamic energy harvesting conditions pose a great challenge to meet the quality of service (QoS) requirement of diverse BAN traffic by energy scheduling. Therefore, it is of theoretical and practical significance to improve the harvesting power and energy utilization efficiency against the background of inefficiencies in energy transfer and low harvesting power in the far-field WPT applications. To this end, this project performs fundamental research on the energy optimization in WPBAN, based on in-depth analysis of human behavior pattern and BAN traffic features. First, the key factors affecting the performance of WPBAN and the effect laws will be investigated. Second, optimized energy scavenging methods that adapt to the crowd behavior pattern will be proposed. Third, to satisfy differentiated QoS requirement of diverse BAN traffic, optimized energy scheduling methods will be designed. Finally, the systematic theories and techniques of the energy optimization for WPBAN are expected to be established.

无线供能技术的突破解决了体域网的持续供能问题，进一步拓宽了体域网的应用前景，同时也产生了能量捕获和调度的优化问题。一方面，由于受人体行为模式影响，无线供能体域网的能量捕获存在不确定性，但又有一定的规律可循，这给能量捕获优化带来一定困难的同时又提供了巨大的提升可能；另一方面，体域网数据业务类型纷繁复杂，在可用能量不确定的情况下，如何优化能量调度满足多样化业务需求，非常具有挑战。因此，在目前远场无线供能的能量转换效率较低、捕获功率较小的背景下，提高能量捕获功率和能量使用效率具有十分重要的理论和现实意义。本项目将在深入分析人体行为模式和体域网业务数据特点的基础上，系统化开展无线供能体域网的能量优化基础研究，目标是探索影响无线供能体域网性能的关键要素及其作用规律，设计适应人体行为模式的能量优化收集方法和满足差异化服务质量要求的能量优化调度方法，形成无线供能体域网能量优化的核心理论和方法体系。

捕获射频能供电是解决无线体域网乃至物联网终端电池续航能力不足问题极具前景的方法之一。典型地，距离射频能量源十几米的设备已能获取数十μW的功率，可以支持其感知、计算和传输任务。但是，这种供能方式存在如下基本问题：1）能量供给不稳定、捕获效率不高，难以满足终端能量不中断需求;2）能量和网络资源受限，难以满足多样化业务需求；3）静态的感知、传输和计算调度策略难以适应捕获能量、信道质量和任务到达量等环境动态变化。本项目从开源节流的视角，首先，提出了经济高效的能量源和节点部署及调度方案，可为提供无线充电服务的运营商定制更有效、更节约成本的能量源部署与功率调整方案，满足移动可穿戴设备持续不间断工作的需求，提升无源物联网节点的能量捕获效率和检测能力；其次，提出了满足多样化业务需求的节点接入控制和数据传递技术，通过合理分配充电时间和数据发送时间，满足不同节点的多样化吞吐量和数据重要性需求；最后，针对静态策略无法适应环境动态变化的问题，基于强化学习、深度强化学习及李雅普诺夫优化等方法设计了节点采能/感知/传输/计算卸载的在线调度策略，能够优化系统长期平均吞吐量、时延等性能。此外，本项目还在商用RFID系统之上研发了基于高能效码本的反向散射感知识别平台，设计的基于前缀码本的数据传输方案比无码本的普通传输方案节省了超过50%的能量，设计的感知识别平台原型系统能够达到超过99%的平均收包率，并且对环境干扰适应性强，在智能家居及冷链物流领域有较大的应用前景。本项目围绕无线供能体域网的能量优化问题在基本理论、方法体系和平台研发方面均取得了丰硕的研究成果，能够为无线供能网络在移动健康监测、智慧物流等领域的应用和产业化提供良好的技术支撑。