无线传感网络自诊断技术研究

In recent years, Wireless Sensor Networks (WSNs) have been widely used in many applications. Due to many unique features of the WSNs, network failures can significantly degrade the system performance. Existing approaches to diagnosing sensor networks are generally sink-based, which rely on actively pulling state information from all sensor nodes so as to conduct centralized analysis. However, the sink-based diagnosis tools incur huge communication overhead to the traffic sensitive sensor networks. Also, due to the unreliable wireless communications, sink often obtains incomplete and sometimes suspicious information, leading to highly inaccurate judgments. To address the above issues, in this project, we plan to study the self-diagnosis approach, which encourages each single sensor to join the fault decision process. We aim to design a novel type of fault detectors through which multiple nodes can cooperate with each other in a diagnosis task. Based on the Finite State Machine model, the fault detectors will encode the diagnosis process to state transitions. Each sensor can participate in the fault diagnosis by transiting the detector's current state to a new one based on local evidences and pass the intermediate results to others. We study the high efficient diagnosis triggering and management strategies. The proposed solution will significantly save bandwidth consumption and improve the diagnosis efficiency.

近年来，无线传感器网络在很多领域得到了广泛应用，但由于其远程部署，无线自组等特性，网络故障会显著降低系统性能并且难于诊断。现有诊断方法都是基于基站的诊断，需要主动收集所有传感节点的状态信息汇报至基站然后再基站端进行集中分析推理故障原因。基站诊断方法会给传感网络带来极大的通讯开销，同时基站收到的信息往往是不完整的，导致错误的诊断结果。为解决以上问题，项目组拟研究无线传感网络自诊断技术，让单个传感器节点直接参与诊断过程。自诊断过程中，多个节点局部协同根据自己的本地状态信息推断故障原因。项目组研究基于有限状态机的故障诊断模型和轻量级诊断算法，多节点的协同诊断机制，以及基于统计变点检测的诊断触发和高效诊断管理技术。无线传感网络自诊断技术将会显著减少网络带宽消耗，提高诊断性能。

无线传感器网络在很多领域得到了广泛应用，但由于其远程部署，无线自组等特性，网络故障会显著降低系统性能并且难于诊断。现有诊断方法都是基于基站的诊断，需要主动收集所有传感节点的状态信息汇报至基站然后再基站端进行集中分析推理故障原因。基站诊断方法会给传感网络带来极大的通讯开销，同时基站收到的信息往往是不完整的，导致错误的诊断结果。为解决以上问题，本项目深入研究了无线传感网络自诊断相关技术，让单个传感器节点直接参与诊断过程，多个节点局部协同根据自己的本地状态信息推断故障原因。项目组提出多项创新性的自诊断算法和技术，包括自主链路状态扫描、数据错误发现、自主故障定位等，实现一套高效自诊断管理组件并与实际系统集成，实验结果证明，项目成果能够显著减少网络带宽消耗，提高诊断性能。