面向无线传感器网络的分布式压缩感知测量矩阵的优化设计

With the application of wireless sensor networks (WSNs) continue to expand, information collection and processing capacity of the systems are restricted by the limited energy, computing and communication capabilities of sensor nodes. By using a small number of random measurements to jointly reconstruct sensor signals, distributed compressive sensing (DCS) provides a new way to solve the above problems. The objectives of the research work are: to unveil the relationship between the joint signal sparse model and reconstruction or classification performance, to expose the impact of sensor energy storage and spectrum resource to DCS information collection, and to derive the optimized sensing matrix design for signal reconstruction and classification applications respectively. This project will lead to appropriate modeling of correlated sensor signals and dictionary learning designs, optimized sensing matrix designs for DCS information collection, and improved WSN systems with sensing matrices which are adapted to match the available the node resource. This project has significant contributions in the improvement of information collection and process capacity of WSNs. In addition, it also has indirect impact on the study of DCS in many other areas within the field of signal and image processing.

随着无线传感器网络的应用不断拓宽，传感器节点有限的能量携带、计算能力和通信能力成为了制约整个系统信息采集和处理能力的瓶颈。分布式压缩感知利用少量的信号随机测量对多个传感器节点信号联合重建，为解决无线传感器网络中的以上问题开辟了一条新的路径。本项目针对信号联合稀疏模型与分布式压缩感知信号重建和分类性能的关系，节点能量供给和频谱资源对分布式压缩感知信息获取的影响，面向信号重建和分类应用的节点测量矩阵的优化设计三个科学问题开展研究。揭示有效反映无线传感器网络多节点信号时空相关性的联合稀疏模型及相应字典学习机制，创新性提出有效提高分布式压缩感知信息获取性能的节点测量矩阵的优化设计方案，建立匹配节点资源的分布式压缩感知自适应信号测量矩阵的系统实现方案。本项目对提高无线传感器网络的信息采集和处理能力、降低布置和运营维护成本具有重大价值，同时对信号和图像处理等许多领域的压缩感知技术研究具有参考意义。

随着无线传感器网络的应用不断拓宽，传感器节点有限的能量携带、计算能力和通信能力成为了制约整个系统信息采集和处理能力的瓶颈。分布式压缩感知利用少量的信号随机测量对多个传感器节点信号联合重建，为解决无线传感器网络中的以上问题开辟了一条新的路径。本项目针对信号联合稀疏模型与分布式压缩感知信号重建和分类性能的关系，节点能量供给和频谱资源对分布式压缩感知信息获取的影响，面向信号重建和分类应用的节点测量矩阵的优化设计三个科学问题开展研究。揭示有效反映无线传感器网络多节点信号时空相关性的联合稀疏模型及相应字典学习机制，创新性提出有效提高分布式压缩感知信息获取性能的节点测量矩阵的优化设计方案，建立匹配节点资源的分布式压缩感知自适应信号测量矩阵的系统实现方案。本项目对提高无线传感器网络的信息采集和处理能力、降低布置和运营维护成本具有重大价值，同时对信号和图像处理等许多领域的压缩感知技术研究具有参考意义。