基于差值信号稀疏模型的分布式压缩感知

Distributed compressed sensing based on differential signal sparse model

利用无线传感器网络的空间相关性,构建了一种差值信号稀疏模型,该模型适用于对同一物理现象或事件进行监测的传感器网络应用。在差值信号稀疏模型的基础上,提出了一种适用于该模型的分布式压缩感知算法,该算法能够在节点间不通信的情况下实现对差值信号的编码。仿真结果表明,与单独重构相比,提出的算法可以用更少的观测值联合重构出信号群,以能量有效的方式满足了无线传感器网络的应用。

This paper proposed a differential signal sparse model by exploiting inter-signal correlation structures.The model was appropriate for the WSNs applications in which multi-node were used to monitor the same physical phenomena or events.Based on differential signal sparse model,this paper proposed a distributed compressed sensing（DCS） algorithm for the model.The proposed algorithm could encode differential signal without inter-node communications.Simulations indicate that,compared with separately reconstruction,the proposed algorithm can joint reconstruct multiple signals with high probability by using significantly fewer measurements per sensor and accommodates requirements of WSNs applications in energy efficient way.