基于压缩感知的无线传感器网络的数据编码

近年来,基于压缩感知的无线传感器网络数据编码的研究取得了一定的进展,但大部分研究是基于"单跳"的传输模型或者没有考虑网络内节点之间的合作。本文提出了一种基于压缩感知和分簇的传感器网络数据编码方法。首先,节点根据自己的编号伪随机的产生一个M维的列向量,把感知的数据xi投影到此向量上,然后把自己的编号和投影的数据一同传输给簇头。簇头把收到的数据进行求和,并且把计算后的结果传送给下一个簇头,直到sink节点。通过仿真实验和理论分析,验证了本文提出的方法比传统的方法能更好地减少网络内数据传输量。

Compressive Sensing-Based Data Coding in Wireless Sensor Networks

Recently,compressive sensing shows high promise in wireless sensor networks.However,most of the studies is based on the model of "single-hop" or no-cooperation network.This paper proposed a new method of data collection in wireless sensor networks based on compressive sensing and clustering.Firstly,node i generated a M dimensional vector by the number of node,then node projected the sensing data xi on the vector and sent the data to the cluster.Cluster summed the data of collection and transmitted the data to next cluster until the fussion node.After simulation and analysis,we can draw the conclusion that the proposed method can reduce the number of communication and save the energy on the nodes compared with the traditional method.