The Research Of Applying Compressed Sensing Technology Into Wireless Sensor Network

It’s an era that computing exists everywhere, the influence of wireless sensor network(WSN) is growing on our daily life and environment. Such networks have potential utility in a wide range of disciplines such as medical, industrial, environmental, and military applications. For long-term monitoring applications, the goal is to cover as large an area as possible while still acquiring high-resolution information about the sensed environment. It’s a challenging problem that how to acquire substantial amount of data efficiently from the wireless sensor network with limited energy and bandwidth.Based on the research of wireless sensor network and data fusion algorithm, the compressed sensing technology is introduced. Addressing the energy-efficiency problem when applying the traditional data acquisition mechanism, we present the data acquisition mechanism based on compressed sensing, and then discusses the advantage of it when compared with the traditional mechanism.For the characteristic of energy-constrained in wireless sensor network, considering routing strategy into the designing of projection matrix, a compressed sensing algorithm based on data fusion tree(DFT-SRP) is proposed. It minimizes communication consumption by means of sparse random projection, and relevance between projection matrix and sparse basis will be decreased in order to guarantee the data reconstruction quality while data fusion tree is generating. Simulation results show that, the proposed algorithm not only achieves a balance between reconstruction quality and energy consumption, but also has high adaptability to variety of data originated from different sparse basis.In addition, in order to verify the applicability of compressed sensing technology into wireless sensor network, detailed analysis and verification is processed from the aspect of system modeling,actual data analyzing and prototype system building. Results show that when applying compressed sensing technology into WSN, energy consumption can be decreased in quantity while high accuracy data can be reconstructed, which means a good applicability in the WSN.