The method of cluster compressive data gathering based on pretreatment

In order to resolve the problems of reducing the data transmission and prolonging the network lifetime of the wireless sensor network (WSN) in the case of the nodes deployed randomly, this paper puts forward a cluster compressive data gathering (CCDG) method based on pretreatment. The method can be realized through the following two steps: Firstly, the original sensor readings received from all nodes of the cluster are rearranged by every cluster head, and the ascending order is adopted. After the linear compressed projection, the information is sent to the sink node. Secondly, in the sink node, the Orthogonal Matching Pursuit (OMP) algorithm is selected to realize the reconstruction. As the network traffic has been proven to inversed proportionally to the sparsity of sensor readings and the sorted data has a good sparsity in the differential domain. So the CCDG method based on pretreatment is able to decrease communication cost and extend the lifetime of the entire sensor network. The simulation experiments over real sensor data are carried out and the results confirm that the method proposed can improve the efficiency of data gathering under the condition of ensuring the accuracy of the reconstructed data.