Abstract: Aiming at the problem that the current data compression algorithm of sensor network had the defects of low compression ratio and serious data deformation, in order to improve the realtime performance of data transmission of sensor networks, we proposed a data compression algorithm based on spatiotemporal correlation. Firstly, the original data of sensor network was collected, and the spatial transformation technology was used to analyze the correlation between the data of sensor network in space, and then the noise was removed to reduce the spatial resource occupied by noise. Secondly, according to the temporal correlation of the data of sensor network, the compression sensing algorithm was introduced to compress the spatial coefficient to reduce the data redundancy of sensor network. Finally, the performance of data compression algorithm of sensor network was analyzed by simulation experiment. The simulation results show that compared with other sensor network data compression algorithms, the proposed algorithm can improve the data compression ratio of sensor network without losing the data information of sensor network, obtain faster data compression speed of sensor network and reduce the communication pressure of sensor network.

Key words: data of sensor network, temporal correlation, spatial correlation, spatial transformation technology