Research On Data Compression Technology Of Safety Monitoring System For Undergound Coal Mine

As we all know, the environment of underground coal mine is very sophisticated and changeable, which leads to increasingly insecure and sometimes fatal accidents in the course of mine exploitation. At present, the situation of safe production of coal mine is very serious in our country, and mining accidents occur frequently in recent years which cause very enormous loss of life and property. The existing coal mine security monitoring system has all sorts of insufficiencies and cannot meet the increasingly stringent demand of safety, so it is urgent to be improved.Introducing wireless sensor network can make up for the weakness of existing systems. In order to improve its practicality, data compression is proposed to reach wireless sensor network. Usually, processing data compression much less power than transmitting data in wireless medium, so it is effective to apply data compression before transmitting data for reducing total power consumption by a sensor node.However, existing compression algorithms are not applicable for sensor nodes because of their limited resource. Therefore, research on data compression algorithm for underground coal mine is the main content of this paper, divided into four parts.Firstly, the concep and key technologiesof wireless sensor networkare presented.Introducing wireless sensor network into security monitoring system is proposed.Second, the existing data compression and some of compression algorithms, which have been specifically designed for WSNs, are presented in this parper.Then, the advantages and disadvantages of them are given, in order to design a better algorithm.Thirdly, based on the thesis, an error with adaptive compression algorithm based on Kalman filter (KL-AEO)are proposed.Through the analysis of the system model, the state equation and observation equation can be created. Based on the adaptive error equation and the implementation steps of basic compression, the improved compression algorithm could be acheived.Finally, existing performance evaluation criteria for data compression algorithm and evaluation designed for WSNs are presented. In addition, an energy loss analysis system is established. Theoretically and experimentally, it is concluded that the proposed algoritms can effectively exploit the temporal correlations on the same sensor node and achieve significant data reduction.