Research And Implementation Of Node Location Technology For Intelligent Minefield

With the appearance and rapid development of the third generation of intelligent mine, which is wide-area and homing, vital function and great development prospect in defense industry of network communication technology, especially wireless sensor network communication technology, has attracted attention of most developed countries including USA. Based on the wireless sensor network communication technology, defense departments of every country have carried out research of the fourth generation of intelligent minefield, that is, networked intelligent minefield. Data exchange is the important premise and means to realize farm-out of each mine node, recognition, capture and attack of target in blockade zone, and then extraordinary military operation performance. Positioning technology of mine node is one of the key technologies in the system research of intelligent mine field network.On the base of deep investigation and research into the latest node positioning technology in intelligent minefield network, several kinds of positioning methods for sensor network has been deeply researched and analyzed combined with the factual positioning requirement. By the method of comparative analysis of application range, positioning precision and requirements for hardware, node positioning method based on RSSI ranging was determined. Ranging method based on kalman-mean filter is proposed to improve the precision and performance of RSSI data pre-processing method. Mean value treatment has been done to the data after kalman filtering, reducing the influence from long-time large disturbances, accelerating the convergence of filter curve and also reducing the variance. Methods of mean filter, Gaussian filter, Kalman filter and Kalman-mean filter have been respectively applied to calculate the A and β of positioning models as well as to establish the radio transmission models. On the base of comparing and analyzing the experimental and simulation results, the feasibility and advantages of Kalman-mean filter have been proved. When it comes to the calculation method of node position, the trilateral centroid positioning method based on the Particle thought has been proposed and its performance has been verified by MATLAB simulation and practical test. It turned out that, comparing with the trilateral positioning method, the trilateral centroid positioning method has smaller error caused by insufficient use of anchor node location parameters and higher precision. Finally, hardware circuit of intelligent node positioning system has been designed on the base of STM32 and CC2530. The feasibility has been verified on this hardware platform.