Research Of Wireless Location Algorithm And Its Application In Underground

An important issue of Wireless Sensor Network concerning is monitoring the location of event happened. For example, the position of forest fires,the moving area of enemy personnel or vehicles,the location of mine accident, the specific location of the natural gas pipeline leak, etc.For these question, the perpetrators find that obtaining location information is helpful to take measures and make decisions in time. So the location information play a key role for wireless sensor network applications.In the actual environment, the wireless signal strength is likely to be affected.For the problem,the thesis studies a localization algorithm that is The Hybrid Localization Algorithm of Maximum Likelihood and Weighted Centroid based on RSSI. Firstly the algorithm estimates roughly the unknown node’s coordinates by using the Maximum Likelihood Estimation, then refines coordinates of unknown node by using the weighted centroid algorithm. The coordinates of the centroid algorithm replace by the rough estimate of the unknown node. So after the solution of the two can obtain a more accurate location information of the unknown nodes. Simulation results show that the algorithm can be greatly improved localization accuracy.The precise positioning of mine workers and locomotives has great influence on improving the productivity of the mining industry, enhancing mine production safety and mine accident rescue. The thesis studies a localization algorithm that is A Localization Algorithm Based on Judgment of RSSI’s Confidence Interval. The algorithm is divided into two phases:Offline training is to establish RSSI location fingerprint database, including the collection of RSSI values and the calculation of RSSI confidence intervals. Online Positioning firstly determine the RSSI value, then uses the K nearest neighbor algorithm to locate the node to be positioned. The algorithm of the thesis define the scope of RSSI values by determining confidence interval of RSSI value. On the one hand the algorithm can reduce the amount of computation, on the other hand it can improve the reliability of the RSSI value. Simulation results show that the proposed algorithm is better than traditional fingerprint recognition algorithm. The algorithm has greatly improved the positioning accuracy.