Wireless Sensor Network Node Self-positioning Technology

The use of wireless sensor networks normally need to know the specific location of the sensor nodes. So it is quite meaningful to research location in wireless sensor networks.At present, many algorithms with complexity of calculating ,always need accurate initial estimator, and even only suite to their given area. Based on the research of monte carlo methords, we proposed new algorithms both in rang-based and rang-free condition, and we also improved the DV-Hop algorithm.As to location based on range, with Monte Carlo method we proposed a new algorithm which doesn't need extra hardware spending. According to anchor node's signal, the algorithm defines random particle's area. Then it gives every particle power value by normal Gaussian function. In the end it converges to the unknown location through the constant particles group updating. The simulation under kinds of environment shows that this algorithm reduces the complexity of calculating, and has high accuracy, well robustness and fault tolerance.As to location based on range-free, we also proposed a new algorithm with Monte Carlo method. By computing the distance between random particles and anchor nodes, then compared to the farthest RF sending distance and filtered according to power value, algorithm ensures the unknown node's position. the algorithm can make full use of sensor node's information, and has low complex of calculating. It realizes rough accuracy localization.At last we improved the DV-Hop algorithm with Monte Carlo method. The result shows that the improved algorithm is better than the DV-Hop algorithm in location accuracy. The algorithms which we proposed are better than other common wireless sensor networks localization algorithms in accuracy location, fault tolerance and low complex of calculating.