Research on APIT and Monte Carlo Method of Localization Algorithm for Wireless Sensor Networks

Traditional approximate point-in-triangulation test (APIT) localization algorithm requiring low equipped hardware, having relatively high location accuracy, is easy to implement, and widely used in wireless sensor network positioning system. However, the location accuracy of unknown node in triangle overlap region should be further improved, especially in the sparse beacons' environment, the location accuracy is seriously affected. In this paper, MC-APIT algorithm is proposed, which implements random sampling using the Monte Carlo method in the overlap region, and filters samples through the target node's RSSI (Received Signal Strength) sequence values, in order that Mathematical expectation of the sample values could converge to that of the target node'. Simulation results show that: the algorithm can reduce the sampling area and the location energy consumption, to a certain extent restrained the propagation error. Compared with APIT algorithm, the location accuracy has been markedly improved.