An Improved Amorphous Localization Algorithm for Wireless Sensor Networks

Node localization is of significant importance for supporting many critical applications in wireless sensor networks. However, traditional Amorphous localization algorithm and its variants still cannot provide sufficient localization accuracy. To address this issue, this paper proposes an improved Amorphous localization algorithm to determine the locations of unknown nodes. To determine the locations of unknown nodes under this algorithm, we need to select these beacon nodes, which are less than a presetting hop count threshold away from unknown nodes. We then adopt two-dimensional hyperbolic function instead of the classic least square method to determine the locations of unknown nodes, which are very close to their actual locations. Simulations are conducted to validate the accuracy of our proposed localization algorithm in wireless sensor networks. Remarkably, the average localization error of our proposed localization algorithm is lower than those of Amorphous algorithm and its improved algorithm, under both the uniform and non-uniform node distributions.