Abstract: Aiming at the problem that the positioning accuracy of wireless sensor network nodes is affected by the environment and the error weight factor, a centroid positioning algorithm was proposed to dynamically correct path loss factor and error weight factor. At earlier stage, a dynamic loss factor was obtained by weighting correction according to the actual measurement and path loss model; at later stage, the weight factor matrix was constructed by dividing rectangular region. Firstly, the location of the unknown node was estimated by introducing dynamic loss factor into the traditional weighted centroid localization algorithm. Then, the error weight factor matrix was queried to determine the optimal weight factor. Finally, the unknown node location was recalculated. The experimental results show that the improved algorithm reduces the average error and the minimum error, and the positioning accuracy is 58% higher than ordinary centroid algorithm, 21% higher than dynamic correction centroid algorithm, and 11% higher than dynamic weighted centroid algorithm.

Key words: Wireless Sensor Network (WSN), positioning, dynamic loss factor, error weight factor, dynamic correction

摘要： 针对无线传感器网络（WSN）节点的定位精度受环境和误差权重因子的影响问题，提出一种对路径损耗因子和误差权重因子动态修正的质心定位算法。前期根据实测和路径损耗模型，加权修正得出动态损耗因子；后期通过划分矩形区域，构造权重因子矩阵。首先，使用动态损耗因子，代入传统加权质心定位算法估算出未知节点的位置；然后，查询误差权重因子矩阵，确定最优权重因子，重新计算出未知节点坐标。实验结果表明，改进的算法降低了平均误差和最小误差，定位精度比普通质心算法提高了58%，比动态修正质心算法提高了21%，比动态加权质心算法提高了11%，定位精度有所提高。

关键词: 无线传感器网络, 定位, 动态损耗因子, 误差权重因子, 动态修正