基于一种改进的无源目标定位算法在WSN中的研究

Research on improved passive target location algorithm in WSN

无源定位作为无线传感中的一种新的定位技术,具有抗干扰能力强、隐蔽性强的特点。针对无源定位中的TDOA算法具有能量消耗大、时间消耗长的特点,对TDOA中的Chan算法进行改进,对算法自身的存在多解的情况,采用极限学习机从解集中选出最优解,并针对最优解采用近似最小似然估计法对定位结果进行修正,能够有效地提高节点定位的精度,减少误差。仿真实验从定位精度与更新次数、TDOA测量噪声方差、节点之间的距离、协作节点数量四个方面来进行比较,该算法相比传统的Chan无源算法能够有效地提高定位精度,具有很好的参考价值。

As a new localization technology in wireless sensor,passive localization has the features of strong capacity of resisting disturbance and strong elusiveness. To solve the large energy consumption and long consumed time of TDOA algorithm,this paper improved the Chan algorithm in TDOA,adopted the extreme learning machine to select the optimal solution from the solution set in response to multiple solution of the algorithm itself,and adopted the approximate minimum likelihood estimation method towards the optimal solution to correct the location results,which could effectively enhance the accuracy of node localization and reduce errors. The simulation experiment compares the localization accuracy with updating frequency,TDOA measurement noise variance,the distance between nodes,and the number of collaborative nodes,the proposed algorithm can effectively improve the localization accuracy and have a good reference value compared with the traditional passive Chan algorithm.