基于聚类和TDOA的无线基站定位算法研究

Wireless Basestation Location Algorithm based on Clustering and TDOA

为提高无线定位的精度,提出了基于聚类和TDOA的基站定位优化算法。首先,建立TDOA定位模型,采用Fang算法对移动终端的位置进行初步估计,并根据多个初步估计值获得的先验信息,去除双解情况下误差较大的估计位置;其次,采用K-means聚类算法进行聚类,进一步筛选出离群点,得到更精确的位置估计数据集,将其取平均后得到精确位置;最后,采用仿真实验验证算法的性能。实验结果表明,相对于经典Fang算法,所提算法具有更优的定位性能,应用前景广阔。

In order to improve the accuracy of wireless positioning,a basestation location optimization algorithm based on clustering and TDOA is proposed.Firstly,the TDOA positioning model is established,and the position of mobile terminal initially estimated by the Fang algorithm.According to the prior information obtained from multiple preliminary estimates,the estimated position with great error in the case of double solutions is eliminated.Then,K-means clustering algorithm is adopted to perform clustering and further screening the outliers to obtain a more accurate position estimation data set,and then it is averaged and the exact position obtained.Finally,the performance of this algorithm is verified by simulation experiments.The experimental results indicate that the proposed clustering algorithm is better in positioning performance than the classical Fang algorithm,and thus has wide application prospects.