Analysis of hyperbolic and circular positioning algorithms using stationary signal-strength-difference measurements in wireless communications
Analysis of hyperbolic and circular positioning algorithms using stationary signal-strength-difference measurements in wireless communications
复制标题
DOI：
10.1109/tvt.2005.863405
复制
发表时间：
2006-03-01
影响因子：
6.8
通讯作者：
Wu, JC
中科院分区：
文献类型：
--
作者：
Liu, BC;Lin, KH;Wu, JC
将接收信号强度(RSS)引入信号传播模型来确定用户位置是无线通信中最有前途的策略之一。本文提出了一种基于相对信号强度测量的简单方法，即在用户位置从多个基站(BTS)测量的平稳信号强度的差异。平稳信号强度是RSS的平均值，也是平稳的高斯过程。在使用该方法时，至关重要的是要确定在将信号传播模型合并到用户定位方法中时，可以消除一些不确定的传播参数。这样，平稳信号强度的差异导致了两种解决方案：一种是BTS对之间的距离差，另一种是用户位置到多个BTS的距离。因此，可以分别提出基于距离差解的双曲线定位算法和基于距离解的圆形定位算法。随后，在真实的传播环境中进行了现场试验，并给出了一些实验结果。结果表明，双曲线和圆形定位算法可以分别将用户定位到约350米和300米以内的67个百分位数。与文献中报道的仅基于RSS的数值结果相比，我们的方法是优越的。尽管结果不符合联邦通信委员会(FCC)的要求，但该方法在负载计算和网络信令负载方面被证明是足够简单和高效的。
Referring received signal strength (RSS) to a signal propagation model to find user location is one of the most promising strategies in wireless communications. This paper develops a simple method based on relative signal-strength measurements, that is, the differences in stationary signal strength measured at the user location from multiple base transceiver stations (BTSs). The stationary signal strength is the averaged RSS and also is a stationary Gaussian process. In using this method, it is vitally important to confirm that some uncertain propagation parameters can be canceled out while a signal propagation model is merged into our method for locating users. In this way, the differences in stationary signal strength lead to two solutions: One is the distance difference between pairs of BTSs, and the other is the distances from the user location to the multiple BTSs. Consequently, the hyperbolic positioning algorithm due to the distance-difference solution and the circular positioning algorithm due to the distances solution can be presented, respectively. Afterward, some experimental results were drawn from a field trial in a real propagation environment. Results show that the hyperbolic and circular positioning algorithms can locate the user to within about 350 and 300 m in 67 percentile, respectively. Compared with the numerical result reported in the literature on existing methods based on RSS only, our method is superior. Despite the result not meeting Federal Communications Commission (FCC) requirements, this method proved to be sufficiently simple and efficient in terms of the computation at burden and network signaling load.