基于改进的质心定位的LSSVR算法的设计

Design of LSSVR Location Algorithm based on Improved Centroid Location

摘要:

针对无线传感中基于质心算法的节点定位存在误差比较大,算法效率低的缺点,提出了一种基于加权的LSSVR的节点定位算法；首先,对未知节点构建节点序列相关度,采用Kendall的Tau指标来估计未知节点的位置,提高了未知节点的定位精度,其次引入了LSSVR概念,构建改进质心算法的LSSVR定位模型,降低了噪声影响,大幅度提高定位精度；仿真实验表明该算法与基本的LSSVR算法在定位精度上有了明显的提高,在锚节点,未知节点所占比例不断增大的情况下该算法定位精度具有很大的提高,降低了算法的计算复杂度,具有较高的应用价值。

Abstract:

Aiming at centroid algorithm’s setback of big errors and low efficiency in node positioning in wireless sensing, a weighted-based LSSVR node positioning algorithm is proposed. First of all, the place of unknown nodes is estimated by establishing node sequence correlation with Kendall’s Tau index, which has improved the positioning accuracy of unknown nodes. Secondly, LSSVR is introduced and LSSVR positioning model of improved centroid location algorithm is constructed to reduce the influence of noise. Simulation experiment shows that compared with basic LSSVR algorithm, this algorithm has significantly improved its positioning accuracy, and with the increasing proportion of anchor nodes and unknown nodes, positioning accuracy of the algorithm has been significantly improved, which has reduced its computational complexity, so this algorithm has relatively high application value.