改进的距离重构三维定位算法

针对三维定位算法中节点坐标转换精度低的问题,在距离重构多维定位算法DR-MDS的基础上,提出了改进的距离重构三维定位算法。该算法在距离重构和MDS-MPA算法的思想下,采用优化的最小均方根偏差几何中心修正算法RMSDGCC(Root Mean Square Deviation-Geometric Center Correction),先计算出坐标转换矩阵,然后利用锚节点的几何中心对所有节点进行修正,实现节点从相对坐标向绝对坐标较高精度的转换。算法可以实现有效的坐标转换,获得较好的定位效果。实验结果显示,与原多维定位算法相比,在不引入测距误差的情况下,改进算法在测距半径为15 m时定位精度提高14%,定位误差缩小至0.63 m,测距半径为35 m时,定位精度提高87%,定位误差几乎为0。该改进算法在三维空间中有更高的节点定位精度。

An Improved Distance Reconstructing Three-dimensional Localization Algorithm

In order to solve the problem of localization algorithm that node transformation accuracy from relative coordinate to absolute coordinate is low in three-dimensional, we proposed a improved distance reconstruct three-dimensional localization algorithm based on the distance reconstruct multi-dimensional localization algorithm. The algorithm based on the distance reconstruct and MDS-MAP algorithm adopted optimized Root Mean Square Deviation-Geometric Center Correction (RMSD-GCC) algorithm , first calculate the coordinate transformation matrix, then use the geometric center of the anchor node correct all nodes position, and realized node transformation from relative coordinate to absolute coordinate. The proposed algorithm can effectively transform coordinate, and get better location performance. Compared with old algorithm, the result shows that positioning accuracy in advanced algorithm increased by 14% and location error reduced to 0.8m in distance radius equal to 15m, positioning accuracy increased by 87% and location error almost reduced to 0 in distance radius equal to 35m under the condition of not introduce ranging error. The improved algorithm in three-dimension space has higher node localization accuracy.