摘要
针对现有室内指纹定位算法时间复杂度高的问题,设计了基于无线传感器网络的快速指纹匹配定位算法。该算法采用二分法的思想,先将未知节点锁定在原定位区域的四分之一区域内,然后通过计算该区域内的主参考点与未知节点的欧氏距离,找出欧氏距离最小的主参考点所在区域,最后将该区域所有次参考点的加权质心作为最终的位置估计。仿真结果表明:与最近邻居定位算法和最近K邻居定位算法相比,快速指纹匹配定位算法的定位精度略高于二者,定位时间约为二者的1/3。该算法在保证定位精度的前提下,减小了算法的时间复杂度。
This paper introduces a fast fingerprint matching algorithm to address the time complexity of the existing fingerprint localization algorithm. This algorithm operates by locking the unknown node in1/4 regions in the original location using the dichotomy; then finding out the minimum Euclidean distance of the main reference point by calculating the Euclidean master reference points and unknown node distance within the region; and ultimately identifying the weighted centroid area of all time as the final position of the reference point estimate. The simulation results show that the fast fingerprint matching algorithm provides a slightly higher positioning accuracy than both the nearest neighbor algorithm and nearest neighbor K positioning algorithm and uses about 1/3 the positioning time of the two. The algorithm could ensure the positioning accuracy while greatly reducing the time complexity.
出处
《黑龙江科技大学学报》
CAS
2017年第3期307-310,共4页
Journal of Heilongjiang University of Science And Technology
基金
黑龙江省自然科学基金项目(F2015019
F2015017)
黑龙江省青年科学基金项目(QC2013C064)
关键词
无线传感器网络
指纹匹配
二分法
时间复杂度
欧式距离
wireless sensor networks
fingerprint matching
dichotomy
time complexity
euclidean distance
