基于动态近邻反馈修正的室内定位算法

针对目前无线传感器网络（WSN）室内接收信号强度（RSSI）测距算法中RSSI易受到信道干扰和传播环境影响从而导致定位精度低的问题，提出一种动态近邻反馈修正的室内定位优化算法FC-DNN，以实现无线传感器室内节点精确定位。首先，通过对环境进行Voronoi图分割确定最小定位区域；然后计算每个区域的路径损耗模型参数得到节点间的精确距离；最后利用Spearman等级相关系数动态选择邻居锚节点，根据邻节点反馈修正进一步提高未知节点的定位精度。仿真结果表明，FC-DNN算法复杂度低、计算开销小、能耗较低，与典型的RSSI测距差分修正定位算法（DDLA）和受限三维空间传感器定位算法（CO-3D）相比，节点的平均定位误差降低了约15个百分点，能够很好地满足室内环境定位要求。

Indoor localization algorithm based on feedback correction of dynamic nearest neighbors

In order to solve the problem that the accuracy of indoor localization algorithm based on Received Signal Strength Indicator (RSSI) for Wireless Sensor Network (WSN) is easy to be influenced by channel interference and environment, a new indoor localization algorithm, namely FC-DNN, was proposed based on Feedback Correction of Dynamic Nearest Neighbors. Firstly, the minimum localization region was determined by partitioning the whole environment based on Voronoi diagram before positioning. Then the parameters of the path loss model for each region were calculated to obtain the precise distance between nodes. Finally, the Spearman rank correlation coefficient was introduced to select neighbor anchor nodes dynamically, and the feedback of neighbor anchor nodes was used to further improve the localization accuracy. The simulation results confirm that the proposed FC-DNN algorithm has low time complexity, small computation and low energy consumption; furthermore, compared with conventional Distance Difference Localization Algorithm (DDLA) based on RSSI and sensor network localization in COnstrained 3-D spaces (CO-3D), the average positioning error is reduced by about 15 percentage points, which can well meet the requirements of indoor localization.