基于多维尺度的矿山无线传感器网络定位算法

Improved MDS localization algorithm for a WSN in a sub-surface mine

无线传感器网络(WSN)应用于具有复杂性,高度动态性以及无线信号的传播不确定性的矿井较为困难。为了获取地下矿井中大量矿工的位置,部署了无线传感器网络用于获取佩戴于矿工身上的无线移动节点的位置信息。引入了基于卡尔曼滤波估计以及弹性粒子的定位算法,用于节点间的距离测量,提高了动态环境下测距模型所获取的距离信息的精度;基于分簇机制减小网络计算复杂度,并使用了Ao A和余弦定理得到网络中无法直接连接的节点间的距离信息,通过簇之间的距离矩阵的整合得到最终网络的距离矩阵;使用多维尺度(MDS)定位算法重构出网络中所有移动节点的最终位置。实验结果表明,算法有效的减小了定位误差,适用于具有高度动态性与复杂结构的矿井无线传感器网络。

Wireless sensor networks( WSNs) are difficult to apply to mines with complex,high dynamic,and propagation uncertainties of wireless signals. In order to obtain the position of a large number of miners underground,this paper deployed a wireless sensor network to obtain the position information of wireless mobile nodes that are worn by miners. First,Kalman Filter estimation and localization algorithm based on Elastic Particles are introduced to measure the distance between nodes,which improves the accuracy of the distance information obtained by the ranging model under dynamic environment. Second,the clustering mechanism is used to reduce the computational complexity of the network,both of the AoA and the cosine theorem are used to obtain the distance information between nodes that cannot be directly connected in the network. The distance matrix of the final network is obtained through the integration of distance matrix between clusters. Last,a multidimensional scaling( MDS) localization algorithm is introduced to reconstruct the final position of all mobile nodes in the network. The experimental results show that the proposed algorithm can effectively reduce the localization errors and is suitable for wireless sensor networks in sub-surface mines with highly dynamic and complex structures.