原DV-Hop( Distance Vector-Hop)方法的定位步骤可归纳为两步：距离估计与位置计算。其中，距离估计精度对网络拓扑敏感，而位置计算算法对距离估计精度敏感，从而导致方法整体对多样性网络拓扑分布的鲁棒性较差。针对这一问题进行分析与改进，在距离估计阶段提出基于1跳内最近邻信标与其余信标的跳数连接关系独立确定未知节点与各信标间平均跳距的策略，以此改善未知节点与信标之间的距离估计误差；在位置计算阶段提出在原有Lateration算法的基础上增加牛顿迭代法优化步骤，以此提高定位精度。实验结果表明，在相同的网络条件下，与原DV-Hop方法和其他典型改进方法相比，改进策略首先在距离估计阶段提高了距离估计精度，进而在位置计算阶段提高了对距离估计误差的鲁棒性，从而整体上可有效提高全网未知节点的定位精度。%The node localization process of the traditional Distance Vector-Hop( DV-Hop) method can be summarized as two steps,the distance estimation and the position calculation. The node-beacon distance estimation accuracy is sensitive to the heterogeneity of network topology,and the location derivation algo-rithm is sensitive to the node-beacon distance estimation accuracy. In this way,the robustness of the meth-od is poor while the network topology is diversiform. To solve such a problem,an improvement strategy is put forward and the main work includes:in the distance estimation step,the average hop distance between the nearest beacon within one jump of an unknown node and every other beacon is used as the average hop distance between the unknown node and the beacon to calculate the node-beacon distance,in order to im-prove the node-beacon distance estimation accuracy;in the position calculation step,a refining process u-sing Newton iterative algorithm is added after the initial position of an unknown node is derived using Later-ation algorithm,in order to improve the node localization accuracy. Simulation shows that under the same network conditions,compared with the original method and another representative improved algorithm,the proposed strategy can obtain higher distance estimation accuracy in the first step,and then,it can improve the robustness to distance estimation error in the second step. Thereby,the strategy can improve the aver-age localization accuracy of the unknown nodes in the network.
展开▼
