传感网络中坐标评估机制的研究

Research on Coordinate Axes Evaluation Mechanism in Sensor Networks

传感网络作为一种融合了计算机分布式处理技术、嵌入式技术、物联网技术于一体的网络体系架构在信息技术应用领域备受欢迎。而节点位置坐标评估作为传感网络核心技术也就成为了研究的关键。论文通过剖析加权算法原理总结了该算法在实际网络中应用的局限性。如:算法对于全局感知节点的均衡分布存在较高程度的依赖,加之感知节点多跳情形下引发的平均跳距计算误差等问题,使得该算法的坐标计算精度较为欠缺。基于加权算法的局限性,论文从降低位置坐标定位误差的角度提出了一种通信半径范围内节点平均跳距评估机制。该机制为多跳情形下的节点重新建立平均跳距评估方案,缩小待测节点位置坐标评估值与实际值的误差距离。同时,为了提高该重新评估机制的可靠性,优化机制下的全局节点均采用了随机投放的方式。最后通过构建实验模型对优化的评估机制开展四项不同的模拟对比实验。不仅证实了论文提出的重评机制在传感网中的定位优势较加权算法显著,同时节点随机布局的拓扑方案也进一步表明了重评机制的可靠性和其在网络中的实用性。

As a network architecture which integrates computer distributed processing technology,embedded technology and Internet of things technology,sensor network is popular in the field of information technology application. As the core technology of sensor network,node localization has become the key point of the research. This paper summarizes the limitations of the algorithm in the actual network by analyzing the principle of weighted localization algorithm. Such as the algorithm has a high degree of dependence on the balanced distribution of the global perception nodes. In addition,the calculation error of average hop distance caused by multi-hop sensing nodes is considered. This makes the positioning accuracy of the algorithm relatively deficient. Based on the limitation of the weighted algorithm,this paper proposes an average hop distance evaluation mechanism in the range of communication radius from the point of view of reducing the positioning error. The mechanism can reconstruct the average hop distance of the nodes in the multi hop case,and reduce the error distance between the coordinate evaluation value and the actual value of the nodes to be measured. What is more,the global nodes in the optimization mechanism use the method of random delivery in order to improve the reliability of the evaluation mechanism. Finally,four different simulation experiments are carried out to optimize the evaluation mechanism by constructing an experimental model. It proves that the localization advantage of the proposed evaluation mechanism in sensor network is more significant than weighted localization algorithm,what is more,the reliability of the evaluation mechanism and its practicability in the network are further demonstrated by the topological scheme of the random placement of nodes.