基于蚁群算法的无线传感器网络节点定位算法研究

发布时间：2018-02-03 20:10

本文关键词： 无线传感器网络节点定位蚁群算法 DV-hop算法自适应　出处：《华中师范大学》2014年硕士论文　论文类型：学位论文

【摘要】：无线传感器网络(WSN)是由大量传感器节点以自组织方式组成的一个监控系统,可以对目标区域的信息进行实时地监控和处理,应用十分广泛。对于大多数WSN来说,未知节点所感知的信息时没有意义的,我们必须了解无线传感器网络中各个节点的位置信息。因此,节点定位在无线传感器网络应用中起着至关重要的作用。目前,节点定位已经成为学术界研究的热点问题。节点定位算法主要分为基于测距(Range-based)定位算法和无需测距(Range-free)定位算法。基于测距定位算法需要给节点配置额外的硬件设备来完成相应的测距任务,该类算法定位的结果精度高,但是增加了网络的成本和能量消耗,影响网络的使用寿命；相比较而言,无需测距定位算法实现起来更加简单方便一些,该类算法不需要额外的硬件设备,通过节点间的通信大致估算出未知节点的位置,但是定位精度不如基于测距定位算法。蚁群算法(ACO)作为人工智能的一个分支,在处理组合优化问题时有较好的效果。本文通过对节点定位问题进行相应的转化,把节点定位问题变成函数优化问题,将蚁群算法应用在节点定位问题上,提出了基于蚁群算法的节点定位算法(ACOL)。由于蚁群算法自身的局限性,容易导致算法早熟或收敛速度过慢。在基本蚁群算法的基础上,我们进行了相应的改进,提出了自适应蚁群算法(AACO),并将该算法应用在节点定位问题上,形成了基于自适应蚁群算法的节点定位算法(AACOL)来避免算法早熟或收敛过慢。最后本文采用MATLAB进行仿真实验,在相同的实验环境下比较了DV-Hop算法、ACOL算法和AACOL算法的定位精度。实验结果表明,ACOL算法和AACOL算法较DV-Hop算法定位精度更高,AACOL算法比ACOL算法结果更加稳定,收敛速度更快。
[Abstract]:Wireless Sensor Network (WSNs) is a self-organizing monitoring system composed of a large number of sensor nodes. It can monitor and process the information of the target area in real time. It is widely used. For most WSN, the information perceived by unknown nodes is meaningless, we must know the location information of each node in WSN. Node location plays an important role in wireless sensor network applications. At present, node location has become a hot topic in academic research. Node localization algorithm is mainly divided into Range-based location algorithm and Range-free-based location algorithm. Location algorithm. Based on the location algorithm, the nodes need to be equipped with additional hardware equipment to complete the corresponding ranging tasks. The accuracy of this algorithm is high, but the cost and energy consumption of the network are increased, and the service life of the network is affected. By comparison, it is more simple and convenient to realize the location algorithm without the need of distance location. This kind of algorithm does not need additional hardware equipment, and the location of unknown nodes can be estimated roughly by the communication between nodes. However, the location accuracy is not as good as the location algorithm based on ranging. As a branch of artificial intelligence, ACO (Ant Colony algorithm) has a good effect in dealing with combinatorial optimization problem. In this paper, the node location problem is transformed accordingly. The problem of node location is turned into a function optimization problem, and the ant colony algorithm is applied to the problem of node location, and a node location algorithm based on ant colony algorithm is proposed. Due to the limitations of ant colony algorithm itself. It is easy to cause premature convergence or slow convergence. Based on the basic ant colony algorithm, we improve the algorithm and propose an adaptive ant colony algorithm (AAC). The algorithm is applied to the problem of node location and a node location algorithm based on adaptive ant colony algorithm (ACO) is developed to avoid premature convergence or slow convergence. Finally, this paper uses MATLAB to carry on the simulation experiment, under the same experimental environment, compares the DV-Hop algorithm ACOL algorithm and the AACOL algorithm localization accuracy. The experimental results show that. The accuracy of ACOL algorithm and AACOL algorithm is higher than that of DV-Hop algorithm. The algorithm is more stable and convergent than ACOL algorithm.
【学位授予单位】：华中师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5;TP18

【参考文献】