基于虚拟力的WSAN定位算法的研究

发布时间：2018-12-24 08:32

【摘要】：无线传感器与执行器网络(Wireless Sensor and Actor Network,WSAN)是在无线传感器网络(Wireless Sensor Network,WSN)的基础上衍生而来,通常由若干传感器节点和执行器节点组成。执行器节点具有较强的处理能力、较高的能量,能够根据传感器节点采集的信息进行分析、决策并采取相应的措施。在某种意义上,WSAN是能够改变物理世界的新型网络。因此,WSAN对于重要的定位技术也提出了更高的要求。另外现有的WSN定位算法不能直接移植到WSAN中使用,故本文针对传统WSN定位算法的缺陷并结合WSAN的定位需求,对WSAN定位算法进行了研究。具体研究内容如下1.总结了国内外关于WSN/WSAN定位算法的研究进展并分析了现有WSN定位算法不适用于WSAN的原因。以WSN为基础,从体系结构、节点组成、网络特征等方面对WSAN进行了全面介绍。同时列举了WSAN定位算法中常见的6种分类方式,并对其中的代表算法进行了详细介绍。2.提出了一种基于正方形区域的移动非测距定位算法(Mobile Range-free Localization Algorithm based on Square Area,MRFS)。算法利用移动的执行器节点代替WSN中的锚节点进行定位,首先通过执行器节点正方形的布局确定未知节点所在区域,然后通过迭代不断缩小该区域,最后计算该区域质心作为未知节点的坐标。利用移动的执行器节点能有效节省网络部署成本,另外与测距技术相比,非测距的定位方式大大降低了硬件成本。仿真实验表明,算法能够取得不错的定位效果。为避免执行器节点分布局部密集或稀疏,使空闲执行器节点的分布均匀合理,在定位的同时引入虚拟力模型,提出了一种基于虚拟力的MRFS非测距定位算法(MRFS Localization Algorithm based on Virtual Force,MRFSVF)。仿真实验证明,虚拟力算法的引入能有效改善执行器节点分布不均的情况,从而优化其覆盖面积,并且减少了定位误差及时间。3.测距定位方面,传统的基于信号传输时间(Time of Arrival,TOA)的定位算法通过计算信号的传输时间来测量节点间距离,将TOA算法与虚拟力模型相结合,促使执行器节点在虚拟力作用下不断移动,有利于执行器节点向请求定位的传感器节点靠近,从而提高定位成功率。仿真实验验证了算法的性能。4.针对各定位算法的特点及适用场合,介绍了各算法的应用。
[Abstract]:Wireless sensor and actuator network (Wireless Sensor and Actor Network,WSAN) is derived from the wireless sensor network (Wireless Sensor Network,WSN) and is usually composed of several sensor nodes and actuator nodes. The actuator node has strong processing ability and high energy. It can analyze, make decision and take corresponding measures according to the information collected by sensor node. In a sense, WSAN is a new network that can change the physical world. Therefore, WSAN also puts forward higher requirements for important positioning technology. In addition, the existing WSN localization algorithm can not be directly transplanted to WSAN, so this paper aims at the shortcomings of the traditional WSN localization algorithm and combined with the needs of WSAN location, the WSAN location algorithm is studied. The specific contents of the study are as follows. This paper summarizes the research progress of WSN/WSAN localization algorithms at home and abroad and analyzes the reasons why existing WSN localization algorithms are not suitable for WSAN. Based on WSN, WSAN is introduced from architecture, node composition and network features. At the same time, six common classification methods in WSAN localization algorithm are listed, and the representative algorithms are introduced in detail. 2. A mobile non-ranging location algorithm (Mobile Range-free Localization Algorithm based on Square Area,MRFS) based on square region is proposed. The algorithm uses moving actuator nodes instead of anchor nodes in WSN to locate. Firstly, the location of unknown nodes is determined by the square layout of actuator nodes, and then the region is continuously reduced by iteration. Finally, the coordinates of the region's centroid as unknown nodes are calculated. The use of mobile actuator nodes can effectively save the network deployment costs, in addition, compared with ranging technology, the location mode of non-ranging greatly reduces the hardware cost. Simulation results show that the algorithm can achieve a good localization effect. In order to avoid the local dense or sparse distribution of actuator nodes and to make the distribution of idle actuator nodes uniform and reasonable, a virtual force model was introduced into the localization process, and a virtual force based MRFS non-ranging localization algorithm (MRFS Localization Algorithm based on Virtual Force, was proposed. MRFSVF). Simulation results show that the introduction of virtual force algorithm can effectively improve the uneven distribution of actuator nodes, thus optimize the coverage area, and reduce the positioning error and time. In the field of ranging and location, the traditional localization algorithm based on signal transmission time (Time of Arrival,TOA) measures the distance between nodes by calculating the signal transmission time, and combines the TOA algorithm with the virtual force model. It is advantageous for the actuator node to approach the sensor node which requests to locate, which can improve the success rate of location because of the continuous movement of the actuator node under the action of virtual force. Simulation results verify the performance of the algorithm. 4. According to the characteristics of each location algorithm and its application, the application of each algorithm is introduced.
【学位授予单位】：南京农业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP212.9;TN929.5

【相似文献】