基于水平集的围猎式水下传感器节点布置研究

Study on Hunting Deployment of Underwater Sensors Based on Level Set Method

水下传感器网络作为一种探索和开发海洋的新方法,在人类不易接触的水下区域的探测和监测中发挥着重要作用,是无线传感器网络领域研究的热点.在水下空间中传感器节点如何自主调整位置实现对“兴趣域”的覆盖和监视是一个重要课题.该文引入水平集理论LSM,提出了一种基于水平集的围猎式水下传感器节点布置方法.在该方法中网关节点根据传感器节点的位置信息和可能感知到的兴趣域信息,以能量泛函计算兴趣域边缘参数,传感器节点以此参数为根据,完成趋向兴趣域边缘的迁移;大量节点的迁移可逐步演化成对兴趣域的围猎式布置.同时,该文建立了结构相似度SSIM和网络能耗均衡度两个性能评价指标,以评估该文方法在实现兴趣域驱动的水下节点布置的综合性能.大量仿真实验结果表明,该方法在不同节点规模、节点密度和兴趣域形状情况下均可实现对兴趣域边缘的动态地、围猎式地覆盖,从而达到对兴趣域的有效跟踪和监视.

As a new method of exploring and developing the ocean,underwater sensor networks play an important role in the detection and monitoring of underwater areas that are not easily accessible to humans.Underwater sensor networks have been widely used in many fields such as marine information collection,geological disaster prevention,resource exploration and military monitoring.It is one of the hot spots in the field of wireless sensor networks.In underwater space,how to autonomously adjust the position of sensor nodes to achieve coverage and monitoring of events is an important topic.However,the existing underwater node development plan does not fully consider the shape of the surveillance target,and lacks the node self-organization arrangement mechanism in the event area.In this paper,the event area is visually defined as the“interest domain”,and the level set theory is introduced to propose a node development method,hunting deployment of underwater sensors based on level set method,for the interest domain.In this method,the gateway node organizes the sensor nodes within the communication range to form a local computing unit and builds a computing model driven by energy functional according to the location information of the sensor node and the concentration information of the interest domain that the sensor node may perceive.The gateway simulates the dynamic evolution mechanism of the level set to obtain the edge information of the interest domain.The sensor node calculates its own dynamic parameters based on its position and the edge information of the interest domain,and autonomously completes the migration toward the edge of the interest domain.The migration of a large number of nodes can gradually evolve into a hunting arrangement on the edge of the interest domain to realize the coverage and monitoring of the interest domain.At the same time,the paper establishes two performance evaluation indicators of structural similarity index and network energy balance to evaluate the comprehensive performance of the proposed method in the implementation of interest domain driven sensor deployment.The structural similarity can reflect the hunting of the interest domain by the sensor nodes.The larger the value of SSIM is,the better the underwater sensor nodes are arranged at the edge of the interest domain,and better tracking and monitoring effects can be exerted.The network energy consumption balance is used to evaluate the energy consumption balance of each sensor node.The smaller the network energy consumption balance,the more balanced the energy consumption of the underwater sensor network can be,and the higher the technical feasibility can be.A large number of simulation experiment results show that this method can achieve dynamic and hunting coverage of the edge of the interest domain under different node scales,node density and shape of the interest domain,and realizes effective tracking and monitoring of the interest domain.The hunting deployment of underwater sensors based on level set method has the advantages of high balance of network energy consumption and short time-consuming algorithm calculation.It can be implemented in water as an underwater sensor node arrangement method.