Localization Algorithms In Wireless Sensor Networks

As a new type of wireless ad hoc network, Wireless Sensor Network has high value and wide applications in military, industry and agricultural, environmental monitoring and other numerous fields. Nowadays, it is becoming a hot topic in current academic circle. In many applications, The location of network nodes is basic and key question in WSN. So lucubrate WSN localization technology is a far-reaching work.Based on wireless sensor network node location technology is studied, the research results obtained as follows:1. We give a method of genetic simulated annealing algorithm by combining with the RSSI ranging technology orientation.Because the node localization is an optimization problem in nature., for the features of wireless sensor networks, a localization algorithm based on GASA(genetic and simulated annealing) is presented to solve the accuracy and efficiency in localization algorithm. This algorithm is used in DV-Hop ranging which does not need range measurement, but DV-Hop ranging technology application limitations is greater. This technique requires uniform distribution of network topology and in such conduction the correction can be reasonably estimate the distance between hop and hop. Obviously, for the actual environment, this assumption is unrealistic. In this paper, In this paper, GA and SA also are discussed deeply.GA has a strong global searching property but it prones to"premature convergence"phenomenon and runnes into local optima.SA has the ability to avoid local optima, but it has the problem of slow evolution velocity. For the two problem above,GASA optimization algorithm is obtained by introducing Metropolis accept criterion SA into GA selection strategy. This can improve the diversity of population and avoids GA falling into the local optima.Finally,GASA is applied to wireless sensor network localization problem and we obtain the localization algorithm. In this paper, the traditional GASA localization method,GASA-HOP localization method and GASA-RSS localization method are simulated and compared respectively by Mat lab simulation tool. The performance of the algorithm is analyzed.2. We give a method of RSSI distributed localization algorithmGenerally speaking, WSN has a huge number of nodes. And the nodes are distribution range and irregular, centralized controlled. This paper also presents a self-subnet mapping the unknown node clustering algorithm, can be used for distributed localization in WSN. First, according to geographical location awareness anchor region, then the unknown node and anchor node information exchange, anchor nodes to achieve the perception of the surrounding network topology, while the unknown node can communicate to accept all the anchor nodes within the RSSI value, RSSI value based on the size of the sorting and space area as do maps, in order to achieve self-clustering of the unknown node method. At the same time the master node of each cluster to obtain cluster node connectivity information and communication between nodes can be RSSI values, and finally with the linear programming method to achieve the positioning of the cluster nodes.