Relative Localization Algorithm Of Wireless Sensor Networks

As a most important support technology in Wireless Sensor Networks (WSN), node localization can greatly influence the performance of WSN. With the development of WSN, the amount of nodes is larger and distribution is wider. As a result, localization algorithm should not only ensure low localization error, but also consume less energy, adapt well to abnormal surroundings and reduce the amount of invalid nodes. With the characteristic of no need to dispose beacon node, relative localization algorithm is fit of the WSN with the restriction of cost, power and node distribution.An Assistant and Distributed Relative Positioning Algorithm (ADRP) is presented to improve energy consumption and complex topology adoption. ADRP is a clustering-based algorithm with forming trilateration and secondary clustering mechanism. The cluster mechanism can improve the localization error with edge nodes; trilateration is used to upgrade new border nodes to assist conversion between node clusters, trilateration can weaken the dependence between single cluster and border node to reduce affection of invalid border nodes; clusters are combined separately by secondary clustering to reduce restriction of conversion between master nodes and number of invalid clusters; meanwhile, the mechanism of secondary clustering can reduces calculation and communication packet number.ADRP can work well in both general and abnormal topology. In stationary scenarios without dynamic nodes, ADRP can ensure high coverage and low localization error, meanwhile; it can obtain low energy consumption and packet number to decrease messages collision. The simulation is made by network simulator NS-2. Compared with the algorithm of clustering-based SPA, the results of simulation show that ADRP performs better in amount of communication packets, node coverage and localization error in irregular topology.