Self-localization Algorithm For Wireless Sensor Networks

Wireless sensor network (WSN) is considered as one of the most important technologies in 21 centuries, which has been widely applied in the aspects of military, environment and industry. The sensor localization technology causes great attention due to its special role in the WSN. This paper focuses on the problem, first quantitatively analyzes the typical localization algorithms, and then proposes an improved algorithm based on DV-Hop.DV-Hop algorithm, which is range-free, has disadvantages of low positioning accuracy and coverage, although it saves the costs of hardware. In order to remedy the deficiency, an improved algorithm is proposed that locates the nodes and updates them to new anchors in terms of the PRI. This algorithm adds a broadcast of the new anchor from second round, then confirms the PRI of unknown nodes according to the number of their received broadcasts from anchors, and chooses the node with the highest PRI and locates it by DV-Hop, and then updates it to a new anchor; meanwhile, it establishes a coordinate algorithmic model based on two anchors and a connection constrain condition. The simulation results show its effectiveness.Then we apply WLSE (Weighted Least Square Estimates) to node localization because of the total cumulative error in the DV-Hop. According to every anchor's position accuracy, the node updating to anchor will be endowed with a reasonable weight which is used to form the weighted matrix, and the node's position will be computed with WLSE during the next round of localization. The experiment indicates that it is really available for the algorithm to decrease the cumulative error.