Research On Target Localization And Tracking In Wireless Sensor Network

The wireless sensor network is a set of nodes with processing and communicating capabilities in targetareas to perform sensing monitoring tasks. During the last decade, the researches in this field are gettingmore and more popular. Determining the location of the sensor nodes and targets in WSNs is critical to itsmonitoring activities. There are many limitations such as price, power, precision and operating environment.Meanwhile, the objective physical phenomena, the signal characteristics and the scope of deployment haveincreased the difficulty in target localization and tracking. Therefore, the research of nodes localization andtarget tracking has important theoretical significance for sensor networks.Target localization and tracking is an important technology in wireless sensor network. The accuracyof localization impacts the availability of application in wireless sensor network. This dissertation focuseson the problems in practical application of target localization and tracking technology. The new algorithmsaiming to solve the existing problems have been proposed. The simulation results demonstrate that newalgorithms have a higher accuracy and a more stable performance. The details are as follows:1、As for low positioning accuracy of traditional Scan localization algorithm with low neighbor nodedensity, this paper proposes a Beacon-Based Grid-Scan localization algorithm. The algorithm introduces amobile beacon to generate plenty of virtual beacons while traversing the whole network and then reducesthe estimative rectangle(ER) of normal nodes. The algorithm takes the center of the new ER as an estimatedlocation of the normal nodes.2、As for the over-reliance on the topology of the network in traditional DV-Hop localizationalgorithm, this paper proposes a improved DV-Hop localization algorithm based on multi-mobile-anchorand DV-Hop algorithm. The algorithm incorporates Range-based and Range-free algorithms by introducingmulti-mobile anchors into the DV-Hop positioning network. The anchors keep a certain relative positionwhile traversing the entire network and ranging with RSSI to provide information for unknown nodes sothat these nodes could estimate locations.3、As for the nodes’ limited communication capabilities and energy constraints in the wireless sensornetworks, a distributed target tracking algorithm is proposed based on domain Dynamic Cluster. Thealgorithm utilizes a dynamic clustering approach to divide the nodes into a number of clusters according tothe domain, and enabled the network connectivity between cluster head and gateway nodes in each cluster,then optimizes updating strategy of the cluster. When all above work has done, the activated member nodesin cluster execute distributed particle filter to track the target.