Minimising coverage holes for prediction-based mobile object target tracking in wireless sensor networks

The use of small, cheap, networked devices to perform a collaborative task presents an attractive opportunity in many scenarios. One such scenario is tracking the coverage hole in sensing area and tracking an object moving through the region of interest in Wireless Sensor Network (WSN). This thesis presents a new framework for tracking coverage hole and mobile object in wireless sensor network. Existing algorithms have been integrated and extended with this framework to perform target tracking for coverage hole in the Region of Interest (ROI) and tracking mobile object whilst managing accuracy of tracking and energy usage, in order to increase the lifetime of a sensor network. The node dies after energy depletion, and such consequence causes holes in the coverage area to be created. In a sensor network with movement capabilities, nodes can be moved to make redeployment. Hence, work presented in this thesis could be useful to WSN algorithms developers, and also to people who plan the deployment of nodes in a region of interest. This thesis focuses on tracking moving objects scenarios, but the research presented is not limited to this. The aim is to increase the coverage environment through smaller mobility of sensor nodes, such that mobility distance is limited to 1-hop, and coverage of the nodes are not increased. By securing full coverage with no coverage hole, accuracy of target tracking can be improved. Novel improvement is presented here in performing target tracking of mobile objects where the target moves in a linear fashion. This thesis proposes a prediction-based scheme called Face-based Target Tracking Technique (FTTT) to minimize energy depletion and prolong the lifetime for sensor node while accurately tracking mobile objects. This strategy of tracking mobile objects will influence accuracy of mobile objects tracking. In addition, we also introduce schemes for object tracking recovery to figure out the problem of object tracking loss. Through analysis from the simulation results, our presented prediction based scheme for tracking movable object can efficiently preserve energy and perfectly attain the aim of tracking mobile objects concurrently and accurately. From the simulation results, the proposed coverage hole algorithm shows a coverage gain of up to 14%. However, on average, the proposed vector space algorithm shows 3% to 4% improvement in the percentage coverage, as compared to the previous algorithm. In the scenario of target tracking, the proposed prediction-based scheme for tracking mobile object can efficiently preserve energy and perfectly attain the aim of tracking moving mobile object concurrently. The proposed prediction-based optimistic object tracking scheme can save up to 24% energy consumption as compared to the related works. The accuracy of FTTT is still higher than 98%, which shows that optimistic design does not influence the tracking accuracy. Thus, FTTT appear to successfully conserve energy and accomplish the objective of moving objects tracking.

Text FK 2020 1 ir.pdf Download (1MB)

Actions (login required)

View Item