Author:  Li, Jingjing 
Title:  Coverage and data aggregation for object tracking in wireless sensor networks 
Degree:  Ph.D. 
Year:  2013 
Subject:  Image processing. Linear programming. Programming (Mathematics) Hong Kong Polytechnic University  Dissertations 
Department:  Dept. of Computing 
Pages:  xviii, 145 p. : ill. ; 30 cm. 
Language:  English 
InnoPac Record:  http://library.polyu.edu.hk/record=b2639098 
URI:  http://theses.lib.polyu.edu.hk/handle/200/7078 
Abstract:  With the development of sensing and wireless communication technologies, more and more applications are designed for wireless sensor networks (WSNs). Object tracking is one of the most important applications, which senses the location and movement of the target and delivers the sensed data to the users. It has been proved challenging to design algorithms, protocols, and systems for object tracking in WSNs. On one hand, wireless sensors have limited sensing range, communication range, storage, energy, and other resources. Therefore, a single sensor node cannot sense the whole area being monitored or store many data. On the other hand, the continuous tracking requires the algorithms and protocols designed for calculating the location of the target and delivering the data to be simple but accurate. Therefore, there is a tradeoff between the accuracy and the complexity of algorithms and protocols. The process of object tracking can be divided into three stages: deployment stage, detection stage, and tracking stage. There are many research problems in the three stages of object tracking such as sleeping schedule, coverage, routing, data aggregation, prediction of the location of target, etc. The reliability and the efficiency of the solution are limited by energy, sensing range, and storage capability. In this thesis, we research on the challenging issues in designing algorithms to improve the performance of coverage, routing, and data aggregation for object tracking in WSNs. The results of this thesis research can be described in three parts as follows. In the first part, the sleeping schedule and coverage maximization are studied, answering questions of how to reduce the number of sensors being used and how to enhance the reliability of the network. We focus on how to find the maximum number of the disjoint cover sets, each can completely cover the given area, and how to identify the corresponding sets of sensors. We develop a localized approach to designing the evolutionary algorithm (EA) in distributed systems. A scheme, named localized evolutionary algorithm (LEA), is proposed to solve optimization problems that can be divided into several subproblems. LEA is different from parallel EA where the global information is required for searching the solution space and consequently, consumes a large amount of time and space resources. In LEA, a set of processors solve the global optimization problem in a cooperative way. First, each processor is assigned to solve a subproblem based on locally collected information. Then the locally optimal solutions obtained by all the processors are combined to obtain the globally optimal or near optimal solution to the whole problem. We analytically discuss the optimal number of subproblems that the problem should be divided and the performance metrics for evaluation the proposed algorithm. Next, we describe the optimization problem that the proposed algorithm that are suitable for being solved by LEA. The characteristics of the suitable optimization problems are also summarized. We apply the proposed LEA to solve the sleeping schedule and coverage in WSNs. Simulations have been carried out to validate the effectiveness of our proposed algorithm in terms of the number of disjoint sets, storage consumed in each node, and calculation time. In the second part, we study the coveragepreserving routing problem, i.e., how to find a routing path in a WSN with the maximum sensing coverage provided by the nodes on the path of object movement subject to the delay constraint. In most of the WSN applications, covering the area of interest and delivering the sensed information to the sink are two fundamental functions. Extensive research associated with these two issues, such as energy efficient coverage and delayconstraint routing, can be found in the literature. However, few works combine these two issues together. Considering the fact that wireless sensors can take the responsibility of both sensing and routing, it is expected that a solution jointly considering these two issues will provide more benefit. We first define the coveragepreserving routing problem and prove this problem is NPhard. Then two coveragepreserving routing algorithms, one centralized and another distributed, are designed. The two algorithms are based on label setting algorithm (LS) and genetic algorithm (GA) respectively. Both algorithms are based on MonteCarlo integration to approximately calculate the sensing area. Then the sensing area between the nodes is regarded as the weight of edges to find the routing path from source node to sink node. Simulation results show that the proposed algorithms achieve better performance than other previous algorithms in terms of sensing area of routing path as well as the number of hops in the routing path. In the third part, we apply our studies on dynamic construction of data aggregation tree to track the location of the target. We first consider the sensor set selection scheme in a fully distributed way. Based on the prediction model, the current active sensors wake up the next set of sensors which are around the predicted location of the target independently. After obtaining the set of sensors to be active, the data aggregation tree from the set of sensors to the sink node is constructed quickly and energy efficiently. When the target is moving, the tree will be reconstructed in real time. Simulation results show that the proposed algorithm achieves much better performance than other methods for constructing data aggregation tree, in terms of the number of active nodes involved and the time of construction of the tree. 
Files  Size  Format 

b2639098x.pdf  1.993Mb 


As a bona fide Library user, I declare that:  


By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms. 