Research On Barrier Coverage In Bistatic Radar Sensor Networks

Coverage is an important performance metric of the service provided by the sensor network. As a fundamental issue of wireless sensor network, the coverage problem of sensor network has being research focusing point in past few years. Depending on the subject should be covered; the coverage problem of wireless sensor network can be divided into three categories:target coverage, area coverage and barrier coverage. The barrier coverage aims at detecting all intruders trying to cross the protected region, which is widely used in homeland security and critical region surveillance applications.The first question to solve all kinds of coverage problem is to determine the coverage model of the sensor. The sensor coverage model of sensor measure the sensing capability of sensor by the geometric relation between space points and the sensor's location, and which mainly depends on the type of sensor. Previous barrier coverage studies are about the wireless sensor network consisting of passive sensors and those passive sensors are based on binary disk coverage model or sector coverage model. In the disk or sector coverage model, the coverage area of passive sensor is the disk region or the sector region centered at sensor location. Compared to the passive sensor, the target detection performance of active sensors like bistatic radar sensor is superior to that of passive sensor, and the superiority is more and more obviously with the development of radar technology. However, previous barrier coverage algorithms can't be applied to the barrier coverage constructed by active sensors since the coverage model of active sensors like bistaitc radar is much different from that of passive sensors.Therefore, for the bistatic radar sensor network based on Cassini Oval coverage model, the barrier coverage of bistatic radar sensor networks is the research focus in this thesis. The main research achievements of this thesis include:The radar sensor nodes deployment issue for perimeter barrier construction is first studied in this thesis. Perimeter barrier coverage has a wide range of applications. In order to construct perimeter barrier coverage, all radar sensors should be deployed on the curve enclosed protected region to detect intruders from either entering its interior or exiting from it. In this thesis, the problem how to use the bistatic radar to construct perimeter barrier coverage has been addressed. For the case when all radars can be determinedly deployed on the boundary, an algorithm has been proposed to compute the number of transmitters and receivers and the optimal locations of those radar nodes on the boundary with objective of minimizing the total deployment cost. We next investigate how to use mobile radar sensors to constuct perimeter barrier coverage. The solution to this problem includes two steps:in the first step, two algorithms are proposed to determine the target locations of all mobile transmitters and receivers on the boundary; in the second step, two bipartite graphs are constructed to decrible the relation between radar's initial locations and target locations. With the objective of minimizing the total moving distance or maximal moving distance to save the total energy, the target location for each radar node is determined by finding the optimal match and max flow of the bipartite graph. The simulation results validate the effectiveness of proposed algorithms.The optimal radar nodes deployment issue for belt barrier coverage construction has been addressed in the second part of the thesis. In order to improve the intruder detection probability, the problem how to construct belt barrier coverage by bistatic radar sensors with barrier breadth no smaller than predefined threshold has been studied. Line-based equipartition placement strategy has been proposed to solve the problem where all transmitters and receivers are placed on the deployment lines parallel to the long side of protected region. Depending on the coverage requirement, an algorithm has been designed to determine the number of deployment lines, the number of transmitters and receivers in each deployment line and the optimal placement strategy of those nodes in each deployment line to statisfy the coverage requirement while the total placement cost can be minimized. The simulation results show that the proposed algorithms can save the total deployment cost according to the unit cost ratio between transmitter and receiver.In the last part of thesis, the radar nodes scheduling problem in randomly deployment network has been invstigated, where the network can provide the barrier coverage. Existing the redundant sensors on the randomly deployment network, two solutions have been designed, nodes non-disjoint and disjoint barrier lifetime maximization solution, to maximize the barrier coverage lifetime of the bistatic radar networks. Note that the nodes non-disjoint barrier solution also considers the energy consumed by sensors switch process. Due to the variety of bistatic radar coverage area, a barrier coverag graph is constructed to decrible the relation among different bsitatic radars'covrage area. Based on barrier coverage graph, an algorithm is proposed to find all possible barriers in the nodes non-disjoint barrier solution, then linear programming method is used to determine the operation time for each non-disjoint barrier. While in the disjoint barrier coverage solution, disjoint barriers are found from the nodes non-disjoint barriers set with the objective that the total lifetime of disjoint barriers can be maximized. For the large-scale network, two heursitc algorithms, greedy algorithm and random algorithm, are proposed to prolong the barrier coverage lifetime, the basic idea of the algorithms is to find one barrier at a time and activate the found barrier for some unit times. The simulation results indicate that the proposed algorithms can improve the barrier coverage lifetime significantly and the linear programming method outperforms the heuristic algorithms.