| Summary: | 碩士 === 國立清華大學 === 資訊工程學系 === 93 === Sensor deployment is an important issue in designing sensor network. If the sensors can be deployed efficiently in a sense field, it will enhance the monitoring ability of overall environment and lower the cost of the sensor network constructing. The goal of this paper is to study how to deploy sensors to maximize the coverage and minimize the sensors used. We propose an incremental algorithm, DT-Score, for sensor deployment with probability sensing model. In DT-Score, we use the Delaunay Triangulation of Computational Geometry to find the weakest position. Our algorithm additionally considers the influence of signals cut off by obstacles and special area with preferential coverage. To evaluate the DT-Score algorithm, we have implemented the DT-score algorithm along with MAX_AVG_COV and MAX_MIN_COV algorithms. We performed a wide range of simulations and case-studies among them. The simulation results show that, in general, the coverage of DT-Score is better than that of MAX_MIN_COV in some cases and vice versa. Both DT-Score and MAX_MIN_COV produce better coverage than MAX_AVG_COV for most of the test cases. Since the coverage and execution time of MAX_AVG_COV and MAX_MIN_COV are limited by number of grid points and sensor used, in some cases, the coverage of MAX_AVG_COV and MAX_MIN_COV will not increase when the number of sensors used over a threshold. In these cases, by applying the DT-Score after the coverage of MAX_AVG_COV and MAX_MIN_COV reached a saturation point, the coverage will increase. This indicates that the DT-score algorithm can be complementary to MAX_AVG_COV and MAX_MIN_COV algorithms.
|
|---|
