Abstract: For achieving boundary coverage of the monitoring region in the WSN, this paper proposes an optimal deployment and selection strategy for sensor node boundary coverage. In the 2D region deployment problem, the coverage region of the sensor node is modeled as a disk and the minimum number of sensor nodes and their locations for border coverage of a given rectangular region are determined. In the 3D region deployment problem, the coverage area of the sensor node is modeled as a closed sphere and the minimum number of sensor nodes and the coverage position of all the points on the surface of the covered cube are determined, so as to find the number and location of the sensor nodes needed for the boundary coverage. In order to obtain better network boundary coverage lifetime, a subset of the resulting sensor nodes is selected to remain active. The simulation results show that the proposed boundary coverage optimal deployment and its selection strategy are superior to random deployment and other boundary coverage algorithms in terms of the required number of sensor nodes and network boundary coverage lifetime.