Recent advancement in wireless communications and electronics has enabled the development of low-cost sensor networks and these networks are finding increased application in the development of intelligent, distributed systems. Distributed deployment of sensor systems has to address issues in power management and efficiency. In this paper, these issues are addressed from the standpoint of optimal deployment of sensors nodes for boundary coverage. Lower bounds on the number of sensors needed for surveillance and target intrusion detection is presented and a “Border Perambulation” technique where each border node can swap between working and sleeping modes and the network only maintains a subset of working nodes is proposed. Mathematical as well as experimental proofs are provided to validate the correctness and efficiency of our algorithms.