Barrier coverage with airdropped wireless sensors

Barrier coverage of a wireless sensor network aims at detecting intruders crossing the network. It provides a viable alternative for monitoring boundaries of battlefields, country borders, coastal lines, and perimeters of critical infrastructures. Early studies on barrier coverage typically assume that sensors are deployed uniformly at random in a large area. This assumption, while theoretically interesting, may be unrealistic in real applications. We take a more realistic approach in this paper. In particular, we consider that sensors are airdropped from an aircraft along its flying route. We note that wind, geographic terrain, and other factors may cause a sensor to land in a location deviating from its targeted landing point with a random offset. Thus, it is more realistic to assume that sensor nodes are distributed with a normal offset along the deployment line. Through extensive simulations, we study how variance of the normal distribution, the number of deployment lines, and the distance between adjacent lines would affect barrier coverage. We then investigate how a strong barrier can be formed efficiently from airdropped sensors and compare it with barrier coverage using uniformly distributed sensors. Our results show that the barrier coverage, with appropriately chosen deployment parameters, can be significantly improved using normally distributed sensors.