On the Gains of Deterministic Placement and Coordinated Activation in Sensor Networks

In this paper we discuss topology design and dimensioning of sensor networks to achieve full sensing coverage. We consider two ways of deploying the sensors, placing them according to some regular pattern or scattering them randomly, and two ways of activating the sensors, optimally according to some predefined schedule or randomly, when each sensor follows a wake-up schedule independently from the other sensors. We provide analytic expressions for the necessary and sufficient number of sensors that guarantee coverage in these scenarios and determine the cases when deterministic sensor placement or optimal sensor activation can achieve significant gains. We consider sensing with bounded delay and show that the number of sensors to be deployed can be decreased significantly even at low sensing delays.