Maximal Lifetime Scheduling for Roadside Sensor Networks With Survivability $k$.

In wireless sensor networks (WSNs), target surveillance has been an important application, particularly in harsh environments. With limited energy of battery-driven sensors, a challenging problem is how to schedule each sensor between active and sleep modes to maximize the network lifetime while meeting surveillance requirements. This problem is often referred to as the wake-up scheduling problem. In this paper, we consider the wake-up scheduling problem for a roadside sensor network that is deployed along a road to monitor the activities on it, such as car accidents, traffic loads and patterns, or enemy activities on battle fields. Previous research has studied a simple case of this problem in which sensors were scheduled so that the entire road could be covered by just one set of active sensors. However, to ensure the reliability of the surveillance tasks, multiple sets of active sensors are often required so that each set can independently cover the entire road. The number of such sets is referred as survivability. This paper studies the wake-up scheduling with survivability k for the road coverage problem. When k >\mbox{1}, the scheduling problem becomes more practical and interesting. We present efficient polynomial algorithms that produce an optimal schedule to satisfy the survivability requirement with the maximum lifetime for both cases k=\mbox{1}$ and k >\mbox{1}. Theoretical proof for the optimality of our algorithms is given, and simulation results further validate the correctness of our algorithms. [ABSTRACT FROM AUTHOR]