Abstract
In this paper, we study a new problem called probabilistic A-weighted coverage placement, which is a generalization of the ß-coverage placement. ß-coverage placement assumes that the monitored area has uniform coverage requirement: events within the monitored area should be detected with probability Q X100% (i.e., detected by Q sensors), while probabilistic A-weighted coverage placement assumes that the monitored area has A-degrees of coverage requirement (i.e., A kinds of detection probability): events within distinct region of the monitored area are detected with distinct detection ability, i.e., one of the A kinds of detection probability. Besides, Q-coverage placement requires that the coverage requirement is integer multiple of 100% detection probability, while probabilistic A-weighted coverage placement allows coverage requirement to be non-integer multiple of 100% detection probability. We derive a lower bound on the number of sensors needed to satisfy the coverage requirement of a probabilistic A-weighted monitored area, and introduce a greedy algorithm to solve the problem.