The deployment strategy for achieving differentiated coverage and probabilistic connectivity in wireless sensor networks is studied in this paper. A novel solution based on elitist non-dominated sorting genetic algorithm (NSGA-II) is proposed. Simulation results show that NSGA-II based strategy can meet the desired coverage requirements and maintain connectivity in a probabilistic manner with a relatively small number of sensors. In addition, for the applications in which the coverage requirement varies in some subareas, a local genetic operation is more time efficient and needs less variation in the original disposal than a renewed global optimization.