The sensor placement problem considered in this paper is to determine optimal sensor configurations satisfying coverage preferences at a minimal cost. This problem is formulated as the binary linear programming problem. The generic formulation is generalised to include several realistic demands, such as a finite sensor supply, single hop wireless sensor communication, and multiple-sensor overlapping coverage. Communication and detection are considered in a probabilistic framework. An exact solution to this problem is computationally demanding (NP-complete) and practically impossible for realistic scenarios. Therefore, a fast algorithm for approximate solutions is used. To evaluate the quality of the approximate solutions, they are compared with exact solutions for cases when the exact solutions are obtainable. The results demonstrate that the approximate solutions are highly satisfactory regardless of the complexity of problem constraints, while an exact algorithm can be impractically slow.