Optimal sensor placement for bacteria detection in water distribution networks

The real-time detection of bacteria and other bio-pollutants in water distribution networks and the real-time control of the water quality is made possible by new biosensors. However, the limited communication capabilities of these sensors, which are placed underground, and their limited number, due to their high cost, pose significant challenges in the deployment and the reliable monitoring. This paper presents a preliminary study concerning the problem of the static optimal sensor placement of a wireless biosensor network in a water distribution network for real-time detection of bacterial contamination. An optimal sensor placement strategy is proposed, which maximizes the probability of detection considering a limited number of sensors while ensuring a connected communication topology. A lightweight algorithm that solves the optimal placement problem is developed. The performance of the proposed algorithm is evaluated through simulations, considering different network topologies using a water pipelines emulator. The results indicate that the proposed optimization outperforms more traditional approaches in terms of detection probability. It is concluded that the availability of a dynamic model of the bacterial propagation along with a spatio-temporal correlation of the process could lead to a more advanced real-time control of the water distribution networks.