Sensor Placement with Multiple Objectives for Structural Health Monitoring in WSNs

Sensor placement plays a vital role in deploying wireless sensor networks (WSNs) for structural health monitoring (SHM) efficiently and effectively. Existing civil engineering approaches do not seriously consider WSN constraints, such as communication load, network connectivity, and fault tolerance. In this paper, we study the methodology of sensor placement optimization for SHM that addresses three key aspects: finding a high quality placement of a set of sensors that satisfies civil engineering requirements; ensuring the communication efficiency and low complexity for sensor placement; and reducing the probability of a network failure. Particularly, after the placement of a subset of sensors, we find some distance-sensitive, but unused, near optimal locations for the remaining sensors to achieve a communication-efficient WSN. By means of the placement, we present a "connectivity tree" by which structural health state or network maintenance can be achieved in a decentralized manner. We then optimize the system performance by considering multiple objectives: lifetime prolongation, low communication cost, and fault tolerance. We validate the efficiency and effectiveness of this approach through extensive simulations and a proof-ofconcept implementation on a real physical structure.