Dependable Wireless Sensor Networks for Prognostics and Health Management: A Survey

Maintenance is an important activity in industry as it reduces costs and enhances availability. This can be done either to revive a system/component or to prevent it from breaking down. The increasing need for reliability has led maintenance strategies to evolve from corrective to condition-based and predictive maintenance. The key process of the latter is prognostics and health management, a tool that predicts the remaining useful life of engineering assets. As plants are requested to offer both safety and reliability, planning a maintenance activity requires accurate information about the system/component health state. Usually, this information is gathered through independent sensors or a wired network of sensors. The use of a wireless sensor network has many advantages. First of all, the absence of wires gives sensor networks the ability to cover a large scale surveillance area. Second, it has become possible to monitor hostile and inaccessible areas by simply dropping the sensors from an aircraft to the monitoring region. Finally, the accuracy of measurements can be improved as the sensors can be placed at specific locations without being wired. Even though the deployment of wireless sensor networks is gaining great importance in monitoring applications, there are some research issues that still need to be studied to provide more accurate and reliable data. Indeed, we strongly believe that a good prognostic process starts with a reliable source of information; the wireless sensor network in our case. For this matter, in this paper, we discuss the Wiem Elghazel et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. dependability of wireless sensor networks, we highlight the attributes that have an impact on data accuracy, and present the state of the art in prognostics.