An integrated system based on wireless sensor networks for patient monitoring

This works describes the system LAURA which provides patient localization, tracking and monitoring services within nursing institutes through a wireless sensor network. The system is composed of three functional blocks: a localization and tracking engine which performs localization out of samples of the received signal strength and tracking through a particle filter; a personal monitoring module based on bi-axial accelerometers which classifies the movements of the patients eventually detecting hazardous situations, and a wireless communication infrastructure to deliver the information remotely. The paper comments on the design and dimensioning of the building blocks. Two approaches are proposed to the implementation of the localization and tracking engine: a centralized implementation where localization is executed centrally out of information collected locally, and a distributed solution where the localization is performed at the mobile nodes and the outcome is delivered to the central controller. Strengths and weaknesses of the two solutions are highlighted from a system’s perspective in terms of localization accuracy, energy efficiency and traffic loads. LAURA modules are finally tested in a real environment using commercial hardware. The main outcomes are an average localization error lower than 2 m in 80% of the cases and a movements classification accuracy as high as 90%.