Duke University

Duke University will develop a residential sensor system that uses a dynamic meta-surface radar antenna design to determine occupancy in residential buildings. Traditional line-of-sight movement sensors suffer from high error rates. To increase accuracy, the Duke team will develop a sensor that monitors electromagnetic waveforms that are scattered both directly and indirectly off a person, eliminating the need for a direct line-of-sight between the sensor and the person. The sensor hardware continuously generates distinct microwave patterns to probe all corners of the house. Once a person enters a room, their motion changes the scattering statistics of the environment, which is used to establish real-time room occupancy. These characteristics are then analyzed using machine-learning techniques to establish human presence. The radar antenna can quickly sample an area and this information can be used to distinguish humans with the sensitivity to detect even stationary human’s micro movements such as breathing. Further, the system operates at microwave frequencies, ensuring minimal concern for human safety. The proposed sensor does not require an internet connection or communication links, ensuring minimal security and privacy concerns. If successful, the system promises detection of occupants and near-zero false negative rate without any complex user interactions.