Sensing platform and object motion detection based on passive UHF RFID tags using a hidden markov model-based classifier

For context-aware systems in indoor work settings, several types of sensors have been applied to capture work activities. We introduce and present a sensing platform and object motion detection system using a hidden Markov Classifier based on a UHF RFID system. Backscattered signal strength of passive UHF RFID tags as a sensor is used for providing information on the movement and identity of work objects. As the read range of passive UHF RFID broadens up to 12 meters compared to 1-meter range of HF RFID, passive tags have been used for many applications such as tracking medical devices and objects of daily living. The RF communication link between the reader antenna and tags for indoors exhibits intermittent loss of signal reception due to antenna orientation mismatch and breakpoints within the antenna coverage area. We propose a design of a sensing platform for tracking objects using a UHF RFID system with passive tags that provides continuous signal reception over the coverage area. We first investigated causes of power loss for passive tags and then designed a sensing platform solution using antenna diversity. The causes of tag’s power loss were eliminated with angle and spatial diversity methods that can cover an arbitrary area of interest. We implemented this design in an indoor setting of a trauma resuscitation room and evaluated it by experimental measurement of signal strength at different points and angles in the area of interest. Our sensing platform supported complete coverage and uninterrupted interrogation of tags as they moved in the area of interest. We conclude that this sensing platform will be suitable for uninterrupted object tracking with UHF RFID technology in generic indoor spaces. In addition to the sensing platform, we design an object motion detection system using passive UHF RFID tags attached on medical objects. To use the signal strength for accurate detection of object movement we propose a novel hidden Markov model with continuous observations, RSSI preprocessor, frame-based data segmentation, and motion-transition finder. We use the change in backscattered signal strength caused by tag’s relocation to reliably detect movement of tagged objects. To maximize the accuracy of movement detection, an HMM-based classifier is designed and trained with dynamic settings, and different object types. We deployed an RFID system in a hospital trauma bay and evaluated our approach with data recorded in the trauma room during 28 simulated resuscitations performed by trauma teams. Our motion detection system shows 89.5% accuracy in this domain.Ph.D.Includes bibliographical referencesby Young Ho Le