RFID-Assisted wireless sensor networks for cardiac tele-healthcare

As the baby boomers head into old age, America will see a dramatic increase in the number of elderly patients admitted to healthcare facilities, such as nursing homes. Due to this rising elderly population, it will be difficult for nursing home personnel to monitor all patients at once. One way to cut down on the amount of supervision by the staff is for patients to administer their own medication. This leads to new problems though, as a patient incorrectly administering one of their many medications could lead to a disastrous end. Technology to wirelessly transmit a patient’s electrocardiogram (ECG) has also been implemented to reduce supervision. Wireless transmissions are infamous for their error rate, but the ECG is a sensitive signal where every second of data matters and cannot tolerate such losses. Additionally, such existing networks employ an expensive communication infrastructure. Due to this healthcare crisis, the ability for a device to remotely monitor a patient’s medication intake and transmit accurate ECG readings, while being cost efficient, is a major innovation. To combat this crisis, this thesis focuses on a multi-hop wireless sensor network (WSN) composed of many wearable sensors, one for each patient, that host a radio frequency identification (RFID) reader and are capable of RF communication. Each wearable device is also assumed to contain an ECG sensor, though this was not implemented in this work. The system is responsible for two distinct features. The first is remotely supervised patient medication intake via RFID and a central workstation/database. The second is the accurate remote transmission of a patient’s ECG using the extended Kalman filter (EKF) for wireless error recovery