SBIR Phase I: Contact-free RF-based tracking and fall sensor for aging-in-place

This Small Business Innovation Research (SBIR) Phase I project will develop a revolutionary new sensor for aging-in-place and assisted living which uses a network of radio frequency (RF) transceivers as a sensor in a home to track a person?s location and detect falls. The sensor is contact-free, i.e., a person does not need to remember to wear it in order to be monitored. The system monitors links for changes in signal strength which indicate a person being near the line between two transceivers. Algorithms use the link information to infer the location of a person. With transceivers at two different heights, algorithms can distinguish whether a person is standing, sitting, or laying on the floor, and use the speed at which someone moves to the floor to detect a dangerous fall. Room-level tracking can also be used to look for unusual patterns which indicate physical or cognitive decline. This project will develop methods to minimize deployment time and expertise. It will develop novel algorithms that allow for a lower density of transceivers in the system, and still achieve room-level tracking and fall detection accuracy. Live demonstrations will be conducted with potential investors and customers. The broader impact/commercial potential of this project is the ability for the sensor to increase the ability of an elder to safely age-in-place in their preferred living environment, benefitting them and reducing the costs of nursing home care. Unfortunately, many older people fall each year, often when alone, resulting in long delays before help arrives. The fear of falling and being unable to get help is a reason many enter nursing care. Older adults at risk for cognitive decline (e.g., dementia) could live independently longer with sensing systems that help caregivers monitor their well-being. Beyond aging-in-place, these technologies have application in hospitals; smart-building systems for automation and energy efficiency; security systems; retail analytics; and generally for context- and location- aware computing applications. Existing technologies use (1) wearable sensors, which are often forgotten, (2) video camera surveillance, which is seen as too invasive of privacy, or (3) infrared motion sensors, which do not detect falls and are prone to false alarms. In contrast, the proposed sensor does not need to be worn, reliably senses falls, and is privacy-preserving. Our sensor provides capabilities currently unavailable, and will be bought by aging-in-place monitoring and system integrator companies.