Changes in the gait characteristics, such as walking speed and stride length, of a person living at home can be used to presage cognitive decline, predict fall potential, monitor long-term changes in cognitive impairment, test drug regimens, and more. This thesis presents a novel approach to gait analysis in a smart-home environment by leveraging new advances in inexpensive sensors and embedded systems to create novel solutions for in-home gait analysis. Using a simple, non-invasive hardware system consisting entirely of wall-mounted infrared and radio frequency sensor arrays, data is collected on the gait of subjects as they pass by. This data is then analyzed and sent to a clinician for further study. The system is non-invasive in that it does not use cameras and could be built into the molding of a home so that it would be nearly invisible. In a finished prototype version, the system presented in this thesis could be used to analyze the gait characteristics of one or more subjects living in a home environment while ignoring the data of visitors and other non-subject cohabitants. The ability to constantly collect data from a home environment could provide thousands of observations per year for clinical analysis. Providing such a robust data set may allow people with gait impairment to live at home longer and more safely before transitioning to a care facility, have a reduced fall risk due to better prediction, and live a healthier life in old age.