How lonely is your grandma? Detecting the visits to assisted living elderly from wireless sensor network data

Existing research on the recognition of Activities of Daily Living (ADL) from simple sensor networks assumes that only a single person is present in the home. In reality, the resident receives visits from family members or professional health care givers. In such cases activity recognition must take into account the presence of multiple persons. Here we investigate the problem of detecting multiple persons in a home environment equipped with a sensor network consisting of 13 binary sensors. We collected data during more than one year in our living labs and used Hidden Markov Model (HMM) for a visitor detection. A cross validation method was used to determine the best set of features from the binary data. Using this set of features the detection rate is approximately 85%.