An ADL Recognition System on Smart Phone

摘要

Multiple kinds of sensors in smart homes have been used successfully and widely on various pattern recognition tasks. In order to detect user's activities of daily living (ADLs), an array of sensors have to be installed in many places in a smart home or armed upon a user's body. Here, we present an approach for collecting and detecting activities data only via a smart phone, which largely reduces the cost of setup in a smart home and energy consumption. To the best of our knowledge, this study represents a pioneering work where a single-point smart phone is used to capture ADLs. The ADLs indoor are recognized by analyzing the data combination of sound, orientation, and Wi-Fi signals. This study engages real-life data collection, and the results from four test environments show that all of the ADL recognition rates are above 90%.