Methodical activity recognition and monitoring of a person through smart phone and wireless sensors

Abstract

Sophisticated Smart phones with ever-growing computing, networking, and sensing powers has been changing the landscape of people's daily life and has opened the doors for many interesting applications, ranging from i) Health and fitness monitoring ii)Personal biometric signature[1], iii) Elder-care and Assistance iv)App Activity Log v) Indoor localization and navigation, etc. Human activity recognition is a core building block behind these applications. It takes the raw sensor reading as inputs and predicts a user's motion activity. Many main stream smart phones are equipped with various sensors, including accelerometers[2], GPS, light sensors, temperature sensors, gyroscope, barometer, etc. These sensors have become a rich data source to measure various aspects of a user's daily life. The typical activities include walking, jogging, sitting, etc. Due to its unobtrusiveness, low/none installation cost, and easy-to-use, smart phones are becoming the main platform for human activity recognition.