RRACE: Robust realtime algorithm for cadence estimation

• Published in: Pervasive and mobile computing Vol. 13; pp. 52 - 66
• Main Authors: Karuei, Idin, Schneider, Oliver S., Stern, Bryan, Chuang, Michelle, MacLean, Karon E.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.08.2014; Elsevier
• Subjects: Applied sciences; Cadence; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Empirical evaluation; Equipments and installations; Exact sciences and technology; Information systems. Data bases; Measurements common to several branches of physics and astronomy; Memory organisation. Data processing; Metrology, measurements and laboratory procedures; Mobile radiocommunication systems; Physics; Radiocommunications; Realtime; Smartphones; Software; Telecommunications; Telecommunications and information theory; Velocity, acceleration and rotation; Smartphones; Cadence; Empirical evaluation; Realtime; System architecture; Energy consumption; Gait; Outdoor installation; Mobile handsets; Posture; Localization; Metering; Accelerometers; Legged locomotion; Breaking in; Frequency domain method; Real time; Integrated management; Tuning; Walking; Software packages; Interactive systems; A priori estimation; Delays; Mobile computing
• ISSN: 1574-1192; 1873-1589
• DOI: 10.1016/j.pmcj.2013.09.006

We present an algorithm which analyzes walking cadence (momentary step frequency) via frequency-domain analysis of accelerometer signals available in common smartphones, and report its accuracy relative to the published state-of-the-art algorithms based on the data gathered in a controlled user study. We show that our algorithm (RRACE) is more accurate in all conditions, and is also robust to speed change and largely insensitive to orientation, location on person, and user differences. RRACE’s performance is suitable for interactive mobile applications: it runs in realtime (∼2 s latency), requires no tuning or a priori information, uses an extensible architecture, and can be optimized for the intended application. In addition, we provide an implementation that can be easily deployed on common smartphone platforms. Power consumption is measured and compared to that of the current commercially available mobile apps. We also describe a novel experiment design and analysis for verification of the best-optimized RRACE’s performance under different conditions, executed outdoors to capture normal walking. The resulting extensive dataset allowed a direct comparison (conditions fully matched) of RRACE variants with a published time-based algorithm. We have made this verification design and dataset publicly available, so it can be re-used for gait (general attributes of walking movement) and cadence measurement studies or gait and cadence algorithm verification.