Use of Inertial Dead Reckoning and Activity Recognition to Enable Always-On, Low-Power Contextually Aware Applications in Mobile Devices

W. Abdelfatah, J. Georgy, Z. Syed, R. Bakshi, and C. Goodall

Abstract:	Location technologies in smart mobile devices are being utilized not only in traditional foreground applications such as maps and navigation, but also now in a new array of background tracking applications. In foreground applications, location and navigation accuracy is key. The power consumption of the foreground application is less of a concern since the screen of the mobile device is on and dominating the battery usage. In background applications, the device screen and many other system resources are powered down, so priorities shift more toward conserving power of any remaining active applications. One important emerging background application is user context, commonly referred to as "contextual awareness". Contextually-aware mobile applications try to identify three key pieces of information: 1) Where am I? (location); 2) What am I doing? (activity); and 3) Who/What am I close to? (proximity) In order to extract the most value from this feature, app developers and service providers seek to leave the context application "always-on". Inertial sensing solutions play a key role in determining the location and activity parameters of context, not only improving accuracy and coverage, but they also help to reduce overall system power as part of a hybrid location solution with WiFi, GNSS, and Bluetooth Low Energy (LE). In contrast to traditional foreground location applications, user context tracking is on all the time and focuses more on coverage versus navigation. In many cases, coarse location (less accuracy) is acceptable in exchange for more efficient power consumption. This paper outlines the characteristics of a background context tracking implementation covering not only robust, low-power inertial dead reckoning, but also user activity classification.
Published in:	Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014) September 8 - 12, 2014 Tampa Convention Center Tampa, Florida
Pages:	117 - 124
Cite this article:	Abdelfatah, W., Georgy, J., Syed, Z., Bakshi, R., Goodall, C., "Use of Inertial Dead Reckoning and Activity Recognition to Enable Always-On, Low-Power Contextually Aware Applications in Mobile Devices," Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014), Tampa, Florida, September 2014, pp. 117-124.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In