Semantics in Mobile Sensing

摘要: The dramatic progress of smartphone technologies has ushered in a new era mobile sensing, where traditional wearable on-body sensors are being rapidly superseded by various embedded our smartphones. For example, typical today, at the very least GPS, WiFi, Bluetooth, triaxial accelerometer, and gyroscope. Alongside, accessories emerging such as proximity, magnetometer, barometer, temperature, pressure sensors. Even default microphone can act an acoustic sensor to track noise exposure for example. These "lens" understand user's context along different dimensions. Data be passively collected from these without interrupting user. As result, this sensing fueled significant interest understanding what extracted data both instantaneously well considering volumes time series GPS logs used determine automatically places associated life (e.g., home, office, shopping areas). may also reveal travel patterns, how user moves one place another driving or using public transport). proactively inform about delays, relevant promotions shops, his "regular" route. Similarly, accelerometer measure average walking speed, compute step counts, gait identification, estimate calories burnt per day. key objective is provide better services end users. book reader methodologies techniques extracting meaningful information (called "semantics") on form cornerstone several application areas utilizing data. We discuss technical challenges algorithmic solutions modeling mining knowledge smartphone-resident streams. This devotes two chapters dive deep into set highly available, commoditized sensors---the positioning (GPS) motion (accelerometer). Furthermore, chapter devoted energy-efficient computation semantics, battery major concern experience. Table Contents: Acknowledgments / Introduction Semantic Trajectories Positioning Sensors Activities Motion Energy-Efficient Computation Semantics Conclusion Bibliography Authors' Biographies