Estimating Contextual Situations Using Indicators from Smartphone Sensor Values

Current context-aware applications often use the location of a user as the only indication of the current situation. These existing applications are therefore limited in their situation awareness, because of the poor indoor resolution of the location sensor and its high resource consumption. In response to these limitations we present an approach to estimate the contextual situation of a user without using resource inefficient location sensors. Our proposed solution utilizes a wide range of low powered sensors, together with two modified machine learning techniques to estimate the situation in a more resource efficient manner. Simulations and a proof-of-concept application show that the situation of a user can be determined within 50 ms at an accuracy above 90%, when only using the low energy sensors available on a smart phone and its limited processing power.

[1]  Erhan Akin,et al.  Estimating driving behavior by a smartphone , 2012, 2012 IEEE Intelligent Vehicles Symposium.

[2]  Fredrik Gustafsson,et al.  Particle filters for positioning, navigation, and tracking , 2002, IEEE Trans. Signal Process..

[3]  Timothy J. Robinson,et al.  Sequential Monte Carlo Methods in Practice , 2003 .

[4]  Yanning Zhang,et al.  Variable kernel density estimation based robust regression and its applications , 2014, Neurocomputing.

[5]  Simon A. Dobson,et al.  Situation identification techniques in pervasive computing: A review , 2012, Pervasive Mob. Comput..

[6]  Mikkel Baun Kjærgaard,et al.  Indoor Positioning Using GPS Revisited , 2010, Pervasive.

[7]  Andy Hopper,et al.  The active badge location system , 1992, TOIS.

[8]  Gregory D. Abowd,et al.  Towards a Better Understanding of Context and Context-Awareness , 1999, HUC.

[9]  Hiroshi Mineno,et al.  A method of estimating outdoor situation for lifelog generation , 2013, 2013 IEEE 2nd Global Conference on Consumer Electronics (GCCE).

[10]  Franco Zambonelli,et al.  Improving Situation Recognition via Commonsense Sensor Fusion , 2011, 2011 22nd International Workshop on Database and Expert Systems Applications.

[11]  Daniel Gatica-Perez,et al.  Discovering places of interest in everyday life from smartphone data , 2011, Multimedia Tools and Applications.

[12]  M. Rosenblatt.  Remarks on Some Nonparametric Estimates of a Density Function , 1956 .

[13]  Gregory D. Abowd,et al.  Cyberguide: A mobile context‐aware tour guide , 1997, Wirel. Networks.

[14]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[15]  Gerasimos G. Rigatos,et al.  Extended Kalman and Particle Filtering for sensor fusion in motion control of mobile robots , 2010, Math. Comput. Simul..

[16]  Kevin P. Murphy,et al.  Machine learning - a probabilistic perspective , 2012, Adaptive computation and machine learning series.

[17]  James J. Little,et al.  A Boosted Particle Filter: Multitarget Detection and Tracking , 2004, ECCV.

[18]  David Garlan,et al.  Context is key , 2005, CACM.

[19]  S. Provost,et al.  A hybrid bandwidth selection methodology for kernel density estimation , 2014 .

[20]  Rahim Tafazolli,et al.  A survey on smartphone-based systems for opportunistic user context recognition , 2013, CSUR.

[21]  Albrecht Schmidt,et al.  There is more to context than location , 1999, Comput. Graph..

[22]  Antonio Iera,et al.  The Internet of Things: A survey , 2010, Comput. Networks.

[23]  C. D. Kemp,et al.  Density Estimation for Statistics and Data Analysis , 1987 .