Prediction of mobile context parameters.

Several human dependent contexts have patterns. Many of these patterns are governed by different factors such daily schedule of the users, mobility of the user.  Learning these patterns several operations in MCS can be managed efficiently and intelligently taking necessary adaptive actions proactively. Various techniques such as machine learning, stochastic modelling, time series methods can be employed.

Perpetrator tracking:

  1.  Perpetrator tracking app involves obtaining images/video data from nearby users as perpetrator moves from one location to another.
  2. This app predicts mobile users location, connectivity, stationary, battery level using various techniques such clustering, stochastic methods, time series analysis, regression analysis.

360 View of Stadium Event

Top view of a stadium showing location of different users located in different parts of the stadium.
  1. This application involves obtaining a better view of stadium using video/ images from users that are located in better seats. 
  2. Prediction of  mobile user contexts  such as location, connectivity,  battery level, and movement helps to improve the selection of the devices. Other context parameters that is considered is the orientation of camera. 
  3. Prediction estimated helps to select a device that will support execution of MCS application with minimal interruption.


Using the brain signal the nervousness  is evaluated for a group of users. The movie parts are blurred based on nervousness signal in Neuro Movie.
  1.  Users in this application are watching a horror movie. The movie scenes are blurred based on the nervousness state of the viewers. 
  2. Users mental state, movement, restlessness is measured and evaluated using machine learning algorithms