Acoustic Background Spectrum
One of the major parts of my PhD thesis was developing a method for smartphones to determine their location by listening to backgound sounds. This work is described in the following paper:
S. P. Tarzia, P. A. Dinda, R. P. Dick, G. Memik. Indoor Localization Without Infrastructure Using the Acoustic Background Spectrum. In Proc. 9th Intl. Conf. on Mobile Systems, Applications, and Services (MobiSys’11). June 2011. pages 155-168. [slides]
Play the room identification listening game. WARNING: not compatible with Internet Explorer
Data and Matlab code
The raw data (recordings) and Matlab code used to produce the above results are available for download here. If you use this code or data in an academic publication, please cite the above paper. Note that our technique may be patented by Northwestern University, so please contact the authors or McCormick's office of technology transfer if you would like to use it in a commercial application.
Matlab batphone scripts and data v1.0 (1.2MB) (may require some toolkits to run). Please report any bugs or problems to me.
Matlab audio scripts v1.0 (0.4MB) (unfortunately, requires several toolkits to run). Please report any bugs or problems to me. The following data is needed for these scripts:
Untar theses in the same folder as the scripts tarball.