A common method used by the indoor location systems (ILS) is the Fingerprinting scheme. This commonly is
composed of two phases 1: training and location determination. First, a radio map of observed signal strength values
from different locations is recorded during a training phase. Then, in the position determination phase, the signal
strength values observed at a user device are compared to the radio map values using proximity matching algorithms,
such as k-NN2 and other classifiers 3,4.
In several research works, a reduction of the fingerprint size has been attempted to improve the performance of
the system, and increase its capacity to storage signal information. For instance, Kamaladas et al. 5 uses a wavelet
transform to extract the main features of audio files in order to increase the capacity of their song recognition system,
while Manjunath et al. 6 use a Fast Fourier Transform (FFT) to extract features from an image to improve the response
time of their system.