A robust phase-reversed tone detection using bispectrum and DFT-based algorithms

This article proposes a novel method and apparatus to determine phase-reversal in a tone corrupted by additive white Gaussian noise. The technique presented here uses time-domain as well as frequency-domain based algorithms for proper estimation of phase change in disabling-tone signal. In this technique, instance of the phase-reversal is detected by a time domain algorithm. This is followed by estimation of the phase by a frequency-domain algorithm using bispectrum and DFT-based approaches. The proposed technique is tested for robustness after optimization of various parameters governing performance of the designs. Both, the bispectrum and DFT-based designs are able to detect successfully the presence of phase-reversed tone signal under noise.