Maximum amplitude sensitive high resolution time-frequency analysis

Various time-frequency analysis methods for digital audio signals proposed in literature are examined. A new high resolution time-frequency analysis method based on Short-Time Fourier Transform (STFT) is proposed. Short-Time Fourier Transform (STFT) has been developed to overcome the time ambiguity problem introduced by Fourier Transform. Unfortunately, STFT offers constant value for the multiplication of the time and frequency resolutions, namely; time resolution degrades the same amount if the frequency resolution is improved. There is a need for better time-frequency analysis when both time and frequency resolution is to be improved which should have better resolution factors than STFT. In this work, Maximum Amplitude Sensitive High Resolution Time-Frequency Analysis method is proposed for improved resolution in both time and frequency domains. It is based on the well known STFT, taking the advantages of STFT in varying time windows with synthesizing technique to present high resolution. The novel method is implemented in MATLAB® software development environment. The performance of method is measured and compared with the methods proposed in literature. The numerical results show that an improvement of %52 is possible in time-frequency analysis.