A Wigner spectral analyzer for nonstationary signals

The authors present computational algorithms for the design of an innovative spectral analyzer using the Wigner distribution. This spectral analyzer uses advanced digital signal-processing chips with associated hardware and software. Algorithms for the kernel generation, kernel transformation, and discrete Hilbert transform are discussed. Design strategies are focused on achieving high time-frequency resolution, precision, flexibility, and computational efficiency. Experiments show that spectral analysis using the Wigner distribution yields significantly improved results for nonstationary signals.<
>