Exploiting ST-Based representation for high sampling rate dynamic signals

Description

In this chapter, dynamic wideband signals with high sampling rate are analysed through the discrete Stockwell transform (ST). To this end, a reduced complexity ST-based transform with independent time and frequency resolution is investigated. We call it dual-resolution approach. It results in a new strategy based on a trade-off between the time–frequency resolution and the ST computational time addressing the problem of large amount of samples in wideband signals. Short-time Fourier transform (STFT) representation is also included to discuss the applicability to feature-based methodologies in a Cognitive Radio context. Real modulated signals are generated by a Software-Defined Radio testbed to validate the dual-resolution technique. Just 11–20% of the time necessary to generate the T matrix without dual-resolution is requested and 3–12% of the ST computational time with respect to the conventional ST. While, the number of samples can be till four times larger.

Additional details