The demand for analyzing nonstationary signals has resulted in the continuous use of time-frequency techniques. Traditional time-frequency techniques such as the short-time Fourier transform (STFT) and the Wigner-Ville distribution (WVD) have such limitations as the fixed resolution and cross-term interference. The latest development of wavelets results in the study of time-scale distributions, whose advantage is the varied time-frequency resolution.; In such DSP applications as the inverse synthetic aperture radar (ISAR) imaging, the conventional Fourier method results in a blurred image due to its inability to deal with the time-varying frequency. To obtain a clear image with a high resolution, we proposed the use of wavelet-packet-based cross-term deleted representation (WPCDR). When this representation is applied to the ISAR imaging, a much sharper image is obtained.; In the detection of multiple linear frequency-modulated (LFM) signal, the Radon-ambiguity transform (RAT) which combines the ambiguity function and the Radon transform is proposed. This method reduces the 2-D problem of the Radon-Wigner transform (RWT) based detection to a 1-D problem, and consequently reduces the computational load.; Motivated by the need to develop a more efficient encoding method for magnetic resonance imaging (MRI), a dynamic wavelet encoding technique is proposed. With this technique, the region of interest (ROI) is determined and then updated by wavelet encoded imaging. Images can be therefore generated at a rate faster than standard phase encoded images, and with no sacrifice in resolution.
展开▼
