| In this study, we present novel applications of time-frequency analysis to spread spectrum based communication and audio watermarking systems. Our objective is to detect and estimate non-stationary signals, such as chirps, that are characterized by directional elements in the time-frequency plane. Towards this goal, we model non-stationary signals using the matching pursuit decomposition algorithm, generate a positive time-frequency representation of the signal model using the Wigner-Ville distribution and estimate the energy varying directional elements using a line detection algorithm based on the Hough-Radon transform.; In this thesis, we develop a new interference excision algorithm for spread spectrum communication systems based on the directional element estimation algorithm.; Watermarking is the process of embedding imperceptible data into the host signal for marking the copyright ownership. The embedded data should be extractable to prove ownership. In this thesis, we develop a spread spectrum audio watermarking scheme, where we embed perceptually shaped linear chirps as watermark messages. (Abstract shortened by UMI.)... |
