非平稳随机信号处理中若干问题的研究

This project studies the common problems in the signal processing for both nonstationary and stationary random signals. The unified solutions for the four kinds of difference equations describing discrete time systems and the unified models for the transformation from the discrete time system to the continuous time system for both stationary and nonstationary random signals are proposed. By studying the theory for the time varying filters, this project makes progress in the linear periodic time varying system and the estimation of its output period, and the linear time varying parameter model. The project also studies the adaptive time delay estimation under the fractional lower order -stable noise conditions, and proposes some new robust algorithms under both Gaussian and non-Gaussian noise conditions.

（1）研究随机与时变信号处理中共性问题，提炼出统一的理论方法；（2）研究时变情况下自适应时延估计和时频分布的自适应核心函数，增强韧性性能和适用范围；（3）研究非平人婊藕诺姆遣问Ｐ偷牟问兰疲唬?）研究非平稳随机信号在脑电信号分析处理中的应用，改善基于脑电的临床诊断技术的准确性和可靠性。