Complex Noise Background Signal Detection And Extraction Technology Research

Signal detection and extraction technology is not only an important part of signal processing and modern information theory, but also one of the underpinnings of information technology. And many novel signal detecting and extracting technology emerges under this kind of mutual influence stimulation, and its application field is expanded as well. Based on the research and work of pre-scholars, the main work and innovations are as follows:1. Introduction of some relative theoriesIntroduce two nonlinear theories, the Stochastic Resonance and Chaos Phase Transformation including their originations, developments, basic principles, applications. And also the definitions and the characteristics of the Higher-order Statistics and Cyclic Statistics Theory are introduced as well as their recent developments, applicability, etc, providing theoretical foundation for the possibility of resolving the problems of signal detection and extraction involved in this thesis.2. Presentation of a nonlinear hybrid method for detecting and extracting weak signal Different from former thesis, which never thought of using the effects of the OddMultiple Frequency phenomenon in the Stochastic Resonance, this thesis got a set of frequency by studying the Odd Multiple Frequency phenomenon, and the real frequency of the object signal must be included in it. The amplitude at each Odd Multiple Frequency decrease at an exponential mode, so the number of the visible wave peaks is very small. To filter the falsehood frequency existing in the Odd Multiple Frequency phenomenon, each frequency included in the group is set as the reference frequency of a chaotic detector. By using the frequency selectivity of the chaotic detector, the real frequency of the object signal is selected out. Experiments show that such weak signal, whose SNR is as small as -42.0412 dB, can be detected by using this method. Additionally, this scheme can be implemented easily via hardware. So it will be a promising method to detect weak signal.3. Presentation of a nonlinear coupling analysis method based on special fourth-order time-average moment spectrumBecause it is not easy to know the statistic characteristics of noise, the problem about the cubic nonlinear coupling analysis hasn't been solved when the additive noise and...