Weak Signal Detection Based On Chaos

This thesis is mainly concerned with problems of weak signal detection based on chaos and chaotic A/D converter.In the area of automation, electronic counter-measure, information reception in the communication engineering, radar signal detection in the department of defense, signal processing in biomedicine, remote signal detection in the earthquake and etc, weak signal detection in the background of strong noise has always been the kernel problem and one of the difficulties. Since chaotic system is of the characteristic of sensitivity to weak signals and immunity to noises, it makes chaos has great application potential in the area of weak signal detection. This paper analyses existing methods of weak signal detection based on chaos and their feasibility. Aiming at their deficiencies, two new detecting methods has been proposed. First, weak signal detection method based on image recognition chaos, which combines image processing with chaos detection method, is a method of quantitative analysis on the basis of qualitative analysis. It avoids the involvement of human beings, improves working efficiency, has certain value of utility and provides a new way of distinguishing characteristics of chaos by computers.Simulation results proved the feasibility and effectiveness of the method. Second, weak periodic signal detection method based on Lyapunov exponent. Firstly, setting the threshold value by the calculating method of Lyapunov exponent based on models, then using the data with noises to do the phase space reconstruction and using the improved method to compute the Lyapunov exponent, according to which to judge the large scale periodic state. The characteristic of this method is the high precision of the threshold value's setting and the realization of recognize automatically. Simulation results prove the correctness and validity of the theoretical analysis.The A/D converter is the bridge from analog signal to digital signal and the key and bottleneck of the development of electronic technology. The existing A/D converter have a high detection threshold to weak signals and the highest resolution is 24 bits but the price is very high. While chaotic A/D converter is of high resolution, high sensitivity and high precision from the point of theory and the measurement range, resolution and sensitivity are variable in the same measurement system, which meet current developing requirement of A/D converter. This thesis studies the structure and working principle of several chaotic A/D converter on the basis of previous work and proves the correctness of theoretical analysis by the simulation of OrCAD software. At last, one circuit model has been chosen to realize by hardware and experiment results have been got to show its practicability, which make a forward step from chaotic A/D converter to practical use.