Research On The Acquisition And Removing Random Noise Technologies For Ultrasonic Echo Signal

Ultrasonic testing and imaging technology is a modern nondestructive testing that is developing rapidly. Because ultrasonic imaging technology has high resolution and it is nondestructive to the object, so the technology obtains a widespread application. Ultrasonic echo data acquisition and denoising is the foundation of ultrasonic testing and imaging technology. The only way to insure the precision of ultrasonic testing and imaging system is having original data which has high SNR. So the ultrasonic echo data acquisition and denoising are always hot spot problems.Aiming at these two problems, this thesis carries out the following research work.Firstly, to solve the ultrasonic signal acquisition problem in low SNR and high dynamic range, a multi-channel data acquisition system which is based on distributed bus structure is proposed. The DSP, which is the controller of the system, combined with the network interface chip are utilized to implement the system. Furthermore, the AGC technology based on pre-sampling is applied in the front of data acquisition system to improve the accuracy. The system is tested and the experimental results show that the system has significant value for practical application, and attains to the demand of high-performance system.Secondly, to solve the problem that the ultrasonic echo signal is a time varying and non-stationary signal and it not only includes much useful information, but also contains various noises, an improving wavelet thresholding denoising method is proposed. The selection of thresholding function is the key to this method. A new thresholding function is proposed. The new thresholding function in this method overcomes not only the constant deviation between the estimated wavelet coefficients and the decomposition wavelet coefficients in the soft thresholding function, but also the discontinuance of the hard thresholding function. The new thresholding function is continuous and higher-order derivative. Simulation results confirm that the method is more effective than soft thresholding function, hard thresholding function and other modified thresholding function. It is also effective in low SNR.Finally, the data which is acquired by the data acquisition system is denoised by the proposed wavelet thresholding method. The experimental results show that the modified wavelet thresholding denoising method could achieve a good performance in processing the actual acquired ultrasonic echo signal.