Subsample Time-delay Estimation Algorithm Research Based On Correlation Of LFM

With the rapid increasing proportion of non-destructive testing technology in the production daily,ultrasonic detection has a wide range applications in the fields of industrial flaw detection,precision instrument detection,underwater ultrasonic positioning,medical testing,etc.due to the advantages of simple requirements of equipment,strong operability,strong penetration,and no harm to human bodies.The part of parameter estimation such as time delay estimation is an important technical field in ultrasonic detection.According to Shannon's theorem,a signal with a large time-width-bandwidth product has higher signal noise immunity,a larger signal duty cycle,and better information carrying efficiency.Therefore,it is a development trend to use linear frequency modulation(LFM)as a detection signal in the field of ultrasonic detection.The most commonly used echo detection method for LFM signals is matched filtering,which is essentially based on cross-correlation functions.However,in the digital echo detection,the discrete sampling process will bring a certain discrete estimation error of the peak of the cross-correlation function.It is found that the cross-correlation function of the LFM is a sinc function by mathematical modeling.On the condition,two new time-delay estimation algorithms finding the true peak of the discrete cross-correlation function of the transmitted and received signals are carried out.The main work of this paper include:1.The study analyzes the existing time-domain and frequency-domain time delay estimation algorithms which based on the direct correlation function method.Performs simulation and comparison to find why the direct correlation method has discrete peak estimation error and cannot obtain the sub-sample delay smaller than the sampling interval.2.Using the essence of the LFM signal cross-correlation function as the sinc function,the correlation sequence obtained by sending and receiving signals is approximated by the least square method in amplitude with the theoretical sinc function,so as to solve the sub-sample time delay and find the true peak of the discrete correlation function.A subsample time delay estimation algorithm(Amplitude-based Subsample Time-delay Estimation,ASTE)based on cross-correlation amplitude approximation is presented.3.Based on the phase of the LFM signal cross-correlation function,the correlation sequence obtained by the receiving signal and the theoretical correlation function are subjected to least squares approximation in phase to solve the sub-sample time delay,and find the true peak of the discrete correlation function.A subsample time delay estimation algorithm(Phase-based Subsample time-delay Estimator,PSTE)based on cross-correlation phase approximation is presented.4.Using MATAB for experimental simulation,the mean square error of the delay estimation is used as the algorithm performance evaluation standard,and the parameters such as signal-to-noise ratio,sampling rate,subsample time delay,least squares calculation points ?n and signal sample points N are analyzed.The impact on the performance of the algorithm,and the proposed two algorithms are compared with the existing time-delay estimation algorithms based on the basic correlation method.The experimental results show that the algorithms ASTE and PSTE in this paper can improve the accuracy of the time-delay estimations with the amount of calculation comparable to the basic correlation method.