Average Plain Gradient Based Indirect Frequency Estimation Using Adaptive Notch Filter

Frequency-modulated continuous wave(FMCW)radar technology is a high-precision radar measurement technology,which is mainly used to determine the thickness of the detected target and its internal structural characteristics.The fast and high-precision processing of the FMCW radar received signal is of great significance to its application in actual measurement.The existing processing algorithms are mostly based on the Fourier transform spectrum analysis method,and the research on the real-time tracking processing of the signal is less.In this paper,the adaptive filtering method is used to track and process the frequency-dependent radar received signal.According to the characteristics of thickness measurement signal from FMCW radar,an improved adaptive notch filter frequency estimation method is proposed in this thesis.The method derives a new average gradient algorithm with a modified error function,which improves the measurement speed,accuracy,and system stability of real-time frequency tracking.In addition,a self-adaptive notch filter system with a cascade structure is constructed to directly estimate the dual-frequency signal.In this thesis,based on the analysis of the characteristics of the FMCW ground penetrating radar thickness measurement signal,an adaptive noise cancellation system is designed at first,which achieve the extraction of weak signals under strong interference,to solve the problem that the lower reflection wave is submerged by the strong surface reflection wave when it is used to detect the thickness of the thin layer target.The theoretical derivation of the principle of adaptive noise cancellation is carried out.The factors affecting the performance of its algorithm are analyzed,and the effectiveness of the method is verified by simulation experiments.Then,for the problem of frequency estimation of frequency-variable signals,the error function in the traditional method is modified,a new average gradient algorithm is derived.An improved frequency estimation method based on an adaptive notch filter is proposed.Compared with other typical algorithms through simulation experiments,it is verified that the improved algorithm has higher convergence speed and estimation accuracy,and solves the problem of limited convergence with better stability.Finally,for the problem of frequency estimation of multiple frequency-variable signals,an adaptive notch system with cascade structure is constructed.A direct frequency estimation algorithm for multi-frequency signals is deduced.and the results of frequency estimation and frequency-tracking of frequency-variable signals are given through simulation experiments,which verifies the feasibility of the algorithm for direct estimation of multi-frequency signals.