Research On The Parameter Estimation Methods Of Hyperbolic Frequency Modulated Signals In Underwater Acoustic Environment

As nation's demand for the deep-sea exploration has gradually increased in recent years,exploration equipment has higher requirements on the performance of sonar systems.On the one hand,it is necessary to improve the detection distance and range resolution,on the other hand,it is important to estimate and analyze the parameter of the received signal with the random nature,in order to reduce the effects of random disturbances from the media and reflections from moving targets.Hyperbolic Frequency Modulation(HFM)signals have good pulse compression characteristics,which solves the contradiction between detection distance and distance resolution.At the same time,HFM signals have the unique Doppler invariance,which can make the echo signal still completely matched with the Matched Filter,and the parameters of signal remain unchanged.Therefore,this paper studies the parameter estimation methods of HFM signal in underwater acoustic environment from the two angles of using signal characteristics and pursuing estimation performance,including:(1)This paper introduces the basic characteristics of the HFM signal in detail,and proves effectiveness of the HFM signal in underwater acoustic environment.At the same time,in order to obtain the characteristics of the HFM signal,this paper introduces several common time-frequency analysis methods,and the time-frequency distribution results of HFM signals under these methods are compared.In addition,the Cramer-Rao lower bound of the HFM signal parameter estimation problem is derived as the evaluation criterion of the parameter estimation performance in this paper.(2)In this paper,using the characteristics of HFM signal,a mapping method based on Hough transform is proposed,so that the hyperbola reflecting the instantaneous frequency of HFM signal in the time-frequency distribution can be detected and the parameters can be estimated.Finally,this paper proposes a screening process based on the time-frequency distribution characteristics of HFM signals,which improves the calculation efficiency and parameter estimation accuracy of Hough transform.The experimental results show that this method can estimate the parameters of the HFM signal well: when the signal-to-noise ratio is 7d B,the mean square error of the parameter estimated values of the frequency modulation and initial frequency can reach-60 d B and-72 d B respectively.(3)In order to pursue the performance of parameter estimation,this paper constructs the likelihood function of the HFM signal,and the estimation accuracy of several common optimization algorithms based on the likelihood function is compared.Finally,according to the optimization process of the algorithm and the characteristics of the likelihood function vectorization,an improved fitness function is proposed to improve the calculation of the algorithm while ensuring the estimation accuracy.The experimental results show that,compared to the method before the improvement,the mean square error of the parameter estimation result of the HFM signal is closer to the Cramer-Rao lower bound when the signal-to-noise ratio is more than-7d B,and the amount of calculation is about one-third of the method before improvement.