Parameter estimation method for chirp signal under alpha non-Gaussian noise

Abstract

The invention relates to a parameter estimation method for a chirp signal under alpha non-Gaussian noise, belongs to the fields of signal detection and information processing, and further relates the fields of signal detection and parameter estimation. Through amplitude limiting preprocessing on a signal including alpha non-Gaussian noise, most interference can be eliminated. In particular for pulse interference, the initial frequency and frequency modulation rate of the chirp signal can be effectively estimated with the energy aggregation feature of FrFT during chirp signal processing. The method has the advantages that most interference can be eliminated through the amplitude-limiting preprocessing; in particular for pulse interference, the signal is processed with fractional order Fourier transform, so that a high-accuracy estimated value can be obtained; the method is low in computation complexity, small in computation amount and suitable for engineering practice; the amplitude-limiting preprocessing is easy to implement and wide in application range, and pulse interference caused by accidental factors can be effectively overcome; and the advantage is more remarkable on the aspect of accuracy judgment. Meanwhile, the method has very high stability. Most importantly, the accuracy is high, and is consistent with the accuracy in the absence of noise.

Description

technical field [0001] The invention belongs to the field of signal detection and information processing, and further relates to the field of signal detection and parameter estimation. Background technique [0002] Chirp signal is linear frequency modulation signal, and chirp signal is widely used in information systems such as communication, medicine and sonar, especially modern new system radar system. Due to the low interception probability of chirp signals, the research on the detection and estimation of chirp signal parameters has attracted extensive attention. In recent years, the detection and estimation of chirp signals are mostly under the condition of Gaussian noise. At present, the commonly used methods include Fractional Fourier Transform (Fractional Fourier Transform, FrFT) method, Wigner-Ville distribution (WVD) time-frequency analysis method, Wigner-Hough transform (WHT) method, etc., wherein FrFT is an extension of the traditional Fourier transform, and its ...

AI Technical Summary

Problems solved by technology

[0004] In the existing methods, it is difficult to estimate the parameters of the chirp signal under the background of α non-Gaussian noise. This is because the non-Gaussian noise has an impulsive characteristic, and its probability density function has a thicker tail than that of the Gaussian distribution. Make most estimation methods degrade or even fail when estimating signals in non-Gaussian noise

Images