CN106357575A - Multi-parameter jointly-estimated interference type identification method - Google Patents

Abstract

The invention provides a multi-parameter jointly-estimated interference type identification method, which relates to the field of monitoring of the satellite navigation interference. According to the method, for the problem of identifying an interference type when the interference of a satellite navigation system is monitored, four characteristic parameters subjected to the interference are used for classifying interference signals, after the interference signals are received, firstly a central frequency of the interference is estimated, then the signal is converted into a zero-middle-frequency baseband signal, and the characteristic parameters of the signal are then extracted and are compared with an identification database, so that seven interference debugging ways can be identified. The method has the characteristics of simple structure and low complexity and is a practical interference type identification method in the satellite navigation interference monitoring field.

Description

A kind of interference type recognition methodss of multiparameter Combined estimator

Technical field

The present invention relates to satellite navigation interference monitoring field, the interference type knowledge of more particularly, to a kind of multiparameter Combined estimator Other method, using some characteristic parameters receiving interference signal, the identification process being designed correctly, to satellite navigation system for the method In the modulation type of common seven kinds of interference signals be identified.

Background technology

The interference effect that satellite navigation system faces is increasingly severe, and navigation interference signal type is identified, for Targetedly eliminating the impact to navigation system for the various interference has positive meaning.The technology of identification of interference signal is in communication In have a wide range of applications, such as interior abroad experts and scholars propose that circulation using signal is general, Higher Order Cumulants, wavelet transformation, The identification of the accomplished in many ways signal of communication debud mode such as planisphere, but these methods all must first carry out complexity to signal Conversion, then extract corresponding characteristic parameter again, structure is complex, computationally intensive, and is not directed to satellite navigation system The application characteristic of interference monitoring is designed, and practical engineering application performance is relatively low.It is therefore proposed that one kind is applied to navigation system doing Disturb knowledge method for distinguishing, solve the practical problem running in navigation interference monitoring, have great significance.

Content of the invention

It is an object of the invention to provide interference type knows method for distinguishing in a kind of satellite navigation system, special using extracting parameter Levy and then and the method compared of priori data lab setting thresholding realize the identification of seven kinds of interference types.For satellite navigation system Interference type identification provides a kind of new technological means.

The object of the present invention is achieved like this: a kind of united interference type recognition methodss of multiparameter, using to be identified The priori features knowledge of interference signal is contrasted with the feature of the current interference signal extracted, and realizes the knowledge of interference type Not, following steps are specifically included:

(1) the current interference signal receiving is carried out digital sample and obtain digital signal；

(2) digital signal is carried out after Fourier transformation, estimate to obtain carrier frequency；

(3) carrier wave lift-off processing is carried out to digital signal according to carrier frequency and obtain zero intermediate frequency baseband complex signal；

(4) choose the zero center normalization Fourth-order moment of zero intermediate frequency baseband complex signal, instantaneous amplitude spectrum density maximum, in zero The absolute value of correlation coefficient of the in-phase component of frequency baseband complex signal envelope and quadrature component and r parameter are joined as identification feature Number, identification feature parameter is contrasted with preset reference thresholding, is realized the identification of interference type；Described zero center normalization Fourth-order moment is the normalized value of zero center Fourth-order moment, and r parameter reflects the intensity of variation of zero intermediate frequency baseband complex signal envelope.

Wherein, described step (4) specifically includes step:

(401) extract zero intermediate frequency baseband complex signal zero center normalization Fourth-order moment, by zero center normalization Fourth-order moment with First preset reference thresholding compares, if zero center normalization Fourth-order moment is more than the first preset reference thresholding, current interference letter Number be FM signal, process ends；Otherwise, execution step (402)；

(402) extract the instantaneous amplitude spectrum density maximum of zero intermediate frequency baseband complex signal, instantaneous amplitude spectrum density is maximum Value is compared with the second preset reference thresholding, if instantaneous amplitude spectrum density maximum is more than the second preset reference thresholding, currently Interference signal is amplitude-modulated signal or single-carrier signal, execution step (403)；Otherwise current interference signal is 2ask signal, 2fsk One of signal, 2psk signal and 4psk signal, proceed to step (404)；

(403) extract the r parameter of zero intermediate frequency baseband complex signal, r parameter is compared with default 3rd thresholding, if r parameter More than default 3rd thresholding, then current interference signal is amplitude-modulated signal, process ends；Otherwise current interference signal is single carrier Signal, process ends；

(404) extract the r parameter of zero intermediate frequency baseband complex signal, r parameter is compared with default 4th thresholding, if r parameter More than default 3rd thresholding, then current interference signal is 2ask signal, process ends；Otherwise, execution step (405)；

(405) extract the absolute value of the in-phase component of zero intermediate frequency baseband complex signal envelope and the correlation coefficient of quadrature component, Absolute value is compared with default 5th thresholding, if absolute value is less than default 5th thresholding, current interference signal is believed for 4psk Number, process ends；Otherwise, execution step (406)；

(406) extract zero intermediate frequency baseband complex signal zero center normalization Fourth-order moment, by zero center normalization Fourth-order moment with 6th preset reference thresholding compares, if zero center normalization Fourth-order moment is more than the 6th preset reference thresholding, current interference letter Number be 2psk signal；Otherwise, current interference signal is 2fsk signal；

Interference signal type identification terminates.

Wherein, described zero center normalization Fourth-order moment is:

${\mathrm{\μ}}_{42}=\frac{e\left\{x_n^4\left(n\right)\right\}}{{\left\{e\right\{x_n^2\left(n\right)\left\}\right\}}^2}$

Wherein, e { } represents the average of zero intermediate frequency baseband complex signal, μ₄₂Represent the zero center normalizing of zero intermediate frequency baseband complex signal Change Fourth-order moment, n is single computing sampled point number, x_nN () makees the zero intermediate frequency base band letter in reply of n sampled point of normalization calculating Number.

Wherein, described r parameter is:

R=σ²/μ²

Wherein, μ is the average of zero intermediate frequency baseband complex signal envelope square, σ²For zero intermediate frequency baseband complex signal envelope square Variance.

Wherein, in step (2), digital signal is carried out after Fourier transformation, estimate obtain carrier frequency, specifically include with Lower step:

(201) digital signal is carried out trying to achieve with the normalization work(of the digital signal after Fourier transformation after Fourier transformation Rate is composed；

(202) normalized power spectrum is scanned for finding power spectrum maximum；

(203) carrier frequency is obtained according to the position calculation of power spectrum maximum.

The technology of the present invention has the advantage that

I () present invention proposes a kind of interference type recognition methodss of multiparameter Combined estimator, be capable of identify that seven kinds common Interference, has the characteristics that parameter extraction is simple, complexity is relatively low；

(ii) present invention is capable of identify that common compacting interference type in satellite navigation system, is dry in navigation neceiver Disturb identification and provide a kind of new means.

Brief description

The interference type recognition methodss flow chart of Fig. 1 present invention；

The interference of Fig. 2 present invention differentiates flow chart；

Fig. 3 is the interference type recognition effect figure of the present invention.

Specific embodiment

The present invention will be further described in detail below in conjunction with the accompanying drawings.

A kind of united interference type recognition methodss of multiparameter, described method goes out signal based on the schema extraction of statistics Characteristic parameter, carries out contrasting the selection realizing interference signal type with the empirical features of signal, specifically includes following steps:

(1) signal receiving is changed into digital intermediate frequency signal through digital sample；

(2) digital signal is carried out Fourier transformation, estimate to receive according to carrier frequency estimating method after calculating power spectrum The carrier frequency of signal；

Carrier frequency is estimated, comprises the following steps:

(201) digital signal is carried out trying to achieve the power spectrum of signal after Fourier transformation；

First Fourier transformation is carried out to signal

$x\left(k\right)=\underset{i=1}{\overset{n}{\σ}}x\left(n\right)e^{-j2{\mathrm{\πf}}_s\frac{nk}{n}}$

Wherein, x (n) is the digital signal of discretization, and n is the points of fft, and x (k) is the frequency-region signal of receipt signal, f_s For sample frequency,Then power spectrum signal is represented by:

$p\left(k\right)=\frac{1}{n}|x\left(k\right){|}^2$

(202) power spectrum is scanned for find power spectrum maximum；

P (k) is carried out with point by point search maximizing p_max, under the corresponding sampled point of maximum, it is designated as n_peak.

(203) mid frequency of the position calculation carrier wave according to power spectrum maximum；

$f=\frac{n_{peak}}{n}\·f_s$

Wherein, f_sFor sample frequency, n is fft points.Estimating of carrier frequency can be improved using the method increasing fft points Meter precision.

(3) carrier wave lift-off processing is carried out to signal according to the carrier frequency estimated, be changed into zero intermediate frequency baseband complex signal；

Single-frequency complex signal is generated according to the signal(-) carrier frequency that estimates in step (2), and receives signal multiplication, to height Frequency obtains the zero intermediate frequency baseband complex signal of carrier wave stripping after being filtered.

(4) choose the zero center normalization Fourth-order moment of zero intermediate frequency baseband complex signal, instantaneous amplitude spectrum density maximum, in zero The absolute value of correlation coefficient of the in-phase component of frequency baseband complex signal envelope and quadrature component and r parameter are joined as identification feature Number, identification feature parameter is contrasted with preset reference thresholding, is realized the identification of interference type；Described zero center normalization Fourth-order moment is the normalized value of zero center Fourth-order moment, and r parameter reflects the intensity of variation of zero intermediate frequency baseband complex signal envelope.Concrete bag Include following steps:

(401) extract zero intermediate frequency baseband complex signal zero center normalization Fourth-order moment, by zero center normalization Fourth-order moment with First preset reference thresholding compares, if zero center normalization Fourth-order moment is more than the first preset reference thresholding, current interference letter Number be FM signal, process ends；Otherwise, execution step (402)；

${\mathrm{\μ}}_{42}=\frac{e\left\{x_n^4\left(n\right)\right\}}{{\left\{e\right\{x_n^2\left(n\right)\left\}\right\}}^2}$

Wherein, e { } represents the average of signal phasor, μ₄₂Represent the normalization Fourth-order moment extracting signal.

(402) extract the instantaneous amplitude spectrum density maximum of interference signal, the instantaneous amplitude spectrum with the interference signal setting Density Detection thresholding γ_th1Relatively, be then identified as (amplitude modulation am or single carrier cw) signal more than thresholding, less than thresholding then think for One of (2ask 2fsk 2psk 4psk), waits further feature parameter to be extracted to be identified.

The maximum of signal zero center normalization instantaneous amplitude spectrum density to be identified is defined as

γ_max=max | dft (a_cn(i))|²

In above formula, dft represents digital fourier transformation, and max represents vector maximization, a_cn(i) be signal to be identified when Carve t=i/f_s(i=1,2 ..., n_s) zero center normalization instant amplitude value, a_cnI () can be tried to achieve in the following way:

$\left\{\begin{array}{c}{m}_a=\frac{1}{n_s}\underset{i=1}{\overset{n}{\σ}}a\left(i\right)\\ a_n\left(i\right)=\frac{a\left(i\right)}{m_a}\\ a_{cn}\left(i\right)=a_n\left(i\right)-1\end{array}\right.$

In formula, a (i) is the instant amplitude value in signal to be identified each moment, m_aIt is the flat of this signal segment signal transient amplitude Average.

(403) as the instantaneous amplitude spectrum density maximum extracting interference signal is more than the instantaneous amplitude of the interference signal setting Spectrum density maximum detection threshold, then extract interference signal r parameter continue identification, with this under the conditions of set r Parameters threshold r_th1Relatively, regard as amplitude-modulated signal more than thresholding, then regard as single-carrier signal, end of identification less than thresholding.

The intensity of variation of r reaction signal envelope, is defined as

R=σ²/μ²

Wherein, μ is the average of signal envelope square, σ²Variance for signal envelope square.

(404) as the instantaneous amplitude spectrum density maximum extracting interference signal is less than the instantaneous amplitude of the interference signal setting Spectrum density maximum detection threshold, then extract interference signal r parameter continue identification, with this under the conditions of set r Parameters threshold r_th2Relatively, then think for 2ask signal more than thresholding, continue to be identified using other parameters less than thresholding.

(405) if the classification of signal still cannot be determined in (404), extract the correlation coefficient of the signal envelope of interference signal In-phase component and quadrature component correlation coefficient absolute value as identification parameter continue identification, with set this parameter identification thresholding l_th1Relatively, if being less than thresholding, it is considered 4psk, if continuing identification more than thresholding；

The absolute value l of the correlation coefficient of complex envelope in-phase component and quadrature component is defined as:

$l=\frac{\left|l_{xy}\right|}{\sqrt{l_{xx}{l}_{yy}}}$

Wherein:

$\begin{array}{cc}{l}_{xx}=\underset{i=0}{\overset{n-1}{\σ}}{x}_{real}^2\left(i\right)-\frac{1}{n}{\[\underset{i=0}{\overset{n-1}{\σ}}{x}_{real}\left(i\right)\]}^2& l_{yy}=\underset{i=0}{\overset{n-1}{\σ}}{x}_{imag}^2\left(i\right)-\frac{1}{n}{\[\underset{i=0}{\overset{n-1}{\σ}}{x}_{imag}\left(i\right)\]}^2\end{array}$

R reflects x under in-phase component_real(i) and quadrature component x_imagThe level of intimate of the linear relationship between (i).

(406) the zero center normalization Fourth-order moment extracting interference signal is as identification parameter, this Parameters threshold with setting μ_th2It is compared, is 2psk more than thresholding, is then considered 2fsk signal less than thresholding, so far whole seven type signals are whole Realize identification.

The effect of identification as shown in figure 3, as can be seen from the figure can achieve interference signal when signal to noise ratio is for 3db More than 95% identification probability, compares some traditional method recognition performances more excellent.

Claims (5)

1. a kind of interference type recognition methodss of multiparameter Combined estimator it is characterised in that: using the elder generation of interference signal to be identified Test feature knowledge and contrasted with the feature of the current interference signal extracted, realize the identification of interference type, specifically include with Lower step:

(1) the current interference signal receiving is carried out digital sample and obtain digital signal；

(2) digital signal is carried out after Fourier transformation, estimate to obtain carrier frequency；

(3) carrier wave lift-off processing is carried out to digital signal according to carrier frequency and obtain zero intermediate frequency baseband complex signal；

(4) the zero center normalization Fourth-order moment of selection zero intermediate frequency baseband complex signal, instantaneous amplitude spectrum density maximum, zero intermediate frequency base The absolute value of the correlation coefficient of the in-phase component with complex signal envelope and quadrature component and r parameter, will used as identification feature parameter Identification feature parameter is contrasted with preset reference thresholding, realizes the identification of interference type；Described zero center normalization quadravalence Square is the normalized value of zero center Fourth-order moment, and r parameter reflects the intensity of variation of zero intermediate frequency baseband complex signal envelope.

2. a kind of multiparameter Combined estimator according to claims 1 interference type recognition methodss it is characterised in that: Described step (4) specifically includes step:

(401) extract the zero center normalization Fourth-order moment of zero intermediate frequency baseband complex signal, by zero center normalization Fourth-order moment and first Preset reference thresholding compares, if zero center normalization Fourth-order moment is more than the first preset reference thresholding, current interference signal is FM signal, process ends；Otherwise, execution step (402)；

(402) extract zero intermediate frequency baseband complex signal instantaneous amplitude spectrum density maximum, by instantaneous amplitude spectrum density maximum with Second preset reference thresholding compares, if instantaneous amplitude spectrum density maximum is more than the second preset reference thresholding, currently disturbs Signal is amplitude-modulated signal or single-carrier signal, execution step (403)；Otherwise current interference signal be 2ask signal, 2fsk signal, One of 2psk signal and 4psk signal, proceed to step (404)；

(403) extract the r parameter of zero intermediate frequency baseband complex signal, r parameter is compared with default 3rd thresholding, if r parameter is more than Default 3rd thresholding, then current interference signal is amplitude-modulated signal, process ends；Otherwise current interference signal is believed for single carrier Number, process ends；

(404) extract the r parameter of zero intermediate frequency baseband complex signal, r parameter is compared with default 4th thresholding, if r parameter is more than Default 3rd thresholding, then current interference signal is 2ask signal, process ends；Otherwise, execution step (405)；

(405) extract the absolute value of the in-phase component of zero intermediate frequency baseband complex signal envelope and the correlation coefficient of quadrature component, will absolutely Value is compared with default 5th thresholding, if absolute value is less than default 5th thresholding, current interference signal is 4psk signal, knot Restraint this flow process；Otherwise, execution step (406)；

(406) extract the zero center normalization Fourth-order moment of zero intermediate frequency baseband complex signal, by zero center normalization Fourth-order moment and the 6th Preset reference thresholding compares, if zero center normalization Fourth-order moment is more than the 6th preset reference thresholding, current interference signal is 2psk signal；Otherwise, current interference signal is 2fsk signal；

Interference signal type identification terminates.

3. the interference type recognition methodss of a kind of multiparameter Combined estimator according to claims 1 or 2, its feature exists In: described zero center normalization Fourth-order moment is:

${\mathrm{\μ}}_{42}=\frac{e\left\{x_n^4\left(n\right)\right\}}{{\left\{e\right\{x_n^2\left(n\right)\left\}\right\}}^2}$

Wherein, e { } represents the average of zero intermediate frequency baseband complex signal, μ₄₂Represent the zero center normalization of zero intermediate frequency baseband complex signal Fourth-order moment, n is single computing sampled point number, x_nN () makees the zero intermediate frequency baseband complex signal of n sampled point of normalization calculating.

4. the interference type recognition methodss of a kind of multiparameter Combined estimator according to claims 1 or 2, its feature exists In: described r parameter is:

R=σ²/μ²

Wherein, μ is the average of zero intermediate frequency baseband complex signal envelope square, σ²Variance for zero intermediate frequency baseband complex signal envelope square.

5. the interference type recognition methodss of a kind of multiparameter Combined estimator according to claims 1 or 2, its feature exists In: in step (2), digital signal is carried out after Fourier transformation, estimate to obtain carrier frequency, specifically include following steps:

(201) digital signal is carried out trying to achieve after Fourier transformation with the normalized power spectrum of the digital signal after Fourier transformation；

(202) normalized power spectrum is scanned for finding power spectrum maximum；

(203) carrier frequency is obtained according to the position calculation of power spectrum maximum.