KR20180055165A - Apparatus and method for removing noise of hrrp using emd - Google Patents

Abstract

The present invention relates to a high resolution range profile (HRRP) noise removing apparatus using empirical mode decomposition (EMD) and a method thereof. According to an embodiment of the present invention, the HRRP noise removing apparatus comprises: a radar signal receiving unit for receiving a radar signal including a Gaussian noise; a noise removing unit for removing noise from the radar signal by using an EMD technique; and an HRRP forming unit for forming an HRRP on the signal from which the noise is removed. Therefore, the HRRP noise removing apparatus improves the target identification performance of a radar using a stepped frequency waveform (SFW).

Description

[0001] APPARATUS AND METHOD FOR REMOVING NOISE OF HRRP USING EMD [0002]

More particularly, the present invention relates to an apparatus and method for removing HRRP using an EMD, and more specifically, to improve the target identification performance of a radar using a SFW waveform by removing noise using EMD technique for a received radar signal, To an apparatus and method for removing HRRP noise using EMD.

A High Resolution Range Profile (HRRP) (hereinafter referred to as "HRRP") of a target is a radar signal mainly used in radar target identification. That is, HRRP is a one-dimensional image showing the intrinsic scattering characteristics of the target. Since HRRP is relatively easy to acquire (or measure) and is a value that expresses the characteristics of the target, HRRP can be used to relatively easily identify the target. However, HRRP has a drawback that it is relatively vulnerable to noise.

Therefore, it is necessary to remove the noise included in the HRRP in order to improve the target discrimination ability. Empirical Mode Decomposition (EMD) technique is used as a representative method for eliminating noise.

Conventionally, a method of eliminating noise included in HRRP by applying EMD technique directly to HRRP has been proposed. That is, in the conventional method, the EMD technique is applied to the source signal of the received radar signal to remove the noise of the radar signal. However, in the conventional method of removing noise by applying the EMD technique to the original signal of the received radar signal, there is a high possibility that the coherence of the raw signal required to form HRRP is broken.

1A and 1B are graphs showing HRRP according to a conventional noise canceling method. FIG. 1A is a graph showing HRRP before noise cancellation, and FIG. 1B is a graph showing HRRP after noise cancellation. As shown in Figs. 1A and 1B, in the conventional method, noise is not well removed from the HRRP, and only the loss of the signal is increased.

Therefore, in order to improve the target identification performance of the radar, there is a need to propose a method that can remove the noise of the HRRP in such a manner that the loss of the signal is small without breaking the consistency of the original signal.

An object of the present invention is to provide an apparatus and method for HRRP noise removal using EMD for improving the target identification performance of a radar using SFW waveform by removing HRRP after eliminating noise using EMD technique for a received radar signal .

It is also an object of the present invention to eliminate HRRP noise in a way that signal loss is minimized without breaking the consistency of the original signal in order to improve the target identification performance of the radar.

An apparatus for removing HRRP noise using EMD according to an embodiment of the present invention includes a radar signal receiving unit for receiving a radar signal including Gaussian noise; A noise eliminator for eliminating noise of the radar signal using an EMD (mpirical Mode Decomposition) technique; And an HRRP forming unit for forming a High Resolution Range Profile (HRRP) with respect to the noise canceled signal.

The noise eliminator may include: a Fourier transformer for performing a Fourier transform on the radar signal to generate a Fourier transform signal; And an EMD processor for removing noise by applying an EMD technique to the Fourier transform signal.

The EMD processing unit applies the EMD technique to each of the real component and the imaginary component of the Fourier transform signal.

The Fourier transform signal is a real signal and is characterized by being Gaussian noise having an average of noise included in the signal is '0'.

And the HRRP forming unit forms an HRRP by taking an absolute value of the noise canceled signal.

The noise canceller is included in a radar using a stepped frequency waveform (SFW).

A method of removing HRRP noise using EMD according to an embodiment of the present invention includes: receiving a radar signal including Gaussian noise; Removing noise of the radar signal using an EMD (mpirical Mode Decomposition) technique; And forming a High Resolution Range Profile (HRRP) for the noise canceled signal.

Wherein the noise removing step comprises: a step of Fourier-transforming the radar signal to generate a Fourier transform signal; And removing the noise by applying an EMD technique to the Fourier transform signal.

The EMD application step is characterized in that an EMD technique is applied to each of a real component and an imaginary component of the Fourier transform signal.

The HRRP forming step is characterized by forming an HRRP by taking an absolute value of the noise canceled signal.

The present invention can enhance the target identification performance of the radar using the SFW waveform by forming the HRRP after removing the noise using the EMD technique for the received radar signal.

In order to improve the target identification performance of the radar, the present invention can remove the HRRP noise in a manner that the loss of the signal is small without breaking the consistency of the original signal.

In addition, the present invention can remove noise while satisfying the condition for removing noise well by applying the EMD technique.

In addition, the present invention can improve the target identification performance of a radar using a stepped frequency waveform (SFW).

1A and 1B are graphs showing HRRP according to a conventional noise canceling method,
FIG. 2 is a block diagram of an apparatus for removing HRRP noise using EMD according to an embodiment of the present invention;
FIGS. 3A and 3B are graphs showing HRRP according to the noise canceling method of the present invention,
4 is a diagram illustrating a noise removal method of HRRP using EMD according to an embodiment of the present invention.

For a better understanding of the present invention, a preferred embodiment of the present invention will be described with reference to the accompanying drawings. The embodiments of the present invention may be modified into various forms, and the scope of the present invention should not be construed as being limited to the embodiments described in detail below. The present embodiments are provided to enable those skilled in the art to more fully understand the present invention. Therefore, the shapes and the like of the elements in the drawings can be exaggeratedly expressed to emphasize a clearer description. It should be noted that in the drawings, the same members are denoted by the same reference numerals. Detailed descriptions of well-known functions and constructions which may be unnecessarily obscured by the gist of the present invention are omitted.

FIG. 2 is a block diagram of an apparatus for removing HRRP noise using EMD according to an embodiment of the present invention. Referring to FIG.

2, an apparatus for removing HRRP noise using EMD according to an embodiment of the present invention (hereinafter referred to as a "noise canceling apparatus") 100 performs a Fourier transform of a received radar signal Then, EMR technique is applied to the Fourier transform signal to form HRRP. Here, since the Fourier transform signal is a complex number signal, the EMD technique is applied to each of the real number component and the imaginary number component of the Fourier transform signal.

The noise eliminator 100 removes the noise before forming the HRRP and forms the HRRP. Also, the noise eliminator 100 removes noise at the Fourier transform signal stage, not at the raw signal stage. In the case of applying the EMD technique to the raw signal, the coherence of the raw signal necessary to form the HRRP is likely to be broken.

Meanwhile, the noise canceller 100 can improve the target identification performance of the radar using the stepped frequency waveform (SFW). That is, the noise canceller 100 can effectively remove the noise included in the HRRP of the target.

The noise removing apparatus 100 includes a radar signal receiving unit 110, a noise removing unit 120, and an HRRP forming unit 130.

이고, 실수 성분은

이고, 허수 성분은

이다.The radar signal receiving unit 110 receives the radar signal including the Gaussian noise. The radar signal can be represented by a real number component and an imaginary number component as shown in the following equation (1). Here, the receiving radar signal is

And the real component is

And the imaginary component is

to be.

The noise removing unit 120 removes the noise of the received radar signal by using the EMD technique. Here, the noise removing unit 120 removes the noise included in the complex signal before forming the HRRP, instead of removing the noise included in the HRRP by applying the EMD technique after forming the HRRP.

As described above, when the noise included in the HRRP is removed by applying the EMD technique after forming the HRRP, the noise is not removed well and the loss of the signal component is large. In this case, the average of the noise included in HRRP does not show a Gaussian noise characteristic of '0'. That is, since HRRP is a signal having an absolute value, it has a positive real value including '0'. Therefore, in this case, it is difficult to satisfy the condition for removing the noise through the EMD technique (that is, the average of the noise included in the signal should be Gaussian noise of '0'). On the other hand, in the case of removing the noise included in the complex signal before HRRP is formed, a high-quality HRRP can be obtained by removing the noise by applying the EMD technique to each of the real and imaginary components of the complex signal. Here, each of the real number component and the imaginary number component of the complex number signal represents a real number value. That is, in this case, the condition for removing the noise through the EMD technique (that is, the signal is a real signal and the average of the noise included in the signal is '0' is Gaussian noise).

Also, the noise removing unit 120 applies the EMD technique to the received radar signal (i.e., the Fourier transform signal) on which the Fourier transform has been performed, instead of directly applying the EMD technique to the received radar signal. That is, when noise is removed by directly applying the EMD technique to the received radar signal, the correlation between the real component and the imaginary component may be broken. In this case, HRRP is not formed properly. However, when the noise is removed by applying the EMD technique to the Fourier transform signal, there is no need to consider the correlation between the real component and the imaginary component. This is because only the square of the magnitude of the real component and the imaginary component is added to form HRRP. In this case, HRRP can be formed with high quality.

To this end, the noise removing unit 120 includes a Fourier transform unit 121 and an EMD processing unit 122.

이고, 실수 성분은

이고, 허수 성분은

이다. 복소수 신호의 실수 성분과 허수 성분 각각은 실수값을 나타낸다. First, the Fourier transform unit 121 performs Fourier transform on the received radar signal. The Fourier transform signal is a complex number signal expressed by a real number component and an imaginary number component as shown in the following equation (2). Here, the Fourier transform signal is

And the real component is

And the imaginary component is

to be. Each of the real number component and the imaginary number component of the complex number signal represents a real number value.

의 실수 성분

과 허수 성분

에 포함된 각각의 잡음은 평균 '0'인 가우스 잡음을 이룬다.Then, the Fourier transform signal

The real component

And imaginary component

≪ / RTI > each of the noise included in the received signal is Gaussian noise having an average of " 0 ".

를 푸리에 변환하면 아래 [수학식 3]과 같이 표현할 수 있다.Specifically, the receiving radar signal

Can be expressed by the following equation (3). &Quot; (3) "

은 수신 레이다 신호

의 선형 결합(linear combination)이다. 따라서, 수신 레이다 신호

가 가우스 잡음을 포함하면, 푸리에 변환 신호

에 포함된 잡음도 가우스 잡음이 된다.That is, the Fourier transform signal

Lt; / RTI >

/ RTI > is a linear combination of < / RTI > Therefore,

Gaussian noise, the Fourier transform signal < RTI ID = 0.0 >

The noise included in the Gaussian noise is also Gaussian noise.

의 평균은 아래 [수학식 4]와 같이 나타낼 수 있다.Here, the Fourier transform signal

Can be expressed by Equation (4) below. &Quot; (4) "

에 포함된 잡음이 평균 '0'인 가우스 잡음이면, 푸리에 변환 신호

에 포함된 잡음도 평균 '0'인 가우스 잡음이 된다. 따라서, 푸리에 변환 신호

의 실수 성분

과 허수 성분

에 포함된 각각의 잡음은 평균 '0'인 가우스 잡음이 된다.In this manner,

Is a Gaussian noise having an average of '0', the Fourier transform signal

The noise included in the Gaussian noise is also zero. Therefore, the Fourier transform signal

The real component

And imaginary component

Each Gaussian noise is an average of " 0 ".

의 실수 성분

과 허수 성분

각각에 대해 EMD 기법을 적용하여 잡음을 제거할 수 있다. EMD 처리부(122)는 특정 문턱값(threshold value)보다 낮은 크기를 갖는 신호를 잡음으로 보고, 이와 같은 잡음을 푸리에 변환 신호의 실수 성분과 허수 성분 각각으로부터 제거한다.Next, the EMD processor 122 removes noise by applying an EMD technique to the Fourier transform signal. That is, the EMD processing unit 122 generates the Fourier transform signal

The real component

And imaginary component

The EMD technique can be applied to each of them to remove noise. The EMD processing unit 122 regards a signal having a magnitude lower than a specific threshold as noise and removes such noise from each of the real and imaginary components of the Fourier transform signal.

Here, the EMD technique is a technique of decomposing a signal, and decomposes a signal x (t) into a basis function called an intrinsic mode fucntion (IMF). At this time, the IMF is automatically determined by the signal x (t), and includes different frequency bands. In the EMD technique, noise is mainly distributed in the high frequency band, and noise is assumed to be Gaussian noise.

The EMD processing unit 122 can remove noise well for signals satisfying the following two conditions. First, the signal is a real signal, and second, the average of the noise included in the signal should be a Gaussian noise of '0'.

In order to satisfy such a condition, the EMD processing unit 122 does not perform the EMD processing that directly applies the EMD technique to the HRRP to remove the noise included in the HRRP after forming the HRRP, And performs EMD processing for applying the EMD technique to each of the real component and the imaginary component of the converted signal. The EMD processing unit 122 can remove the noise well without loss of the signal component with respect to the Fourier transform signal satisfying the above two conditions.

If the EMD processing unit 122 removes the noise included in the HRRP, HRRP has a positive real value including 0 because it is the absolute value of the signal. In this case, the EMD processing unit 122 does not satisfy the condition of '0', which is the second condition that the noise is well removed. Its statistical characteristics do not satisfy the condition of Gaussian noise. Therefore, in this case, the EMD processing unit 122 hardly removes the noise well, and the loss of the signal becomes large.

에 절대값을 취한 신호

를 형성한다. The HRRP forming unit 130 forms the HRRP for the Fourier transform signal from which the noise is removed. That is, the HRRP forming unit 130 generates the noise-removed Fourier transform signal < RTI ID = 0.0 >

≪ / RTI >

.

3A and 3B are graphs showing HRRP according to the noise canceling method of the present invention. FIG. 3A is a graph showing HRRP before noise cancellation, and FIG. 3B is a graph showing HRRP after noise cancellation. The signal-to-noise ratio (SNR) for HRRP in FIGS. 3A and 3B can be expressed as shown in Table 1 below.

enforcement SNR before noise cancellation SNR after noise cancellation SNR gain One -2.8530 5.5555 8.4085 2 -0.7570 6.7608 7.5178 3 2.7670 10.1754 7.4084

Here, the raw signal used the measurement data obtained for the reduced model of F117 in the compact range, and the experimental data was constructed by adding noise to the raw signal. The noise cancellation performance of the noise canceller 100 can be evaluated through comparison of the SNR (see Table 1) of the HRRP (see Figs. 3A and 3B) before and after the noise cancellation. That is, the noise canceller 100 can effectively remove HRRP noise. Thus, the noise canceller 100 can improve radar target identification performance.

4 is a diagram illustrating a noise removal method of HRRP using EMD according to an embodiment of the present invention.

The noise eliminator 100 receives the radar signal and performs Fourier transform (S101). That is, the noise eliminator 100 generates the received radar signal as a Fourier transform signal. Thereafter, the noise eliminator 100 applies the EMD technique to each of the real and imaginary components of the Fourier transform signal to remove noise (S102). Then, the noise removal apparatus 100 forms an HRRP with respect to the Fourier transform signal from which the noise is removed (S103). That is, the noise removing apparatus 100 takes the absolute value of the noise-removed Fourier transform signal to form HRRP.

It will be apparent to those skilled in the art that various modifications and equivalent arrangements may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Accordingly, it is to be understood that the present invention is not limited to the above-described embodiments. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims. It is also to be understood that the invention includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.

110: Radar signal receiving unit 120: Noise removing unit
121: Fourier transform unit 122: EMD processing unit
130: HRRP forming part

Claims (11)

A radar signal receiving unit for receiving a radar signal including Gaussian noise;
A noise eliminator for eliminating noise of the radar signal using an EMD (mpirical Mode Decomposition) technique; And
An HRRP forming unit for forming a HRRP (High Resolution Range Profile) with respect to the noise canceled signal;
A noise canceling device of HRRP using EMD including
The method according to claim 1,
The noise-
A Fourier transform unit for Fourier transforming the radar signal to generate a Fourier transform signal;
An EMD processor for removing noise by applying an EMD technique to the Fourier transform signal;
A noise canceling device of HRRP using EMD including
3. The method of claim 2,
The EMD processing unit,
Wherein an EMD technique is applied to each of a real component and an imaginary component of the Fourier transform signal.
The method of claim 3,
Wherein the Fourier transform signal is a real signal and the average of the noise included in the signal is Gaussian noise of '0'.
The method according to claim 1,
The HRRP forming unit includes:
Wherein the HRRP is formed by taking an absolute value of the noise-removed signal.
6. The method according to any one of claims 1 to 5,
The noise cancellation device includes:
Wherein the signal is included in a radar using a stepped frequency waveform (SFW).
Receiving a radar signal including Gaussian noise;
Removing noise of the radar signal using an EMD (mpirical Mode Decomposition) technique; And
Forming a High Resolution Range Profile (HRRP) on the noise canceled signal;
A method for removing HRRP noise using an EMD.
8. The method of claim 7,
Wherein the noise removing step comprises:
Generating a Fourier transform signal by Fourier transforming the radar signal; And
Removing noise by applying an EMD technique to the Fourier transform signal;
A method for removing HRRP noise using an EMD.
9. The method of claim 8,
In the EMD application step,
Wherein the EMD technique is applied to the real and imaginary components of the Fourier transform signal.
10. The method of claim 9,
Wherein the Fourier transform signal is a real signal and the average of the noise included in the signal is Gaussian noise of '0'.
8. The method of claim 7,
The HRRP forming step includes:
Wherein the HRRP is formed by taking an absolute value of the noise-removed signal.

Images