KR20210087344A - Apparatus and method for target estimation - Google Patents

Abstract

A target estimation method includes: a step of transmitting radar signals using a step frequency waveform to a target; a step of generating a time-distance image based on the radar signals received by the target reflecting the radar signals; a step of extracting the upper and lower envelopes of the time-distance image; a step of generating a sinusoidal signal based on the upper and lower envelopes; and a step of estimating at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal by applying the sinusoidal signal to nonlinear least square.

Description

Target estimation apparatus and method {APPARATUS AND METHOD FOR TARGET ESTIMATION}

The present invention relates to a target estimation apparatus and a target estimation method.

In order to accurately respond to the fragmentation fragments of missiles with high-speed and high mobility characteristics, and the rotating fragments generated by explosions in SAAM (Surface-to-Air Anti-Missile), which intercepts the opponent's aircraft by firing missiles from ships, In order to do this, it is necessary to grasp the motion information and physical characteristics of the targets (the fragments) by using the micro-Doppler characteristics.

In this regard, when a target (eg, the target may be rotating) is detected through the radar, the radar reception signal hit and reflected by the target includes motion information and physical characteristic information of the target, at this time, Detailed information such as their posture, length and width could not be obtained.

A high-resolution image or image is required to obtain detailed information such as the posture, length, and width of these targets. The value obtained from the high-resolution image or image of the radar reception signal hit and reflected by the target is the rotation vector of the target. and the value changed by the relationship between the radar and the LOS vector between the target and the target, so it was not possible to estimate the actual size of the target.

Therefore, there is a need for a technology capable of accurately estimating the size of the actual target by accurately estimating the rotation vector of the target, the width of the target, and the length of the target with respect to the rotating target.

Korean Patent Laid-Open Publication No. 10-2018-0091372 (published on August 16, 2018)

An object of the present invention is to provide a target estimation apparatus and a target estimation method.

In addition, accurately estimating information on the rotation vector of the target and the width and length of the target through such an apparatus and method may be included in the problems to be solved by the present invention.

However, the problems to be solved of the present invention are not limited to those mentioned above, and other problems to be solved that are not mentioned can be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be.

A target estimation method according to an embodiment comprises the steps of: transmitting a plurality of radar signals using a step frequency waveform to a target; the plurality of radar signals are reflected by the target, and a time-distance based on the received radar signals generating an image, extracting a maximum envelope and a lower envelope of the time-distance image, and generating a sinusoidal signal based on the maximum envelope and the lowest envelope; Applying the sinusoidal signal to a nonlinear least square method (NLS) to estimate at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal. Can include have.

Calculating vector values for the plurality of radar signals based on at least one estimated value of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the estimated value of the current value of the sinusoidal signal; Vector values for a plurality of radar signals are applied to the nonlinear least squares (NLS) method to obtain a rotation angle with respect to one axis of the target, a rotation angle for an axis different from the one axis of the target, a width of the target and The method may further include estimating at least one of the lengths of the target.

In addition, the step of estimating at least one of a rotation angle of the target with respect to one axis, a rotation angle of the target about an axis different from the one axis, a width of the target, and a length of the target may include: At least one of a rotation angle of the target about one axis, a rotation angle of the target about an axis different from the one axis, a width of the target, and a length of the target can be estimated using a line of sight (LOS) vector between targets have.

In addition, the generating of the time-distance image includes performing image processing on the generated time-distance image, wherein the image processing compares each of the plurality of radar signals with a preset value, and the comparison result Accordingly, the plurality of radar signals may be converted into setting data.

A target estimation apparatus according to an embodiment includes: a radar transmitter configured to transmit a plurality of radar signals using a step frequency waveform to a target; a time-distance image generator configured to reflect the plurality of radar signals to the target and generate a time-distance image based on the received radar signals; An envelope extractor for extracting a maximum envelope and a lower envelope of the time-distance image, and a sinusoidal signal generator for generating a sinusoidal signal based on the maximum envelope and the lowest envelope; A first target data estimator for estimating at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal by applying a signal to a nonlinear least square method (NLS) can do.

In addition, based on at least one estimated value of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the estimated value of the current value of the sinusoidal signal estimated by the first target data estimator, a vector value calculator for calculating a vector value for the radar signal; It may further include a second target data estimator for estimating at least one of a rotation angle with respect to an axis different from , a width of the target, and a length of the target.

In addition, the second target data estimator, using a line of sight (LOS) vector between the plurality of radar signals and the target, a rotation angle with respect to one axis of the target, a rotation angle of the target with respect to an axis different from the one axis , at least one of a width of the target and a length of the target may be estimated.

In addition, the time-distance image generator performs image processing on the generated time-distance image, wherein the image processing compares each of the plurality of radar signals with a preset value, and according to the comparison result, the A plurality of radar signals can be converted into setting data.

A computer-readable recording medium according to an embodiment is a computer-readable recording medium storing a computer program, and the computer program, when executed by a processor, transmits a plurality of radar signals using a step frequency waveform to a target. generating a time-distance image based on the received radar signals by reflecting the plurality of radar signals to the target, and an upper envelope and a lower envelope of the time-distance image. envelope), generating a sinusoidal signal based on the maximum envelope and the minimum envelope, and applying the sinusoidal signal to a Nonlinear Least Square (NLS) angular velocity of the target, the target The method may include instructions for causing the processor to perform a method including estimating at least one of a phase of , a magnitude of the sinusoidal signal, and a current value of the sinusoidal signal.

A computer program according to an embodiment is a computer program stored in a computer-readable recording medium, wherein the computer program, when executed by a processor, comprises the steps of: transmitting a plurality of radar signals using a step frequency waveform to a target; generating a time-distance image by reflecting the plurality of radar signals on the target, based on the received radar signals, and generating an upper envelope and a lower envelope of the time-distance image extracting, generating a sinusoidal signal based on the maximum envelope and the minimum envelope, and applying the sinusoidal signal to a Nonlinear Least Square (NLS) method to determine the angular velocity of the target, the phase of the target, The method may include instructions for causing the processor to perform a method including estimating at least one of a magnitude of the sinusoidal signal and a current value of the sinusoidal signal.

The apparatus for estimating a target according to an embodiment may accurately estimate information about a rotation vector of a target and a width and length of the target.

In addition, when a plurality of radar signals (eg, using a step frequency waveform) are reflected by a target and the target is estimated based on the received radar signals, a measurement value for the received radar signals requires phase information. Therefore, when transmitting a plurality of radar signals, a synchronization operation for the radar signals may not be required.

1 is a block diagram of an apparatus for estimating a target according to an embodiment.
2 is a diagram illustrating a target according to an embodiment.
3 is a diagram illustrating a time-distance image before image processing is performed on a time-distance image according to an exemplary embodiment.
4 is a diagram illustrating a time-distance image obtained by performing image processing on a time-distance image according to an exemplary embodiment.
5 is a graph for confirming performance of an apparatus for estimating a target according to an exemplary embodiment.
6 is a graph illustrating a root mean square error value (RMSE: Root-) of estimated values when white Gaussian noise is considered with respect to estimated values of a phase, a rotation angle, a width, and a length of a target estimated by the target estimation apparatus according to an embodiment; Mean-Squared-Error) and a graph showing the Cramer-Rao bound (CRB) value for the square root.
7 is an exemplary flowchart of a target estimation method according to an embodiment.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the art to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims.

In describing the embodiments of the present invention, if it is determined that a detailed description of a well-known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, the terms to be described later are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification.

1 is a block diagram of an apparatus for estimating a target according to an embodiment.

The target estimation apparatus 100 according to an embodiment transmits a plurality of radar signals to a target (eg, the target may be rotating), and based on the radar signals reflected by the target and received, the posture (eg, the target) , which may include the rotation vector of the target, etc.) and the size of the target (eg, it may include the width of the target and the length of the target), and the like.

More specifically, the target estimation apparatus 100 according to an embodiment includes a radar transmitter 110 , a time-distance image generator 120 , an envelope extractor 130 , a sinusoidal signal generator 140 , and a first target. It may include a data estimator 150 , a vector value calculator 160 , and a second target data estimator 170 , but is not limited thereto. In addition, since the target estimation apparatus 100 of FIG. 1 is only an embodiment of the present invention, the spirit of the present invention is not interpreted as being limited by FIG. 1 .

In this case, according to an embodiment, the target estimation apparatus 100 may not include at least one of the components shown in FIG. 1 and/or may additionally include a component not shown in FIG. 1 . In addition, each of the components included in the target estimation apparatus 100 may be implemented in the form of a software module or a hardware module, or a combination of a software module and a hardware module, for example, a computer or a smart device, etc. can be electrically connected.

The radar transmitter 110 may transmit a plurality of radar signals using the step frequency waveform to the target.

The step frequency waveform (s(t))) of the radar signal transmitted to the target has, for example, M burst signals and N pulses per one burst signal. can be expressed together.

는

이며, m과 n은 각각 계단 주파수 파형(s(t)))의 m번째 버스트(burst) 신호, 계단 주파수 파형(s(t)))의 n번째 펄스를 의미하며, 여기서

,

의 값을 가질 수 있다.here,

is

where m and n are the m-th burst signal of the step frequency waveform (s(t))) and the n-th pulse of the step frequency waveform (s(t))), respectively, where

,

can have a value of

는 계단 주파수 파형(s(t)))의 펄스 폭을 나타내고,

는 계단 주파수 파형(s(t)))의 펄스 반복 간격을 나타낸다.Also,

denotes the pulse width of the step frequency waveform (s(t))),

denotes the pulse repetition interval of the step frequency waveform s(t)).

라고 표현하면, 수신된 레이더 신호(

)는 하기 수학식 2와 같이 나타낼 수 있다.On the other hand, the time delay for the received radar signal by hitting the target and being reflected

When expressed as , the received radar signal (

) can be expressed as in Equation 2 below.

는 표적으로부터 반사되어, 수신된 레이더 신호에 대한 크기이다.here

is the magnitude of the received radar signal reflected from the target.

)를 복조(demodulation) 시키기 위해 사용하는 기준 신호(reference signal,

)를 정의하면 하기 수학식 3과 같이 나타낼 수 있다.Further, the received radar signal (

) used to demodulate the reference signal (reference signal,

), it can be expressed as in Equation 3 below.

)은 하기의 수학식 4와 같이 나타낼 수 있다.In addition, through Equation 3, a sample for the m-th burst signal and the demodulated received radar signal for the n-th pulse (

) can be expressed as in Equation 4 below.

Hereinafter, the target will be described in detail with reference to FIG. 2 .

2 is a diagram illustrating a target according to an embodiment.

이고, 높이가

이며, x축에 대해서

, y축에 대해서

만큼 회전한 축에 대하여

만큼의 각속도를 가지고 회전하고 있다고 가정한다. The target 50 shown in FIG. 2 has a width

and the height

and for the x-axis

, about the y-axis

about an axis rotated by

Assume that it is rotating with an angular velocity of

For example, the target 50 shown in FIG. 2 may be a plank-shaped target, but this is only an example and the target may have various shapes.

In addition, the target estimation apparatus according to an embodiment uses an LOS vector between a target and a radar signal (eg, may be the radar transmitter 110 ) in order to estimate rotation variables, and in this case, the target and the radar so that the LOS vector does not change. It is assumed that the distance between the signal radar signals (eg, may be the radar transmitter 110 ) is sufficiently large.

라고 정의한다.Here, the reference coordinate system is defined as (X, Y, Z), and the coordinate system for the target 50 is parallel to the reference coordinate system and

define it as

)는 도 2에 도시된 바와 같이 기울어져 있으며, 고려하는 표적(50)이 4개의 모서리를 포함하여 Q개의 산란점들로 이루어져 있다고 가정하면, 표적의 q번째 산란점의 위치(

)는 표적 (50)에 대한 중력 중심(center of gravity)

에 대해서 수평 방향으로

, 수직 방향으로

의 범위에 위치하게 되고, 하기의 수학식 5와 같이 표현할 수 있다.At this time, the rotation unit vector for the target 50 (

) is inclined as shown in FIG. 2, and assuming that the target 50 under consideration consists of Q scattering points including four corners, the position of the q-th scattering point of the target (

) is the center of gravity for the target 50

in the horizontal direction for

, in the vertical direction

It is located in the range of , and can be expressed as in Equation 5 below.

는 표적에 대한 초기 위상이며, 여기서

와

는 각각 x축, y축에 대한 방향 코사인 행렬(Direction Cosine Matrix)이다.In Equation 5 above

is the initial phase for the target, where

Wow

is a direction cosine matrix for the x-axis and y-axis, respectively.

)은 하기의 수학식 6과 같이 나타낼 수 있다.At this time, the time delay of the radar reception signal reflected by the q-th scattering point of the target (

) can be expressed as in Equation 6 below.

는 레이더 신호에 대한 표적의 위치 벡터를 나타낸다.In Equation 6, c represents the speed of light,

denotes the target's position vector with respect to the radar signal.

At this time, assuming that the distance between the target 50 and the radar signal (for example, the radar transmitter 110) for transmitting the radar signal is far apart, Equation 6 is a simple equation as in Equation 7 below. can indicate

는 레이더 신호에 대한 LOS 벡터이며,

와 같이 계산할 수 있다.In Equation 7 above

is the LOS vector for the radar signal,

can be calculated as

In addition, by substituting Equation 5 into Equation 7, Equation 8 below can be derived.

, A, B, C,

는 하기의 수학식 9와 같이 계산할 수 있다.At this time, in Equation 8

, A, B, C,

can be calculated as in Equation 9 below.

,

는 각각

,

로 나타내면, 이는 레이더 신호와 표적(50)간의 기하학적 관계와 표적(50)의 회전 벡터에 대해서 스칼라 값이 곱해져 있는 형태가 된다.At this time, in order to express it conveniently in terms of notation, in Equation 8

,

are each

,

When expressed as , this is in a form in which a scalar value is multiplied by the geometric relationship between the radar signal and the target 50 and the rotation vector of the target 50 .

)은 하기의 수학식 10과 같이 나타낼 수 있다.Therefore, through this, the time delay for the q-th scattering point of the target 50 (

) can be expressed as in Equation 10 below.

)은 하기의 수학식 11과 같이 나타낼 수 있다.At this time, through Equation (10), a sample for the m-th burst signal and the demodulated received radar signal for the n-th pulse (

) can be expressed as in Equation 11 below.

Therefore, the time delay of the received radar signal that hits the target 50 and is reflected through Equations 10 and 11 is not only the rotation speed and the size of the target 50, but also the geometric relationship between the radar signal and the target 50. , it can be seen that the rotation vector of the target 50 is also affected.

In contrast, the target estimating apparatus 100 according to an embodiment may generate a time-distance image based on the received radar signals in order to accurately estimate information on the rotation vector and the width and length of the target.

In more detail, the time-distance image generator 120 may generate a time-distance image based on a plurality of radar signals reflected by the target 50 and received radar signals.

)과 표적(50)의 길이(

)를 추정하고, 표적(50)에 대한 회전 단위 벡터와 실제 표적(50)에 대한 형상 관련 변수들을 계산할 수 있다.At this time, the target estimating apparatus 100 according to an embodiment is a time-distance image of a target 50 that is multiplied by an angular velocity and a scalar value based on High Resolution Range Profiles (HRRP). width of

) and the length of the target 50 (

) may be estimated, and a rotation unit vector for the target 50 and shape-related variables for the actual target 50 may be calculated.

Hereinafter, the estimation of values related to the angular velocity and shape of the target 50 will be described in more detail.

When using a step frequency waveform (which may be a step frequency signal), the distance from the target 50 to the target estimation apparatus 100 according to an embodiment has N different center frequencies in each burst signal of the radar transmission signal. It can be estimated by finding a peak value in the result of applying an inverse discrete Fourier transform (IDFT) to samples obtained from pulses.

However, when the target 50 rotates with an angular velocity rather than in a stationary state, since the peak value may deviate from the existing position, blur may occur.

)들을

이라고 정의하면, 레이더 송신 신호에서 m번째 버스트(burst) 신호에서 q번째 산란점에 의한 첨두치의 위치(

)는 하기의 수학식 12와 같이 나타낼 수 있다.In contrast, the absolute value of the result of the inverse discrete Fourier transform (IDFT) in the m-th burst signal in the radar transmission signal (

) to hear

If defined as , the position of the peak by the q-th scattering point in the m-th burst signal in the radar transmission signal (

) can be expressed as in Equation 12 below.

는 0부터 N-1까지의 범위를 가지며,

와 같이 나타낼 수 있다.In Equation 12 above

has a range from 0 to N-1,

can be expressed as

과 같게 되고, 이는 도 3에 도시된 바와 같이 시간-거리 영상을 생성할 수 있다.At this time, if the absolute values of the results of the inverse discrete Fourier transform (IDFT) of the burst signal of each radar transmission signal are expressed in a matrix,

, which can generate a time-distance image as shown in FIG. 3 .

In addition, as shown in Equation 12, Q sine wave signals having different amplitudes and current values (for example, including DC components) of the radar signal are obtained in the time-distance image by the rotating target having Q scattering points. can do.

는 음수와 양수 둘 다 사용될 수 있으므로 예컨대, 위상차이가 180도 차이 나는 두 종류의 정현파 신호가 생성될 수도 있다.At this time,

Since both negative and positive numbers can be used, for example, two types of sinusoidal signals having a phase difference of 180 degrees may be generated.

으로부터 가장 멀리 떨어져 있을 때의 거리는

가 될 수 있으며, 이때,

,

가 되므로 Q개의 산란점 중 가장 큰

는

가 될 수 있다.On the other hand, the sinusoidal signal by the q-th scattering point is

the distance at the farthest from

can be, in which case

,

becomes the largest among the Q scattering points.

is

can be

,

와

를 추정할 수 있게 된다.These values are values directly related to the size of the target 50 , and the angular velocity of the target 50 to estimate the rotation vector of the target 50 and the size of the target 50 .

,

Wow

can be estimated.

Hereinafter, estimation of the angular velocity and size of the target 50 for estimating the size of the target 50 will be described in detail later.

), 표적의 폭(

), 표적의 길이(

))을 추정하기 위해서는 시간-거리 영상을 이용하게 되는데, 관심 있는 변수들의 추정 성능을 높이기 위해서 시간-거리 영상에 영상 처리를 수행할 수 있다.First, the values related to the angular velocity and shape of the target (the angular velocity of the target (

), the width of the target (

), the length of the target (

)), a time-distance image is used, and image processing can be performed on the time-distance image to increase the estimation performance of the variables of interest.

의 데이터를 획득할 수 있다.In this case, by performing image processing on a time-distance image,

of data can be obtained.

은 하기의 수학식 13과 같이 나타낼 수 있다.At this time, the value of the nth range bin in the mth burst signal with respect to the received radar signal

can be expressed as in Equation 13 below.

는 가드(guard) 셀(cell)을 나타내는 것으로 1로 설정하고, 시간-거리 영상 생성부(120)는 dB 스케일(scale)을 임계값(thresholding)에 기초하여 처리하는 방법으로 영상 처리를 수행할 수 있다.In Equation 13 above

is set to 1 to indicate a guard cell, and the time-distance image generator 120 performs image processing by processing a dB scale based on a threshold value. can

More specifically, the time-distance image generator 120 may compare each of the plurality of radar signals with a preset value, and convert the plurality of radar signals into setting data according to the comparison result.

For example, the time-distance image generator 120 may set values smaller than a preset value (eg, may be 3 dB) to 0 for each burst signal of a plurality of radar signals, as shown in FIG. By performing image processing together to reduce unnecessary noise, estimation performance for variables of interest can be improved.

The envelope extractor 130 may extract an upper envelope and a lower envelope of the time-distance image.

More specifically, the envelope extractor 130 may extract a maximum envelope and a lower envelope with respect to all burst signals from the image-processed time-distance image, but is limited thereto. it's not going to be

,

라고 정의하면, 하기의 수학식 14 및 수학식 15와 같이 나타낼 수 있다.At this time, the maximum envelope (upper envelope) and the lowest envelope (lower envelope) for the m-th burst signal in the radar signal of the time-distance image, respectively

,

, it can be expressed as Equations 14 and 15 below.

In this case, when values satisfying the conditions of Equations 14 and 15 are found, the maximum envelope and the lower envelope may be expressed by Equations 16 and 17 below, respectively.

이며,

는

,

는

로 나타낼 수 있다.At this time,

is,

is

,

is

can be expressed as

The sinusoidal signal generator 140 may generate a sinusoidal signal based on an upper envelope and a lower envelope of the time-distance image.

만큼의 각속도를 가지고, 정현파 신호의 크기가

, 정현파 신호의 전류값(예컨대, 정현파 신호의 DC 성분일 수 있음)이

인 절대값이 씌워진 정현파 신호

을 하기의 수학식 18과 같이 획득할 수 있다.More specifically, the sinusoidal wave signal generator 140 subtracts Equation 16 from Equation 17 to

With an angular velocity equal to, the magnitude of the sinusoidal signal is

, the current value of the sinusoidal signal (for example, may be the DC component of the sinusoidal signal) is

A sinusoidal signal overlaid with an absolute value of

can be obtained as in Equation 18 below.

와

값을 계산하기 위해서는

와

에 대한 값을 필요로 하게 되며, 상기 수학식 18을 벡터 형태로 표현하면 하기 수학식 19 및 수학식 20과 같이 나타낼 수 있다.Meanwhile,

Wow

To calculate the value

Wow

is required, and when Equation 18 is expressed in a vector form, it can be expressed as Equation 19 and Equation 20 below.

The first target data estimator 150 applies the sinusoidal signal to a nonlinear least square method (NLS) to estimate at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal. can

), 표적의 위상(

), 정현파 신호의 크기(

), 정현파 신호의 전류값(

)을 추정할 수 있다.More specifically, the first target data estimator 150 assumes that 1 in Equation 20 is a column vector having a size of M x 1 in which all elements are 1, and in Equations 19 and 20, When a nonlinear least square (NLS) method is applied to

), the phase of the target (

), the magnitude of the sinusoidal signal (

), the current value of the sinusoidal signal (

) can be estimated.

On the other hand, since the objective function of Equation 21 has a myriad of local minimum/maximum values, in the present invention, grid search, simulated annealing, particle swarm optimization (PSO), quasi -You can obtain an initial estimate using an algorithm such as Newton, etc., and obtain a more accurate estimate through a gradient based method.

Hereinafter, a method of estimating variable values for determining the posture and shape of the target will be described in detail later.

,

,

,

)들을 측정값을

,

로 표현하면 각각 수학식 22 및 수학식 23과 같이 나타낼 수 있다.The value estimated by the first target data estimator 150 (

,

,

,

) to take the measurements

,

It can be expressed as Equation 22 and Equation 23, respectively.

), 표적의 상기 일축과 상이한 축에 대한 회전 각도(

), 표적의 폭(

), 표적의 길이(

)로서, 총 4개인데 상기 수학식 21 및 상기 수학식 22는 비결정 시스템 (underdetermined system)이므로 해를 구할 수 없게 된다. At this time, as can be seen from the equations of Equation 21 and Equation 22, the variables to be estimated are the rotation angle (

), the angle of rotation about an axis different from the one axis of the target (

), the width of the target (

), the length of the target (

), there are 4 in total, and since Equation 21 and Equation 22 are an underdetermined system, a solution cannot be obtained.

Therefore, two more equations are required, and accordingly, at least two radar signals must be transmitted, but four variables can be estimated.

That is, the target estimating apparatus 100 according to an embodiment transmits at least two or more radar signals using the step frequency waveform to the target, and estimates the posture and size of the target based on the received radar signals reflected by the target. can do.

와

를 각각

,

라고 정의하면, 각각 하기의 수학식 24 및 수학식 25와 같이 나타낼 수 있다.On the other hand, the measured value in the p-th radar signal

Wow

each

,

, it can be expressed as Equation 24 and Equation 25 below, respectively.

,

,

,

는 상기 수학식 9에서 정의한 p번째 레이더에서의 이전에 정의한 값들이며, P는 고려하는 레이더의 총 개수를 의미한다. Here, in Equations 24 and 25,

,

,

,

are previously defined values of the p-th radar defined in Equation 9, and P denotes the total number of radars to be considered.

On the other hand, it can be seen through the above equations that phase information is not included in the measured values for a plurality of radars, which means that synchronization is not required for radars when using the algorithm proposed in the present invention. that can be checked

,

,

,

) 중 적어도 하나의 추정값에 기초하여 복수의 레이더 신호에 대한 벡터값을 계산할 수 있다.The vector value calculation unit 160 estimates the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal (

,

,

,

), a vector value for a plurality of radar signals may be calculated based on at least one estimated value.

를

로 변환하고, 상기 수학식 25에서

를

로 변환 한다고 가정하여, 총 P개의 레이더로부터의 전체 측정 벡터 z를 하기의 수학식 26과 같이 나타낼 수 있다.More specifically, the vector value calculation unit 160, in Equation 24,

to

, and in Equation 25 above

to

Assuming that the conversion is performed to , the total measurement vector z from the total P radars can be expressed as in Equation 26 below.

,

,

,

로 나타낼 수 있다. In this case, Equation 26 is

,

,

,

can be expressed as

), 표적의 상기 일축과 상이한 축(예컨대, 표적의 x,y,z축 중 y축일 수 있음)에 대한 회전 각도(

), 표적의 폭(

), 표적의 길이(

)중 적어도 하나를 추정할 수 있다.The second target data estimator 170 applies a vector value z for a plurality of radar signals to a nonlinear least square method (NLS) on one axis of the target (eg, one of the x, y, and z axes of the target). angle of rotation (which can be the x-axis)

), the angle of rotation about an axis different from the one axis of the target (eg, which may be the y axis of the x, y, z axis of the target) (

), the width of the target (

), the length of the target (

) can be estimated.

), 표적의 상기 일축과 상이한 축에 대한 각도(

), 표적의 폭(

) 및 표적의 길이(

))를 차례로 추정할 수 있다.More specifically, the second target data estimator 170 uses a nonlinear least square method (NLS) with respect to Equation 26 to obtain four variables (the angle of the target with respect to one axis) as shown in Equation 27 below. (

), the angle with respect to an axis different from the one axis of the target (

), the width of the target (

) and the length of the target (

)) can be estimated sequentially.

On the other hand, the vector value z for the plurality of radar signals is information obtained through estimation of the magnitude of the sinusoidal signal and the current value of the sinusoidal signal (for example, it may be a DC component of the sinusoidal signal) without phase information of the signal for each radar signal. Therefore, it is not necessary to perform the synchronization operation of the radar signal.

5 is a graph for confirming performance of an apparatus for estimating a target according to an exemplary embodiment.

)에 따른 실제 신호는 검은색 실선으로 표시하고, 시간-거리 영상에서 추출한 신호는 파란색 점섬으로 표시하고, 시간-거리 영상에서 추출한 신호를 기반으로 추정한 신호는 빨간색 점으로 표시하였다.Referring to FIG. 5 , the angle formed by the LOS vector between the target and the radar signal (eg, may be a radar transmitter) and the rotation vector of the target (

) is indicated by a solid black line, the signal extracted from the time-distance image is indicated by a blue dotted line, and the signal estimated based on the signal extracted from the time-distance image is indicated by a red dot.

More specifically, as shown in the graph shown in FIG. 5 , it can be seen that the amplitude and current value of the sinusoidal signal to which the absolute value is covered varies according to the angle (0, 30, 60, 90) formed by the rotation vector of the target, and the sinusoidal signal Even if there is an error, it can be confirmed that the signal estimated based on the signal extracted from the time-distance image is estimated similarly to the actual signal through the nonlinear least squares method (NLS).

6 is a graph illustrating a root mean square error value (RMSE: Root-) of estimated values when white Gaussian noise is considered with respect to estimated values of a phase, a rotation angle, a width, and a length of a target estimated by the target estimation apparatus according to an embodiment; Mean-Squared-Error) and a graph showing the Cramer-Rao bound (CRB) value for the square root.

,

,

,

) 들에 대하여 정확도를 고려하기 위해 평균 제곱근 오차 수치(RMSE: Root-Mean-Squared-Error) 및 제곱근에 대한 크래머-라오 하한 수치(CRB: Cramer-Rao bound) 값과 비교해본 결과, 레이더 신호가 많을수록 표적의 추정값들과 평균 제곱근 오차 수치(RMSE: Root-Mean-Squared-Error) 및 제곱근에 대한 크래머-라오 하한 수치(CRB: Cramer-Rao bound) 값이 근접하게 나타났으며, 이는 레이더 신호가 많을수록 표적의 추정값들에 대한 정확도가 높아지는 것을 의미한다.6, the estimated value of the target (

,

,

,

), compared with the root-mean-squared-error value (RMSE: Root-Mean-Squared-Error) and the Cramer-Rao bound value for the square root (CRB: Cramer-Rao bound) to consider the accuracy, the radar signal As the number increases, the target estimates, Root-Mean-Squared-Error (RMSE) and Cramer-Rao bound (CRB: Cramer-Rao bound) values for square root appear closer. It means that the more signals, the higher the accuracy of the estimates of the target.

,

,

,

) 들에 대한 정확도가 향상될 수 있다.Therefore, as mentioned above, the target estimation apparatus 100 according to an embodiment may use at least two or more radar signals, and accordingly, the target estimation value (

,

,

,

) can be improved.

7 is an exemplary flowchart of a target estimation method according to an embodiment. The target estimation method illustrated in FIG. 7 may be performed by the target estimation apparatus 100 illustrated in FIG. 1 . In addition, the target estimation method illustrated in FIG. 7 is merely exemplary.

Referring to FIG. 7 , the radar transmitter 110 may transmit a plurality of radar signals using a step frequency waveform to a target (step S10 ).

In this case, the plurality of radar signals may be at least one radar signal, preferably two or more radar signals, but is not limited thereto.

The time-distance image generator 120 may generate a time-distance image based on the received radar signals by reflecting a plurality of radar signals on the target (step S20 ).

More specifically, the time-distance image generator 120 may generate a time-distance image and check a high resolution range profile (HRRP) from the time-distance image, but is not limited thereto.

The envelope extractor 130 may extract a maximum envelope and a lower envelope of the time-distance image (step S30 ).

The sinusoidal signal generator 140 may generate a sinusoidal signal based on an upper envelope and a lower envelope (step S40).

The first target data estimator 150 applies the sinusoidal signal to a nonlinear least square method (NLS) to estimate at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal. can be (step S50).

The vector value calculator 160 may calculate vector values for the plurality of radar signals based on at least one estimated value among the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the estimated value of the current value of the sinusoidal signal ( step S60).

), 표적의 상기 일축과 상이한 축(예컨대, x,y,z축 중 y축일 수 있음)에 대한 회전각도(

), 표적의 폭(

) 및 표적의 길이(

) 중 적어도 하나를 추정할 수 있다(단계 S70).The second target data estimator 170 applies the vector values for the plurality of radar signals to the nonlinear least squares method (NLS), and the rotation angle with respect to one axis of the target (eg, may be the x axis among the x, y, and z axes). (

), an angle of rotation about an axis different from the one axis of the target (eg, the y axis may be one of the x, y, and z axes)

), the width of the target (

) and the length of the target (

) can be estimated at least one of (step S70).

As described above, the target estimating apparatus according to an embodiment may accurately estimate information on a rotation vector of a target and information on a width and length of a target.

In addition, when a plurality of radar signals (eg, using a step frequency waveform) are reflected by a target and the target is estimated based on the received radar signals, a measurement value for the received radar signals requires phase information. Therefore, when transmitting a plurality of radar signals, a synchronization operation for the radar signals may not be required.

The above description is merely illustrative of the technical idea of the present invention, and various modifications and variations will be possible without departing from the essential quality of the present invention by those skilled in the art to which the present invention pertains. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

50: target
100: target estimation device
110: radar transmitter
120: time-distance image generation unit
130: envelope extraction unit
140: sinusoidal signal generator
150: first target data estimator
160: vector value calculator
170: second target data estimator

Claims (10)

Transmitting a plurality of radar signals using a step frequency waveform to a target;
generating a time-distance image by reflecting the plurality of radar signals on the target based on the received radar signals;
Extracting the maximum envelope (upper envelope) and the lowest envelope (lower envelope) of the time-distance image;
generating a sinusoidal signal based on the maximum envelope and the minimum envelope;
estimating at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal by applying the sinusoidal signal to a nonlinear least square method (NLS)
Target estimation method. The method of claim 1,
calculating vector values for the plurality of radar signals based on at least one estimated value of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the estimated value of the current value of the sinusoidal signal;
Vector values of the plurality of radar signals are applied to the nonlinear least square method (NLS) to apply a rotation angle of the target about one axis, a rotation angle of the target about an axis different from the one axis, and the width of the target and estimating at least one of the length of the target.
Target estimation method. 3. The method of claim 2,
estimating at least one of a rotation angle of the target with respect to one axis, a rotation angle of the target about an axis different from the one axis, a width of the target, and a length of the target,
Using a Line of Sight (LOS) vector between the plurality of radar signals and the target, a rotation angle about one axis of the target, a rotation angle of the target about an axis different from the one axis, a width of the target, and a length of the target to estimate at least one of
Target estimation method. The method of claim 1,
The step of generating the time-distance image comprises:
Perform image processing on the generated time-distance image,
The image processing is
comparing each of the plurality of radar signals with a preset value, and converting the plurality of radar signals into set data according to the comparison result
Target estimation method. a radar transmitter for transmitting a plurality of radar signals using a step frequency waveform to a target;
a time-distance image generator configured to reflect the plurality of radar signals to the target and generate a time-distance image based on the received radar signals;
And an envelope extractor for extracting the maximum envelope (upper envelope) and the lowest envelope (lower envelope) of the time-distance image;
a sinusoidal signal generator configured to generate a sinusoidal signal based on the maximum envelope and the minimum envelope;
First target data estimation for estimating at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal by applying the sinusoidal signal to a nonlinear least square method (NLS) including wealth
target estimation device. 6. The method of claim 5,
Vector for the plurality of radar signals based on the estimated value of at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the estimated value of the current value of the sinusoidal signal estimated by the first target data estimator a vector value calculator that calculates a value;
Vector values of the plurality of radar signals are applied to the nonlinear least square method (NLS) to apply a rotation angle of the target about one axis, a rotation angle of the target about an axis different from the one axis, and the width of the target and a second target data estimator for estimating at least one of the lengths of the target.
target estimation device. 7. The method of claim 6,
The second target data estimator,
Using a Line of Sight (LOS) vector between the plurality of radar signals and the target, a rotation angle about one axis of the target, a rotation angle of the target about an axis different from the one axis, a width of the target, and a length of the target to estimate at least one of
target estimation device. 6. The method of claim 5,
The time-distance image generating unit,
Perform image processing on the generated time-distance image,
The image processing is
comparing each of the plurality of radar signals with a preset value, and converting the plurality of radar signals into set data according to the comparison result
target estimation device. As a computer-readable recording medium storing a computer program,
The computer program, when executed by a processor,
Transmitting a plurality of radar signals using a step frequency waveform to a target;
generating a time-distance image by reflecting the plurality of radar signals on the target based on the received radar signals;
Extracting the maximum envelope (upper envelope) and the lowest envelope (lower envelope) of the time-distance image;
generating a sinusoidal signal based on the maximum envelope and the minimum envelope;
Applying the sinusoidal signal to a nonlinear least square method (NLS) to estimate at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal A method comprising the step of estimating including instructions for causing the processor to perform
computer readable recording medium. As a computer program stored in a computer-readable recording medium,
The computer program, when executed by a processor,
Transmitting a plurality of radar signals using a step frequency waveform to a target;
generating a time-distance image by reflecting the plurality of radar signals on the target based on the received radar signals;
Extracting the maximum envelope (upper envelope) and the lowest envelope (lower envelope) of the time-distance image;
generating a sinusoidal signal based on the maximum envelope and the minimum envelope;
Applying the sinusoidal signal to a nonlinear least square method (NLS) to estimate at least one of the angular velocity of the target, the phase of the target, the magnitude of the sinusoidal signal, and the current value of the sinusoidal signal A method comprising the step of estimating including instructions for causing the processor to perform
computer program.

Images