JP5933330B2 - Pulse compressor - Google Patents

Description

  The present invention relates to a pulse compression apparatus that increases distance resolution.

In the pulse compression device disclosed in the following Patent Document 1, a carrier signal is pulse-modulated at a predetermined time interval, and a carrier signal after pulse modulation is modulated in a pulse to generate a transmission signal.
Then, the transmission signal is radiated to the space, while the transmission signal is reflected by the target and returned, and the received signal is correlated with the reference signal (the signal having the same modulation component as the transmission signal). By performing the above, linear chirp pulse compression is performed.
This improves the distance resolution, but weights the amplitude of the transmission signal by multiplying the transmission signal by a window function in order to suppress side lobes after linear chirp pulse compression.

JP-A-5-27019 (paragraph number [0009])

Since the conventional pulse compression apparatus is configured as described above, the distance resolution can be increased. However, in order to increase the distance resolution, the compression ratio TB calculated from the product of the pulse width of the transmission signal and the modulation bandwidth is set so that the shape of the signal spectrum in the signal after linear chirp pulse compression becomes the shape of a sinc function. Although it is necessary to make it sufficiently large, when the compression ratio TB is low, the shape of the signal spectrum in the signal after linear chirp pulse compression is different from the shape of the sinc function, and there is a problem that the distance resolution may deteriorate (see FIG. 15).
Further, when the compression rate TB is low, there is a problem that the sidelobe suppression performance may be deteriorated by multiplying by the window function (see FIG. 16).

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a pulse compression device capable of increasing the distance resolution even when the signal compression ratio after linear chirp pulse compression is low. .

The pulse compression device according to the present invention pulse-modulates a carrier signal at a predetermined time interval, and generates a transmission signal by performing intra- pulse modulation on the carrier signal after pulse modulation using a linear chirped intra-pulse modulation signal. Transmission signal generating means for performing transmission, and pulse compression means for outputting a signal after pulse compression by performing signal processing on the reception signal of the transmission / reception means, and the pulse compression means modulates the transmission signal within the pulse. A correlation signal that generates a reference signal that is the same signal as the component and performs a correlation process between the received signal of the transmission / reception means and the reference signal; and an auto-correlation of the intra-pulse modulation signal used by the transmission signal generation means the inverse of the amplitude of the signal spectrum in the signal as a regular stove function, normal stove function processing of multiplying a regular stove function for the signal after the correlation processing by correlation processing unit And a regular stove function processing unit to implement regular stove function processing unit calculates the signal-to-noise ratio in the signal after the pulse compression in the case of performing normal stove function processing, the signal-to-noise ratio The width of the normalization window function is set according to .

According to this invention, the pulse compression means generates a reference signal that is the same signal as the intra-pulse modulation component of the transmission signal, and performs a correlation process between the reception signal of the transmission / reception means and the reference signal ; Using the inverse of the signal spectrum amplitude in the signal after autocorrelation of the intra-pulse modulated signal used by the transmission signal generation means as the normalized window function, the signal after correlation processing by the correlation processing unit is multiplied by the normalized window function Since the normalization window function processing unit for performing the normalization window function processing is provided, the distance resolution can be improved even when the signal compression rate after linear chirp pulse compression is low.
Further, the normalization window function processing unit calculates a signal-to-noise ratio in the signal after pulse compression when the normalization window function processing is performed, and sets the width of the normalization window function according to the signal-to-noise ratio With this configuration, it is possible to prevent a decrease in signal-to-noise ratio in a signal after pulse compression by suppressing increase in noise at both ends of the frequency modulation band of the transmission signal in accordance with the normalization window function processing. There is an effect that can be done.

It is a block diagram which shows the pulse compression apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the relationship between the frequency and time in a transmission signal and a received signal. It is explanatory drawing which shows the signal spectrum of signal _{Fv * Ex} (n, _kr ) after the correlation process in the case of a high compression rate and a low compression rate. It is an explanatory view showing that the signal after the correlation processing by the regular stove function processing _{F v · Ex (n, k} r) signal strength is constant. It is explanatory drawing which shows that the sinc shape and distance resolution of the signal after linear chirp pulse compression of a low compression rate improve by performing a normalization window function process. It is explanatory drawing which shows that the sidelobe suppression performance of the signal after the linear chirp pulse compression of a low compression rate improves by performing a normalization window function process and a window function process. It is explanatory drawing which shows the attention cell, guard cell, and sample cell regarding a CFAR process. It is a block diagram which shows the pulse compression apparatus by Embodiment 2 of this invention. It is explanatory drawing which shows the example which noise increases with a normalization window function process. It is explanatory drawing which shows the adjustment width | variety of a normalization window function. It is explanatory drawing which shows selection of the width | variety of the normalization window function by the normalization window function process part 21 other than the relationship between the adjustment width | variety of a normalization window function and the signal-to-noise ratio after pulse compression. It is explanatory drawing which shows the improvement of the signal to noise ratio after the pulse compression by the normalization window function process part. It is a block diagram which shows the pulse compression apparatus by Embodiment 3 of this invention. It is explanatory drawing which shows that the sidelobe suppression performance of the signal after the linear chirp pulse compression of a high compression rate improves by performing a normalization window function process and a window function process. It is explanatory drawing which shows the shape of the signal spectrum in the signal after linear chirp pulse compression. It is explanatory drawing which shows the example which the suppression performance of a side lobe deteriorates by multiplying a window function when the compression rate TB is low.

Embodiment 1 FIG.
1 is a block diagram showing a pulse compression apparatus according to Embodiment 1 of the present invention, and FIG. 2 is an explanatory diagram showing the relationship between frequency and time in a transmission signal and a reception signal.
In FIG. 1, a transmission signal generator 1 includes a local oscillator 2, a pulse modulator 3, an intra-pulse modulation signal generator 4, and a transmitter 5. A local oscillation signal L ₀ (t) ( The carrier signal is subjected to pulse modulation, and the local oscillation signal L ′ ₀ (t) after pulse modulation is modulated in the pulse to generate the transmission RF signal Tx (t). The transmission signal generator 1 constitutes transmission signal generation means.

The local oscillator 2 of the transmission signal generator 1 generates a local oscillation signal L ₀ (t) having a constant frequency with a pulse repetition period T _pri , and the local oscillation signal L ₀ (t) is transmitted to the pulse modulator 3 and the receiver 8. Output.
The pulse modulator 3 performs pulse modulation on the local oscillation signal L ₀ (t) output from the local oscillator 2, and outputs the pulse-modulated local oscillation signal L ′ ₀ (t) to the transmitter 5.
The intra-pulse modulation signal generator 4 generates a linearly chirped intra-pulse modulation signal φ (t) and outputs the intra-pulse modulation signal φ (t) to the transmitter 5.
The transmitter 5 uses the intra-pulse modulation signal φ (t) output from the intra-pulse modulation signal generator 4 to intra-pulse modulate the pulsed modulated local oscillation signal L ′ ₀ (t). The local oscillation signal is output to the transmission / reception switch 6 as a transmission RF signal Tx (t).

The transmission / reception switch 6 outputs the transmission RF signal Tx (t) output from the transmitter 5 of the transmission signal generator 1 to the antenna 7, while receiving the reflected RF signal Rx _tgt (n, t) output from the antenna 7. Output to the receiver 8.
The antenna 7 radiates the transmission RF signal Tx (t) output from the transmission / reception switch 6 to the space, while the transmission RF signal Tx (t) is reflected by the target and returned to the reflected RF signal Rx _tgt (n, t) is received.

The receiver 8 uses the local oscillation signal L ₀ (t) output from the local oscillator 2 to down-convert the reflected RF signal Rx _tgt (n, t) output from the transmission / reception switch 6, and after down-conversion The received video signal S (n, m) is obtained by performing amplification processing and phase detection processing on the reflected RF signal, and the received video signal S (n, m) is output to the signal processor 9.
The transmission / reception switch 6, the antenna 7 and the receiver 8 constitute transmission signal generation means.

The signal processor 9 includes a pulse compression unit 10, a target candidate detection unit 17, and a target relative distance calculation unit 18, and a predetermined signal with respect to the received video signal S (n, m) output from the receiver 8. By performing the process, a process of calculating the relative distance R ′ ₀ to the target candidate is performed.
In the example of FIG. 1, the pulse compression unit 10, the target candidate detection unit 17, and the target relative distance calculation unit 18 that are components of the signal processor 9 are, for example, a semiconductor integrated circuit on which a CPU is mounted, or a one-chip Although it is assumed that it is configured by dedicated hardware such as a microcomputer, the signal processor 9 may be configured by a computer.
When the signal processor 9 is composed of a computer, a program describing the processing contents of the pulse compression unit 10, the target candidate detection unit 17, and the target relative distance calculation unit 18 is stored in the memory of the computer, and the CPU of the computer May execute a program stored in the memory.

The pulse compression unit 10 includes a frequency domain conversion unit 11, a correlation processing unit 12, a normalization window function selection unit 13, a normalization window function processing unit 14, a window function processing unit 15, and a time domain conversion unit 16. The signal R _PC (n, l) after linear chirp pulse compression is output to the target candidate detection unit 17 by performing predetermined signal processing on the received video signal S (n, m) output from the machine 8. Perform the process. The pulse compression unit 10 constitutes a pulse compression means.

The frequency domain transform unit 11 of the pulse compression unit 10 performs fast Fourier transform (FFT) on the received video signal S (n, m) output from the receiver 8 to receive the received video signal F _v (Fast Fourier Transform). n, k _r ) is generated, and the received video signal F _v (n, k _r ) in the frequency domain is output to the correlation processing unit 12.
The correlation processing unit 12 generates the same A / D-converted reference signal Ex (m _τ ) as the intra-pulse modulation component of the transmission RF signal Tx (t) and fast Fourier transforms the reference signal Ex (m _τ ). Thus, the frequency domain reference signal F _Ex (k _r ) is generated, and the correlation processing between the frequency domain received video signal F _v (n, k _r ) and the frequency domain reference signal F _Ex (k _r ) is performed. The correlation-processed signal F _{v · Ex} (n, k _r ) is output to the normalization window function selection unit 13.

The normalization window function selection unit 13 uses the compression rate TB of the signal F _{v · Ex} (n, k _r ) after the correlation processing output from the correlation processing unit 12 as a preset compression rate for selecting the normalization window function. If α or more, the signal F _{v · Ex} (n, k _r ) after correlation processing is output to the window function processing unit 15, and the compression rate TB of the signal F _{v · Ex} (n, k _r ) after correlation processing is output. Is lower than the compression ratio α for selecting the normalization window function, the signal F _{v · Ex} (n, k _r ) after the correlation processing is output to the normalization window function processing unit 14.
The normalized window function processing unit 14 performs signal F _{v · Ex} (n, k _r after correlation processing) so that the shape of the signal spectrum in the signal R _PC (n, l) after pulse compression matches the shape of the sinc function. ) Using the normalized window function w _Norm (k _r ) that makes the signal intensity (amplitude) of the signal spectrum substantially constant in FIG. 4), the normalized window function processing for the correlation signal F _{v · Ex} (n, k _r ) To implement.

Here, the window function in the window function processing performed by the window function processing unit 15 to be described later is, for example, a window function widely used for suppressing side lobes such as a Hamming window (the signal F _{v. In} contrast to the window function designed assuming that the signal intensity of the signal spectrum in _Ex (n, k _r ) is constant, the normalized window function w _Norm (k _r in the normalized window function processing). ) Is a window function having a different shape depending on the compression rate in order to make the signal intensity of the signal spectrum constant in the signal F _{v · Ex} (n, k _r ) after the correlation processing, and the window functions of both are different. .

The window function processing unit 15 outputs the correlation-processed signal F _{v · Ex} (n, k _r ) output from the normalized window function selection unit 13 or the normalized window function processed by the normalization window function processing unit 14. The window function processing is performed on the signal F ′ _{v · Ex} (n, k _r ) to suppress the side rope, and the signal F _{v · Ex · w} (n, k _r ) after the window function processing is converted into the time domain conversion unit 16. Output to.
The time domain conversion unit 16 receives a video signal S (n, m) in greater number than the number of samples, the signal _{F v · Ex · w (n} , k r) after the window function processing inverse fast Fourier transform (IFFT: Generate a signal R _PC (n, l) after pulse compression, which is a time domain signal, and output the signal R _PC (n, l) after pulse compression to the target candidate detection unit 17 by performing an Inverse Fast Fourier Transform. To do.

The target candidate detection unit 17 performs CFAR (Constant False Alarm Rate) processing on the pulse-compressed signal R _PC (n, l) output from the time domain conversion unit 16 of the pulse compression unit 10 to thereby generate the target candidate R _{PC. , CFAR} (n, l) is detected.
The target relative distance calculation unit 18 calculates a relative distance R ′ ₀ to the target candidate R _{PC, CFAR} (n, l) detected by the target candidate detection unit 17.
The display 19 displays the relative distance R ′ ₀ calculated by the target relative distance calculator 18 and the signal R _PC (n, l) after pulse compression on the screen as target information.

Next, the operation will be described.
The transmission signal generator 1 performs pulse modulation on the local oscillation signal L ₀ (t) at predetermined time intervals, and modulates the local oscillation signal L ′ ₀ (t) after pulse modulation in the pulse, thereby transmitting the RF signal Tx ( t).
The processing contents of the transmission signal generator 1 will be specifically described with reference to FIG.
In FIG. 2, f ₀ is the center frequency of the transmission signal, T ₀ is the pulse width of the transmission signal, B ₀ is the frequency modulation bandwidth of the transmission signal, and T _pri is the pulse repetition period.
Here, description will be made assuming that up-chirp modulation is performed as frequency modulation of a transmission signal.

First, the local oscillator 2 of the transmission signal generator 1 generates a local oscillation signal L ₀ (t) having a constant frequency with a pulse repetition period T _pri as shown in FIG. 2, and the local oscillation signal L ₀ (t). Are output to the pulse modulator 3 and the receiver 8.
When the pulse modulator 3 receives the local oscillation signal L ₀ (t) from the local oscillator 2, the pulse modulator 3 performs pulse modulation on the local oscillation signal L ₀ (t) and outputs the local oscillation signal L ′ ₀ (t) after the pulse modulation. Output to the transmitter 5.

On the other hand, the intra-pulse modulation signal generator 4 generates a linearly chirped intra-pulse modulation signal φ (t) and outputs the intra-pulse modulation signal φ (t) to the transmitter 5.
When the transmitter 5 receives the pulse-modulated local oscillation signal L ′ ₀ (t) from the pulse modulator 3 and receives the intra-pulse modulation signal φ (t) from the intra-pulse modulation signal generator 4, the intra-pulse modulation is performed. Using the signal φ (t), the pulse-modulated local oscillation signal L ′ ₀ (t) is modulated in the pulse, and the pulse-modulated local oscillation signal is transmitted to the transmission / reception switch 6 as the transmission RF signal Tx (t). Output.
As shown in FIG. 2, the transmission RF signal Tx (t) is a signal having a pulse width T ₀ and a center frequency f ₀ .

When the transmission / reception switch 6 receives the transmission RF signal Tx (t) from the transmitter 5 of the transmission signal generator 1, the transmission / reception switch 6 outputs the transmission RF signal Tx (t) to the antenna 7.
Thus, the transmission RF signal Tx (t) is radiated from the antenna 7 to the space.
After the transmission RF signal Tx (t) is radiated to the space, a part of the transmission RF signal Tx (t) is reflected to the target and returns to the antenna 7 as a reflected RF signal Rx _tgt (n, t).
The antenna 7 receives the reflected RF signal Rx _tgt (n, t) that is reflected back by the target and outputs the reflected RF signal Rx _tgt (n, t) to the transmission / reception switch 6.
When the transmission / reception switch 6 receives the reflected RF signal Rx _tgt (n, t) from the antenna 7, it outputs the reflected RF signal Rx _tgt (n, t) to the receiver 8.

When the receiver 8 receives the reflected RF signal Rx _tgt (n, t) from the transmission / reception switch 6, the receiver 8 uses the local oscillation signal L ₀ (t) output from the local oscillator 2 and uses the reflected RF signal Rx _tgt ( down-convert n, t).
When the receiver 8 down-converts the reflected RF signal Rx _tgt (n, t), the receiver 8 performs amplification processing and phase detection processing on the reflected RF signal after down-conversion, as shown in the following formula (1). A received video signal S (n, m) is obtained, and the received video signal S (n, m) is output to the signal processor 9.

ここで、Ａ_Sは受信ビデオ信号の振幅、ｎはパルス番号、Ｎはパルス数、ｍは１ＰＲＩ内のサンプリング番号、Ｍは１ＰＲＩ内のサンプリング点数、Δｔは１ＰＲＩ内のサンプリング時間、ｃは光速、Ｒ₀は目標相対距離（図２を参照）、ｖは目標相対速度を表している。

Here, A _S is the amplitude of the received video signal, n is the pulse number, N is the number of pulses, m is the sampling number within 1 PRI, M is the number of sampling points within 1 PRI, Δt is the sampling time within 1 PRI, c is the speed of light, R ₀ is a target relative distance (see FIG. 2), and v is a target relative speed.

When the signal processor 9 receives the received video signal S (n, m) from the receiver 8, the signal processor 9 performs predetermined signal processing on the received video signal S (n, m) to obtain a target candidate. The relative distance R ′ ₀ is calculated.
Hereinafter, the processing content of the signal processor 9 will be specifically described.

式（２）において、ｌはＰＲＩ内のサンプリング番号、Ｌ’はＦＦＴ点数を表している。ただし、Ｌ’＞ＭのときはＳ（ｎ，ｍ）に０が代入される。 When the pulse compressor 10 of the signal processor 9 receives the received video signal S (n, m) from the receiver 8, it performs predetermined signal processing on the received video signal S (n, m). The signal R _PC (n, l) after linear chirp pulse compression is output to the target candidate detection unit 17.
That is, the frequency domain transform unit 11 of the pulse compression unit 10 performs fast Fourier transform on the received video signal S (n, m) output from the receiver 8 as shown in the following equation (2), and thereby performs frequency domain video signal F _v (n, k _r) of the received, and outputs the received video signal F _v (n, k _r) in the frequency domain to the correlation processing unit 12.

In Expression (2), l represents a sampling number in the PRI, and L ′ represents the number of FFT points. However, when L ′> M, 0 is substituted for S (n, m).

式（３）において、Ａ_Eは参照信号の振幅、ｍ_τはサンプリング番号、Ｍ_τは１ＰＲＩのサンプリング点数（式（４）を参照）、Ｆ_sampはサンプリング周波数、ｆｌｏｏｒ（Ｘ）は変数Ｘを越えない最大の整数を表している。
ここでは、相関処理部１２が参照信号Ｅｘ（ｍ_τ）を近似計算している例を示しているが、パルス内変調信号発生器４により生成されたパルス内変調信号φ（ｔ）を参照信号Ｅｘ（ｍ_τ）として用いるようにしてもよい。 The correlation processing unit 12 generates a reference signal Ex (m _τ ) after A / D conversion that is the same as the intra-pulse modulation component of the transmission RF signal Tx (t), as shown in the following equation (3).

In Equation (3), A _E is the amplitude of the reference signal, m _τ is the sampling number, M _τ is the number of 1 PRI sampling points (see Equation (4)), F _samp is the sampling frequency, and floor (X) is the variable X. Represents the largest integer that does not exceed.
Here, an example is shown in which the correlation processing unit 12 approximates the reference signal Ex (m _τ ), but the intra-pulse modulation signal φ (t) generated by the intra-pulse modulation signal generator 4 is used as the reference signal. It may be used as Ex (m _τ ).

When generating the reference signal Ex (m _τ ), the correlation processing unit 12 performs fast Fourier transform on the reference signal Ex (m _τ ), as shown in the following equation (5), to thereby generate the frequency domain reference signal F _Ex. (K _r ) is generated.

ただし、＊は複素共役を表している。 When the correlation processing unit 12 generates the frequency domain reference signal F _Ex (k _r ), the frequency domain received video signal F _v (n, k _r ) output from the frequency domain conversion unit 11 and the frequency domain reference signal. Correlation processing with F _Ex (k _r ) is performed, and the signal F _{v · Ex} (n, k _r ) after correlation processing is output to the normalization window function selection unit 13.
That is, as shown in the following equation (6), the correlation processing unit 12 calculates the complex conjugate of the frequency domain received video signal F _v (n, k _r ) and the frequency domain reference signal F _Ex (k _r ). To generate a signal F _{v · Ex} (n, k _r ) after the correlation processing.

However, * represents a complex conjugate.

Upon receiving the signal F _{v · Ex} (n, k _r ) after correlation processing from the correlation processing unit 12, the normalization window function selection unit 13 receives the pulse width T ₀ of the transmission signal and the frequency modulation bandwidth B _{0 of the} transmission signal. Is equal to or greater than a preset compression ratio α for normalization window function selection, the signal F _{v · Ex} (n, k _r ) after correlation processing is used as the window function processing section. If the compression rate TB calculated from the product of the pulse width T ₀ of the transmission signal and the frequency modulation bandwidth B ₀ of the transmission signal is lower than the compression rate α for selecting the normalized window function, The signal F _{v · Ex} (n, k _r ) is output to the normalization window function processing unit 14.
Hereinafter, the processing content of the normalization window function selection part 13 is demonstrated concretely.

FIG. 3 is an explanatory diagram showing a signal spectrum of the signal F _{v · Ex} (n, k _r ) after the correlation processing in the case of the high compression rate and the low compression rate.
As shown in FIG. 3A, the signal spectrum of the signal F _{v · Ex} (n, k _r ) after the correlation processing has a substantially constant signal strength as shown in FIG. As shown in FIG. 3B, the signal strength is not constant, and the signal strength changes greatly. The lower the compression rate, the greater the variation in signal strength.
Therefore, in the case of a high compression rate, the shape of the signal spectrum after linear chirp pulse compression matches the shape of the sinc function, but in the case of a low compression rate, the shape of the signal spectrum after linear chirp pulse compression is the shape of the sinc function. In contrast, the distance resolution is degraded (see FIG. 15B) and the sidelobe suppression performance by the window function processing is degraded (see FIG. 16B).

In the first embodiment, in order to prevent the degradation of the distance resolution and the sidelobe suppression performance, the normalized window function processing is performed on the signal F _{v · Ex} (n, k _r ) after the correlation processing.
However, a regular stove function processing for the signal after the correlation process _{F v · Ex (n, k} r) , noise to weight the signal after correlation _{F v · Ex (n, k} r) increases, There is a possibility that the signal-to-noise ratio (SNR) in the signal after pulse compression is reduced.
On the other hand, in the case of a high compression ratio, sufficient pulse compression results are obtained, and it is less necessary to perform normalization window function processing. It is desirable to avoid a reduction in SNR.

Therefore, the normalization window function selection unit 13 uses the compression ratio TB calculated from the product of the pulse width T ₀ of the transmission signal and the frequency modulation bandwidth B ₀ of the transmission signal according to the following equation as the compression ratio for selecting the normalization window function. If it is lower than α, the signal F _{v · Ex} (n, k _r ) after correlation processing is output to the normalization window function processing unit 14 in order to perform normalization window function processing.
On the other hand, when the compression rate TB of the signal F _{v · Ex} (n, k _r ) after the correlation processing is higher than the compression rate α for selecting the normalized window function, in order to avoid performing the normalization window function processing, The signal F _{v · Ex} (n, k _r ) is output to the window function processing unit 15.
TB = T ₀ B ₀

ここで、正規化窓関数ｗ_Norm（ｋ_r）は、正規化窓関数処理部１４が下記の式（８）にしたがって生成する窓関数である。

When the normalized window function processing unit 14 receives the signal F _{v · Ex} (n, k _r ) after correlation processing from the normalized window function selecting unit 13, the signal in the signal R _PC (n, l) after pulse compression A normalized window function w _Norm (which makes the signal intensity (amplitude) of the signal spectrum in the correlated signal F _{v · Ex} (n, k _r ) substantially constant so that the shape of the spectrum matches the shape of the sinc function. k _r ) is used to perform normalized window function processing on the signal F _{v · Ex} (n, k _r ) after correlation processing, and the signal after normalized window function processing as shown in the following equation (7) F ′ _{v · Ex} (n, k _r ) is generated.

Here, the normalized window function w _Norm (k _r ) is a window function generated by the normalized window function processing unit 14 according to the following equation (8).

As described above, the normalized window function processing unit 14 performs the normalized window function processing on the signal F _{v · Ex} (n, k _r ) after the correlation processing, and as shown in FIG. Since the signal intensity of the signal spectrum in the signal F _{v · Ex} (n, k _r ) can be made substantially constant, the signal spectrum in the signal R _PC (n, l) after pulse compression is shown in FIG. The shape can be matched with the shape of the sinc function.

In Expression (8), the normalized window function processing unit 14 performs frequency domain reference signal F _Ex (k _r ) generated by FFT of the reference signal Ex (m _τ ) that is the intra-pulse modulation signal φ (t). using the normalized regular stove function w _Norm (k _r) the inverse of the amplitude of the signal spectrum in the signal after the autocorrelation is shown what is generated (a pulse in the modulated signal phi (t) a As shown in the following equation (9), the reference signal F _Ex (k _r ) that is the intra-pulse modulation signal φ (t) in the frequency domain may be obtained by approximation calculation.

式（９）で周波数領域の参照信号Ｆ_Ex（ｋ_r）を近似計算する場合、大規模なＦＦＴを行うことなく求められるため、装置の処理負荷の低減が可能になる。
また、参照信号Ｅｘ（ｍ_τ）として、実信号であるパルス内変調信号発生器４により生成されたパルス内変調信号φ（ｔ）を用いずに、設定パラメータに基づいて数値計算を行う場合、事前に容易に正規化窓関数を計算することができるため、装置の処理負荷の低減が可能になる。 In the equation (9), Z (u) is Fresnel integral, and u ₁ and u ₂ can be obtained by the following equations (10) and (11).

When the frequency domain reference signal F _Ex (k _r ) is approximated by Equation (9), it is obtained without performing a large-scale FFT, so that the processing load on the apparatus can be reduced.
Further, when the numerical calculation is performed based on the setting parameter without using the intra-pulse modulation signal φ (t) generated by the intra-pulse modulation signal generator 4 which is an actual signal as the reference signal Ex (m _τ ), Since the normalization window function can be easily calculated in advance, the processing load of the apparatus can be reduced.

式（１２）において、ｗ_Ham（ｋ_r）はハミング窓関数である。
式（１２）では、窓関数として、ハミング窓関数を用いている例を示しているが、例えば、サイドローブの抑圧効果があるハミング窓関数以外のハニング窓関数、ブラックマン窓関数など、相関処理後の信号における信号スペクトルの信号強度の一定を想定して設計された窓関数を用いてもよい。
窓関数処理部１５により窓関数処理が実施されることで、図１６（ａ）に示すように、高圧縮率のパルス圧縮後の信号のサイドローブを抑圧することができる（特に、サイドローブの両端付近を抑圧することができる）。 When the window function processing unit 15 receives the signal F _{v · Ex} (n, k _r ) after correlation processing from the normalized window function selection unit 13, the signal after correlation processing is _expressed as shown in the following equation (12). The window function processing for F _{v · Ex} (n, _kr ) is performed to suppress the side rope, and the signal _{Fv · Ex · w} (n, _kr ) after the window function processing is output to the time domain conversion unit 16 To do.

In Expression (12), w _Ham (k _r ) is a Hamming window function.
Expression (12) shows an example in which a Hamming window function is used as the window function. For example, a correlation process such as a Hanning window function other than a Hamming window function having a sidelobe suppression effect, a Blackman window function, or the like. A window function designed on the assumption that the signal intensity of the signal spectrum in the later signal is constant may be used.
By performing the window function processing by the window function processing unit 15, as shown in FIG. 16A, it is possible to suppress the side lobe of the signal after pulse compression with a high compression rate (particularly, the side lobe of the side lobe). It is possible to suppress the vicinity of both ends).

ここで、図６は正規化窓関数処理と窓関数処理が実施されることで、低圧縮率のリニアチャープパルス圧縮後の信号のサイドローブ抑圧性能が向上することを示す説明図である。
図６に示すように、正規化窓関数処理と窓関数処理が実施されることで、低圧縮率の場合でも、サイドローブの抑圧が可能になる。 When the window function processing unit 15 receives the signal F ′ _{v · Ex} (n, k _r ) after the normalization window function processing from the normalization window function processing unit 14, the correlation processing is performed from the normalization window function selection unit 13. The same window function processing as that when the subsequent signal F _{v · Ex} (n, k _r ) is output is performed.
That is, the window function processing unit 15 performs window function processing on the signal F ′ _{v · Ex} (n, k _r ) after the normalized window function processing, as shown in the following equation (13), and calculates the side rope. The signal F _{v · Ex · w} (n, k _r ) after the window function processing is output to the time domain conversion unit 16.

Here, FIG. 6 is an explanatory diagram showing that the sidelobe suppression performance of the signal after linear chirp pulse compression with a low compression ratio is improved by performing the normalization window function processing and the window function processing.
As shown in FIG. 6, by performing the normalized window function process and the window function process, side lobes can be suppressed even in the case of a low compression rate.

When the time domain conversion unit 16 receives the signal F _{v · Ex · w} (n, k _r ) after the window function processing from the window function processing unit 15, the window at the L point as shown in the following equation (14). The signal F _{v · Ex · w} (n, k _r ) after function processing is subjected to inverse fast Fourier transform to generate a signal R _PC (n, l) after pulse compression, which is a time domain signal, and after pulse compression The signal R _PC (n, l) is output to the target candidate detection unit 17.

In addition, when the time domain conversion unit 16 samples the signal R _PC (n, l) after pulse compression with higher accuracy than the sampling interval of the received video signal S (n, m), the time domain conversion unit 16 uses the following equation (15). Set “0”.
Here, L is a point of inverse fast Fourier transform, and is represented by the following formula (16). However, q is an integer greater than or equal to 0. When q = 0, the sampling interval is the same as the sampling interval of the received video signal S (n, m).

式（１７）において、Ｒ_PC,CFAR（ｎ，ｌ）は、ＣＦＡＲ処理による目標候補を表しており、目標候補は０が設定される。
また、ＣＦＡＲ閾値であるＣＦＡＲ＿ｔｈ（ｎ，ｌ）は、下記の式（１８）で算出される。

式（１８）において、ＣＦＡＲ＿ｃｏｒはＣＦＡＲ係数、Ｓａｍｐ＿ｃｅｌｌ（ｎ，ｌ）はサンプルセル、ａｖｅ（Ｚ（ｐ））は配列Ｚ（ｐ）の平均値を表している。
ここで、図７はＣＦＡＲ処理に関わる注目セル、ガードセル、サンプルセルを示す説明図である。 When the target candidate detection unit 17 receives the signal R _PC (n, l) after the pulse compression from the time domain conversion unit 16 of the pulse compression unit 10, as shown in the following equation (17), the signal after the pulse compression A target candidate is detected by performing CFAR processing on R _PC (n, l).

In equation (17), R _{PC, CFAR} (n, l) represents a target candidate by the CFAR process, and 0 is set as the target candidate.
Also, CFAR_th (n, l) which is a CFAR threshold value is calculated by the following equation (18).

In Expression (18), CFAR_cor represents a CFAR coefficient, Samp_cell (n, l) represents a sample cell, and ave (Z (p)) represents an average value of the array Z (p).
Here, FIG. 7 is an explanatory view showing a target cell, a guard cell, and a sample cell related to the CFAR process.

式（１９）において、Δｒ_PCはパルス圧縮後のサンプリング距離間隔である。
目標相対距離算出部１８は、目標候補までの相対距離Ｒ’₀を算出すると、目標候補までの相対距離Ｒ’₀とパルス圧縮後の信号Ｒ_PC（ｎ，ｌ）を表示器１９に出力する。
表示器１９は、目標相対距離算出部１８から目標候補までの相対距離Ｒ’₀とパルス圧縮後の信号Ｒ_PC（ｎ，ｌ）を受けると、目標候補までの相対距離Ｒ’₀とパルス圧縮後の信号Ｒ_PC（ｎ，ｌ）を画面上に表示する。 Upon receiving the target candidate detection result R _{PC, CFAR} (n, l) by the target candidate detection unit 17, the target relative distance calculation unit 18 receives the relative distance R to the target candidate as shown in the following equation (19). 'Calculate ₀ .

In Equation (19), Δr _PC is the sampling distance interval after pulse compression.
Target relative distance calculating unit 18 'when calculating the _0, the relative distance R to the target candidate' relative distance R to the target candidate output signal after ₀ and pulse compression R _PC (n, l) on the display 19 a .
Indicator 19 'receives the ₀ and pulse compression signal after R _PC (n, l), the relative distance R to the target candidate' relative distance from the target relative distance calculation section 18 to the target candidate R ₀ and pulse compression The later signal R _PC (n, l) is displayed on the screen.

As is apparent from the above, according to the first embodiment, the correlation for performing the correlation process between the frequency domain received video signal F _v (n, k _r ) and the frequency domain reference signal F _Ex (k _r ). A processing unit 12 is provided, and the normalized window function processing unit 14 performs signal F _v after correlation processing so that the shape of the signal spectrum in the signal R _PC (n, l) after pulse compression matches the shape of the sinc function. A normalization with respect to the signal F _{v · Ex} (n, k _r ) after the correlation processing by using a normalized window function w _Norm (k _r ) that makes the signal intensity of the signal spectrum at _Ex (n, k _r ) substantially constant. Since the conversion window function processing is performed, the distance resolution can be increased even when the compression rate of the signal R _PC (n, l) after pulse compression is low.

Further, according to the first embodiment, if the compression rate TB output from the correlation processing unit 12 is equal to or greater than a preset compression rate α for selecting a normalized window function, the signal F _v after the correlation processing is used. _Ex (n, k _r ) is output to the window function processing unit 15, and if the compression rate TB is lower than the compression rate α for selecting the normalized window function, the signal F _{v · Ex} (n, k after correlation processing) _Since the normalization window function selection unit 13 that outputs _r ) to the normalization window function processing unit 14 is provided, in the case of a high compression rate, the SNR of the signal R _PC (n, l) after pulse compression is reduced. On the other hand, in the case of a low compression rate, there is an effect that the distance resolution can be increased.

Further, according to the first embodiment, when the compression rate of the signal R _PC (n, l) after the pulse compression is low, the normalization with respect to the signal F _{v · Ex} (n, k _r ) after the correlation processing is performed. Since the window function processing and the window function processing are performed, the distance resolution and the sidelobe suppression performance are improved even when the signal compression rate of the signal R _PC (n, l) after pulse compression is low. There is an effect that can.
Note that the normalization window function w _Norm (k _r ) can be generated by approximation calculation in addition to the case where it is generated by signal processing such as FFT. When generating by approximate calculation, it is not necessary to perform large-scale FFT, and the processing load of the apparatus can be reduced.

Embodiment 2. FIG.
FIG. 8 is a block diagram showing a pulse compression apparatus according to Embodiment 2 of the present invention. In the figure, the same reference numerals as those in FIG.
The second embodiment is different from the first embodiment only in that a normalized window function processing unit 21 is mounted instead of the normalized window function processing unit 14 of FIG.
That is, the pulse compression unit 20 includes a frequency domain conversion unit 11, a correlation processing unit 12, a normalized window function selection unit 13, a normalized window function processing unit 21, a window function processing unit 15, and a time domain conversion unit 16. Then, by performing predetermined signal processing on the received video signal S (n, m) output from the receiver 8, the signal R _PC (n, l) after linear chirp pulse compression is converted into the target candidate detection unit 17 Execute the process to output to. The pulse compression unit 20 constitutes a pulse compression means.

Similar to the normalization window function processing unit 14 in FIG. 1, the normalization window function processing unit 21 is configured so that the shape of the signal spectrum in the signal R _PC (n, l) after pulse compression matches the shape of the sinc function. using regular stove function signal _{F v · Ex (n, k} r) after correlation processing the signal strength of the signal spectrum at a substantially constant, for the signal after the correlation process _{F v · Ex (n, k} r) The normalization window function processing is performed, but the width of the normalization window function is set in accordance with the SNR of the signal R _PC (n, l) after pulse compression when the normalization window function processing is performed. 1 normalization window function processing unit 14.

Next, the operation will be described.
Except for the normalized window function processing unit 21, the processing is the same as in the first embodiment, and only the processing content of the normalized window function processing unit 21 will be described here.
FIG. 9 is an explanatory diagram showing an example in which noise increases with normalization window function processing.

In the normalized window function processing unit 14 of FIG. 1, as shown in FIG. 9, the signal intensity of the signal spectrum in the signal F _{v · Ex} (n, k _r ) after the correlation processing is constant. Weighting is performed by normalization window function processing over the entire frequency modulation bandwidth B ₀ .
For this reason, there is a concern that noise increases at both ends of the frequency modulation band of the transmission signal, and the SNR of the signal R _PC (n, l) after pulse compression decreases.

Therefore, in the second embodiment, the normalized window function processing unit 21 performs the normalized window function processing so that the SNR of the signal R _PC (n, l) after pulse compression is not greatly reduced. For this purpose, the width of the normalized window function is adjusted as shown in FIG.
Hereinafter, the process of adjusting the width of the normalized window function by the normalized window function processing unit 21 will be specifically described.

式（２０）において、Ｎ_wは変調帯域幅の正規化窓関数のサンプリング数、ｎ_wは正規化窓関数の調整幅のサンプリング数を表している。 First, as shown in the following equation (20), the normalized window function processing unit 21 performs signal-to-noise ratio SNR of the signal R _PC (n, l) after pulse compression when the normalized window function processing is performed. _Norm (n _w ) is calculated.

In the formula (20), N _w is the number of samples of the normal stove function of the modulation bandwidth, n _w represents the sampling number of the adjustment range of the normal stove functions.

Here, FIG. 11 is an explanatory diagram showing selection of the width of the normalized window function by the normalized window function processing unit 21 in addition to the relationship between the adjustment width of the normalized window function and the signal-to-noise ratio after pulse compression. .
When the normalized window function processing unit 21 calculates the signal-to-noise ratio SNR _Norm (n _w ) of the signal R _PC (n, l) after pulse compression, the adjustment width of the normalized window function shown in FIG. And the signal-to-noise ratio after pulse compression, the signal-to-noise ratio SNR _Norm (n _w ) of the signal R _PC (n, l) after pulse compression when the normalized window function processing is performed. Select the width n _{w, max} of the normalized window function that maximizes.

式（２１）において、Δｆ_FFTは周波数領域のサンプリング間隔を表している。 When the normalization window function processing unit 21 selects the width n _{w, max} of the normalization window function that maximizes the signal-to-noise ratio SNR _Norm (n _w ), the normalization window function w is expressed according to the following equation (21). 'to generate the _Norm (k _r).

In Expression (21), Δf _FFT represents a sampling interval in the frequency domain.

Regular stove function processing unit 21 'when generating the _Norm (k _r), normal stove function w' normal stove function w with _Norm (k _r), output from the normal stove function selecting section 13 signal F _{v · Ex} post correlation processing (n, k _r) normalized stove function processing performed for the signal F _{'v · Ex} (n after normalization stove function processing shown in equation (22) below , K _r ).

Here, FIG. 12 is an explanatory diagram showing the improvement of the signal-to-noise ratio after the pulse compression by the normalized window function processing unit 21.
The normalized window function processing unit 21 uses the normalized window function w ′ _Norm (k _r ) that maximizes the signal-to-noise ratio SNR _Norm (n _w ), and uses the signal F _{v · Ex} (n, Since the normalized window function processing for k _r ) is performed, as shown in FIG. 12, the normalized window function processing unit 14 of FIG. The SNR is improved and the target detection performance is improved.

As is apparent from the above, according to the second embodiment, the normalized window function processing unit 21 sets the SNR of the signal R _PC (n, l) after pulse compression when the normalized window function processing is performed. Accordingly, since the width of the normalized window function is set, the SNR can be improved and the target detection performance can be enhanced.

Embodiment 3 FIG.
FIG. 13 is a block diagram showing a pulse compression apparatus according to Embodiment 3 of the present invention. In the figure, the same reference numerals as those in FIGS.
In the third embodiment, a normalization window function processing unit 31 is mounted instead of the normalization window function processing unit 21 of FIG. 8, and the normalization window function selection unit 13 is not mounted. This is different from the second embodiment.
That is, the pulse compression unit 30 includes a frequency domain conversion unit 11, a correlation processing unit 12, a normalized window function processing unit 31, a window function processing unit 15, and a time domain conversion unit 16, and is output from the receiver 8. By performing predetermined signal processing on the received video signal S (n, m), processing for outputting the signal R _PC (n, l) after linear chirp pulse compression to the target candidate detection unit 17 is performed. The pulse compression unit 30 constitutes a pulse compression unit.

Regardless of the compression rate TB, the normalization window function processing unit 31 generates the normalization window function w ′ _Norm (k _r ) according to the above equation (21) (the normalization window function w ′ _Norm (k _r )). The generation process is the same as the normalization window function processing unit 21 in FIG. 8), and the normalization for the signal F _{v · Ex} (n, k _r ) after the correlation process is performed using the normalization window function w ′ _Norm (k _r ). Perform the conversion window function processing.

Next, the operation will be described.
In the second embodiment, when the compression rate TB of the signal F _{v · Ex} (n, k _r ) after the correlation processing is lower than the compression rate α for selecting the normalized window function, the normalized window function processing unit of FIG. 21 generates a normalized window function w ′ _Norm (k _r ) that maximizes the signal-to-noise ratio SNR _Norm (n _w ), and normalizes the correlation signal F _{v · Ex} (n, k _r ) If the compression ratio TB is higher than the compression ratio α for selecting the normalization window function, the normalization window function for the signal F _{v · Ex} (n, k _r ) after the correlation processing is performed. Processing is not performed.

In this way, when the compression ratio is high, when the window function processing unit 15 performs the window function processing without performing the normalized window function processing, as shown in FIG. Although suppression is performed by the window function processing, the signal intensity of the signal after the correlation processing is not constant near the center of the side lobe, so that the correlation is disturbed, and the side lobe suppression performance by the window function processing may deteriorate.
On the other hand, when the compression ratio is high, when normalized window function processing and window function processing are performed on the signal F _{v · Ex} (n, k _r ) after the correlation processing, as shown in FIG. Since the signal intensity of the signal after the correlation processing is constant, the correlation is not disturbed, and the sidelobe suppression performance by the window function processing is improved.

Therefore, in the third embodiment, the normalized window function processing unit 31 generates the normalized window function w ′ _Norm (k _r ) according to the above equation (21) regardless of the compression rate TB, and the normalization function The normalized window function processing is performed on the signal F _{v · Ex} (n, k _r ) after the correlation processing using the normalized window function w ′ _Norm (k _r ).
Thereby, even when the compression ratio is high, the sidelobe suppression performance by the window function processing is improved, and the proximity target detection performance is improved.

  In the present invention, within the scope of the invention, any combination of the embodiments, or any modification of any component in each embodiment, or omission of any component in each embodiment is possible. .

  DESCRIPTION OF SYMBOLS 1 Transmission signal generator (transmission signal generation means), 2 Local oscillator, 3 Pulse modulator, 4 In-pulse modulation signal generator, 5 Transmitter, 6 Transmission / reception switch (transmission signal generation means), 7 Antenna (Transmission signal generation) Means), 8 receiver (transmission signal generation means), 9 signal processor, 10, 20, 30 pulse compression section (pulse compression means), 11 frequency domain conversion section, 12 correlation processing section, 13 normalization window function selection section , 14, 21, 31 Normalized window function processing unit, 15 window function processing unit, 16 time domain conversion unit, 17 target candidate detection unit, 18 target relative distance calculation unit, 19 display.

Claims (6)

A transmission signal generating means for generating a transmission signal by pulse-modulating a carrier signal at a predetermined time interval and using a linearly chirped intra-pulse modulation signal to modulate the carrier signal after pulse modulation within the pulse;
A transmission / reception means for radiating the transmission signal generated by the transmission signal generation means, while receiving the signal reflected by the target and returning;
By performing signal processing on the reception signal of the transmission / reception means, pulse compression means for outputting a signal after pulse compression,
The pulse compression means includes
And generates a reference signal is the same signal as the pulse in the modulated component of the transmission signal, a correlation processing section for performing the correlation processing between the received signal and the reference signal of said transmitting and receiving means,
The normalized window for the signal after the correlation processing by the correlation processing unit using the inverse of the amplitude of the signal spectrum in the signal after autocorrelation of the intra-pulse modulated signal used by the transmission signal generating means as a normalized window function and a regular stove function processing unit for implementing a normal stove function processing for multiplying function,
The normalized window function processing unit calculates a signal-to-noise ratio in the signal after pulse compression when the normalized window function processing is performed, and sets the width of the normalized window function according to the signal-to-noise ratio. A pulse compression device characterized by setting . The normalization window function processing unit performs normalization window function processing for multiplying the signal after correlation processing by the correlation processing unit by the normalization window function only when the compression rate is lower than a predetermined compression rate. The pulse compression apparatus according to claim 1, wherein: The pulse compression unit performs window function processing on the signal after correlation processing by the correlation processing unit or the signal after normalization window function processing by the normalization window function processing unit and suppresses side lobes. pulse compression apparatus according to claim 1, further comprising a. The pulse compression means includes
A frequency domain converter for converting the received signal of the receiving means into a frequency domain signal,
A time domain conversion unit that converts a signal after the window function processing by the window function processing unit into a time domain signal with a score that is greater than the sampling number of the received signal ,
The correlation processing unit converts the reference signal into a frequency domain signal, and performs correlation processing between the frequency domain received signal converted by the frequency domain conversion unit and the frequency domain reference signal. The pulse compression apparatus according to claim 3 . The regular stove function processing unit, by carrying out the regular stove function processing on the signal after the correlation processing by the correlation processing unit, to the shape of the signal spectrum in the signal after the pulse compression in the shape of the sinc function The pulse compression device according to any one of claims 1 to 4, wherein The regular stove function processing unit, pulse compression apparatus as claimed in claim 1, wherein the Ru determined by an approximate calculation of the pulse in a modulated signal used by the transmitting signal generation means.

Images