JP2008107242A - Delay time estimator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a calculation quantity by making smaller the number Md of dimensions of a correlation matrix without causing a deterioration in estimation accuracy on delay time. <P>SOLUTION: Received digital signals s(1), s(2), ..., and s(N<SB>d</SB>) are divided into a plurality of partial signals in a previously set delay time estimation zone #l (1≤l≤L). A decimation processing part 7 is provided for performing decimation processing on the plurality of partial signals in the estimation zone #l. Delay time estimation parts 8-1 to 8-L perform hyper-resolution delay time estimation processing on a plurality of decimation processing signals x<SB>1</SB>(1), x<SB>1</SB>(2), ..., and x<SB>1</SB>(M) to estimate delay time on a target signal existing in the estimation zone #l. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a delay time estimation device mounted on a radar device that detects a target, for example.

A conventional delay time estimation apparatus is disclosed in, for example, Non-Patent Document 1 below, and a schematic configuration of the conventional delay time estimation apparatus is as follows.
(1) Transmitter A that generates radio waves
(2) Transmit antenna B that transmits the radio wave generated by transmitter A toward the target
(3) Receiving antenna C that receives radio waves reflected by the target
(4) A receiver D that generates a reception signal by subjecting the reception wave of the reception antenna C to band limitation and phase detection.
(5) A / D converter E that performs analog / digital conversion on the received signal generated by receiver D and outputs a digital signal
(6) An FFT processing unit F that performs FFT (Fast Fourier Transform) processing, which is fast Fourier transform processing, on the digital signal output from the A / D converter E to obtain the frequency spectrum of the received signal
(7) A division signal generation unit G that generates a division signal by dividing the frequency spectrum of the reception signal obtained by the FFT processing unit F by the frequency spectrum of the transmission wave.
(8) A super-resolution estimation processing unit H that performs super-resolution delay time estimation processing on the division signal generated by the division signal generation unit G to estimate the delay time of the target signal.

Next, the operation will be described.
When the transmitter A generates radio waves, the transmission antenna B transmits the radio waves generated by the transmitter A toward the target.
When the reception antenna C receives the radio wave transmitted from the transmission antenna B and reflected by the target, the receiver D performs band limitation or phase detection on the reception wave of the reception antenna C to generate a reception signal.
Here, the time from when the radio wave is transmitted from the transmitting antenna B until the radio wave is received by the receiving antenna C is defined as a delay time τ _d .

When the receiver D generates a reception signal, the A / D converter E performs analog / digital conversion on the reception signal to obtain digital reception signals s (1), s (2),. _d ) is output to the FFT processing unit F.
Here, the received signals s (1), s (2),..., S (N _d ) are signals composed of target signal components and noise components of the receiver D.

When receiving the received signals s (1), s (2),..., S (N _d ) from the A / D converter E, the FFT processing unit F receives the received signals s (1), s (2). , ..., performs FFT processing on s (N _d), the frequency spectrum y (1) of the received signal, y (2), ..., and calculates the y (N _d).
When the FFT processing unit F calculates the frequency spectrum y (1), y (2),..., Y (N _d ) of the received signal, the division signal generation unit G is as shown in the following equation (1). , the frequency spectrum y of the received signal (1), y (2) , ···, y (N d) a frequency spectrum of the transmitted wave Γ (1), Γ (2 ), ···, Γ (N d ) To generate division signals x (1), x (2),..., X (N _d ).
x (i) = y (i) / Γ (i) (1)
However, 1 ≦ i ≦ N _d

When the division signal generation unit G generates the division signals x (1), x (2),..., X (N _d ), the super-resolution estimation processing unit H generates the division signals x (1), x (2 ,..., X (N _d ) is subjected to super-resolution delay time estimation processing to estimate the delay time of the target signal.
Hereinafter, the processing content of the super-resolution estimation processing unit H will be specifically described.

式（２）において、Ｍ_dは相関行列の次元数、ｘ^* _mはベクトルｘ_mの共役転置を表している。 When the division signal generation unit G generates division signals x (1), x (2),..., X (N _d ), the correlation matrix generation unit H1 of the super-resolution estimation processing unit H ) To generate a correlation matrix R representing the correlation between the divided signals x (1), x (2),..., X (N _d ).

In Equation (2), M _d represents the number of dimensions of the correlation matrix, and x ^* _m represents the conjugate transpose of the vector x _m .

When the correlation matrix generation unit H generates the correlation matrix R, the eigenvector calculation unit H2 of the super-resolution estimation processing unit H generates eigenvalues v _e (1), v _e (2) _,. (M _d) eigenvectors and _{(v e (1)> v} e (2)>···> v e (M d)), corresponding to the eigenvalue _{v e (i) (1 ≦} i ≦ M d) e (I) is calculated.
When the eigenvector calculation unit H2 of the super-resolution estimation processing unit H calculates the eigenvector e (i) corresponding to the eigenvalue v _e (i) (1 ≦ i ≦ M _d ) as described above, the target number is set to K. , E (K + 1), e (K + 2),..., E (M _d ) are output to the MUSIC processing unit H3.

そして、ＭＵＳＩＣ処理部Ｈ３は、固有ベクトルｅ（Ｋ＋１），ｅ（Ｋ＋２），・・・，ｅ（Ｍ_d）の全てに直交するｋ種類のステアリングベクトルａ（τ）に対応する遅延時間τ_kを求める。 When the MUSIC processing unit H3 of the super-resolution estimation processing unit H receives the eigenvectors e (K + 1), e (K + 2),..., E (M _d ) from the eigenvector calculation unit H2, the eigenvectors e (K + 1), e MUSIC (Multiple Signal Classification) processing using (K + 2),..., E (M _d ) as a noise space is performed.
That is, the MUSIC processing unit H3 calculates the steering vector a (τ) from the following equation (3), where T is the sampling interval and τ _k is the kth target delay time.

Then, the MUSIC processing unit H3 determines the delay time τ _k corresponding to k types of steering vectors a (τ) orthogonal to all of the eigenvectors e (K + 1), e (K + 2),..., E (M _d ). Ask.

In the conventional delay time estimation apparatus, in order to obtain sufficient delay time estimation accuracy, the correlation value calculation width M _d / (N _d T) needs to be larger than the transmission / reception band 1 / T. There is.
For this reason, the dimension number M _d of the correlation matrix increases, and the amount of calculation of the MUSIC process may increase.
For example, if half the number N _d of the input signal dimensionality M _d of the correlation matrix, M _d = N _d / 2, and the operation amount of the MUSIC processing, on the order of M _d ³ = N _d ^3/8 .

"Adaptive signal processing by array antenna" by Kikuma 1998 Science and Technology Publishing

Since the conventional delay time estimation apparatus is configured as described above, if the correlation value calculation width M _d / (N _d T) is larger than the transmission / reception band 1 / T, a sufficient delay time can be obtained. Estimation accuracy can be obtained. However, when the correlation value calculation width M _d / (N _d T) is increased, the number of dimensions M _d of the correlation matrix increases, and there is a problem that the amount of calculation of the MUSIC processing becomes enormous.

The present invention has been made in order to solve the above-described problems, and it is possible to reduce the number of computations by reducing the dimension number M _d of the correlation matrix without deteriorating the estimation accuracy of the delay time. It is an object of the present invention to obtain a delay time estimation device that can be used.

  A delay time estimation apparatus according to the present invention divides a reception signal generated by a radio wave receiving means into partial signals of a plurality of preset delay time estimation sections, and performs decimation processing on the partial signals of the plurality of delay time estimation sections. The decimation means is provided, and the delay time estimation means performs a super-resolution delay time estimation process on a plurality of decimation processing signals by the decimation means to estimate the delay time of the target signal existing in the delay time estimation section. Is.

According to this invention, the decimation means that divides the reception signal generated by the radio wave reception means into partial signals of a plurality of preset delay time estimation sections and performs decimation processing on the partial signals of the plurality of delay time estimation sections. The delay time estimation means is configured to perform a super-resolution delay time estimation process on a plurality of decimation processing signals by the decimation means to estimate the delay time of the target signal existing in the delay time estimation section. without degrading the time of estimation accuracy, by reducing the number of dimensions M _d of the correlation matrix, there is an effect that it is possible to reduce the calculation amount.

Embodiment 1 FIG.
1 is a block diagram showing a delay time estimation apparatus according to Embodiment 1 of the present invention.
In the figure, the transmitter 1 generates radio waves to be transmitted toward the target 3.
The transmission antenna 2 transmits the radio wave generated by the transmitter 1 toward the target 3. The transmitter 1 and the transmission antenna 2 constitute radio wave transmission means.
The receiving antenna 4 receives the radio wave transmitted from the transmitting antenna 2 and reflected by the target 3.
The receiver 5 generates a reception signal by subjecting the radio wave received by the reception antenna 4 to band limitation and phase detection.
A / D converter 6 received signal generated by the analog / digital conversion by the receiver 5, the received signal s (1) of the digital, s (2), ···, s a (N _d) decimating Output to unit 7.
The receiving antenna 4, the receiver 5, and the A / D converter 6 constitute radio wave receiving means.

The decimation processing unit 7 uses a plurality of delay time estimation intervals # in which the digital reception signals s (1), s (2),..., S (N _d ) output from the A / D converter 6 are set in advance. divided into partial signals of l (1 ≦ l ≦ L), and subjected to decimation processing on the partial signals of a plurality of delay time estimation sections #l to obtain a plurality of decimation processing signals x _l (1), x _l (2), .., X _l (M) (1 ≦ l ≦ L) is output. Note that the decimation processing unit 7 constitutes a decimation means.

The delay time estimation units 8-1 to 8-L are a plurality of decimation processing signals x _l (1), x _l (2),..., X _l (M) (1 ≦ l ≦ L) by the decimation processing unit 7. Are subjected to super-resolution delay time estimation processing to estimate the delay time τ of the target signal existing in the delay time estimation section # 1. The delay time estimation units 8-1 to 8-L constitute delay time estimation means.

FIG. 2 is a block diagram showing the decimation processing unit 7 of the delay time estimation apparatus according to Embodiment 1 of the present invention.
In the figure, the received signal dividing unit 11 is a plurality of delays set in advance for digital received signals s (1), s (2),..., S (N _d ) output from the A / D converter 6. The signal is divided into partial signals of time estimation interval # 1 (1 ≦ l ≦ L), and partial signals of a plurality of delay time estimation intervals # 1 are output to reception signal spectrum generation units 12-1 to 12-L.
That is, the received signal dividing unit 11 divides the required maximum delay time N _d T into L equal parts, and adds the received signal length to the sampled part to generate an N-sampling partial signal. For example, the reception signal spectrum generation unit 12-1 has partial signals s (1 + M (l−1)), s (2 + M (l−1)),..., S (N + M) where M = N _d / L. (L-1)) is output.

The reception signal spectrum generation units 12-1 to 12-L are partial signals s (1 + M (l−1)), s (2 + M (l−1)),. N + M (l−1)) is subjected to FFT processing to generate received signal spectra y _l (1), y _l (2),..., Y _l (N). However, 1 ≦ l ≦ L.
Division signal generating section 13-1 to 13-L is the received signal spectrum generating unit 12-1 to 12-L received signal spectrum generated by _{y l (1), y l} (2), ···, y l ( N) is divided by the transmission wave spectrum Γ (1), Γ (2),..., Γ (N), and divided signals x ′ _l (1), x ′ _l (2),. _l Generate (N). However, 1 ≦ l ≦ L.
The decimation filters 14-1 to 14-L are divided signals x ′ _l (1), x ′ _l (2),..., X ′ _l (N ) Is subjected to decimation processing, and decimation processing signals x _l (1), x _l (2),..., X _l (M) are output. However, 1 ≦ l ≦ L.

FIG. 3 is a block diagram showing the delay time estimation unit 8-l (1 ≦ l ≦ L) of the delay time estimation apparatus according to Embodiment 1 of the present invention.
In the figure, a correlation matrix generation unit 21 represents a correlation between decimation processing signals x _l (1), x _l (2),..., X _l (M) (1 ≦ l ≦ L) by the decimation processing unit 7. generating a correlation matrix R _l.
The eigenvector calculation unit 22 uses the eigenvalues v _{e, l} (1), v _{e, l} (2),..., V _{e, l} (M _d ) (v) of the correlation matrix R ₁ generated by the correlation matrix generation unit 21. _{e, l} (1)> ve _{, l} (2)>...> ve _{, l} (M _d )) and the eigenvalue ve _{, l} (i) (1 ≦ i ≦ M _d ) calculating the eigenvector e _l (i) corresponding to the target number as K _l, eigenvector _{e l (K l +1),} e l (K l +2), ···, e l a (M _d) MUSIC treatment To the unit 23.
The MUSIC processing unit 23 performs MUSIC processing using the eigenvectors e _l (K _l +1), e _l (K _l +2),..., E _l (M _d ) output from the eigenvector calculation unit 22 as noise spaces. The delay time τ of the target signal included in the delay time estimation section # 1 is estimated.

Next, the operation will be described.
When the transmitter 1 generates radio waves, the transmission antenna 2 transmits the radio waves generated by the transmitter 1 toward the target 3.
When the reception antenna 4 receives the radio wave transmitted from the transmission antenna 2 and reflected by the target 3, the receiver 5 performs band limitation or phase detection on the reception wave of the reception antenna 4 to generate a reception signal.
Here, the time from when the radio wave is transmitted from the transmission antenna 2 until the radio wave is received by the reception antenna 4 is defined as a delay time τ _d .

When the receiver 5 generates a reception signal, the A / D converter 6 performs analog / digital conversion on the reception signal to obtain digital reception signals s (1), s (2),. _d ) is output to the decimation processing unit 7.
Here, the received signals s (1), s (2),..., S (N _d ) are signals composed of target signal components and noise components of the receiver 5.

When the decimation processing unit 7 receives the digital reception signals s (1), s (2),..., S (N _d ) from the A / D converter 6, the digital reception signals s (1), s (2),..., S (N _d ) is divided into partial signals of a plurality of preset delay time estimation intervals # 1 (1 ≦ l ≦ L), and portions of the plurality of delay time estimation intervals # 1 The signal is subjected to decimation processing, and a plurality of decimation processing signals x _l (1), x _l (2),..., X _l (M) (1 ≦ l ≦ L) are output.
Specific processing contents of the decimation processing unit 7 are as follows.

When the received signal dividing unit 11 of the decimation processing unit 7 receives the digital received signals s (1), s (2),..., S (N _d ) from the A / D converter 6, it is shown in FIG. As described above, the digital reception signals s (1), s (2),..., S (N _d ) are converted into partial signals of a plurality of preset delay time estimation intervals # 1 (1 ≦ l ≦ L). To divide.
That is, the reception signal dividing unit 11 divides the required maximum delay time N _d T into L equal parts (divided into the number of reception signal spectrum generation units 12-1 to 12-L), and the signal of the reception signal is divided into the sampling amount. By adding the length, a partial signal of N sampling is generated.
Thereby, for example, the received signal spectrum generation unit 12-1 has partial signals s (1 + M (l−1)), s (2 + M (l−1)),..., As M = N _d / L. s (N + M (l−1)) is output.

The reception signal spectrum generation units 12-1 to 12-L of the decimation processing unit 7 receive partial signals s (1 + M (l−1)), s (2 + M (l−1)),. , S (N + M (l-1)), the partial signals s (1 + M (l-1)), s (2 + M (l-1)), ..., s (N + M (l-1)) Are subjected to FFT processing to generate reception signal spectra y _l (1), y _l (2),..., Y _l (N). However, 1 ≦ l ≦ L.

The division signal generation units 13-1 to 13-L of the decimation processing unit 7 are different from the reception signal spectrum generation units 12-1 to 12-L in the reception signal spectrums y ₁ (1), y ₁ (2),. When y _l (N) is generated, the received signal spectrums y _l (1), y _l (2),..., y _l (N) are converted into the transmission wave spectrum Γ as shown in the following equation (4). (1), Γ (2),..., Γ (N) are divided to generate division signals x ′ _l (1), x ′ _l (2) _,. .
x ′ _l (i) = y _l (i) / Γ (i) (4)
However, 1 ≦ i ≦ N and 1 ≦ l ≦ L.

The decimation filters 14-1 to 14 -L of the decimation processing unit 7 are divided by the division signal generation units 13-1 to 13 -L by the division signals x ′ _l (1), x ′ _l (2),. _{When l} (N) is generated, the division signals x ′ _l (1), x ′ _l (2),..., x ′ _l (N) are subjected to decimation processing, and the decimation processing signal x _l (1). , X _l (2),..., X _l (M) are output. However, 1 ≦ l ≦ L.

Here, FIG. 5 is a block diagram showing the inside of the decimation filter 14-1 (1 ≦ l ≦ L).
In the decimation filter 14-1, the weight w (m) (1 ≦ m ≦ M ′) is set by the following equation (5) with M ′ = L / M, and the following equation (6) is calculated. Thus, the decimation process is performed to generate a decimation process signal x _l (i) (1 ≦ i ≦ M, 1 ≦ l ≦ L).

When the decimation processing unit 7 generates the decimation processing signal x _l (i) (1 ≦ i ≦ M, 1 ≦ l ≦ L), the delay time estimation units 8-1 to 8-L receive the decimation processing signal x _l ( A super-resolution delay time estimation process is performed on i), and the target signal delay time τ _d is estimated with the target number existing in the delay time estimation section #l as K _l .
Specific processing contents of the delay time estimation units 8-1 to 8-L are as follows.

式（７）において、Ｍ_dは相関行列の次元数、ｘ^* _mはベクトルｘ_mの共役転置を表している。 Correlation matrix generation unit 21 of the delay time estimation unit 8-l (1 ≦ l ≦ L) is decimation unit 7 is decimated signal _{x l (1), x l} (2), ···, x l (M ) Is calculated, the following equation (7) is calculated to obtain a correlation matrix R _l representing the correlation between the decimation processing signals x _l (1), x _l (2),..., X _l (M). Generate.

In Equation (7), M _d represents the number of dimensions of the correlation matrix, and x ^* _m represents the conjugate transpose of the vector x _m .

When the correlation matrix generation unit 21 generates the correlation matrix R _l , the eigenvector calculation unit 22 of the delay time estimation unit 8-1 generates eigenvalues v _{e, l} (1), v _{e, l} (2) of the correlation matrix R _l. ,..., Ve _{, l} ( _Md ) (ve _{, l} (1)> ve _{, l} (2)>...> ve _{, l} ( _Md )) and its eigenvalues ve _, Calculate the eigenvector e _l (i) corresponding to _l (i) (1 ≦ i ≦ M _d ).
When the eigenvector calculation unit 22 calculates the eigenvector e _l (i) corresponding to the eigenvalue v _{e, l} (i) (1 ≦ i ≦ M _d ) as described above, the eigenvector e is set with the target number as K _{l. l} (K + 1), e _l (K + 2),..., e _l (M _d ) are output to the MUSIC processing unit 23.

そして、ＭＵＳＩＣ処理部２３は、固有ベクトルｅ_l（Ｋ_l＋１），ｅ_l（Ｋ_l＋２），・・・，ｅ_l（Ｍ_d）の全てに直交するK_l種類のステアリングベクトルａ（τ₁）, ａ（τ₂）,…, ａ（τ_Kl）を求める。これより、ｋ（1≦k≦K_l）番目の目標の遅延時間τ_kが求まる。 When the MUSIC processing unit 23 of the delay time estimation unit 8-1 receives the eigenvectors e _l (K _l +1), e _l (K _l +2),..., E _l (M _d ) from the eigenvector calculation unit 22, MUSIC processing is performed with the eigenvectors e _l (K _l +1), e _l (K _l +2),..., E _l (M _d ) as noise spaces.
That is, the MUSIC processing unit 23 calculates the steering vector a (τ) from the following equation (8) using T as a sampling interval.

Then, the MUSIC processing unit 23 outputs K _l types of steering vectors a (τ ₁ ) orthogonal to all of the eigenvectors e _l (K _l +1), e _l (K _l +2),..., E _l (M _d ). ), A (τ ₂ ),..., A (τ _Kl ). Thus, the delay time τ _{k of} the kth (1 ≦ k ≦ K _l ) th target is obtained.

In the first embodiment, since the dimension number M _d of the correlation matrix can be reduced as compared with the above-described conventional example, the amount of processing can be reduced.
For example, if the dimension number M _d of the correlation matrix is half of the point M of the input signal, M _d = M / 2 = N _d / 2L, and the amount of computation of the MUSIC processing is LM _d ³ = N _d ³ / 8L ² It becomes an order.
Calculation of MUSIC processing in the conventional example are the N _d ^3/8, the amount of calculation MUSIC processing is reduced to 1 / L ^2.

As is apparent from the above, according to the first embodiment, the digital received signals s (1), s (2),..., S (N _d ) output from the A / D converter 6 are converted. The signal is divided into partial signals of a plurality of delay time estimation sections # 1 (1 ≦ l ≦ L) set in advance, and the decimation processing is performed on the partial signals of the plurality of delay time estimation sections # 1 to obtain a plurality of decimation processing signals x _l (1), x _l (2),..., x _l (M) (1 ≦ l ≦ L) are provided, and the delay time estimation units 8-1 to 8-L are decimated. The plurality of decimation processing signals x _l (1), x _l (2),..., X _l (M) (1 ≦ l ≦ L) by the processing unit 7 are subjected to super-resolution delay time estimation processing to obtain a delay time. Since the delay time of the target signal existing in the estimation section # 1 is estimated, Without causing the estimation accuracy degradation, to reduce the number of dimensions M _d of the correlation matrix, the effect that it is possible to reduce the calculation amount.

Embodiment 2. FIG.
FIG. 6 is a block diagram showing the decimation processing unit 7 of the delay time estimating apparatus according to the second embodiment of the present invention. FIG. 7 shows the inside of the estimation section cut-out type decimation units 15-1 to 15-L in the decimation processing unit 7. FIG.
The overall configuration of the delay time estimation apparatus according to Embodiment 2 of the present invention is the same as that of the delay time estimation apparatus of FIG.
6 is the same as the decimation processing unit 7 of FIG. 2 except for the estimated section cut-out decimation units 15-1 to 15-L.

In FIG. 7, the IFFT processing unit (inverse Fourier transform processing unit) 31 of the estimated section cut-out decimation unit 15-1 (1 ≦ l ≦ L) is a division signal x ′ _l ( 1), x ′ _l (2),..., X ′ _l (N) are subjected to IFFT (inverse Fourier transform) processing to obtain demodulated signals u _l (1), u _l (2),. _l Generate (N).
The estimation section cut-out processing section 32 which is a signal cut-out processing section is a delay time estimation section from the demodulated signals u _l (1), u _l (2), ..., u _l (N) generated by the IFFT processing section 31. # 1 signals u _l (1), u _l (2),..., U _l (M) are cut out.
The FFT processing unit 33, which is a Fourier transform processing unit, outputs signals u _l (1), u _l (2),..., U _l (M) of the delay time estimation section #l extracted by the estimation section extraction processing unit 32. Are subjected to FFT processing, and decimation processing signals x _l (1), x _l (2),..., X _l (M) are output.

In the first embodiment, the decimation processing unit 7 has been mounted with the decimation filters 14-1 to 14-L. However, as shown in FIG. 6, instead of the decimation filters 14-1 to 14-L. Even if the estimated section cut-out type decimation units 15-1 to 15-L are mounted, the number of dimensions M _d of the correlation matrix can be reduced as in the first embodiment, so The amount can be reduced.

Hereinafter, the processing content of the estimation section cut-out decimation units 15-1 to 15-L will be specifically described.
IFFT processing unit 31 of the estimation interval cutout type decimation unit 15-l (1 ≦ l ≦ L) , similarly as in the first embodiment, the division signal generating section 13-l division signal x _'l (1), _{x 'l (2), ···} , x' when generating the _l (N), the division signal _{x 'l (1), x} ' l (2), ···, the x _'l (N) IFFT Processing is performed to generate demodulated signals u _l (1), u _l (2),..., U _l (N).

When the IFFT processing unit 31 generates demodulated signals u _l (1), u _l (2),..., U _l (N), the estimated section cutout processing unit 32 generates the demodulated signals u _l (1), u. _{l (2), ···, u} l (N) of the delay time estimation section #l from signal _{u l (1), u l} (2), ···, cuts out u _l (M) (Fig. 4).
When the estimated interval cutout processing unit 32 cuts out the signals u _l (1), u _l (2),..., U _l (M) of the delay time estimation interval #l, the FFT processing unit 33 delays the delay time. The signals u _l (1), u _l (2),..., U _l (M) in the estimation interval #l are subjected to FFT processing, and decimation processed signals x _l (1), x _l (2) _,. .., x _l (M) is output.
Hereinafter, since it is the same as that of the said Embodiment 1, description is abbreviate | omitted.

Embodiment 3 FIG.
FIG. 8 is a block diagram showing a decimation processing unit 7 of the delay time estimation apparatus according to the third embodiment of the present invention. The entire configuration of the delay time estimation apparatus according to the third embodiment of the present invention is the delay time estimation of FIG. Identical to the device.
In FIG. 8, the same reference numerals as those in FIG.
Similarly to the division signal generation units 13-1 to 13-L in FIG. 2, the zero division prevention type division signal generation units 16-1 to 16-L are generated by the reception signal spectrum generation units 12-1 to 12-L. The received signal spectrum y _l (1), y _l (2),..., Y _l (N) is divided by the transmission wave spectrum Γ (1), Γ (2),. To generate divided signals x ′ _l (1), x ′ _l (2),..., X ′ _l (N), but transmit wave spectra Γ (1), Γ (2),. When the amplitude value of (N) is smaller than the preset minimum value Γ _min , the transmission wave spectrum Γ (1), Γ (2),..., Γ (N) is corrected to the minimum value Γ _min . The corrected transmission wave spectrums Γ ′ (1), Γ ′ (2),..., Γ ′ (N) are received signal spectra y _l (1), y _l (2) _,. Divide N). However, 1 ≦ l ≦ L.

In the first and second embodiments, the reception signal spectrums y ₁ (1), y _l (2) generated by the division signal generation units 13-1 to 13-L by the reception signal spectrum generation units 12-1 to 12-L. _{), ···, y l (N} ) of the transmitted wave spectrum Γ (1), Γ (2 ), ···, by dividing the gamma (N) dividing the signal _{x 'l (1), x} ' l ( 2),..., X ′ _l (N) are generated, but the amplitude values of the transmission wave spectra Γ (1), Γ (2),. If it is small, the division signals x ′ _l (1), x ′ _l (2),..., X ′ _l (N) may suddenly have large errors.

Therefore, in the third embodiment, it is possible to prevent the occurrence of a sudden large error due to the small amplitude value of the transmission wave spectrum Γ (1), Γ (2),..., Γ (N). I have to.
That is, the zero-dividing prevention division signal generators 16-1 to 16-L set in advance the minimum value Γ _min of the transmission wave spectrum, and receive signal spectrums y ₁ (1), y ₁ (2),. _.. , Y _l (N) divided by the transmission wave spectrum Γ (1), Γ (2),..., Γ (N), the received signal spectrum y _l (1), y _l (2) ,..., Y _l (N) are compared with the minimum value Γ _min .

Then, zero percent prevent type division signal generating section 16-1 to 16-L is the received signal spectrum _{y l (1), y l} (2), ···, y l minimum amplitude value of (N) gamma If it is smaller than _min , the transmission wave spectrum Γ (1), Γ (2),..., Γ (N) is corrected to the minimum value Γ _min as shown in the following equation (9).

The zero-dividing prevention division signal generators 16-1 to 16-L have the amplitude values of the received signal spectra y ₁ (1), y ₁ (2),..., Y ₁ (N) from the minimum value Γ _min . If it is not small, the received signal spectrums y ₁ (1), y ₁ (2),..., Y ₁ (N) are converted into transmission wave spectra Γ (1), Γ (2),. , X ′ _l (1), x ′ _l (2),..., X ′ _l (N) are generated, but the received signal spectrums y _l (1), y _l (2), If the amplitude value of..., Y _l (N) is smaller than the minimum value Γ _min , reception is performed with the corrected transmission wave spectrums Γ ′ (1), Γ ′ (2),. The signal spectra y _l (1), y _l (2),..., Y _l (N) are divided to divide signals x ′ _l (1), x ′ _l (2) _,. (N) is generated.

Thus, according to the third embodiment, even when the amplitude values of the transmission wave spectrums Γ (1), Γ (2),..., Γ (N) are smaller than the values assumed in advance, the division signal There is an effect that it is possible to prevent x ′ _l (1), x ′ _l (2),..., x ′ _l (N) from suddenly having a large error.

Embodiment 4 FIG.
FIG. 9 is a block diagram showing a delay time estimation unit 8-l (1 ≦ l ≦ L) of the delay time estimation device according to the fourth embodiment of the present invention. The overall configuration is the same as that of the delay time estimation apparatus of FIG.
In FIG. 9, the same reference numerals as those in FIG.
The target signal determination type eigenvector calculation unit 24 which is a target determination unit includes eigenvalues v _{e, l} (1), v _{e, l} (2),..., V of the correlation matrix R _l generated by the correlation matrix generation unit 21. _{e, l} (M _d ) (ve _{, l} (1)> ve _{, l} (2)>...> ve _{, l} (M _d )) is calculated, and its eigenvalue ve _{, l} (i ) (calculates 1 ≦ i ≦ M _d) eigenvectors corresponding to e _l (i), if there is a predetermined threshold value Th larger eigenvalues, eigenvectors and certified as a target signal is present e _l: (i) While outputting to the MUSIC processing unit 23, if there is no eigenvalue greater than the predetermined threshold Th, it is determined that the target signal does not exist.

Except for the delay time estimation unit 8-1 (1 ≦ l ≦ L) of the delay time estimation apparatus, the delay time estimation unit 8-1 (1 ≦ l ≦ L) is the same as in the first to third embodiments. ) Will be described in detail.
The correlation matrix generation unit 21 of the delay time estimation unit 8-1 performs the above-described implementation when the decimation processing unit 7 generates the decimation processing signals x _l (1), x _l (2),..., X _l (M). As in the first embodiment, a correlation matrix R _l representing the correlation between the decimation processing signals x _l (1), x _l (2),..., X _l (M) is generated.

Target signal determined type eigenvector computing section 24 of the delay time estimation unit 8-l, when the correlation matrix generation unit 21 generates a correlation matrix R _l, eigenvalues v _e of the correlation matrix _{R l, l (1),} v e, _l (2), ..., ve _{, l} ( _Md ) (ve _{, l} (1)> ve _{, l} (2)>...> ve _{, l} ( _Md )) and its The eigenvector e _l (i) corresponding to the eigenvalue v _{e, l} (i) (1 ≦ i ≦ M _d ) is calculated.
When the target signal determination type eigenvector calculation unit 24 calculates the eigenvector e _l (i) corresponding to the eigenvalue v _{e, l} (i) (1 ≦ i ≦ M _d ) as described above, a preset threshold value is obtained. Th and eigenvalues ve _{, l} (1), ve _{, l} (2),..., Ve _{, l} (M _d ) are compared, and an eigenvalue greater than or equal to the threshold Th is obtained.

The target signal determination type eigenvector calculation unit 24, when all eigenvalues v _{e, l} (1), v _{e, l} (2),..., V _{e, l} (M _d ) are smaller than the threshold Th, It is recognized that does not exist, and a series of processing is terminated. In this case, the eigenvector e _l (i) is not output to the MUSIC processing unit 23 and the delay time of the target signal is not estimated.
The target signal determination type eigenvector calculation unit 24 has eigenvalues larger than the threshold Th among the eigenvalues v _{e, l} (1), v _{e, l} (2),..., V _{e, l} (M _d ). If the target signal is present, it is recognized that the target signal exists.
For example, if ve _{, l} (1)> ve _{, l} (2)>...> Ve _{, l} (K)> Th, the target number is recognized as K, and the eigenvector e _l ( K + 1), e ₁ (K + 2),..., E ₁ (M _d ) are output to the MUSIC processing unit 23.

When the MUSIC processing unit 23 of the delay time estimation unit 8-1 receives the eigenvectors e ₁ (K + 1), e ₁ (K + 2),..., E ₁ (M _d ) from the target signal determination type eigenvector calculation unit 24, As in the first embodiment, the MUSIC process using the eigenvectors e _l (K + 1), e _l (K + 2),..., E (M _d ) as a noise space is performed, and the delay time of the target signal is determined. presume.

As is apparent from the above, according to the fourth embodiment, the eigenvalues v _{e, l} (1), v _{e, l} (2),... Of the correlation matrix R _l generated by the correlation matrix generation unit 21. , V _{e, l} (M _d ), eigen vector e _l (i) corresponding to the eigen value v _{e, l} (i) (1 ≦ i ≦ M _d ) is calculated, and is larger than a predetermined threshold Th. If there is an eigenvalue, it is determined that the target signal exists and the eigenvector e _l (i) is output to the MUSIC processing unit 23. On the other hand, if there is no eigenvalue greater than the predetermined threshold Th, the target signal does not exist. Since it is configured to be certified, it is possible to perform super-resolution delay time estimation processing even when the number of target signals is unknown, and to eliminate unnecessary estimation processing when there is no target signal. Play.

Embodiment 5. FIG.
10 is a block diagram showing a delay time estimation apparatus according to Embodiment 5 of the present invention. In the figure, the same reference numerals as those in FIG.
The decimation processing section 9 receives a plurality of digital delay signals s (1), s (2),..., S (N _d ) output from the A / D converter 6 and sets a plurality of preset delay time estimation intervals # is divided into partial signals of l (1 ≦ l ≦ L), it is determined whether or not the target signal is included in the partial signals of the plurality of delay time estimation sections # 1, and the partial signal including the target signal is determined. Decimation processing is performed, and decimation processing signals x _h (1), x _h (2),..., X _h (M) are output. Note that the decimation processing unit 9 constitutes a decimation means.
The delay time estimation unit 10 performs super-resolution delay time estimation processing on the decimation processing signals x _h (1), x _h (2),..., X _h (M) by the decimation processing unit 9 to delay the target signal. Estimate time τ. The delay time estimation unit 10 constitutes a delay time estimation unit.

FIG. 11 is a block diagram showing a decimation processing unit 9 of the delay time estimating apparatus according to Embodiment 5 of the present invention. In the figure, the same reference numerals as those in FIG.
The target signal detector 17 receives the received signal spectrums y _l (1), y _l (2),..., Y _l (N) (1 ≦ l) generated by the received signal spectrum generators 12-1 to 12 -L. ≦ L) average power P _l is calculated, and if the average power P _l is greater than the reference power Th _r , the received signal spectrum y _l (1), y _l (2),..., Y _l (N ) was determined to contain a target signal, the received signal spectrum contains target signal _{y h (1), y h} (2), ···, the division signal generating section y _h (N) 18 is output.

The division signal generation unit 18 converts the reception signal spectrums y _h (1), y _h (2),..., Y _h (N) output from the target signal detection unit 17 into transmission wave spectra Γ (1), Γ ( 2), ···, Γ (N ) divided by to divide the signal _{x 'h (1), x} ' h (2), to produce a _{···, x 'h (N)} .
The decimation filter 19 performs a decimation process on the division signals x ′ _h (1), x ′ _h (2),..., X ′ _h (N) generated by the division signal generation unit 18 to obtain a decimation processing signal x. _h (1), x _h (2),..., x _h (M) are output.

Next, the operation will be described.
When the A / D converter 6 outputs the digital received signals s (1), s (2),... S (N _d ), as in the first embodiment, the decimation processing unit 9 The received signals s (1), s (2),..., S (N _d ) are divided into partial signals of a plurality of preset delay time estimation intervals #l (1 ≦ l ≦ L). It is determined whether or not the target signal is included in the partial signal of the delay time estimation section # 1 of the delay time, the decimation process is performed on the partial signal including the target signal, and the decimation process signal x _h (1), x _h (2),..., x _h (M) are output to the delay time estimation unit 10.

That is, when the received signal dividing unit 11 of the decimation processing unit 9 receives the digital received signals s (1), s (2),..., S (N _d ) from the A / D converter 6, the above-described implementation is performed. , S (N _d ) of the digital received signals s (1), s (2),..., S (N _d ) in the same manner as in the first embodiment, the partial signal s (1 + M (l− 1)), s (2 + M (l−1)),..., S (N + M (l−1)), and the partial signals s (1 + M (l−1)) and s (2 + M (l− 1)),..., S (N + M (l−1)) are output to the reception signal spectrum generation units 12-1 to 12-L. However, 1 ≦ l ≦ L.

The reception signal spectrum generation units 12-1 to 12 -L of the decimation processing unit 9 receive partial signals s (1 + M (l−1)), s (2 + M (l−1)),. , S (N + M (l−1)), the partial signals s (1 + M (l−1)), s (2 + M (l−1)),. s (N + M (l−1)) is subjected to FFT processing to generate received signal spectra y ₁ (1), y ₁ (2),..., y ₁ (N). However, 1 ≦ l ≦ L.

ただし、１≦ｌ≦Ｌである。 Target signal detection unit 17 of the decimation processor 9 receives the signal spectrum generation unit 12-1 to 12-L is the received signal spectrum _{y l (1), y l} (2), ···, y l a (N) When generated, the average power P _l of the received signal spectrums y _l (1), y _l (2),..., Y _l (N) is calculated as shown in the following equation (10).

However, 1 ≦ l ≦ L.

The target signal detector 17 compares the average power P _l of the received signal spectrums y _l (1), y _l (2),..., Y _l (N) with a preset reference power Th _r .
If the average power P _l of the received signal spectrums y _l (1), y _l (2),..., Y _l (N) is smaller than the reference power Th _{r, the} target signal detector 17 receives the received signal spectrum y. _It is determined that the target signal is not included in _l (1), y _l (2),..., y _l (N), and the received signal spectrums y _l (1), y _l (2) _,. .., Y _l (N) is not output to the division signal generator 18.
On the other hand, if the average power P _{l of} the received signal spectrum y _l (1), y _l (2),..., Y _l (N) is larger than the reference power Th _r , the received signal spectrum y _l (1), It is determined that y ₁ (2),..., y ₁ (N) includes a target signal, and the received signal spectrums y ₁ (1), y ₁ (2) _,. (N) is output to the division signal generator 18 as a received signal spectrum y _h (1), y _h (2),..., Y _h (N) including the target signal.

Upon receiving the received signal spectrum y _h (1), y _h (2),..., Y _h (N) including the target signal from the target signal detecting unit 17, the division signal generating unit 18 receives the received signal spectrum. The signal spectrum y _h (1), y _h (2),..., Y _h (N) is divided by the transmission wave spectrum Γ (1), Γ (2),. Generate signals x ′ _h (1), x ′ _h (2),..., X ′ _h (N).
When the division signal generator 18 generates the division signals x ′ _h (1), x ′ _h (2),..., X ′ _h (N), the decimation filter 19 generates the division signal x ′ _h (1). _{, x 'h (2),} ···, x' is subjected to decimation processing _h (N), decimated signal _{x h (1), x h} (2), ···, x h a (M) Output.

When the delay time estimation unit 10 receives the decimation processing signals x _h (1), x _h (2),..., X _h (M) of the delay time estimation interval #h from the decimation processing unit 9, the delay time estimation unit 10 performs the above-described implementation. as with the delay time of the first estimation unit 8-1 to 8-L, the decimated signal _{x h (1), x h} (2), ···, super-resolution delay time estimating process in x _h (M) And delay time τ of the target signal existing in delay time estimation section #h is estimated.

As is apparent from the above, according to the fifth embodiment, the digital received signals s (1), s (2),..., S (N _d ) output from the A / D converter 6 are converted. Dividing into partial signals of a plurality of preset delay time estimation sections # 1 (1 ≦ l ≦ L) and determining whether or not the target signal is included in the partial signals of the plurality of delay time estimation sections # 1 , Decimation processing is performed on the partial signal including the target signal, and decimation processing signals x _h (1), x _h (2),..., X _h (M) are output to the delay time estimation unit 10. In addition to the effects similar to those of the first embodiment, the super-resolution delay time estimation process can be performed even when the delay time estimation section where the target signal exists is unknown. Play.

Embodiment 6 FIG.
12 is a block diagram showing a delay time estimation apparatus according to Embodiment 6 of the present invention. In the figure, the same reference numerals as those in FIG.
The receiving antennas 4-1 and 4-2 receive the radio waves transmitted from the transmitting antenna 2 and reflected by the target 3.
The receivers 5-1 and 5-2 perform band limitation and phase detection on the radio waves received by the reception antennas 4-1 and 4-2 and generate reception signals.

The A / D converters 6-1 and 6-2 perform analog / digital conversion on the reception signals generated by the receivers 5-1 and 5-2, thereby obtaining digital reception signals s (1), s (2), , S (N _d ), s ′ (1), s ′ (2),..., S ′ (N _d ) are output to the maximum ratio composition processing unit 41.
The receiving antennas 4-1 and 4-2, the receivers 5-1 and 5-2, and the A / D converters 6-1 and 6-2 constitute radio wave receiving means.
The maximum ratio composition processing unit 41 receives digital received signals s (1), s (2),..., S (N _d ), s ′ (output from the A / D converters 6-1 and 6-2. 1), s ′ (2),..., S ′ (N _d ) are combined at the maximum ratio, and the received signals s ″ (1), s ″ (2),. , S ″ (N _d ) are output to the decimation processing unit 7. The maximum ratio composition processing unit 41 constitutes a maximum ratio composition processing means.

Next, the operation will be described.
When the transmitter 1 generates radio waves, the transmission antenna 2 transmits the radio waves generated by the transmitter 1 toward the target 3.
When the reception antennas 4-1 and 4-2 receive the radio waves transmitted from the transmission antenna 2 and reflected by the target 3, the receivers 5-1 and 5-2 receive signals from the reception antennas 4-1 and 4-2. A reception signal is generated by subjecting the reception wave to band limitation and phase detection.
When the receivers 5-1 and 5-2 generate reception signals, the A / D converters 6-1 and 6-2 perform analog / digital conversion on the reception signals to generate digital reception signals s (1), s (2), ..., s (N _d ), s' (1), s' (2), ..., s' (N _d ) are output to the maximum ratio composition processing unit 41.

The maximum ratio combining processing unit 41 receives digital received signals s (1), s (2),..., S (N _d ), s ′ (1) from the A / D converters 6-1 and 6-2. , s' (2), · · ·, s' if (N _d) the receiving, the received signal s (1), s (2 ), ···, and s (N _d), the received signal s' ( 1), s ′ (2),..., S ′ (N _d ) are combined at the maximum ratio, and the received signals s ″ (1), s ″ (2) after the maximum ratio combination are combined. ., S ″ (N _d ) is output to the decimation processing unit 7.
Specific processing contents of the maximum ratio combining processing unit 41 are as follows.

Maximum ratio combining processing unit 41, A / D converter 6-1, 6-2 from the digital received signal s (1), s (2 ), ···, s (N d), s' (1) , S ′ (2),..., S ′ (N _d ), the received signals s (1), s (2),. N _d ) and a correlation matrix R _r between the received signals s ′ (1), s ′ (2),..., S ′ (N _d ). However, in Formula (11), matrix [ _sr (1) ... _sr ( _Nd )] ^H represents Hermitian conjugate of matrix [ _sr (1) ... _sr ( _Nd )]. ing.

Maximum ratio combining processing unit 41, when calculating the correlation matrix R _r of the two received signals, calculates the eigenvectors e _r corresponding to the maximum eigenvalue of the correlation matrix R _r, as shown in the following equation (12), using the eigenvectors e _r, the maximum ratio combining processing signal s '' (1), s '' (2), ···, s '' (N d) the calculated, maximum ratio combining processing signal s '' (1 , S ″ (2),..., S ″ (N _d ) are output to the decimation processing unit 7.
s ″ (i) = _er ^H s _r (i) (12)
Since the processing contents of the decimation processing unit 7 and the delay time estimation units 8-1 to 8-L are the same as those in the first embodiment, description thereof is omitted.

As is apparent from the above, according to the sixth embodiment, the maximum ratio composition processing unit 41 receives the digital received signals s (1), s (output from the A / D converters 6-1 and 6-2. 2),..., S (N _d ), s ′ (1), s ′ (2),..., S ′ (N _d ) are combined at the maximum ratio, and the received signal s after the maximum ratio combining. Since '' (1), s '' (2),..., S '' (N _d ) are output to the decimation processing unit 7, the same effects as those of the first embodiment can be obtained. In addition, even if multipath fading or the like occurs, the power of the received signal can be improved, and the delay time estimation accuracy can be prevented from deteriorating.

  In the sixth embodiment, two sets of receiving antennas 4-1 and 4-2, receivers 5-1 and 5-2, and A / D converters 6-1 and 6-2 are prepared. Although what was shown about performing the maximum ratio composition of a received signal was shown, the received signal with the larger electric power is selected from the received signals output from two A / D converters 6-1 and 6-2, A selectivity diversity system that outputs a received signal to the decimation processing unit 7 may be adopted.

Embodiment 7 FIG.
FIG. 13 is a block diagram showing a delay time estimation apparatus according to Embodiment 7 of the present invention. In the figure, the same reference numerals as those in FIG.
When the oversampling A / D converter 42 performs analog / digital conversion on the reception signal generated by the receiver 5, if the radio wave transmitted from the transmission antenna 2 is code-modulated in units of pulses, The radio wave is sampled at a time interval shorter than the chip width to generate digital received signals s (1), s (2),..., S (N ′ _d ). The oversampling A / D converter 42 constitutes radio wave receiving means.

Next, the operation will be described.
The transmitter 1 code-modulates the radio wave in units of pulses, and outputs the radio wave after the code modulation to the transmission antenna 2.
When receiving the code-modulated radio wave from the transmission antenna 2, the transmission antenna 2 transmits the radio wave toward the target 3.
When the reception antenna 4 receives the radio wave transmitted from the transmission antenna 2 and reflected by the target 3, the receiver 5 performs band limitation or phase detection on the reception wave of the reception antenna 4 to generate a reception signal.

When the receiver 5 generates a reception signal, the oversampling A / D converter 42 performs analog / digital conversion on the analog reception signal in the same manner as the A / D converter 6 in FIG. .., S (N ′ _d ) are generated by sampling the analog reception signal at a time interval shorter than the chip width.
Here, FIG. 14 is an explanatory diagram showing a difference between normal sampling in the A / D converter 6 and oversampling in the oversampling A / D converter 42 in a multipath fading environment.

In a multipath fading environment, if normal sampling is performed as in the A / D converter 6, the fading location is sampled as shown in FIG. 14, and sufficient received signal power is obtained. It may not be possible.
On the other hand, when oversampling is performed as in the oversampling A / D converter 42, not only fading but also non-fading can be sampled as shown in FIG. The power of the received signal can be obtained.
Since the processing contents of the decimation processing unit 7 and the delay time estimation units 8-1 to 8-L are the same as those in the first embodiment, description thereof is omitted.

As is apparent from the above, according to the seventh embodiment, when the radio wave transmitted from the transmission antenna 2 is code-modulated in units of pulses, the radio wave is sampled at a time interval shorter than the chip width of the pulse. Since the digital reception signals s (1), s (2),..., S (N ′ _d ) are generated, the same effects as in the first embodiment can be obtained. Even if path fading or the like occurs, the power of the received signal can be improved, and the delay time estimation accuracy can be prevented from deteriorating.

Embodiment 8 FIG.
FIG. 15 is a block diagram showing a delay time estimation apparatus according to Embodiment 8 of the present invention. In the figure, the same reference numerals as those in FIG.
The memory circuit 43 receives the digital received signals s _np (1), s _np (2),..., S _np (N _d ) (1 ≦ n _p ≦ N _p ) from the A / D converter 6. The received signals are stored side by side in the pulse direction, and the received signals are stored side by side in the time direction. Note that the memory circuit 43 constitutes reception signal storage means.

The pulse direction FFT units 44-1 to 44-N _d are received signals s ₁ (n _d ), s ₂ (n _d ),..., S _Np (n _d ) stored in the memory circuit 43 in the pulse direction. ) Is subjected to FFT processing, and pulse direction FFT processing signals z ₁ (n _d ), z ₂ (n _d ),..., Z _Np (n _d ) are output. However, 1 ≦ n _d ≦ N _d , 1 ≦ n _p ≦ N _p .
The pulse direction FFT unit 44-1 to 44-N _d constitute a Fourier transform processing means.
The delay time estimation interval detector 45 outputs the pulse direction FFT processing signals z ₁ (n _d ), z ₂ (n _d ),..., Z _Np (from the pulse direction FFT units 44-1 to 44-N _d ( n _d ) is analyzed to detect a delay time estimation interval #h in which the target signal exists. Note that the delay time estimation interval detection unit 45 constitutes a delay time estimation interval detection means.

Decimation processing section 46-1 to 46-N _p received signal are stored side by side in the time direction in the memory circuit _{43 s np (1), s} np (2), ···, delayed from s _np (N _d) Extracting the partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} (N) of the delay time estimation section #h detected by the time estimation section detection unit 45; The partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} (N) of the delay time estimation interval #h are subjected to decimation processing to obtain a decimation processing signal x _{h, np} (1), x _{h, np} (2),..., x _{h, np} (M) are output. However, 1 ≦ n _p ≦ N _p . Incidentally, decimation processing section 46-1 to 46-N _p constitute a decimation means.
The delay time estimation unit 47 exceeds the decimation processing signals x _{h, np} (1), x _{h, np} (2),..., X _{h, np} (M) by the decimation processing units 46-1 to 46-N _p. Resolution delay time estimation processing is performed to estimate the delay time of the target signal existing in the delay time estimation section #h. The delay time estimation unit 47 constitutes a delay time estimation unit.

FIG. 16 is a block diagram showing a decimation processing unit 46-n _p (1 ≦ n _p ≦ N _p ) of the delay time estimating apparatus according to the eighth embodiment of the present invention.
In the figure, a signal extraction unit 51 detects a delay time estimation interval from received signals s _np (1), s _np (2),..., S _np (N _d ) stored side by side in the memory circuit 43. The partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} (N) of the delay time estimation interval #h detected by the unit 45 are extracted. However, 1 ≦ n _p ≦ N _p .
The received signal spectrum generation unit 52 outputs the partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} (N ) Is subjected to FFT processing to generate received signal spectrums y _{np, h} (1), y _{np, h} (2),..., Y _{np, h} (N). However, 1 ≦ n _p ≦ N _p .

The division signal generation unit 53 converts the reception signal spectrum y _{np, h} (1), y _{np, h} (2),..., Y _{np, h} (N) generated by the reception signal spectrum generation unit 52 into the transmission wave spectrum. Divided by Γ (1), Γ (2),..., Γ (N) and divided signals x ′ _{np, h} (1), x ′ _{np, h} (2) _{,. h} (N) is generated. However, 1 ≦ n _p ≦ N _p .
The decimation filter 54 performs decimation processing on the division signals x ′ _{np, h} (1), x ′ _{np, h} (2),..., X ′ _{np, h} (N) generated by the division signal generation unit 53. , X _{np, h} (1), x _{np, h} (2),..., X _{np, h} (M) are output. However, 1 ≦ n _p ≦ N _p .

FIG. 17 is a block diagram showing a delay time estimation unit 47 of the delay time estimation apparatus according to the eighth embodiment of the present invention. In the figure, the same reference numerals as those in FIG.
The pulse direction average correlation matrix generation unit 25 is decimated by the decimation processing unit 46-n _p (1 ≦ n _p ≦ N _p ) x _{np, h} (1), x _{np, h} (2) _,. A correlation matrix R representing the correlation between x _{np, h} (M) (1 ≦ n _p ≦ N _p ) is generated.

Next, the operation will be described.
The transmitter 1 repeatedly generates radio waves N _p times, and the transmitting antenna 2 repeatedly transmits the radio waves generated by the transmitter 1 toward the target 3 N _p times.
When the reception antenna 4 receives the radio wave transmitted from the transmission antenna 2 and reflected by the target 3, the receiver 5 performs band limitation or phase detection on the reception wave of the reception antenna 4 to generate a reception signal.
When the receiver 5 generates a reception signal, the A / D converter 6 performs analog / digital conversion on the reception signal to obtain digital reception signals s _np (1), s _np (2) _{,. np} (N _d ) (1 ≦ n _p ≦ N _p ) is output to the memory circuit 43.

The memory circuit 43 receives digital received signals s _np (1), s _np (2),..., S _np (N _d ) (1 ≦ n _p ≦ N _p ) from the A / D converter 6. In addition, the received signals are stored side by side in the pulse direction, and the received signals are stored side by side in the time direction.
Note that the memory circuit 43, when the transmission and reception of N _p times is completed, the received signal is stored side by side in the pulse direction _{s 1 (n d), s} 2 (n d), ···, s Np (n _d) outputting a pulse direction FFT unit 44-1 to 44-N _d. However, 1 ≦ n _d ≦ N _d , 1 ≦ n _p ≦ N _p .

The pulse direction FFT units 44-1 to 44-N _d receive signals s ₁ (n _d ), s ₂ (n _d ),..., S _Np (n _d ) arranged in the pulse direction from the memory circuit 43. ), The received signals s ₁ (n _d ), s ₂ (n _d ),..., S _Np (n _d ) are subjected to FFT processing to obtain a pulse direction FFT processed signal z ₁ (n _d ). , Z ₂ (n _d ),..., Z _Np (n _d ) are output.
Delay time estimation section detecting unit 45, the pulse direction FFT unit 44-1 to 44-N _d Pulse direction FFT processed signal from _{z 1 (n d), z} 2 (n d), ···, z Np (n d ), The pulse direction FFT processing signals z ₁ (n _d ), z ₂ (n _d ),..., Z _Np (n _d ) are analyzed to estimate the delay time in which the target signal exists. Section #h is detected.
The specific processing contents of the delay time estimation interval detection unit 45 are as follows.

Delay time estimation section detecting unit 45, the pulse direction FFT unit 44-1 to 44-N _d Pulse direction FFT processed signal from _{z 1 (n d), z} 2 (n d), ···, z Np (n d ), The power P _{np of the} pulse direction FFT processing signals z ₁ (n _d ), z ₂ (n _d ),..., Z _Np (n _d ) as shown in the following equation (13). (L) is calculated.

Delay time estimation section detecting unit 45, calculating the power P _np pulses direction FFT processed signal (l), compares the power P _np (l) with a preset reference power Th _rd of the pulse direction FFT processed signal .
If the power P _np (l) of the pulse direction FFT processing signal is smaller than the reference power Th _{rd, the} delay time estimation interval detector 45 determines that the target signal does not exist in the delay time estimation interval, but the pulse direction If the power P _np (l) of the FFT processing signal is larger than the reference power Th _rd, it is determined that the target signal exists in the delay time estimation section.

また、遅延時間推定区間検出部４５は、遅延時間推定区間＃ｈに目標信号が存在している旨をデシメーション処理部４６−１〜４６−Ｎ_pに通知する。 For example, when the power P _np0 (h) of the pulse direction FFT processing signal when n _p = n _p0 and l = h is larger than the reference power Th _{rd, the} delay time estimation interval detection unit 45 performs the pulse direction FFT processing signal z. _np0 (1 + M (h−1)), z _np0 (2 + M (h−1)),..., z _np0 (N + M (h−1)) include a target signal component and The Doppler frequency _fd is obtained from (14), and the Doppler frequency _fd is output.

The delay time estimation section detecting unit 45 notifies the target signal is present in the delay time estimation section #h to the decimation processing section 46-1 to 46-N _p.

Decimation processing section 46-1 to 46-N _p receives the notification that the target signal to the delay time estimation section #h from the delay time estimation section detecting unit 45 is present, arranged from the memory circuit 43 in the time direction , Received signals s _np (1), s _np (2),..., S _np (N _d ) are input, and the received signals s _np (1), s _np (2) _,. , S _np (N _d ), the partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} (N) of the delay time estimation interval #h are extracted. However, 1 ≦ n _p ≦ N _p .
That is, the signal extraction unit 51 of the decimation processing units 46-1 to 46-N _p receives the received signals s _np (1) and s _np (2) arranged in the time direction as shown in the following equation (15). _{), ···, s np (N} d) partial signals of the delay time estimation section #h from _{s np, h (1),} s np, h (2), ···, s np, h (N) is Extract (see FIG. 18).

Received signal spectrum generating unit 52 of the decimation unit 46-1 to 46-N _p is the partial signal of the signal extraction unit 51 is a delay time estimation section _{#h s np, h (1)} , s np, h (2), , S _{np, h} (N) is extracted, the partial signals s _{np, h} (1), s _{np, h} (2),..., S _{np, h} ( N) is subjected to FFT processing to generate received signal spectrums y _{np, h} (1), y _{np, h} (2),..., Y _{np, h} (N). However, 1 ≦ n _p ≦ N _p .
The division signal generation unit 53 of the decimation processing units 46-1 to 46-N _p is received by the reception signal spectrum generation unit 52 by the reception signal spectrum y _{np, h} (1), y _{np, h} (2) _{,. When np, h} (N) is generated, the received signal spectrums y _{np, h} (1), y _{np, h} (2),..., y _{np, h} (N) are converted into transmission wave spectra Γ (1), Divided by Γ (2),..., Γ (N) and divided signals x ′ _{np, h} (1), x ′ _{np, h} (2) _,. Generate. However, 1 ≦ n _p ≦ N _p .
Decimation processing section 46-1 to 46-N _p decimation filter 54, the division signal generating section 53 divides the signal _{x 'np, h (1)} , x' np, h (2), ···, x 'np _and generates the h (N), the division signal _{x 'np, h (1)} , x' np, h (2), ···, x 'np, subjected to decimation processing h (N), the decimation Processing signals x _{np, h} (1), x _{np, h} (2),..., X _{np, h} (M) are output. However, 1 ≦ n _p ≦ N _p .

Delay time estimating unit 47 decimating the signal from the decimation processor _{46-1~46-N p x np, h} (1), x np, h (2), ···, x np, h (M) is Then, the decimation processing signals x _{h, np} (1), x _{h, np} (2),..., X _{h, np} (M) are subjected to super-resolution delay time estimation processing, and the delay time estimation interval # Estimate the delay time of the target signal present in h.
Specific processing contents of the delay time estimation unit 47 are as follows.

The pulse direction average correlation matrix generation unit 25 of the delay time estimation unit 47 receives the decimation processing signals x _{np, h} (1), x _{np, h} (2) from the decimation processing units 46-1 to 46-N _{p. ..} , X _{np, h} (M) (1 ≦ n _p ≦ N _p ), the decimation processing signals x _{np, h} (1), x _{np, h} (2 ,..., X _{np, h} (M) A correlation matrix R representing the correlation between (1 ≦ n _p ≦ N _p ) is generated.

The eigenvector calculation unit 22 of the delay time estimation unit 47 generates eigenvalues v _e (1), v _e (2),... Of the correlation matrix R when the pulse direction average correlation matrix generation unit 25 generates the correlation matrix R. , v _e (M) eigenvectors and _{(v e (1)> v} e (2)>···> v e (M)), corresponding to the eigenvalue _{v e (i) (1 ≦} i ≦ M) e (I) is calculated.
When the eigenvector calculation unit 22 calculates the eigenvector e (i) corresponding to the eigenvalue v _e (i) (1 ≦ i ≦ M) as described above, the eigenvector e (K + 1), e (K + 2),..., E (M) are output to the MUSIC processing unit 23.

  When the MUSIC processing unit 23 of the delay time estimation unit 47 receives the eigenvectors e (K + 1), e (K + 2),..., E (M) from the eigenvector calculation unit 22, the eigenvectors e (K + 1) and e (K + 2) are received. ),..., E (M) is performed as a noise space, and the delay time of the target signal existing in the delay time estimation section #h is estimated.

  As apparent from the above, according to the eighth embodiment, the super-resolution delay time estimation process is performed only in the delay time estimation section #h where the target signal exists. In addition to the same effects as those of the first embodiment, the super-resolution delay time estimation process can be performed with a low load.

Embodiment 9 FIG.
FIG. 19 is a block diagram showing a delay time estimation apparatus according to Embodiment 9 of the present invention. In the figure, the same reference numerals as those in FIGS.
The memory circuit 61 receives the digital received signals s _np (1), s _np (2),..., S _np (N _d ) (1 ≦ n _p ≦ N _p ) from the A / D converter 6. The received signals are stored side by side in the pulse direction. Note that the memory circuit 61 constitutes reception signal storage means.
The delay time estimation interval detection unit 62 performs the same detection process as the delay time estimation interval detection unit 45 in FIG. 15 and outputs the delay time estimation interval #h in which the Doppler frequency f _d and the target signal exist. a pulsed direction FFT processed signal z _npO of the delay time estimation section #h (1 + M (h- 1)), z np0 (2 + M (h-1)), ···, z np0 (N + M (h-1) ) Is output. Note that the delay time estimation section detector 62 constitutes a delay time estimation section detection means.

The received signal spectrum generation unit 63 outputs the pulse direction FFT processing signals z _np0 (1 + M (h−1)) and z _np0 (2 + M (h−1) of the delay time estimation interval #h output from the delay time estimation interval detection unit 62. _{), ···, z np0 (N} + M (h-1)) received signal by performing FFT processing on the spectrum _{y h (1), y h} (2), ···, and generates a y _h (N).
The division signal generation unit 64 converts the reception signal spectrums y _h (1), y _h (2),..., Y _h (N) generated by the reception signal spectrum generation unit 63 into transmission wave spectra Γ (1), Γ. (2), ···, Γ ( N) divided by to divide the signal _{x 'h (1), x} ' h (2), to produce a _{···, x 'h (N)} .

The decimation filter 65 performs decimation processing on the division signals x ′ _h (1), x ′ _h (2),..., X ′ _h (N) generated by the division signal generation unit 64 to obtain a decimation processing signal x. _h (1), x _h (2),..., x _h (M) are output.
The reception signal spectrum generation unit 63, the division signal generation unit 64, and the decimation filter 65 constitute decimation means.
The delay time estimation unit 66 performs super-resolution delay time estimation processing on the decimation processing signals x _h (1), x _h (2),..., X _h (M) output from the decimation filter 65 to obtain a delay time. The delay time of the target signal existing in the estimation section #h is estimated. The delay time estimation unit 66 constitutes a delay time estimation unit.

Next, the operation will be described.
The transmitter 1 repeatedly generates radio waves N _p times, and the transmitting antenna 2 repeatedly transmits the radio waves generated by the transmitter 1 toward the target 3 N _p times.
When the reception antenna 4 receives the radio wave transmitted from the transmission antenna 2 and reflected by the target 3, the receiver 5 performs band limitation or phase detection on the reception wave of the reception antenna 4 to generate a reception signal.
When the receiver 5 generates a reception signal, the A / D converter 6 performs analog / digital conversion on the reception signal to obtain digital reception signals s _np (1), s _np (2) _{,. np} (N _d ) (1 ≦ n _p ≦ N _p ) is output to the memory circuit 43.

The memory circuit 61 receives digital received signals s _np (1), s _np (2),..., S _np (N _d ) (1 ≦ n _p ≦ N _p ) from the A / D converter 6. In addition, the received signals are stored side by side in the pulse direction.
Note that the memory circuit 61, N by _p times when the reception is completed, the received signal s ₁ which have accumulated alongside the pulse direction _{(n d), s 2 (} n d), ···, s Np (n _d) outputting a pulse direction FFT unit 44-1 to 44-N _d. However, 1 ≦ n _d ≦ N _d , 1 ≦ n _p ≦ N _p .

The pulse direction FFT units 44-1 to 44-N _d receive signals s ₁ (n _d ), s ₂ (n _d ),..., S _Np (n _d ) arranged from the memory circuit 61 in the pulse direction. ), The received signals s ₁ (n _d ), s ₂ (n _d ),..., S _Np (n _d ) are subjected to FFT processing in the same way as in the eighth embodiment, and the pulse direction FFT processing signals z ₁ (n _d ), z ₂ (n _d ),..., Z _Np (n _d ) are output.
Delay time estimation section detecting unit 62, the pulse direction FFT unit 44-1 to 44-N _d Pulse direction FFT processed signal from _{z 1 (n d), z} 2 (n d), ···, z Np (n d ), A detection process similar to that of the delay time estimation section detection unit 45 in FIG. 15 is performed to detect a delay time estimation section #h in which the target signal exists.
Similarly to the delay time estimation interval detection unit 45 in FIG. 15, the delay time estimation interval detection unit 62 outputs the delay time estimation interval #h in which the Doppler frequency f _d and the target signal exist, as well as the delay time thereof. Pulse direction FFT processing signals z _np0 (1 + M (h−1)), z _np0 (2 + M (h−1)),..., Z _np0 (N + M (h−1)) in the estimation interval #h are received signal spectra. The data is output to the generation unit 63.

The received signal spectrum generation unit 63 sends the pulse direction FFT processing signals z _np0 (1 + M (h−1)), z _np0 (2 + M (h−1)) of the delay time estimation interval #h from the delay time estimation interval detection unit 62, , Z _np0 (N + M (h−1)), the pulse direction FFT processing signal z _np0 (1 + M (h−1)), z _np0 (2 + M (h−1)) _,. z _np0 (N + M (h−1)) is subjected to FFT processing to generate reception signal spectra y _h (1), y _h (2),..., y _h (N).
When the reception signal spectrum generation unit 63 generates the reception signal spectrum y _h (1), y _h (2),..., Y _h (N), the division signal generation unit 64 receives the reception signal spectrum y _h (1 _{), y h (2),} ···, y h (N) of the transmitted wave spectrum Γ (1), Γ (2 ), ···, by dividing the gamma (N) dividing the signal x _'h (1 , X ′ _h (2),..., X ′ _h (N).

When the division signal generation unit 64 generates the division signals x ′ _h (1), x ′ _h (2),..., X ′ _h (N), the decimation filter 65 generates the division signal x ′ _h (1). _{, x 'h (2),} ···, x' is subjected to decimation processing _h (N), decimated signal _{x h (1), x h} (2), ···, x h a (M) Output.
When receiving the decimation processing signals x _h (1), x _h (2),..., X _h (M) from the decimation filter 65, the delay time estimation unit 66 receives the decimation processing signals x _h (1), x _h (2),..., x _h (M) is subjected to super-resolution delay time estimation processing to estimate the delay time of the target signal existing in the delay time estimation section #h.

  As apparent from the above, according to the ninth embodiment, the super-resolution delay time estimation process is performed only in the delay time estimation section #h where the target signal exists. In addition to the same effects as those of the first embodiment, the super-resolution delay time estimation process can be performed with a low load.

Embodiment 10 FIG.
FIG. 20 is a block diagram showing a delay time estimation unit 8-l (1 ≦ l ≦ L) of the delay time estimation device according to the tenth embodiment of the present invention. The delay time estimation device according to the tenth embodiment of the present invention is shown in FIG. The overall configuration is the same as that of the delay time estimation apparatus of FIG.
In FIG. 20, the same reference numerals as those in FIG.
The ESPRIT processing unit 26 uses the eigenvectors e ₁ (K + 1), e ₁ (K + 2),..., E ₁ (M _d ) calculated by the eigenvector calculation unit 22 to perform ESPRIT, which is a low-load super-resolution estimation process. (Estimation of Signal Parameters via Rotational Invariance Techniques) processing is performed to estimate the delay time of the target signal.

In the first embodiment, the MUSIC processing unit 23 of the delay time estimation unit 8-1 (1 ≦ l ≦ L) performs the eigenvectors e ₁ (K + 1), e ₁ (K + 2),..., E ₁ (M _d ) As shown in FIG. 20, the ESPRIT processing unit 26 performs the MUSIC processing using the noise space to estimate the delay time τ of the target signal included in the delay time estimation interval # 1. The target signal delay time τ may be estimated by performing ESPRIT processing using the eigenvectors e _l (K + 1), e _l (K + 2),..., E _l (M _d ).

That, ESPRIT processing unit 26 of the delay time estimation unit 8-l (1 ≦ l ≦ L) is eigenvector computing section 22 is the eigenvector _{e l (K + 1),} e l (K + 2), ···, e l (M d ) Is calculated, eigenvectors e _l (1), e _l (2), corresponding to upper K types of eigenvalues having a large value among eigenvalues of correlation matrix R _l , as shown in equation (17) below. ..., matrix E _l is generated from e _l (K _l ).
_El = [ _el (1)... _El ( _Kl )] (17)

式（１８）において、行列Ｅ^' _l ^Hは行列Ｅ^' _lのエルミート共役を表している。
また、行列Ｊ₁は（Ｍ_d−１）行Ｍ_d列の行列であり、Ｊ₁（ｊ，ｋ）は行列Ｊ₁のｉ行ｋ列成分を表している。
また、行列Ｊ₂は（Ｍ_d−１）行Ｍ_d列の行列であり、Ｊ₂（ｊ，ｋ）は行列Ｊ₂のｉ行ｋ列成分を表している。 When the ESPRIT processing unit 26 generates the matrix E _l from the eigenvectors e _l (1), e _l (2),..., E _l (K _l ), as shown in the following equation (18), the matrix A matrix Ψ _l is calculated from E _l .

In Expression (18), the matrix E ^′ _l ^H represents the Hermitian conjugate of the matrix E ^′ _l .
The matrix J ₁ is a matrix of (M _d −1) rows and M _d columns, and J ₁ (j, k) represents the i row and k column components of the matrix J ₁ .
The matrix J ₂ is a matrix of (M _d −1) rows and M _d columns, and J ₂ (j, k) represents the i row and k column components of the matrix J ₂ .

式（１９）において、ｖ_Ψ（ｋ_l）は行列Ψ_lのｋ_l番目の固有値、ａｒｇ［ｖ_Ψ（ｋ_l）］は固有値ｖ_Ψ（ｋ_l）の偏角を表している。 ESPRIT processing unit 26, calculating the matrix [psi _l from the matrix E _l, by the following equation (19), obtains the delay time of the target signal tau _k (l) hat.

In Expression (19), v _Ψ (k _l ) represents the k _l- th eigenvalue of the matrix Ψ _l , and arg [v _Ψ (k _l )] represents the declination of the eigenvalue v _Ψ (k _l ).

  As is apparent from the above, according to the tenth embodiment, the ESPRIT process is performed to estimate the delay time of the target signal, so that the delay time estimation unit 8-1 (1 ≦ l ≦ L The effect of reducing the load of the super-resolution estimation process in FIG.

Embodiment 11 FIG.
FIG. 21 is a block diagram showing a delay time estimation unit 8-l (1 ≦ l ≦ L) of the delay time estimation apparatus according to Embodiment 11 of the present invention. The delay time estimation apparatus according to Embodiment 11 of the present invention is shown in FIG. The overall configuration is the same as that of the delay time estimation apparatus of FIG.
In FIG. 21, the same reference numerals as those in FIG.
The maximum likelihood estimation processing unit 27 performs maximum likelihood estimation using the eigenvectors e _l (K + 1), e _l (K + 2),..., E _l (M _d ) calculated by the eigenvector calculation unit 22, Estimate the delay time of the target signal.

In the first embodiment, the MUSIC processing unit 23 of the delay time estimation unit 8-1 (1 ≦ l ≦ L) performs the eigenvectors e ₁ (K + 1), e ₁ (K + 2),..., E ₁ (M _d ) In FIG. 21, the MUSIC process using the noise space as the noise space to estimate the delay time τ of the target signal included in the delay time estimation section # 1 is shown. As shown in FIG. 27 performs estimation of the maximum likelihood using the eigenvectors e _l (K + 1), e _l (K + 2),..., E _l (M _d ), thereby estimating the delay time τ of the target signal. May be.

式（２０）において、ａ（τ）は、上記の式（３）で定義しているステアリングベクトルを表している。 That is, in the maximum likelihood estimation processing unit 27 of the delay time estimation unit 8-1 (1 ≦ l ≦ L), the eigenvector calculation unit 22 uses the eigenvectors e _l (K + 1), e _l (K + 2) _,. When M _d ) is calculated, an evaluation function Θ (τ ₁ , τ ₂ ,... With the delay times τ ₁ , τ _{2 ,.}・ ・ 、 Τ _Kl ) is set.

In equation (20), a (τ) represents the steering vector defined in equation (3) above.

Maximum likelihood estimation processing unit 27, the evaluation function _{Θ (τ 1, τ 2,} ···, τ Kl) Setting, the evaluation function theta and the maximum _{(τ 1, τ 2, ···} , τ Kl) the Τ ₁ = τ ₁ (l) Hat,..., Τ _Kl = τ _K (l) Find the hat.
Note that the estimated value of the delay time of the target k included in the estimation interval l is τ _k (l) hat.

  As is apparent from the above, according to the eleventh embodiment, the maximum likelihood estimation process is performed and the delay time of the target signal is estimated. Therefore, the delay time estimation unit 8-1 (1 ≦ l The effect of reducing the load of the super-resolution estimation process in ≦ L) is achieved.

Embodiment 12 FIG.
FIG. 22 is a block diagram showing a delay time estimation unit 8-l (1 ≦ l ≦ L) of the delay time estimation device according to the twelfth embodiment of the present invention. The delay time estimation device according to the twelfth embodiment of the present invention is shown in FIG. The overall configuration is the same as that of the delay time estimation apparatus of FIG.
In FIG. 22, the same reference numerals as those in FIG.
The memory circuit 28 serving as a signal storage unit receives the decimation processing signals x _l (1), x _l (2),..., X _l (N _d ) (1 ≦ l ≦ L) output from the decimation processing unit 7. When accumulating and receiving a determination result indicating that there is a target signal from the adjacent signal correlation target signal determination unit 29, the decimation processing signals x _l (1), x _l (2),..., X _l (N _d ) (1 ≦ l ≦ L) is output to the correlation matrix generator 21.
The adjacent signal correlation type target signal determination unit 29, which is a presence / absence determination unit, determines the presence / absence of a target signal based on the correlation value between adjacent decimation processing signals among the decimation processing signals stored in the memory circuit 28. judge.

Next, the operation will be described.
The memory circuit 28 of the delay time estimation unit 8-1 (1 ≦ l ≦ L) receives the decimation processing signals x ₁ (1), x ₁ (2), from the decimation processing unit 7 in the same manner as in the first embodiment. .., X _l (N _d ) (1 ≦ l ≦ L), the decimation processing signals x _l (1), x _l (2),..., X _l (N _d ) are accumulated. .

式（２１）において、ｘ（ｉ）^*はｘ（ｉ）の複素共役を表している。 The adjacent signal correlation type target signal determination unit 29 of the delay time estimation unit 8-1 (1 ≦ l ≦ L) receives the decimation processing signals x ₁ (1), x ₁ (2) _,. ..., X _l (N _d ), the presence / absence of a target signal is determined based on the correlation value between adjacent decimation processing signals.
That is, the adjacent signal correlation type target signal determination unit 29 calculates a correlation value cor (l) between adjacent decimation processing signals as shown in the following equation (21).

In formula (21), x (i) ^* represents the complex conjugate of x (i).

When the adjacent signal correlation type target signal determination unit 29 calculates a correlation value cor (l) between adjacent decimation processing signals, the correlation value cor (l) is compared with a preset threshold.
If the correlation value cor (l) is smaller than the threshold, the adjacent signal correlation type target signal determination unit 29 determines that there is no target signal and outputs the determination result to the memory circuit 28.
On the other hand, if the correlation value cor (l) is larger than the threshold, it is determined that there is a target signal, and the determination result is output to the memory circuit 28.

When the memory circuit 28 of the delay time estimation unit 8-1 (1 ≦ l ≦ L) receives the determination result that the target signal is present from the adjacent signal correlation type target signal determination unit 29, the accumulated decimation processing signal x _l (1), x _l (2),..., x _l (M) (1 ≦ l ≦ L) are output to the correlation matrix generation unit 21.
Since the processing contents after the correlation matrix generation unit 21 are the same as those in the first embodiment, description thereof is omitted.

  As apparent from the above, according to the twelfth embodiment, the presence / absence of the target signal is determined, and only when the target signal is present, the MUSIC process is performed to estimate the delay time of the target signal. Since it comprised, there exists an effect which can reduce the load of the super-resolution estimation process in the delay time estimation part 8-1 (1 <= l <= L).

It is a block diagram which shows the delay time estimation apparatus by Embodiment 1 of this invention. It is a block diagram which shows the decimation processing part 7 of the delay time estimation apparatus by Embodiment 1 of this invention. It is a block diagram which shows the delay time estimation part 8-1 (1 <= l <= L) of the delay time estimation apparatus by Embodiment 1 of this invention. It is explanatory drawing which shows the processing content of the decimation processing part. It is a block diagram which shows the inside of decimation filter 14-1 (1 <= l <= L). It is a block diagram which shows the decimation processing part 7 of the delay time estimation apparatus by Embodiment 2 of this invention. It is a block diagram which shows the inside of the estimation area cut-out decimation part 15-1 (1 <= l <= L) in the decimation processing part 7. FIG. It is a block diagram which shows the decimation processing part 7 of the delay time estimation apparatus by Embodiment 3 of this invention. It is a block diagram which shows the delay time estimation part 8-1 (1 <= l <= L) of the delay time estimation apparatus by Embodiment 4 of this invention. It is a block diagram which shows the delay time estimation apparatus by Embodiment 5 of this invention. It is a block diagram which shows the decimation processing part 9 of the delay time estimation apparatus by Embodiment 5 of this invention. It is a block diagram which shows the delay time estimation apparatus by Embodiment 6 of this invention. It is a block diagram which shows the delay time estimation apparatus by Embodiment 7 of this invention. It is explanatory drawing which shows the difference between the normal sampling in the A / D converter 6, and the oversampling in the oversampling A / D converter 42 in a multipath fading environment. It is a block diagram which shows the delay time estimation apparatus by Embodiment 8 of this invention. Is a block diagram illustrating a decimation processor 46-n _p of the delay time estimation apparatus according to Embodiment 8 of the present invention _{(1 ≦ n p ≦ N p} ). It is a block diagram which shows the delay time estimation part 47 of the delay time estimation apparatus by Embodiment 8 of this invention. It is an explanatory view showing a process of decimation processing section 46-1 to 46-N _p. It is a block diagram which shows the delay time estimation apparatus by Embodiment 9 of this invention. It is a block diagram which shows the delay time estimation part 8-1 (1 <= l <= L) of the delay time estimation apparatus by Embodiment 10 of this invention. It is a block diagram which shows the delay time estimation part 8-1 (1 <= l <= L) of the delay time estimation apparatus by Embodiment 11 of this invention. It is a block diagram which shows the delay time estimation part 8-1 (1 <= l <= L) of the delay time estimation apparatus by Embodiment 12 of this invention.

Explanation of symbols

1 transmitter (radio wave transmission means), 2 transmission antenna (radio wave transmission means), 3 target, 4,4-1, 4-2 reception antenna (radio wave reception means), 5,5-1,5-2 receiver ( Radio wave receiving means), 6,6-1,6-2 A / D converter (radio wave receiving means), 7 decimation processing section (decimation means), 8-1 to 8-L delay time estimating section (delay time estimating means) ), 9 decimation processing unit (decimation means), 10 delay time estimation unit (delay time estimation means), 11 received signal division unit, 12-1 to 12-L received signal spectrum generation unit, 13-1 to 13-L division Signal generation unit, 14-1 to 14-L decimation filter, 15-1 to 15-L estimation section cut-out type decimation unit, 16-1 to 16-L zero division prevention type division signal generation unit, 17 target signal detection unit, 18 Division signal Generator, 19 decimation filter, 21 correlation matrix generator, 22 eigenvector calculator, 23 MUSIC processor, 24 target signal determination type eigenvector calculator (target discrimination unit), 25 pulse direction average correlation matrix generator, 26 ESPRIT process 27, maximum likelihood estimation processing unit, 28 memory circuit (signal storage unit), 29 adjacent signal correlation type target signal determination unit (presence / absence determination unit), 31 IFFT processing unit (inverse Fourier transform processing unit), 32 estimation section cutout processing Unit (signal extraction processing unit), 33 FFT processing unit (Fourier transform processing unit), 41 maximum ratio combining processing unit (maximum ratio combining processing unit), 42 oversampling A / D converter (radio wave receiving unit), 43 memory circuit (received signal storage unit), 44-1 to 44-N _d pulse direction FFT unit (Fourier transform processing unit), 45 a delay time Constant section detection unit (the delay time estimation section detecting means) 46-1 to 46-N _p decimation unit (decimation means), 47 a delay time estimating unit (delay time estimating means) 51 signal extraction unit, 52 received signal spectrum Generator, 53 division signal generator, 54 decimation filter, 61 memory circuit (received signal storage means), 62 delay time estimation section detector (delay time estimation section detector), 63 received signal spectrum generator (decimation means), 64 division signal generation unit (decimation means), 65 decimation filter (decimation means).

Claims (14)

  A radio wave transmitting means for transmitting a radio wave toward the target, a radio wave receiving means for receiving a radio wave transmitted from the radio wave transmitting means and reflected by the target and generating a reception signal, and generated by the radio wave receiving means. The received signal is divided into partial signals of a plurality of preset delay time estimation sections, and decimation means for performing decimation processing on the partial signals of the plurality of delay time estimation sections, and more than a plurality of decimation processing signals by the decimation means. A delay time estimation apparatus comprising: delay time estimation means for performing a resolution delay time estimation process to estimate a delay time of a target signal existing in the delay time estimation section.   The decimation means includes a received signal dividing section that divides the reception signal generated by the radio wave receiving means into partial signals of a plurality of preset delay time estimation sections, and Fourier transforms into the partial signals divided by the received signal dividing section A plurality of reception signal spectrum generation units that generate a reception signal spectrum by performing processing, and a plurality of division signal generations that generate a division signal by dividing the reception signal spectrum generated by the reception signal spectrum generation unit by a transmission wave spectrum 2. The delay time estimation apparatus according to claim 1, wherein the delay time estimation apparatus comprises: a division unit; and a plurality of decimation filters for performing decimation processing on the division signal generated by the division signal generation unit.   The decimation means includes a received signal dividing section that divides the reception signal generated by the radio wave receiving means into partial signals of a plurality of preset delay time estimation sections, and Fourier transforms into the partial signals divided by the received signal dividing section A plurality of reception signal spectrum generation units that generate a reception signal spectrum by performing processing, and a plurality of division signal generations that generate a division signal by dividing the reception signal spectrum generated by the reception signal spectrum generation unit by a transmission wave spectrum A plurality of inverse Fourier transform processing units for generating a demodulated signal by performing an inverse Fourier transform process on the division signal generated by the division signal generating unit, and a delay from the demodulated signal generated by the inverse Fourier transform processing unit Multiple signal extraction processing units that extract signals in the time estimation section, and delay times extracted by the signal extraction processing unit Delay time estimating apparatus according to claim 1, characterized in that it is composed of a plurality of Fourier transform processing unit for performing Fourier transform processing on a signal of a constant interval.   The division signal generation unit corrects the transmission wave spectrum to the minimum value when the amplitude value of the transmission wave spectrum is smaller than a preset minimum value, and divides the reception signal spectrum by the corrected transmission wave spectrum. 4. The delay time estimation apparatus according to claim 2, wherein the delay time estimation apparatus is characterized in that:   The delay time estimation unit calculates a correlation matrix generation unit that generates a correlation matrix representing a correlation between a plurality of decimation processing signals by the decimation unit, and calculates an eigenvalue and an eigenvector of the correlation matrix generated by the correlation matrix generation unit, If there is an eigenvalue larger than the threshold value, the target signal is recognized as being present and the eigenvector is output. On the other hand, if there is no eigenvalue larger than the predetermined threshold value, 2. The delay according to claim 1, further comprising a MUSIC processing unit that performs a MUSIC process using the eigenvector output from the target determination unit as a noise space and estimates a delay time of the target signal. Time estimation device.   A radio wave transmitting means for transmitting a radio wave toward the target, a radio wave receiving means for receiving a radio wave transmitted from the radio wave transmitting means and reflected by the target and generating a reception signal, and generated by the radio wave receiving means. The received signal is divided into partial signals of a plurality of preset delay time estimation sections, and it is determined whether or not the target signals are included in the partial signals of the plurality of delay time estimation sections. Delay time estimation comprising: decimation means for performing decimation processing on a partial signal, and delay time estimation means for performing delay resolution estimation processing on the decimation processing signal by the decimation means to estimate a delay time of a target signal apparatus.   The decimation means includes a received signal dividing section that divides the reception signal generated by the radio wave receiving means into partial signals of a plurality of preset delay time estimation sections, and Fourier transforms into the partial signals divided by the received signal dividing section A plurality of reception signal spectrum generation units that perform processing to generate a reception signal spectrum, and among reception signal spectra generated by the plurality of reception signal spectrum generation units, a reception signal spectrum whose average power is larger than a reference power is a target. A target signal detection unit that determines that a signal is included and outputs the received signal spectrum, and a division that generates a division signal by dividing the reception signal spectrum output from the target signal detection unit by the transmission wave spectrum A decimation unit that performs decimation processing on the signal generation unit and the division signal generated by the division signal generation unit. Delay time estimating apparatus according to claim 6, characterized in that it is composed of a down filter.   In addition to mounting two radio wave receiving means, there is provided a maximum ratio combining processing means for combining the received signals generated by the two radio wave receiving means with a maximum ratio and outputting the received signal after the maximum ratio combining to the decimation means. The delay time estimation apparatus according to any one of claims 1 to 7, wherein the delay time estimation apparatus according to any one of claims 1 to 7 is provided.   The radio wave receiving means is characterized in that when the radio wave transmitted from the radio wave transmitting means is code-modulated in units of pulses, the radio wave sampling means generates a reception signal by sampling the radio wave at a time interval shorter than the chip width of the pulse. The delay time estimation apparatus according to any one of claims 1 to 8.   A radio wave transmitting means for transmitting a radio wave toward a target; a radio wave receiving means for receiving a radio wave transmitted from the radio wave transmitting means and reflected by the target; and generating a received signal; and The reception signal storage means for storing the reception signal in the pulse direction and storing the reception signal in the time direction, and the reception signal stored in the pulse direction in the reception signal storage means A plurality of Fourier transform processing means for performing Fourier transform processing on the signal, and a delay time estimation section detection for detecting a delay time estimation section in which the target signal exists by analyzing a Fourier transform processing signal by the plurality of Fourier transform processing means And a delay time detected by the delay time estimation interval detection means from the received signals stored side by side in the time direction in the received signal storage means A decimation unit that extracts a received signal in the estimation interval and performs decimation processing on the received signal, and a target that exists in the delay time estimation interval by performing super-resolution delay time estimation processing on the decimation processing signal by the decimation unit A delay time estimation device comprising delay time estimation means for estimating a delay time of a signal.   A radio wave transmitting means for transmitting a radio wave toward a target; a radio wave receiving means for receiving a radio wave transmitted from the radio wave transmitting means and reflected by the target; and generating a received signal; and And a plurality of Fourier transform processing means for performing Fourier transform processing on the received signals stored in the pulse direction in the received signal storage means. And a delay time estimation section detecting means for detecting a delay time estimation section in which a target signal exists and analyzing the Fourier transform processing signals by the plurality of Fourier transform processing means, and detecting by the delay time estimation section detection means Decimation means for performing decimation processing on the Fourier transform processing signal in the estimated delay time estimation section, and decimation by the decimation means Perform super-resolution delay time estimation processing Shon processed signal, the delay time estimation device that includes a delay time estimating means for estimating a delay time of the target signal present in the delay time estimation section.   The delay time estimation means calculates a correlation matrix generation unit that generates a correlation matrix representing a correlation between a plurality of decimation processing signals by the decimation means, and calculates an eigenvalue of the correlation matrix generated by the correlation matrix generation unit, An eigenvector calculation unit that calculates a corresponding eigenvector and an ESPRIT processing unit that performs ESPRIT processing using the eigenvector calculated by the eigenvector calculation unit and estimates a delay time of the target signal. The delay time estimation apparatus according to claim 1.   The delay time estimation means calculates a correlation matrix generation unit that generates a correlation matrix representing a correlation between a plurality of decimation processing signals by the decimation means, and calculates an eigenvalue of the correlation matrix generated by the correlation matrix generation unit, An eigenvector calculation unit that calculates a corresponding eigenvector, and a maximum likelihood estimation processing unit that performs a maximum likelihood estimation process using the eigenvector calculated by the eigenvector calculation unit and estimates a delay time of the target signal. The delay time estimation apparatus according to claim 1, wherein:   The delay time estimation means is based on the correlation value between adjacent decimation processing signals among the signal accumulation section for accumulating a plurality of decimation processing signals by the decimation means and the decimation processing signals accumulated in the signal accumulation section. Thus, if the presence / absence determination unit for determining the presence / absence of the target signal and the determination result of the presence / absence determination unit indicate that the target signal is present, a plurality of decimation processing signals accumulated in the signal accumulation unit A correlation matrix generation unit that generates a correlation matrix that represents correlation, an eigenvalue of the correlation matrix generated by the correlation matrix generation unit, an eigenvector calculation unit that calculates an eigenvector corresponding to the eigenvalue, and the eigenvector calculation unit MU that performs a MUSIC process using the calculated eigenvector as a noise space and estimates a delay time of the target signal Delay time estimating apparatus according to claim 1, characterized in that it is constituted by an IC unit.

Images