JP2010025576A - Wave number estimating apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To overcome the problem that a conventional wave number estimating apparatus erroneously estimates a wave number if a difference between a target characteristic value and a noise characteristic value are reduced by a low SNR. <P>SOLUTION: A beam synthesis of a reflection signal from a target is implemented so as to obtain a high gain in the beam orientation direction of an antenna. An inter-pulse integration of a beam synthesis signal is implemented. If a target signal is detected from an inter-pulse integration signal, a correlation matrix of the reflection signal from the target is calculated by a correlation matrix calculating means. A characteristic value and a characteristic vector are calculated by a characteristic value and characteristic vector calculating means. A beam is formed by a characteristic beam forming means using the characteristic vector. A coherent integration of a beam formation output is implemented by a coherent integration means. A coherent integration value is extracted by a predetermined reference. The wave number is estimated by a wave number estimating means. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a wave number estimating apparatus for estimating a target number included in a received signal in a sensor array system such as a radar or a sonar that transmits and receives a plurality of pulses.

  As a conventional wave number estimation device of this type, for example, non-patent literature “M. Wax, T. Kailath,“ Detection of signals by information theoretic criteria, ”IEEE Trans. Acoustics, Speech, and Signal Processing, Vol. 33, No. 2, pp. 387-392, Apr. 1985 ”.

  In the above document, eigenvalues and eigenvectors are obtained from the correlation matrix of the received signal, a normative value based on AIC (Akaike Information Criterion) is obtained from the eigenvalues arranged in descending order, and the wave number is estimated from the minimum normative value. . In the above-mentioned document, similarly, a normative value based on the minimum description length (MDL) is obtained from the eigenvalue, and the wave number is estimated from the smallest normative value.

M. Wax, T. Kailath, “Detection of signals by information theoretic criteria,” IEEE Trans. Acoustics, Speech, and Signal Processing, Vol. 33, no. 2, pp. 387-392, Apr. 1985

In the conventional wave number estimating device, any of the above methods has a problem of performing an erroneous wave number estimation when the signal to noise ratio (SNR) is small so that the difference between the target eigenvalue and the noise eigenvalue is small. there were.
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a wave number estimation device that improves the wave number estimation performance at a low SNR.

  The wave number estimating apparatus according to the present invention transmits and receives a plurality of pulse waves by a plurality of sensors, and forms a sum beam with respect to the output of the target signal acquiring means for acquiring the reflected signal from the target and the target signal acquiring means Sum beam forming means, pulse integration means for integrating the output of the sum beam, amplitude detection means for detecting amplitude of the output of the pulse integration means, and signal detection of the output of the amplitude detection means. When a signal is detected by the signal detection means and the signal detection means, a correlation matrix calculation means for calculating a correlation matrix of the target signal from the output of the target signal acquisition means, and eigenvalues / eigenvectors are calculated from the correlation matrix. Eigenvalue / eigenvector calculation means, eigenbeam forming means for performing eigenbeam formation from the eigenvector and the output of the target signal acquisition means, A coherent integration means for coherently integrating each beam output by the means, an integration value extraction means for extracting a predetermined coherent integration value from the coherent integration value, and a wave number for estimating the wave number of the target signal from the output of the integration value extraction means And estimation means.

  According to the wave number estimating apparatus of the present invention, the reflected signal from the target is beam-synthesized so as to obtain a high gain in the beam directing direction, the beam synthesized signal is integrated between pulses, and the target signal is detected from the inter-pulse integrated signal. When the target signal is detected, the correlation matrix of the reflected signal from the target is calculated, eigenvalues / eigenvectors are calculated, beam forming is performed using the eigenvectors, and coherent integration is performed on the beam forming output. Since the wave number is estimated by extracting the coherent integration value according to a predetermined reference, the wave number estimation performance at a low SNR can be improved.

Embodiment 1 FIG.
1 to 3 are configuration diagrams showing Embodiment 1 of a wave number estimating apparatus of the present invention incorporated in a radar system.
In FIG. 1, a transmitter 1 generates a transmission wave based on preset specifications and outputs the transmission wave to a plurality of circulators 2.
The circulator 2 outputs the transmission waves output from the transmitter 1 to the corresponding plurality of phase shifters 4, reflects the target from the phase shifter 4, and receives it by the corresponding antennas 5 as the plurality of sensors. The received signals are output to the corresponding receivers 6.
The phase shift value calculating means 3 calculates a phase shift value corresponding to the beam directing direction of the antenna 5 as a preset sensor, and outputs the phase shift value to the plurality of phase shifters 4.
The phase shifter 4 shifts the phase of the transmission wave output from the circulator 2 by the phase shift value set by the phase shift value calculating means 3, and shifts the phase of the reception wave received by the antenna 5. To do. Note that the phase of the received wave received from the antenna 5 may not be shifted.

The antenna 5 as a sensor is an element antenna or a subarray antenna, and radiates the transmission wave output from the phase shifter 4 in a preset beam directing direction and receives the transmission wave reflected by the target as a target signal. . At this time, clutter and interference waves that are unnecessary waves may be received.
The receiver 6 receives the reception wave output from the circulator 2, down-converts the frequency of the reception wave, and converts it into a baseband reception signal.
The AD (Analog to Digital) converter 7 converts the analog reception signal output from the receiver 6 into a digital reception signal.
The transmitter 1, the plurality of circulators 2, the phase shift value calculation means 3, the plurality of phase shifters 4, the plurality of antennas 5, the plurality of receivers 6, and the plurality of AD converters 7 constitute target signal acquisition means. Has been.

The clutter suppression unit 8 suppresses the clutter power included in the reception signal output from the AD converter 7 using a Doppler filter having a null in the Doppler frequency of the clutter. Specific Doppler filters include MTI (Moving Target Indicator) and AMTI (Adaptive Moving Target Indicator).
The interference wave suppression means 9 suppresses interference wave power included in the received signal after clutter suppression output from the clutter suppression unit 8 using a spatial filter having a null in the arrival direction of the interference wave. Specific spatial filters include SMI (Sample Matrix Inverse) and PCI (Principal Component Inverse).

The sum beam forming unit 10 performs beam synthesis on the reception signal after the interference wave suppression output from the interference wave suppression unit 9 so as to have a high gain in a preset beam directing direction.
The inter-pulse integration unit 11 performs inter-pulse integration of the reception signal output from the sum beam forming unit 10. Specific integration between pulses includes NCI (Noncohernet Integration) and CI (Coherent Integration).
The signal detection means 12 performs signal detection by threshold processing after detecting the amplitude of the reception signal output from the interpulse integration means 11. Specific threshold processing includes CFAR (Constant False Alarm Ratio) processing.

  In the radar described above, when a plurality of target signals are received, they can be detected separately if they are separated from each other by the distance resolution or the Doppler resolution of the radar. It becomes. However, when a plurality of target signals are equal to or smaller than the distance resolution and the Doppler resolution, there is a problem that the target number is erroneously handled if the detected number is handled as it is. Treating it as a single target despite multiple targets causes the operator to misrecognize the situation around the radar, and makes it impossible to correctly estimate the target angle by monopulse angle measurement or super-resolution angle measurement. Develop into a serious problem. For this reason, it is an important issue to correctly know the distance and the multiple target numbers below the Doppler resolution.

In the present invention, in preparation for the case where it is difficult to separate the target signal in the distance direction and the Doppler direction, the radar configuration includes a correlation matrix calculation unit, an eigenvalue / eigenvector calculation unit, an eigenbeam formation unit, a coherent integration unit, an integration value By adding extraction means and wave number estimation means, a wave number estimation function of the target signal is given.
When there is signal detection, the correlation matrix calculation means 13 calculates the correlation matrix of the target signal output from the interference wave suppression means 9 in the detection range bin.
The eigenvalue / eigenvector calculation means 14 calculates the eigenvalue / eigenvector of the correlation matrix output from the correlation matrix calculation means 13.

The eigen beam forming unit 15 performs beam forming on the target signal output from the interference wave suppressing unit 9 using the eigen vector calculated by the eigen value / eigen vector calculating unit 14.
The coherent integration unit 16 performs coherent integration on the beam output from the eigen beam forming unit 15.
The integral value extracting unit 17 extracts a predetermined integral value from the coherent integral value that is the output of the coherent integrating unit 16.
The wave number estimating means 21 estimates the target number from the integrated value that is the output from the integrated value extracting means 17.

FIG. 2 is a block diagram showing the integral value extracting means 17 according to the first embodiment of the present invention. The integral value extracting means 17 is formed by a first maximum value extracting means 18, and the first maximum value extracting means 18 is shown. Extracts the maximum value of the coherent integration value that is the output of the coherent integration means 16.
FIG. 3 is a block diagram showing another example of the integral value extracting means 17 according to Embodiment 1 of the present invention. The integral value extracting means 17 is formed by a moving average means 19 and a second maximum value extracting means 20. Is done.
The moving average means 19 performs a moving average of the coherent integration values that are outputs from the plurality of coherent integration means 16, and the second maximum value extraction means 20 extracts the maximum value from the output from the moving average means 19.

The correlation matrix calculation means 13 obtains a target signal correlation matrix R _XX as in the following equation (1).

  Where N (n = 1,..., N) is the number of pulse hits, x (n) is the target signal vector for the array antenna with the number of antennas M (m = 1,..., M), and H is the Hermitian transpose. It is.

  Here, the target signal vector x (n) is expressed as shown in Equation (2).

However, A is a matrix having K steering vectors a (θ _k ) as column vectors, and A is expressed as equation (3), and a (θ _k ) is expressed as equation (4).

g _m (θ _k ) is the complex gain of the n-th antenna with respect to the arrival direction θ _k of the k-th wave, and Ψ _m (θ _k ) is the phase of the n-th antenna with respect to the arrival direction θ _k of the k-th wave.
Further, s (n) in equation (2) is a complex amplitude vector of K target signals that are not completely correlated with each other, and is represented by the following equation (5).

  Similarly, n (n) in Equation (2) is a receiver noise vector that is uncorrelated between the receiving channels after the antenna, and is represented by Equation (6).

The eigenvalue / eigenvector calculation means 14 obtains the eigenvalue / eigenvector of the correlation matrix R _XX of the target signal as shown in the following equation (7).

Here, since the correlation matrix R _XX is a Hermitian matrix, Λ is a diagonal matrix with M eigenvalues and is expressed as in Equation (8). Incidentally, E is a matrix with column vectors eigenvectors e _m corresponding to the eigenvalue lambda _m. Further Equation (9) holds for the eigenvector e _m. Note that the norm of the eigenvector e _m shall meet the equation (10).

The eigen beam forming means 15 obtains the eigen beam output y _m (n) for the target signal vector x (n) as follows.
First, when the K target signals have different normalized Doppler frequencies, the target signal s _k (n) can be expressed as the following equation (11).

Where A _k is the amplitude, f _k is the normalized Doppler frequency, and φ _k is the initial phase.
Therefore, the target signal vector x (n) is expressed by the following equation (12).

Given the inherent beam of the M to the eigenvector e _m which is obtained from the correlation matrix R _XX of the target signal vector x (n) and the beam weight, each beam output can be expressed as follows.

First, if the eigenvector e _m corresponds to the target eigenvalue lambda _m, since the subspace eigenvectors corresponding to the target eigenvalues lambda _m e _m spanned target signal steering vector a (θ _k) are also included, e _m ^H a It can be expressed as (θ _k ) = c _m (θ _k ). That is, as will be described later, the beam output y _m (n) is a combination of each target Doppler frequency signal and receiver noise.

On the other hand, the eigenvector if e _m corresponding to the noise eigenvalues lambda _m, since the noise eigenvalues lambda eigenvectors corresponding to _m e _m of the target signal subspace spanned by the steering vectors a (θ _k) are orthogonal, e _m ^H a ( θ _k ) = 0. That is, as shown below, the beam output y _m (n) is a combination of only receiver noise.

  Here, coherent integration has better integration efficiency than non-coherent integration. For this reason, when the target signals are represented by different Doppler frequency signals, the difference between the coherent integral value due to the target signal and the coherent integral value due to noise is obtained by obtaining the integral value when the eigenbeam output is coherently integrated. Become clearer. Therefore, by estimating the target number using this coherent integration value, the target number estimation accuracy at a low SNR can be improved.

  Therefore, the coherent integration means 16 performs coherent integration on the m-th eigenbeam output as follows.

Here, L is the number of Doppler frequency bins, and l (l = 0,..., L-1) is a Doppler frequency bin number.
Here, if the difference between the Doppler frequencies of the k waves is equal to or less than the Doppler frequency resolution and exists in the same Doppler bin, the signal component is integrated into at most one Doppler bin of the coherent integration value Z _m (l).

Therefore, the signal component in the integration value extracting means 17 as larger than the noise components, by any of the means shown later, to extract particular the integrated value from the coherent integration value Z _m (l).

FIG. 2 is a block diagram of an embodiment of the integrated value extracting means 17.
The first maximum value extraction means 18 in FIG. 2 receives the coherent integral value Z _m (l) and outputs the maximum value Z _max (m) as shown in the following equation.

FIG. 3 is a block diagram of another embodiment of the integrated value extracting means 17.
The moving average means 19 of FIG. 3 receives the coherent integral value Z _m (l) as an input, and outputs the moving average value Z _m-MA (l) as shown in the following equation (21).

The second maximum value extraction means 20 receives the moving average value Z _m-MA (l) as input, and outputs the maximum value Z _max (m) as shown in the following equation (22).

The wave number estimation means 21 applies the AIC or MDL to the maximum value Z _max (m) output from the integral extraction means 17 by rearranging the threshold test or the maximum value Z _max (m) in descending order. Estimate the target number.

  As described above, in the wave number estimating apparatus according to the present embodiment, the correlation matrix calculating unit 13, the eigenvalue / eigenvector calculating unit 14, the eigen beam forming unit 15, the coherent integrating unit 16, the integral value extracting unit 17, and the wave number estimating unit 21. The effect of improving the wave number estimation performance at a low SNR is added.

Embodiment 2. FIG.
FIG. 4 is a block diagram showing Embodiment 2 of the wave number estimating apparatus of the present invention incorporated in a radar system.
The second embodiment is different from the first embodiment in the configuration of the wave number estimating means 21, and the other configurations are the same as those in the first embodiment. Therefore, only the wave number estimating means 21 of the second embodiment will be described.

In the wave number estimating means 21 in the second embodiment, Z _max (m) which is the output of the integral extracting means 17 is rearranged in descending order, and the outputs λ ₁ ,..., Λ _k ,. .. Estimating the target number by applying AIC or MDL to the value obtained by adding λ _M

  As described above, in the wave number estimating apparatus according to the present embodiment, the correlation matrix calculating unit 13, the eigenvalue / eigenvector calculating unit 14, the eigen beam forming unit 15, the coherent integrating unit 16, the integral value extracting unit 17, and the wave number estimating unit 21. The effect of improving the wave number estimation performance at a low SNR is added.

Embodiment 3 FIG.
FIG. 5 is a block diagram showing Embodiment 3 of the wave number estimating apparatus of the present invention incorporated in a radar system.
In the figure, the configuration excluding the angle measuring means 22 is the same as that of the first embodiment. Therefore, only the angle measuring means 22 of the present embodiment will be described.

The angle measurement means 22 measures the angle according to the wave number estimation value of the wave number estimation means, and outputs the angle measurement value. As the angle measurement algorithm, there are a monopulse angle measurement method or a super-resolution angle measurement method represented by MUSIC (Multiple Signal Classification) and MLE (Maximum Likelihood Estimator).
For example, when the wave number estimation value is 1, a highly accurate angle measurement value can be obtained with a smaller amount of calculation than the super-resolution angle measurement method by performing the monopulse angle measurement method. When there are a plurality of wave number estimation values, a plurality of highly accurate angle measurement values that match the wave number estimation values can be obtained by using the super-resolution angle measurement method.
Of course, when there is no restriction on the amount of calculation, even if the wave number estimation value is 1, the angle measurement value can be obtained using the super-resolution angle measurement method.

  As described above, in the wave number estimating apparatus according to the present embodiment, the correlation matrix calculating unit 13, the eigenvalue / eigenvector calculating unit 14, the eigen beam forming unit 15, the coherent integrating unit 16, the integral value extracting unit 17, and the wave number estimating unit 21. By adding the angle measurement means 22, angle measurement can be performed based on the wave number estimation value with improved wave number estimation performance at a low SNR, and the angle measurement performance can be improved.

Embodiment 4 FIG.
FIG. 6 is a block diagram showing Embodiment 4 of the wave number estimating apparatus of the present invention incorporated in a radar system.
In the figure, the configuration excluding the angle measuring means 22 is the same as that of the second embodiment. Further, the function and performance of the angle measuring means 22 are as described in the third embodiment.

  As described above, in the wave number estimating apparatus according to the present embodiment, the correlation matrix calculating unit 13, the eigenvalue / eigenvector calculating unit 14, the eigen beam forming unit 15, the coherent integrating unit 16, the integral value extracting unit 17, and the wave number estimating unit 21. By adding the angle measurement means 22, angle measurement can be performed based on the wave number estimation value with improved wave number estimation performance at a low SNR, and the angle measurement performance can be improved.

  The wave number estimation apparatus according to the present invention relates to estimating a target number included in a received signal in a sensor array system that transmits and receives a plurality of pulses, and is used for target number estimation by a radar or a sonar.

It is a block diagram which shows Embodiment 1 of the wave number estimation apparatus of this invention. It is a block diagram which shows one Example of the integral value extraction means in Embodiment 1 of this invention. It is a block diagram which shows the other Example of the integral value extraction means in Embodiment 1 of this invention. It is a block diagram which shows Embodiment 2 of the wave number estimation apparatus of this invention. It is a block diagram which shows Embodiment 3 of the wave number estimation apparatus of this invention. It is a block diagram which shows Embodiment 4 of the wave number estimation apparatus of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1; Transmitter, 2; Circulator, 3; Phase shift value calculation means, 4; Phase shifter, 5; Antenna, 6; Receiver, 7; AD converter, 8: Clutter suppression means, 9: Interference wave suppression means 10; sum beam forming means, 11; inter-pulse integrating means, 12; signal detecting means, 13; correlation matrix calculating means, 14; eigenvalue / eigenvector calculating means, 15; eigen beam forming means, 16; coherent integrating means, 17 Integrated value extraction means, 18; maximum value extraction means, 19; moving average means, 20; second maximum value extraction means, 21; wave number estimation means, 22;

Claims (5)

  A target signal acquisition unit that transmits and receives a plurality of pulse waves by a plurality of sensors to acquire a reflected signal from a target, a sum beam forming unit that forms a sum beam with respect to an output of the target signal acquisition unit, and the sum Inter-pulse integration means for integrating the output of the beam, amplitude detection of the output of the inter-pulse integration means, signal detection means for detecting the output signal after amplitude detection, and signal detection by the signal detection means A correlation matrix calculation means for calculating a correlation matrix of the target signal from an output of the target signal acquisition means, an eigenvalue / eigenvector calculation means for calculating an eigenvalue / eigenvector from the correlation matrix, the eigenvector and the target An eigen beam forming unit that performs eigen beam formation from the output of the signal acquisition unit, and a coherent integration of each beam output by the eigen beam forming unit -Integral means, integral value extracting means for extracting a predetermined coherent integral value from the coherent integration value, and wave number estimating means for estimating the wave number of the target signal from the output of the integral value extracting means. A characteristic wave number estimation device.   The wave number estimation unit according to claim 1, wherein the wave number estimation unit is configured to estimate the wave number of the target signal using the output of the eigenvalue / eigenvector calculation unit in addition to the output of the integral value extraction unit. apparatus.   3. The wave number estimating apparatus according to claim 1, further comprising angle measuring means for obtaining an arrival angle of the target signal based on a wave number estimation result obtained by the wave number estimating means.   The wave number estimation device according to any one of claims 1 to 3, wherein the integral value extracting means is constituted by first maximum value extracting means for extracting the maximum value of the coherent integral value. .   The integral value extracting means comprises a moving average means for performing a moving average of the coherent integral value, and a second maximum value extracting means for extracting the maximum value of the output of the moving average means. The wave number estimation apparatus according to any one of claims 1 to 3.

Images