JP2010038744A - Radar system for target identification - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve a target identification performance by utilizing a polarization wave characteristic of a clutter and an artifact, in a radar system for target identification which conducts identification of an observed target. <P>SOLUTION: The radar system for target identification includes a transmitter which generates and transmits a wideband pulse, a receiver which receives a plurality of polarization wave signals by using a plurality of antennas different in the polarization wave characteristic from each other and outputs a scattering matrix range profile showing the amplitude and the phase of the polarization wave signal, a background identifying part which calculates a polarization wave feature of a background area of the target by using the scattering matrix range profile, a reference cell selection calculating part which selects an optimum filter, based on the polarization feature, a target detecting part which suppresses the clutter by the filter and detects the target, a feature amount extracting part which calculates a plurality of feature amounts by using the scattering matrix range profile near the detected target, and a target identifying part which conducts the identification of the target by using the feature amounts. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a radar apparatus for identifying a target using information on a target range profile and polarization in the field of radar.

  As a conventional technique related to a radar apparatus for performing target identification, there is a technique for identifying a target using a target range profile observed in a radar apparatus having a high range resolution. A library pre-stored as a reference range profile and means for calculating the correlation between the observed target range profile and the reference range profile stored in the library, and a reference highly correlated with the observed target range profile In the case where there is a range profile, it is disclosed that an observed target is identified as a target corresponding to the reference range profile (see Non-Patent Document 1, for example).

“Correlation Filters for Aircraft Identification From Radar Range Profiles”, IEEE Transactions on Aerospace and Electronic Systems, vol.29, no.3, pp.741-748, July 1993

  The conventional target identification radar apparatus is configured as described above. However, when there is a building in the vicinity of the target, or when the clutter around the target has a complicated shape, When the frequency band with a short wavelength such as the millimeter wave band is used, there is a problem that the target is erroneously identified.

  The present invention has been made to solve the above-described problems. When there is a building in the vicinity of the target, the clutter around the target has a complicated shape, or the clutter state around the target (according to wind) A radar system for target identification that can improve target identification performance even when a frequency band with a short wavelength relative to the target size, such as a millimeter wave band, is used when the grassland fluctuations are likely to change. The purpose is to obtain.

  A radar device for target identification according to the present invention receives a plurality of polarization signals using a transmitter that generates and transmits a wideband pulse and a plurality of antennas having different polarization characteristics from each other. A receiver that outputs a scattering matrix range profile indicating the amplitude and phase of the signal, a background identification unit that calculates the polarization characteristics of the target background location using the scattering matrix range profile, and an optimum based on the polarization characteristics A reference cell selection calculation unit that selects a filter, a clutter is suppressed by the filter, a target detection unit that detects the target, and a plurality of feature amounts are calculated using the detected scattering matrix range profile near the target. A feature amount extraction unit and a target identification unit that identifies a target using the feature amount are provided.

  According to this invention, the clutter generated based on the polarization signal and the filter that suppresses the main component of the clutter according to the polarization characteristics of the artifact are constructed, the signal to be handled is selected, and the range of each polarization channel is selected. By calculating multiple types of indicators that indicate the rough shape of the profile and multiple types of indicators that indicate the relationship between the range profiles of multiple polarization channels, the target and artifacts are identified by combining features. This has the effect of improving the probability of correctly detecting and identifying the.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a configuration of a target identifying radar apparatus according to Embodiment 1 of the present invention. Since the polarization characteristics of clutter vary greatly depending on the shape of the clutter and the state of time conversion, the polarization characteristics observed at every aspect angle and every elevation angle (radio wave incident angle) for every possible clutter environment. Must be prepared as reference data, and all of these must be stored in the candidate target polarization characteristics library.
On the other hand, because the target range profile depends greatly on the target aspect angle (the angle between the target center line and the radar line-of-sight direction), the range profile observed at every aspect angle is prepared as reference data for each candidate target. All of these must be stored in a library of candidate target range profiles.

  However, this method requires that reference data be created and stored for each candidate clutter, artifact, and target by changing the elevation angle and aspect angle in very fine increments. There arises a problem that a library of clutter, artifacts and target polarization characteristics becomes enormous. In order to make the candidate clutter, artifact, and target polarization characteristics library as small as possible, it is desirable to increase the increments of the elevation angle and aspect angle stored as reference data. Is a problem that the identification performance of artifacts and targets deteriorates when the polarization characteristics of candidate clutters, artifacts, and targets do not include data that completely matches the elevation angle and aspect angle at the time of observation. Will occur. Therefore, in this embodiment, the problem is solved by performing target identification by combining feature amounts.

  Therefore, focusing on the polarization characteristics depending on the elevation angle with respect to the target, clutter, etc., in addition to the three-component decomposition method, the three-component scattering model decomposition method is used to increase the ability to classify clutter, artifacts, and targets.

For example, it is possible to grasp the polarization characteristics of clutter and artifacts in an urban area, construct a filter that suppresses the clutter principal component, and construct a target and artifact classification filter. Filter selection becomes appropriate, such as a city filter, a grassland filter, a desert filter, and an asphalt filter, and the target detection probability is improved.
In the evaluation of the polarization characteristics in urban areas, the contribution ratio of the odd-numbered reflection component and the even-numbered reflection component is almost constant at 50% regardless of the elevation angle in the evaluation by the three-component decomposition method. However, when the three-component scattering model decomposition method is used, the two-time reflection component tends to increase as the elevation angle increases (closer to the target). Utilizing these, classify buildings in various clutters, classify targets and buildings, construct a filter suitable for suburban clutter subtraction, and improve target extraction classification ability.
For target classification by the three-component decomposition method, it is desirable to extract a feature quantity with low aspect angle sensitivity. Therefore, when categorizing the target and the building group, the feature quantity other than the standard deviation of the moment of reflection of the odd-numbered reflection component and the even-numbered reflection component contribution rate, and the multiple reflection component contribution rate, which is considered effective for each target category. Comparison is effective.

Hereinafter, FIG. 1 will be described.
101 is a transmitter for transmitting a pulse signal, 102 is a transmission / reception switch for switching transmission / reception, 103 is a polarization switch, and 104 is a first polarization characteristic orthogonal to the polarization characteristic of the second polarization transmission / reception antenna 105. A single polarization transmission / reception antenna 105 is a second polarization transmission / reception antenna having a polarization characteristic orthogonal to the polarization characteristic of the first polarization transmission / reception antenna 104.

  Note that as the combinations in which the polarization characteristics of the first polarization transmitting / receiving antenna 104 and the second polarization transmitting / receiving antenna 105 are orthogonal, for example, a combination of vertical polarization and horizontal polarization, right-handed circular polarization and left-handed circular polarization, and the like. A combination of waves can be considered.

  A digital reception unit 106 performs phase detection and A / D conversion processing on the reception signals received by the first polarization transmission / reception antenna 104 and the second polarization transmission / reception antenna 105, and indicates the amplitude and phase of the polarization signal. This is a receiver that outputs a polarization signal (hereinafter referred to as a scattering matrix range profile, which will be described later).

  107 is a received polarization signal storage unit that accumulates the scattering matrix range profile output from the receiver 106, and 108 is a background identification unit that identifies clutters and artifacts from the scattering matrix range profile sent from the received polarization signal storage unit. , 109 is a reference cell selection for selecting a reference cell and a target cell for calculating a filter coefficient for suppressing the clutter principal component based on a scattering matrix range profile for each range cell sent from the clutter and the artifact identification unit. The calculation unit 110 applies a filter to the target cell (target) sent from the reference cell selection calculation unit, suppresses clutter, detects the target, and cuts out the scattering matrix range profile near the detected target position. The target detection unit 111 includes a plurality of feature amounts from the scattering matrix range profile cut out by the target detection unit 110. A target identifying unit that identifies a target by calculating and comparing the obtained feature amount with a library of feature amounts relating to known targets accumulated in advance, and 112 displays a detection result and an identification result of the target It is a display unit.

  Reference numeral 108a denotes a polarization basis conversion processing unit that performs polarization basis conversion processing on the scattering matrix range profile sent from the received polarization signal storage unit 107 to generate a range profile of a plurality of polarization channels, and 108b denotes 108a. 108c is a three-component decomposition method and three-component scattering model decomposition obtained by 108b, which calculates a plurality of feature amounts used to identify clutter and artifacts from the plurality of polarization channel range profiles obtained by This is a scattering contribution rate calculation unit that calculates the contribution rate of the odd-numbered reflection component and the even-numbered reflection component based on the characteristic amount by the method.

  Note that 111a is a polarization basis conversion processing unit that performs polarization basis conversion processing on the scattering matrix range profile sent from the target detection unit 110 to generate a range profile of a plurality of polarization channels, and 111b is obtained by 111a. 111c is a feature quantity extraction unit that includes a polarization basis conversion processing unit 111a and a feature quantity calculation unit 111b, and 111d is known. The reference target feature amount library 111e that stores feature amounts related to the target of the target is determined in advance from the feature amounts extracted by the feature amount extraction unit 111c and the feature amounts stored in the reference target feature amount library 111d. It is a judgment part that performs target identification according to established rules.

Next, the operation will be described.
When the transmitter 101 generates a pulse signal as a broadband pulse, the transmission / reception switch 102 sends the pulse signal to the polarization switch 103, and the polarization switch 103 drives the first polarization transmitting / receiving antenna 104, The pulse signal is radiated from the first polarization transmitting / receiving antenna 104 to the space. Hereinafter, a case will be described in which the polarization characteristics of the first polarization transmitting / receiving antenna 104 and the second polarization transmitting / receiving antenna 105 are horizontal polarization and vertical polarization, respectively.

The pulse signal radiated into the space from the first polarization transmitting / receiving antenna 104 is scattered by the observation target.
The polarization switching unit 103 drives both the first polarization transmission / reception antenna 104 and the second polarization transmission / reception antenna 105, so that the first polarization transmission / reception antenna 104 and the second polarization transmission / reception antenna 105 are scattered by the observation target. When the scattered waves are received, the first polarization transmitting / receiving antenna 104 and the second polarization transmitting / receiving antenna 105 send the received signal to the receiver 106, and the receiver 106 receives the first polarization transmitting / receiving antenna 104 and the second polarization. Digital detection signals S ₁₁ (m) and S ₂₁ (m) indicating the amplitude and phase of each received signal by performing phase detection processing and A / D conversion processing on each received signal received by the transmitting / receiving antenna 105. Is output.

S _ij (m) is the m-th (m = 1, 2,..., M) sample value of the received signal transmitted by the j-th polarization transmitting / receiving antenna and received by the i-th polarization transmitting / receiving antenna. It is. Here, M is the number of samples. Similarly, the broadband pulse generated by the transmitter 101 is sent to the polarization switching unit 103 via the transmission / reception switch 102, and the target is irradiated from the second polarization transmitting / receiving antenna 105 to the first polarization transmitting / receiving antenna 105. And the received signal S ₁₂ (m), S ₂₂ (m) are obtained by repeating the same processing as the received signal received by the second polarization transmitting / receiving antenna 106.

The received polarization signal storage unit 107 temporarily stores the obtained received signals S ₁₁ (m), S ₂₁ (m), S ₁₂ (m), and S ₂₂ (m). Is a scattering matrix range profile having a value of the scattering matrix Sm (m = 1, 2,..., M) in each resolution cell, and is expressed as in equation (1).
Hereinafter, the combination of the polarization states of the transmission antenna and the reception antenna may be referred to as a polarization channel.

The basis conversion processing unit 108a applies an expression to the scattering matrix range profile Sm (m = M ₀ , M ₀ +1,..., M ₀ + M _T −1) accumulated in the received polarization signal storage unit 107. Polarization component decomposition is performed using the three-component decomposition method and static component scattering model decomposition method exemplified in (2).

For example, using a polarization component resolution represented by the formula (2), of the received wave, an odd number of times reflection component reaches the receiving antenna after being reflected an odd number of times in the target K _o (m), the even number of reflections in the target After that, the even-numbered reflection component K _e (m) reaching the transmitting / receiving antenna and the polarization component (cross component) K _c (m) orthogonal to the incident wave generated when reflected by the target are obtained.

As described above, it is possible to obtain a range profile for any type of polarization channel by performing various polarization basis conversions and polarization component decomposition.
The set of range profiles obtained here is hereinafter referred to as clutter and artifact multiple polarization channel range profiles. Here, the “polarization channel” includes not only a combination of transmission polarization and reception polarization but also polarization components such as K _o , K _e , and K _c (the same applies hereinafter).
Next, the feature quantity calculation unit 108b calculates the feature quantity defined below from the multiple polarization channel range profile obtained by the above-described base conversion process.

  Next, the scattering contribution rate calculation unit 108c calculates the composition ratio of the odd-numbered reflection component, the even-numbered reflection component, the multiple reflection component, and the like based on the feature amount calculated by the three-component decomposition method and the three-component scattering model resolution method. Analyze and classify clutter as well as artifacts and targets.

  Next, the reference cell selection calculation unit 109 calculates and determines a cell of interest based on the result of classifying the artifact and the target from the scattering contribution rate calculation unit, and uses it to construct a filter for suppressing clutter. The reference cell to be selected is selected and determined.

Next, the target detection unit 110 detects a target from the scattering matrix range profile output from the reference cell selection calculation unit. The target detection unit 110 cuts out a sample Sm (m = M ₀ , M ₀ +1,..., M ₀ + M _T −1) near the detected target position.

The basis conversion processing unit 111a applies an expression (for the target scattering matrix range profile Sm (m = M ₀ , M ₀ +1,..., M ₀ + M _T −1) cut out by the target detection unit 110). The polarization component decomposition exemplified in 2) is performed.

  Next, the feature quantity calculation unit 111b calculates the feature quantity defined below from the multiple polarization channel range profile obtained by the above-described base conversion process.

In Equation (3), P (m) represents the power value of a signal of a certain polarization channel. For example, a polarization channel for horizontal polarization transmission and horizontal polarization reception is defined as in Equation (4). Is done. r _g is the center of gravity is defined as Equation (5). Xp represents the Fourier spectrum of the range profile of the polarization channel p.

  In Equation (3), the second-order and third-order cross moment spectra are exemplified by the polarization channels 1 and 2 and the polarization channels 1, 2, and 3, respectively. It is possible to calculate for all combinations of polarization channels.

FIG. 2 is an explanatory diagram showing definitions of the average, standard deviation, maximum value, and moment around the center of gravity among the feature amounts calculated by the feature amount calculation unit.
The feature amount is calculated by the feature amount calculation unit 111b, the average value is the average reflection intensity of each polarization channel, the maximum value is the maximum reflection intensity, the standard deviation is the complexity of the range profile shape of each polarization channel, and the center of gravity It can be seen that the surrounding moment is an index mainly representing the length of the range profile.

  Furthermore, from equation (3), the normalized moment around the center of gravity and bispectrum show the distribution of points with strong scattering intensity on the range profile, and the cross moment spectrum shows the relationship between the range profiles of multiple polarization channels. It can be said that each is an indicator. Summarizing the above, the feature quantity shown in Equation (3) can be broadly divided into an index indicating the rough shape of the range profile of each polarization channel and an index indicating the relationship between the range profiles of a plurality of polarization channels.

  In the reference target feature amount library 111d, feature amounts extracted from the scattering matrix range profiles of N types of targets observed in advance in the same manner are stored. Smn can be obtained from theoretical calculations (for example, GTD: Geometrical Theory of Diffraction) in addition to prior observations.

  The determination unit 111e performs comparison with feature amounts related to N types of targets stored in the reference target feature amount library 111d, and identifies a target according to a predetermined rule.

For example, when the number of feature quantities used for identification is L, L feature quantities relating to each target are accumulated in the reference target feature quantity library 111d in the form of L-dimensional feature quantity vectors and extracted from the received signal. After constructing an L-dimensional feature quantity vector from the L feature quantities, the Euclidean distance between the L-dimensional feature quantity vector of the received signal and the feature quantity vector of each reference target is calculated, and the reference distance is the shortest. A method of determining that the type is the same as the target is conceivable.
Finally, the display unit 112 displays the outputs of the target detection unit 110 and the determination unit 111e.

  According to the first embodiment, the polarization information processing method is further increased for clutters and artifacts, the feature amount is grasped, and the artifact and the target are identified based on the scattering contribution rate. The target profile is improved by being able to scrutinize the range profile used to construct the clutter suppression filter required for detection, and multiple types of indices indicating the relationship between the range profiles of multiple polarization channels are calculated. Since the target identification is configured by combining the two, the probability of correctly identifying the target can be improved and the fluctuation of the target due to the movement of the target in the distance direction and the change in the aspect angle can be determined from the scattering matrix range profile. By using multiple types of indicators that indicate the rough shape of the range profile of each polarization channel as a small feature quantity, Li amount, there is an effect that the calculation processing amount can be reduced.

Embodiment 2. FIG.
FIG. 3 is a block diagram showing the configuration of a target identifying radar apparatus according to Embodiment 2 of the present invention. 108d is a reference background feature quantity library for storing feature quantities relating to known clutters and artifacts, and 108e is a reference background feature quantity library. The spatially and temporally varying characteristics of the received polarization profile for each cell having the clutter and artifact features calculated by the scattering contribution rate calculation unit 108c are stored in advance in the reference background feature library 108d. A comparison is made in accordance with a predetermined library of known clutters and artifacts and a predetermined rule, and what kind of clutter is output to the scattering contribution rate calculation unit 108c. Based on the result of the characteristic change evaluation unit 108d, the scattering contribution rate calculation unit 108c outputs a received polarization profile for selecting a reference cell and a target cell to the reference cell selection calculation unit 109.

  In general, since the clutter varies spatially and temporally for each range cell, it can be expected that the characteristic amount of the polarization characteristic of the clutter is different from the reference background characteristic amount acquired in advance. Accordingly, the clutter suppression filter is further constructed by sequentially comparing the result of the scattering contribution rate calculation unit 108c with the feature amount data of the reference background feature amount library 108d, evaluating the characteristic change, and identifying whether the target is the clutter or the target. It is possible to scrutinize the reference cells required for

  According to the second embodiment, a clutter suppression filter suitable for a grassland or the like in which fluctuations occur due to the influence of wind is selected, or an asphalt filter that does not change in space and time at all, and fine particles such as sand fluctuate. By detecting the filter, it is possible to set a threshold with improved accuracy according to the clutter to improve the probability of correctly detecting and identifying the target, such as a filter suitable for sandy areas such as coastal areas, and the influence of clutter This has the effect of reducing the probability of misidentification of the target.

It is a block diagram which shows the structure of the radar apparatus for target identification by Embodiment 1 of this invention. It is explanatory drawing which shows the definition of an average, a standard deviation, a maximum value, and the moment around a gravity center among the feature-values calculated in a feature-value calculation part. It is a block diagram which shows the structure of the radar apparatus for target identification by Embodiment 2 of this invention.

Explanation of symbols

  101 transmitter, 102 transmission / reception switch, 103 polarization switch, 104 first polarization transmission / reception antenna, 105 second polarization transmission / reception antenna, 106 receiver, 107 reception polarization signal storage unit, 108 background identification unit, 108a base Conversion processing unit, 108b feature amount calculation unit, 108c scattering contribution rate calculation unit, 108d reference target feature amount library, 108e characteristic change evaluation unit, 109 reference cell selection calculation unit, 110 target detection unit, 111 target identification unit, 111a base Conversion processing unit, 111b feature quantity calculation unit, 111c feature quantity extraction unit, 111d reference feature quantity library for reference, 111e determination unit, 112 display unit

Claims (3)

A transmitter that generates and transmits wideband pulses; and
A receiver that receives a plurality of polarization signals using a plurality of antennas having different polarization characteristics from each other, and outputs a scattering matrix range profile indicating the amplitude and phase of the polarization signal;
Using the above scattering matrix range profile, a background identifying unit that calculates the polarization characteristics of the target background location,
A reference cell selection calculator that selects an optimal filter based on the polarization characteristics;
A target detection unit that suppresses clutter by the filter and detects the target;
A feature amount extraction unit that calculates a plurality of feature amounts using the detected scattering matrix range profile in the vicinity of the target;
A target identifying unit for identifying a target using the feature amount;
A target identification radar apparatus comprising: The target identification radar apparatus according to claim 1, wherein the background identification unit calculates a plurality of polarization channel signals from a plurality of received polarization signals received, and calculates a background feature amount. 3. The background identifying unit categorizes the types of the clutter, the artifact, and the target by comparing the polarization characteristics depending on the incident angle of the radio wave with respect to the clutter, the artifact, and the target. Target identification radar device.

Images