JP3664816B2 - Clutter suppression radar device - Google Patents

Description

【００２６】
【数６】

【００２７】
【数７】

【００２８】
ここでＮは入力データ数、ｍを近似関数の最大次数とする。マトリックスＨの行数はＮとなる。あらかじめＨ^T ・Ｈを計算しておき、入力データ行列ｙを得て、Ｈ^T ・ｙを計算し、数式８の解ベクトルｕを多元連立方程式の解を反復法で計算する方法によって計算する。
【００２９】
【数８】
（Ｈ^T Ｈ）・ｕ＝Ｈ^T ・ｙ
【００３０】
ここで、Ｈ^T Ｈの計算結果はＮの値に関わらず、ｍ行ｍ列のマトリックスとなっている。Ｈ^T ・ｙもｍ行のベクトルとなっている。数式８を計算し、数式５のｕ₀、ｕ₁、ｕ₂、……、ｕ_m が得られる。
ここで、（Ｈ^T Ｈ）・ｕ＝Ｈ^T ・ｙの解ベクトルｕを反復法で計算するかわりに数式９と数式８を変形し、掃き出し法等で逆行列（Ｈ^T Ｈ）^-1を計算する方法も考えられる。Ｑチャネルの計算も全く同様となる。
【００３１】
【数９】
ｕ＝（Ｈ^T Ｈ）^-1Ｈ^T ｙ
【００３２】
こうして得られた数式３は、受信データ列ｙ_i に基づくトレンド曲線を表すことになり、これが時間軸におけるクラッタ波形を表すことになる。こうして数式３ができたならば、次は第２の演算手段５において、数式３を用いて、ｔが０、Ｔ、２Ｔ、……、ＮＴの各値の時のｕ（ｉＴ）、即ちｕ（０）、ｕ（Ｔ）、ｕ（２Ｔ）、……、ｕ（ＮＴ）を算出する。こうして得られた算出結果列ｕ（ｉＴ）はＩチャネルの第３のメモリ６へ一旦格納される。
【００３３】
次に、この格納された算出結果列ｕ（ｉＴ）はＩチャネル減算器７へ入力される。この他Ｉチャネル減算器７へは第２のメモリ２からデータ列ｙ_i が入力される。そして、ここでｙ_i の各データから対応する算出結果列ｕ（ｉＴ）を減算する。この減算により受信データからクラッタ波形が除去されたことになる。こうしてクラッタ波形が除去されたデータは複素演算器１５へ送られる。
【００３４】
以上はＩチャネルについて述べて来たが、Ｑチャネルについても全く同様の動作が行われクラッタ波形の除去されたデータが複素演算器１５へ送られる。複素演算器１５では、２つの入力信号から移動目標の振幅情報を取り出したり、ドップラー周波数スペクトラムを計算して指示器２５へ送る。複素演算器１５へ入力されるＩチャネルのデータは複素表示の実部と見ることができ、Ｑチャネルのデータは虚部と見ることができるので、両チャネル入力の２乗和を求めればそれは信号の電力振幅となるし、２乗和の平方根を求めればそれは目標信号ベクトルの電圧振幅を表すことになる。従ってＰＰＩ表示の信号として用いることができる。
【００３５】
また、ＩチャネルデータをＦＦＴ(Fast Fourier Transform)回路の実部入力データとし、ＱチャネルデータをＦＦＴ回路の虚部入力データとしてＦＦＴを行わせることによりドップラー周波数スペクトラムを得ることができる。従って指示器２５に周波数スペクトラムの表示機能があれば、これを表示させ観測することができる。
以上の説明は方位方向に連なる入力データの場合であるが、レンジ方向に連続したデータの場合も同様にして実施できる。
【００３６】
【発明の効果】
以上説明したように、本発明のレーダ装置では、周波数ゼロの近辺に見られるクラッタスペクトルは、時間軸波形では、ゼロボルトからオフセットされた低周波の波形（トレンド波形）に対応するものであることに着眼し、受信レーダビデオの時間軸波形からトレンド波形を減算することによりクラッタスペクトルを無くするようにしたので、クラッタスペクトルの広がりが大きくなっても、従来の周波数特性が固定的であるＭＴＩの消去器に較べて、クラッタ抑圧能力が劣化しない即ち消え残りが少ないという利点がある。
【図面の簡単な説明】
【図１】本発明のクラッタ抑圧レーダ装置の発明主要部の構成を示すブロック図である。
【図２】従来のディジタルＭＴＩレーダ装置の構成を示すブロック図である。
【図３】従来のディジタルＭＴＩの単消去器の回路構成とその周波数特性を示す図である。
【図４】従来のディジタルＭＴＩの二重消去器の回路構成とその周波数特性を示す図である。
【符号の説明】
１ 第１のメモリ
２ 第２のメモリ
３ トレンド計算回路
４ 第１の演算手段
５ 第２の演算手段
６ 第３のメモリ
７ Ｉチャネル減算器
８ 第１のメモリ
９ 第２のメモリ
１０ トレンド計算回路
１１ 第１の演算手段
１２ 第２の演算手段
１３ 第３のメモリ
１４ Ｑチャネル減算器
１５ 複素演算器
１６ 空中線
１７ 送受切替器
１８ 送信機
１９ 受信機
２０ ＡＤ変換器
２１ ＡＤ変換器
２２ 消去器
２３ 消去器
２４ 合成器
２５ 指示器[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a radar receiver for detecting a moving target such as an aircraft or a weather echo, and in particular, an input signal composed of a signal waveform, a clutter waveform that is sufficiently low in frequency with respect to the signal frequency, and a general noise waveform. Belongs to a clutter removal circuit for removing a clutter waveform from
[0002]
[Prior art]
When detecting a moving object such as an aircraft, it is necessary to suppress a reflected wave from a fixed object such as a ground clutter. The Doppler frequency of a fixed object can be considered almost zero, and its frequency spectrum is almost in the vicinity of zero. When the clutter spectrum is caused by a reflected wave from a cloud moving in the wind, the clutter spectrum is wider than a clutter spectrum caused by a reflected wave from rain, waves, or the like. A technique related to clutter removal is the MTI technique.
[0003]
FIG. 2 is a block diagram showing the configuration of a conventional digital MTI radar apparatus.
Two types of video signals (generally called Inphase I channel and Qvandrature phase Q channel) differ in phase by 90 degrees in the local oscillation signal mixed with the received signal in the receiver 19 and converted to digital signals respectively. Then, it is sent to the erasers 22 and 23.
[0004]
The erasers 22 and 23 are linear digital filters having a constant coefficient for suppressing a reflected wave having a Doppler frequency near zero, and the configuration thereof is as shown in FIG. 3 (a) or FIG. 4 (a). A delay circuit or FIFO memory and a subtracting circuit are used to subtract the received signal of the previous cycle and the received signal of the next cycle to eliminate the clutter and output the moving target.
[0005]
The frequency characteristic H _s (f) of the configuration shown in FIG. 3A is Equation 1, and this is shown in FIG. 3B. Also, the frequency characteristic H _d (f) of the configuration of FIG. 4A is Equation 2, which is shown in FIG. 4B.
[0006]
[Expression 1]
H _s (f) = | sinπfT |
However, T is the same delay time as the transmission repetition period.
[Expression 2]
H _d (f) = sin ² πfT
[0008]
In either case, since it becomes zero when πfT = nπ (where n is 0, 1, 2, 3,...), It becomes zero at the point of f = n / T. That is, it becomes zero when the target Doppler frequency is zero, 1 / T, 2 / T, 3 / T,.
Accordingly, since the output of the aircraft flying in the radial direction centering on the radar apparatus at a speed that generates such a Doppler frequency is also zero, such a speed is called a blind (blind) speed.
[0009]
By the way, when viewed on the frequency axis, the frequency spectrum of the received signal reflected back from a complete fixed target for continuous transmission of period T is 0, 1 / T, 2 / T, 3 / T,. Stand in the position. Accordingly, when a digital filter having frequency characteristics as shown in FIGS. 3B and 4B is passed, the output becomes zero. That is, it is suppressed (erased). Therefore, reflection from a fixed target such as a ground clutter is suppressed, and a moving target such as an aircraft that is not blind is output and detected.
[0010]
[Problems to be solved by the invention]
However, the conventional MTI is a filter whose frequency characteristics are fixed so that the blind speed becomes an appropriate value.
On the other hand, the so-called ground clutter that is reflected and returned from a target that is actually said to be a fixed target such as land is that the aerial is rotating, the vegetation is swaying in the wind, etc. For this reason, it is not a complete fixed target, and its frequency spectrum is not a line spectrum of 0, 1 / T, 2 / T, 3 / T,. Then, as is clear from FIG. 3B and FIG. 4B, the output of the eraser does not become 0 but is output as “unerased”.
[0011]
Other clutters, such as clouds, rain, and sea surface, are originally not moving completely but moving, so their Doppler frequencies deviate from 0, 1 / T, 2 / T, 3 / T, etc. The extent of the deviation and the extent of the spread are not fixed but fluctuate.
[0012]
On the other hand, the frequency characteristics of the conventional MTI filter are fixed, and it is impossible to change the frequency at which the output becomes zero or to widen the width of the portion close to zero. For this reason, a complete fixed component (also referred to as a DC component) can be suppressed and removed. However, there is a problem that the “clutter” of the clutter occurs for the clutter having a broad spectrum.
[0013]
On the other hand, the clutter spectrum seen near the zero frequency corresponds to the low frequency waveform offset from zero volts (this is called the trend waveform) in the time axis waveform, so if the trend is removed from the time axis waveform, A method of focusing on the fact that the clutter spectrum is also eliminated (that is, the trend waveform and the clutter waveform are the same) can be considered.
[0014]
In order to solve the above-described problems, an object of the present invention is to calculate a clutter waveform by calculating a trend every time input data is input on the time axis of a radar reception signal, and then calculate the clutter waveform. An object of the present invention is to provide a clutter suppression radar apparatus in which the suppression removal characteristic does not deteriorate even if the spread of the clutter spectrum changes by subtracting the clutter waveform data sequence from the data sequence.
[0015]
[Means for Solving the Problems]
In order to achieve the above object, the first means configuration of the clutter suppression radar apparatus of the present invention includes the following components (a) to (h).
(A) Radar receiver circuit having two systems (I channel and Q channel) of received outputs whose phase of the local oscillation signal mixed with the received signal differs by 90 degrees (b) Each of the I channel and Q channel of the radar receiver circuit output The first I-channel memory and the Q-channel first memory (C) that store the received output for each transmission azimuth individually for each range bin that is subdivided in the distance direction are the same in the I-channel first memory. An I channel second memory for extracting and storing N data strings y _i (i = 1, 2, 3,..., N and so on) continuous in the azimuth direction at a distance, and the Q channel first memory Q channel second memory for extracting and storing N data strings z _i continuous in the azimuth direction at the same distance (d) When the radar transmission cycle is T, a variable t whose range is from 0 to NT is Defined If, and said t, real _{u 0, u 1, u 2} , ......, u 0 so that the polynomial equation 3 with u _m a least square approximation function of the data sequence y _i, u _1, u ₂ , ..., I channel first calculation means for calculating u _m ,
Wherein t and the real v _0, v _1, v _2, ......, so that v polynomial equation 4 using _m is minimum square approximation function of the data sequence z _i v _0, v _1, v _2, ... ..., Q-channel first calculating means for calculating the v _m [0016]
[Equation 3]
u (t) = u ₀ + u ₁ t + u ₂ t ² + …… + u _m t ^m
[0017]
[Expression 4]
v (t) = v ₀ + v ₁ t + v ₂ t ² + v ₃ t ³ + …… + v _m t ^m
[0018]
(E) I channel second calculation means for calculating u (t) of Formula 3 at t = iT (i = 1, 2, 3,..., N), and v (t) of Formula 4 is calculated. Q channel second computing means (f) I channel third memory for storing u (t) calculation result sequence T _i by I channel second computing means, and v (t by Q channel second computing means Q channel third memory storing the calculation result string L _i of ( _i ) Subtracting the calculation result string T _i of the I channel third memory from the data string y _i of the I channel second memory (y _i- T _i ), an I channel subtractor that outputs a data string;
A Q channel subtractor (H) _I that subtracts the calculation result sequence L _i of the Q channel third memory from the data sequence z _i of the Q channel second memory and outputs a data sequence of (z _i −L _i ) I The sum of squares of the output from the channel subtractor (y _i −T _i ) and the output from the Q channel subtractor (z _i −L _i ) or the square root of the sum of squares is calculated, and this is received with the clutter removed. Complex arithmetic unit that outputs as a target
Further, in the second means configuration of the present invention, in addition to the first means configuration, the output from the I channel subtracter (y _i −T _i ) and the output from the Q channel subtractor (z _i −L _i ). And a Doppler frequency calculating means for calculating a reception target Doppler frequency.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
According to the present invention, each of the I channel video output and the Q channel video output of the radar receiver is converted into a digital signal, and then the position of the column is determined based on a finite number of data columns in the azimuth direction of the same distance of the received signal. A trend curve (waveform) on the time axis is created by creating a least square approximation function as follows, and a variable corresponding to each position in the data string is calculated and values corresponding to individual data in the data string are calculated. By subtracting this value from the corresponding individual data, a signal from which the trend is removed from the received signal, that is, a signal from which the clutter spectrum is removed is obtained for each of the I channel and the Q channel, and this is complex-combined again. , To obtain a clutter suppression radar apparatus in which clutter is removed and a moving target such as an aircraft can be clearly obtained. .
[0021]
【Example】
Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of the clutter suppression radar apparatus of the present invention. However, since the antenna to the receiver is not different from the conventional one, the illustration is omitted. The AD converters 20 and 21 and the indicator 25 are the same as the conventional ones. The I channel subtractor 7 and the Q channel subtractor 14 have two systems of I channel and Q channel, but the operation itself is the same, so the I channel will be described.
[0022]
The I channel radar video output from the receiver 19 (FIG. 2) is converted into a digital signal by the AD converter 20 as in the prior art. The digitized radar video is stored in the first memory 1. The first memory 1 is composed of a large number of storage elements corresponding to range bins obtained by subdividing the radar detection space in the azimuth direction and subdividing in the distance direction, and the radar signal (received data) has its azimuth and distance. According to the above, it is stored in the corresponding storage element.
[0023]
Next, N data strings y _i (i = 1, 2, 3,..., N) having the same distance and continuous in the azimuth direction are extracted from the received data stored in the first memory 1. And stored in the second memory 2. The N data strings are sent to the first calculation means 4 of the trend calculation circuit 3. In the first calculation means 4, when the radar transmission cycle is T, and a variable t whose range is from 0 to NT is defined, t and real numbers u ₀ , u ₁ , u ₂ ,. , U _m is a polynomial (Formula 3)
u (t) = u ₀ + u ₁ t + u ₂ t ² + …… + u _m t ^m
There the data sequence y _i coefficients u ₀ such that least square approximation _{function, u 1, u 2, ...} , and calculates a u _m.
[0024]
Here, how to calculate u ₀ , u ₁ , u ₂ ,..., U _m is as follows. If the time interval between the input data connected in the azimuth direction of the radar is T, T is often one sweep time. A vector having a coefficient of the polynomial as an element is represented by u in the following Equation 5, and an input data string connected in the azimuth direction of the radar is represented by y in the following Equation 6.
In this case, since it is a system in which y≈H · u, H needs to be in the matrix form of Equation 7 below.
[0025]
[Equation 5]

[0026]
[Formula 6]

[0027]
[Expression 7]

[0028]
Here, N is the number of input data, and m is the maximum order of the approximate function. The number of rows of the matrix H is N. H ^T · H is calculated in advance, the input data matrix y is obtained, H ^T · y is calculated, and the solution vector u of Equation 8 is calculated by a method of calculating the solution of the simultaneous simultaneous equations by an iterative method.
[0029]
[Equation 8]
(H ^T H) · u = H ^T · y
[0030]
Here, the calculation result of H ^T H regardless of the value of N, and has a matrix of m rows and m columns. H ^T · y is also a vector of m rows. Equation 8 to _{calculate, u 0, u 1, u} 2 of Equation 5, ......, u _m can be obtained.
Here, instead of calculating the solution vector u of (H ^T H) · u = H ^T · y by an iterative method, Equations 9 and 8 are modified, and an inverse matrix (H ^T H) ⁻¹ is obtained by a sweeping method or the like. A calculation method is also conceivable. The calculation of the Q channel is exactly the same.
[0031]
[Equation 9]
u = (H ^T H) ⁻¹ H ^T y
[0032]
Equation 3 thus obtained represents a trend curve based on the received data string y _i , which represents a clutter waveform on the time axis. If Equation 3 is thus obtained, then the second computing means 5 uses Equation 3 to obtain u (iT) when t is 0, T, 2T,. (0), u (T), u (2T),..., U (NT) are calculated. The calculation result sequence u (iT) thus obtained is temporarily stored in the third memory 6 of the I channel.
[0033]
Next, the stored calculation result sequence u (iT) is input to the I channel subtractor 7. In addition, the data string y _i is input from the second memory 2 to the I channel subtractor 7. Then, the corresponding calculation result sequence u (iT) is subtracted from each data of y _i here. By this subtraction, the clutter waveform is removed from the received data. The data from which the clutter waveform has been removed is sent to the complex computing unit 15.
[0034]
Although the I channel has been described above, the same operation is performed for the Q channel, and the data from which the clutter waveform has been removed is sent to the complex computing unit 15. The complex computing unit 15 extracts amplitude information of the moving target from two input signals, calculates a Doppler frequency spectrum, and sends it to the indicator 25. The I channel data input to the complex computing unit 15 can be viewed as the real part of the complex display, and the Q channel data can be viewed as the imaginary part. If the square root of the sum of squares is obtained, it represents the voltage amplitude of the target signal vector. Therefore, it can be used as a signal for PPI display.
[0035]
Also, the Doppler frequency spectrum can be obtained by performing the FFT using the I channel data as the real part input data of the FFT (Fast Fourier Transform) circuit and the Q channel data as the imaginary part input data of the FFT circuit. Therefore, if the indicator 25 has a function of displaying a frequency spectrum, it can be displayed and observed.
The above explanation is for input data that continues in the azimuth direction, but the same can be applied to data that is continuous in the range direction.
[0036]
【The invention's effect】
As described above, in the radar apparatus of the present invention, the clutter spectrum seen in the vicinity of the zero frequency corresponds to the low frequency waveform (trend waveform) offset from zero volts in the time axis waveform. Since the clutter spectrum is eliminated by subtracting the trend waveform from the time axis waveform of the received radar video, the conventional MTI elimination is fixed even if the spread of the clutter spectrum increases. There is an advantage that the clutter suppression capability is not deteriorated, that is, the remaining unerased is less than that of the vessel.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a main part of an invention of a clutter suppression radar apparatus of the present invention.
FIG. 2 is a block diagram showing a configuration of a conventional digital MTI radar apparatus.
FIG. 3 is a diagram showing a circuit configuration of a conventional digital MTI single eraser and its frequency characteristics;
FIG. 4 is a diagram showing a circuit configuration and frequency characteristics of a conventional digital MTI double eraser.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 1st memory 2 2nd memory 3 Trend calculation circuit 4 1st calculating means 5 2nd calculating means 6 3rd memory 7 I channel subtractor 8 1st memory 9 2nd memory 10 Trend calculating circuit 11 First computing means 12 Second computing means 13 Third memory 14 Q channel subtractor 15 Complex computing unit 16 Antenna 17 Transmission / reception switch 18 Transmitter 19 Receiver 20 AD converter 21 AD converter 22 Eraser 23 Eraser 24 Synthesizer 25 Indicator

Claims (2)

A clutter suppression radar apparatus having the following configurations.
(A) Radar receiver circuit having two systems (I channel and Q channel) of received outputs whose phase of the local oscillation signal mixed with the received signal differs by 90 degrees (b) Each of the I channel and Q channel of the radar receiver circuit output The first I-channel memory and the Q-channel first memory (C) that store the received output for each transmission azimuth individually for each range bin that is subdivided in the distance direction are the same in the I-channel first memory. An I channel second memory for extracting and storing N data strings y _i (i = 1, 2, 3,..., N and so on) continuous in the azimuth direction at a distance, and the Q channel first memory Q channel second memory for extracting and storing N data strings z _i continuous in the azimuth direction at the same distance (d) When the radar transmission cycle is T, a variable t whose range is from 0 to NT is Defined If, and said t, real _{u 0, u 1, u 2} , ......, polynomial u (t) with _{u m = u 0 + u 1} t + u 2 t 2 + ...... + u m t m
I channel first computing means for calculating u ₀ , u ₁ , u ₂ ,..., U _{m so} that is a least square approximation function of the data string y _i ;
Polynomial v (t) using the above t and real numbers v ₀ , v ₁ , v ₂ ,..., V _m = v ₀ + v ₁ t + v ₂ t ² + v ₃ t ³ + ... + v _m t ^m
Q channel first computing means (e) t = iT (i = 1, i) for calculating v ₀ , v ₁ , v ₂ ,..., V _{m so} that is a least square approximation function of the data sequence z _i 2, 3... N) I channel second computing means for calculating u (t) of the polynomial, and Q channel second computing means (f) I for calculating v (t) of the polynomial. I channel third memory for storing the calculation result string T _i of u (t) by the channel second calculation means, and Q for storing the calculation result string L _i of v (t) by the Q channel second calculation means Channel third memory (g) Subtract calculation result sequence T _i of I channel third memory from data sequence y _i of I channel second memory to output a data sequence of (y _i −T _i ). An I channel subtractor;
A Q channel subtractor (H) _I that subtracts the calculation result sequence L _i of the Q channel third memory from the data sequence z _i of the Q channel second memory and outputs a data sequence of (z _i −L _i ) I The sum of squares of the output from the channel subtractor (y _i −T _i ) and the output from the Q channel subtractor (z _i −L _i ) or the square root of the sum of squares is calculated, and this is received with the clutter removed. Complex operator that outputs as target 2. A Doppler frequency calculation means for calculating a Doppler frequency of a reception target using an output (y _i −T _i ) from an I channel subtractor and an output (z _i −L _i ) from a Q channel subtractor. The clutter suppression radar device described.

Images