JPS63171380A - Apparatus for detecting target signal - Google Patents

Abstract

PURPOSE:To detect a target signal from a noise signal showing arbitrary noise distribution, by constituting the title apparatus of a logarithmic amplifying/cologarithmic amplifying means, a subtraction circuit, a non-linear amplifying circuit and a target signal detection circuit. CONSTITUTION:A logarithmic amplifying/cologarithmic amplifying means 10 subtracts the average value <Y> of a logarithmic amplified signal Y, obtained by logarithmically amplifying a receiving signal X having amplitude (x), from the logarithmic amplified signal Y and further respectively outputs a logarithmic amplified signal Z obtained by cologarithmic amplification to a square average value calculation means 20 and a means 30 for calculating the square value of the average value. Next, a subtraction circuit 40 calculates a dispersion value sigma<2> on the basis of a square average value <Z<2>> and the square value <Z2> of the average value to add the same to a non- linear amplifying circuit 50 which in turn raises the dispersion value sigma<2> to k-th power (k is the arbitrary real number) to output a non-linear signal U. Further, a target signal detection circuit 60 compares the non-linear signal U outputted from the circuit 50 with a preset threshold value TH manually or automatically and outputs the non- linear signal U larger than the threshold value TH as one corresponding to a target signal OS.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a target signal detection device for detecting a target signal from a noise signal whose amplitude characteristics exhibit an arbitrary distribution.

[Prior art] In radar observation, in order to detect a target object (hereinafter referred to as a target), it is necessary to suppress unnecessary reflected waves from the object (hereinafter referred to as clutter) and to suppress signals from the target object (hereinafter referred to as clutter). The problem is how to detect the target signal (hereinafter referred to as target signal).

Clutter includes ground clutter caused by reflection from the ground surface, and g-clutter caused by reflection from the sea surface.

Conventional target signal detection devices define an appropriate noise distribution,
The target signal was detected based on this noise distribution.

[Problems to be Solved by the Invention] By the way, the conventional target signal detection method has a problem in that the target signal cannot be detected when the noise distribution is outside the prescribed range.

The present invention has been made to solve the above problems,
It is an object of the present invention to provide a target signal detection device that detects a target signal from a noise signal exhibiting an arbitrary noise distribution.

[Means for Solving the Problems] Therefore, in the present invention, a signal obtained by subtracting the average value of the logarithmically amplified signal Y from a logarithmically amplified signal Y obtained by logarithmically amplifying an input signal X containing noise having an arbitrary distribution. Logarithmic amplification/anti-logarithm amplification means for outputting the anti-logarithmically amplified logarithmically amplified/anti-logarithm amplified signal 2; variance value calculation means for outputting the variance value σ2 of the anti-logarithmically amplified signal Z; An object is detected from a nonlinear calculation means that outputs a signal U, and a target signal detection means that compares the nonlinear signal U with a preset threshold TH and outputs a nonlinear signal U that is larger than the threshold TH as a target signal O8. Configure a target signal detection device.

[Function] In the target signal detection device having the above configuration, the logarithmic amplification/antilogarithm amplification means antilogarithmically amplifies the signal obtained by subtracting the average value of the logarithmically amplified signal Y from the logarithmically amplified signal Y obtained by logarithmically amplifying the input signal X. outputting the logarithmically amplified/antilogarithmically amplified signal Z, the variance value calculation means calculates the variance value σ2 of the logarithmically amplified/antilogarithmically amplified signal Z,
The nonlinear calculation means outputs a nonlinear signal U multiplied by the variance value σ2, and the target signal detection means compares the nonlinear signal U with a preset threshold TH, and detects a nonlinear signal U larger than the threshold TH as a target signal. Output as .

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the accompanying drawings.

First, the signal processing theory of the target signal detection device according to the present invention will be explained.

The target signal detection device according to the present invention detects a target signal from a noise signal whose amplitude characteristic exhibits an arbitrary distribution.
1 distribution) and a majority normal distribution (Log-norma 1 distribution) will be described.

The amplitude intensity X follows the Weibull distribution, and its probability density function P
C(X) is (where b is a scale parameter and C is a shape parameter.) The received signal is logarithmically amplified and the logarithmically amplified signal Y (Y-ag 'kn (b6 The signal V (V-Y-<
When antilogarithmically amplifying Y)), the initial S/N ratio is maintained. Amplification characteristics of anti-logarithmic amplifier circuit Z-CQ EXP (DQ V) (
2) (Here, if Go and d (1 is a constant that determines the gain of the anti-logarithmic amplifier circuit) are given as follows: ag −do −1 (3), then the average value of the output Z of the anti-logarithmic amplifier circuit <z> and the root mean square value 〉 are respectively 〈Z〉 −f
Z P (X)dXC -CEXP(-r<-+1) (4)OC
C -c Exp(P)r ('-+l>C...(5) Therefore, the variance value σ2 is σ 2 Proverb <72>-4>2 2 2γ 2 21 - C.EXP(τ-) (r (7+1>-r' (
Te41) )...(6) becomes. The target signal O8 is detected by multiplying this variance value σ.

Next, FIG. 1 is a block diagram of a target signal detection device according to the present invention. In FIG. 1, the logarithmic amplification/anti-logarithm amplification means 10 converts the logarithmically amplified signal Y obtained by logarithmically amplifying the signal X1 shown in FIG. The logarithmically amplified/antilogarithmically amplified signal Z which has been further antilogarithmically amplified is outputted to the mean square value calculating means 20 and the mean square value calculating means 30, respectively. It is comprised of a delay circuit 12, an integration circuit 13, a division circuit 14, a subtraction circuit 15, and an antilogarithmic amplification circuit 16, which are configured by a shift register or the like, and outputs the above-mentioned logarithmically amplified/antilogarithmically amplified signal Z. That is, the logarithmic amplifier circuit 11 logarithmically amplifies the input signal X and outputs the logarithmically amplified signal Y, and the delay circuit 12 samples and quantizes the logarithmically amplified signal Y to obtain n corresponding to the amplitude y of the logarithmically amplified signal Y. sample values Y1, Y2,...
・, Yn is taken out, and the integrating circuit 13 collects n sample values Y1.
, Y2, . . . , Yn, and the division circuit 14
divides the integrated value by n to calculate the average value of the integrated values, that is, the average value <Y> of the logarithmically amplified signal Y. Further, the subtracting circuit 15 subtracts the average value (Y> of the logarithmically amplified signal Y from the logarithmically amplified signal Y, and the antilogarithm amplifier circuit 1B subtracts the average value (Y>) of the logarithmically amplified signal Y from the logarithmically amplified signal Y. is inverse logarithmically amplified,
A logarithmically amplified/antilogarithmically amplified signal Z (Z-CQ EXP (DQ V)) is output.

The mean square value calculation means 20 is composed of a square calculation circuit 21, a delay circuit 22 constituted by a shift register, etc., an integration circuit 23, and a division circuit 24. The root mean square value <22> of the logarithmically amplified/antilogarithmically amplified signal Z which is the output of the amplifier circuit 1o is calculated. That is, the square calculation circuit 21 squares the amplitude 2 of the logarithmically amplified/antilogarithmically amplified signal Z to obtain the amplitude z2, and the delay circuit 22 samples and quantizes the logarithmically amplified/antilogarithmically amplified signal Z to obtain the logarithmically amplified/antilogarithmically amplified signal Z. n (n
is a natural number) sample values zi, zl, ..., ZM are taken out, and the integrating circuit 23 extracts n sample values zf, z? 1. ...
.

The average square value calculation means 3o is composed of a delay circuit 3L integration circuit 32, a division circuit 33, and a square value calculation circuit 34, and calculates the average value of the logarithmically amplified/antilogarithmically amplified signal Z shown in Equation 7. Calculate the square value <z>2 of the value.

That is, the delay circuit 31 samples and quantizes the logarithmically amplified/antilogarithmically amplified signal Z to obtain the amplitude 2 of the logarithmically amplified/antilogarithmically amplified signal Z.
n sample values Z11Z2 (n is a natural number) corresponding to
..., Zn is taken out, and the integrating circuit 33 collects n sample values Z.
ISZ2, ..., integrated value Σ Z of Zn
(11) i-11 is calculated, and the division circuit 34 divides the integrated value by n to obtain the average value of the integrated values, that is, the average value of the logarithmically amplified/logarithmically amplified signal Z <z> -Σ Z
(12) nt"t t is calculated, and the square value calculation circuit 35 squares the average value to calculate the square value <z>2 of the average value of the anti-logarithmically amplified signal Z.

Next, the subtraction circuit 40 calculates the variance value σ 2 −<22>−<Z>2 (14) based on the root mean square value <22> and the square value of the mean value <z>.

FIG. 2 is a diagram showing the variance value σ2 obtained as described above. In Fig. 2, logarithmic amplification/antilogarithmic amplification circuit 1
Even if the amplitude of the target signal O8 and the amplitude of the noise signal NS are almost the same (see FIG. 2(a)), the signal The value of the part corresponding to the target signal O8 is large, and the noise signal NS
Since the value of the part corresponding to is almost constant and the fluctuation range is small (
(see Figure 2(b)), the variance value σ corresponding to the target signal O8
2 can be easily detected.

In this embodiment, in order to more easily detect the dispersion value σ2 corresponding to the target signal, the dispersion value σ2 is added to the nonlinear amplification circuit 50 (nonlinear amplification means).

This nonlinear amplifier circuit 50 has a dispersion value σ2 raised to the power of (however,
k is a real number) and outputs a nonlinear signal U U-(σ2k) (15). Note that, as a specific example of the nonlinear amplification circuit 50, a square amplification circuit with k-2 is known. As shown in FIG. 2(C), the nonlinear signal U output from the nonlinear amplifier circuit 50 has fluctuations in the portion corresponding to the noise signal NS suppressed compared to the variance value σ2 output from the subtraction circuit 40. The part corresponding to the target signal O8 is greatly amplified, and the dispersion value σ2 corresponding to the target signal O8 is
Detection, that is, threshold setting becomes easier.

Furthermore, the target signal detection circuit 60 (target signal detection means) compares the nonlinear signal U output from the nonlinear amplifier circuit 50 with a preset threshold manually or automatically, and selects a nonlinear signal U larger than the threshold. It is output as a signal corresponding to the target signal O8.

The threshold TH mentioned above is set manually or automatically,
In the case of manual operation, the setting is made while observing the nonlinear signal U with an A scope, a B scope, a PPI scope, or the like. In addition, in the case of automatic detection, based on the average value <u> of the nonlinear signal U and the variance value Mar around the average value, the false alarm probability that the noise signal NS is mistakenly detected as the target signal O8 or the target signal O8 The detection probability is set to a predetermined value.

The average value <IJ> and the root mean square value <U2> of the nonlinear signal U are expressed as < if the probability density function of the nonlinear signal U is p (u).
1> Kazu f U −P (U) dU
(16), and the variance value Var around the average value <ll> of the nonlinear signal U is Var = <U”>−<U>2
(1g). Also, the false alarm probability Pf'a is ~Pra - J P (U) dU
(19)H, and the detection probability Pa of the target signal O8 when the amplitude t of the target signal O8 is superimposed on the nonlinear signal U is Pa−J
' P (U+t) d(U+t)
(20) becomes I. Therefore, the threshold value TH is set so that the false alarm probability Pfa or the detection probability Pa becomes a predetermined value.

Next, a specific value of the threshold value T11 calculated as described above will be explained. That is, the m power of the variance value σ2 which is the output of the subtraction circuit 4° is M, the average value of M is <1>, M
The root mean square value of M is <M2>, and the variance value of M is (<M2> −
〈M〉2), and let A be a constant, it will be one of the following values.

K・〈M〉
(21) K, J'3π
(22) K ・ (〈M2〉
-〈M)2) (
23)〈M〉+K ・ (cM2>−〈M〉2)
(24) K ・ (〈M〉
+ (<)42>-<river>2)
(25) K ・〈M〉 + A
(2B)K
-f1C+ A (27)K
・ (<M2>-<1>2) + A
(28)〈M〉+K・(
<M2>-<M>2) 10 A (29)・
K ・(〈M〉 + ＜%”＞−＜@X) +A
(30) Next, the amplitude intensity X has a lognormal distribution (Log
-Normal) will be described.

The lognormal distribution for the received signal X is P(X, m, ρ). Here, m is the median value of the received signal A logarithmically amplified signal Y (Y=aok (bo
Signal V obtained by subtracting the average value <Y> of the logarithmically amplified signal Y from X)
When (V-Y-<Y>) is anti-logarithmically amplified, the initial S
/N ratio is maintained. Amplification characteristics of anti-logarithmic amplifier circuit Z = COEXP (dOV) ' (
33') (here, co and do are constants that determine the gain of the anti-logarithm amplifier circuit), ao'' dO=1 (34
), then the average value <z> and the root mean square value of the output Z of the antilogarithmic amplifier circuit, which is the output of the antilogarithmic amplifier circuit.
z−〉 is respectively <'l> −f Z−P(X; m, ρ) dX
−CQ ρ (35
)2 ■ 2 Kuz ＞=f Z P (X: m, ρ
) ax″Coρ. Therefore, the variance value σ2 is (36)σ
2-<Z2>-<z>2-Co (ρ-ρ). In an actual circuit, the root mean square value of logarithmically amplified/antilogarithmically amplified signal 2 obtained by logarithmically amplifying/antilogarithmically amplifying the received signal <2
2> and the square value of the average value <z>2 are the values mentioned above, so their explanation will be omitted.

In this embodiment, a case has been described in which the target signal O8 is detected from a noise signal exhibiting a Weibull distribution and a lognormal distribution. However, even if the target signal O8 is a noise signal exhibiting an arbitrary distribution,
Target signal O3 can be detected.

[Effects of the Invention] As explained above, according to the present invention, the average value of the logarithmically amplified signal Y is subtracted from the logarithmically amplified signal Y obtained by logarithmically amplifying the input signal Z containing the noise signal whose amplitude characteristics exhibit an arbitrary distribution. By outputting the dispersion value of the logarithmically amplified/antilogarithmically amplified signal Z (Z-CQ EXP (dQ V)) obtained by antilogarithmically amplifying the signal V, and further nonlinearly amplifying the dispersion value, it corresponds to the target signal. Since the magnitude of the dispersion value or the nonlinearly amplified dispersion value is larger than the magnitude of the dispersion value corresponding to the noise signal, by setting an appropriate threshold value, the amplitude of the target signal is smaller than the amplitude of the noise signal. target signal can be easily detected.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a target signal detection device according to the present invention,
FIG. 2 is an explanatory diagram of changes in the received signal over time, a dispersion value, and a new dispersion value multiplied by the dispersion value, and FIG. 3 is an explanatory diagram showing changes in the received signal over time. lO... Logarithmic amplification/anti-logarithmic amplification means, 11... Logarithmic amplification circuit, 12.22.31... Delay circuit, 13.2
3.32... Integration circuit, 14.24.33... Division circuit, 15... Subtraction circuit, 1B... Anti-logarithm amplifier circuit, 20... Mean square value calculation means, 21... - Square calculation circuit, 30...mean value square value calculation means, 34...
Square calculation circuit, 40... Subtraction circuit, 50... Nonlinear amplifier circuit, 60... Target signal detection circuit.

Claims (1)

[Scope of Claims] (1) In a target signal detection device that detects a target signal from an input signal X mixed with noise exhibiting an arbitrary distribution, from a logarithmically amplified signal Y obtained by logarithmically amplifying the input signal X. A logarithmically amplified/antilogarithmically amplified signal Z (Z=C_0EXP) is obtained by further antilogarithmically amplifying the signal V obtained by subtracting the average value of the logarithmically amplified signal Y.
(D_0V)); a variance calculation means that outputs the variance value σ^2 of the logarithmically amplified/antilogarithmically amplified signal Z; K is an arbitrary real number) and a nonlinear calculation means for outputting a nonlinear signal U;
and target signal detection means for comparing the nonlinear signal U with a preset threshold TH and outputting a nonlinear signal U larger than the threshold TH as a target signal OS. (2) the logarithmic amplification/anti-logarithm amplification means outputs a logarithmically amplified signal Y obtained by logarithmically amplifying the input signal X;
An average value calculation means for calculating the average value <Y> of the logarithmically amplified signal Y, a subtraction means for subtracting the average value <Y> from the logarithmically amplified signal Y, and a subtraction means for subtracting the average value <Y> from the logarithmically amplified signal Y; 2. The target signal detection device according to claim 1, comprising antilogarithm amplification means for antilogarithmically amplifying the difference. (4) The threshold value in the target signal detection means is based on the average value of the value M, which is the m-th power of the variance value σ^2 obtained by the variance value calculation means, and the variance value around the average value of M.
The target signal detection device according to claim 1, wherein the false alarm probability or detection probability of the target signal is set to a predetermined value. (5) The threshold value is the value K·<M> obtained by multiplying the average value <M> of the value M, which is the mth power of the variance value σ^2 obtained by the variance value calculating means, by a constant K. The target signal detection device according to item 1. (6) The threshold value is the value k obtained by multiplying the square root √<M^2> of the root mean square value <M^2> of the value M, which is the mth power of the variance value σ^2 obtained by the variance value calculation means, by the constant K. - √<M^2> The target signal detection device according to claim 1. (7) The threshold value is the variance value (〈M^2〉−〈M〉^2) around the average value 〈M〉 of the value M which is the mth power of the variance value σ^2 obtained by the above-mentioned variance value calculation means. A patent claim in which the value K·√(<M^2>−<M>^2) is calculated and further multiplies the square root of the variance value √(<M^2>−<M>^2) by a constant K. The target signal detection device according to item 1. (8) The threshold value is the variance value (〈M^2〉−〈M〉^2) around the average value 〈M〉 of the value M which is the mth power of the variance value σ^2 obtained by the above-mentioned variance value calculation means. Then, the square root of the variance value √(〈M^2〉−〈M〉^2) is multiplied by a constant K, and the value obtained by adding the mean value〈M〉〈M〉+K・√(〈M㼾2>-<M>^2) The target signal detection device according to claim 1. (9) The threshold value is the variance value (〈M^2〉−〈M〉^2) around the average value 〈M〉 of the value M which is the mth power of the variance value σ^2 obtained by the above-mentioned variance value calculating means. Calculate the mean value〈M〉 to the square root √(〈M^2〉−〈M〉^2) of the variance value.
and further multiplied by a constant K, which is the value K・(<M>+√(<M^2>−<M>^2)). (10 ) The threshold value is a value obtained by multiplying the average value <M> of the value M obtained by raising the variance value σ^2 obtained by the variance value calculating means by a constant K, and further adding a constant A, K・〈M〉+A. The target signal detection device according to claim 1. (11) The threshold value is the root mean square value <M^2> of the value M obtained by raising the variance value σ^2 obtained by the variance value calculating means to the m power. The target signal detection device according to claim 1, which is a value obtained by multiplying the square root √〈M^2〉 by a constant K and further adding a constant A to the value K・√〈M̂2〉+A. (12) Threshold value calculates the variance value (〈M^2〉−〈M〉^2) around the average value <M> of the value M obtained by raising the variance value σ^2 obtained by the variance value calculation means to the m power, and The square root of the variance √(〈M^2〉−〈M〉^2) is multiplied by a constant K, and the constant A is further added to the value K・√(〈M^2〉−〈M〉^2)+A. The target signal detection device according to claim 1. (13) The threshold value is the variance around the average value <M> of the value M obtained by raising the variance value σ^2 obtained by the variance value calculation means to the m power. Calculate the value (〈M^2〉−〈M〉^2), multiply the square root of the variance value √(〈M^2〉−〈M〉^2) by a constant K, and calculate the average value〈M〉. In addition, the value 〈M
〉+K·√(〈M^2〉−〈M〉^2)+A. The target signal detection device according to claim 1. (14) The threshold value is the variance value (〈M^2〉−〈M〉^2) around the average value 〈M〉 of the value M which is the mth power of the variance value σ^2 obtained by the above-mentioned variance value calculation means. Calculate the mean value 〈M
〉, the sum of the square root of the variance value and the average value is multiplied by a constant K, and the constant A is further added to the value K・(〈M〉+√〈M^2〉−〈M〉^2)+A. A target signal detection device according to claim 1.