JPH03156396A - Method and apparatus for processing radar signal - Google Patents

Abstract

PURPOSE:To positively suppress a surface-of-sea reflecting signal by respectively multiplying moment signals of plural orders, calculated from linear signals obtained through antilogarithmic conversion of difference signals, by coefficients and setting the sum thereof as a threshold value and subtracting the sum from the linear signals. CONSTITUTION:From a logarithmic conversion signal Y obtained by amplifying a radar signal X through a logarithmic amplifier 100, an moving average value signal (Y) of (N) signals Y is subtracted to produce a difference signal V from the first difference circuit 12. This signal V is subjected to antilogarithmic conversion through an antilogarithmic operation circuit 14 to obtain a linear signal Z which is, in turn, inputted to a threshold value operation part 50. In the operation part 50, moment signals from the first order to the n-th order (n is 4 in this example) of the signal Z are calculated by first, second, third and fourth order moment operation circuits 16-1 - 16-4 and multiplied by coefficients predetermined for every moment by first, second, third and fourth order moment coefficient circuits 18-1 - 18-4 and the sum of the respective moment signals subjected to coefficient multiplication is calculated as a threshold signal T by a summing circuit 20 and the signal T is subtracted from the signal Z to obtain a final output signal. By this method, a signal causing constant erroneous alarm probability can be calculated with respect to a radar signal containing a surface-of-sea reflecting signal according to Weibull distribution and the surface-of-sea reflecting signal is suppressed to make it possible to clearly display a ship image.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a radar signal processing method and apparatus for suppressing sea surface reflection signals and detecting target target signals such as ships.

[Prior Art] Log-CFAR is conventionally known as a signal processing method for suppressing clutter signals in marine radars.

In Log-CFAR, the clutter signal is a Rayleigh distribution signal (R171eilh dislrivation si
This method uses the property that the deviation of the Rayleigh signal around the average value of the log amplifier output signal is constant when the output signal is 100%, and the false alarm is kept at a constant value.

That is, as shown in FIG. 8, the clutter signal W is input to the Log amplifier 100 to obtain the output signal Y, and the average value <y> is calculated from the Log amplifier output signal Y by the averaging circuit 20G. By subtracting the average value <y> from the Log amplifier output signal Y in the difference circuit 300, a CFAR signal V with constant false alarms can be obtained.

[Problems to be Solved by the Invention] However, it has been confirmed that the amplitude distribution characteristic of the sea surface reflection signal, which is one of the clutter signals, is not a Rayleigh distribution but a Weibull distribution. Log-C for the Rayleigh distribution signal shown in Figure 8 for the sea surface reflection signal
Even if the FAR method is applied, the deviation around the average value <y> will not be constant due to the difference in statistical properties between the two, increasing the probability of false alarms and causing sea surface reflections to remain undisappeared.

Particularly with ship radars, it becomes difficult to distinguish between sea surface reflection signals and ship images, which may lead to an extremely dangerous situation for ship navigation.

The present invention has been made in view of such conventional problems, and it is an object of the present invention to calculate a signal with a constant probability of false alarm for radar signals including sea surface reflection signals according to the Weibull distribution received by radar devices on ships, etc. An object of the present invention is to provide a radar signal processing method and device that can reliably suppress sea surface reflected signals.

[Means for Solving the Problem] In order to achieve this object, the present invention first calculates the moving average of logarithmically transformed radar signals as a radar signal processing method including sea surface reflection signals according to Weibull distribution. A difference signal is obtained by subtraction, this difference signal is inverse logarithmically converted to a linear signal, and moment signals from the 1st to the nth order of this linear signal are calculated, and a predetermined coefficient is used for each moment signal. A threshold signal is calculated as the sum of the moment signals multiplied by the coefficient, and finally the threshold signal is subtracted from the linear signal to obtain an output signal.

Further, in the present invention, as a radar signal processing device including a sea surface reflection signal according to a Weibull distribution, a moving average means for calculating a moving average of a logarithmically transformed radar signal; a moving average signal from the moving average means; a first difference means that subtracts the difference signal from the logarithmically transformed radar signal and outputs a difference signal; an antilogarithm transform means that performs an antilogarithm transform on the difference signal from the first difference means and outputs a linear signal; a moment calculation means for generating in parallel moment signals from the first order to the nth order of the linear signal from the anti-logarithmic conversion means; moment coefficient means for multiplying by a coefficient using a predetermined coefficient; summing means for calculating a threshold signal as the sum of the first to nth moment moment signals multiplied by a coefficient by the moment coefficient means; a second difference means for obtaining an output signal by subtracting the obtained threshold signal from the linear signal converted by the anti-logarithmic conversion means;

Furthermore, as the radar signal processing method and apparatus, the threshold signal obtained as the sum of the moment signals multiplied by the coefficient may be used as the output signal as it is.

[Operation] According to the radar signal processing method and device of the present invention having such a configuration, a CFA is achieved in which the probability of false alarm is constant by signal processing for sea surface reflection signals according to Weibull distribution.
It is possible to obtain an R output signal, reliably suppress sea surface reflection signals so that they do not remain, and ensure navigation safety by clearly displaying images of targets such as ships even when sea surface reflections occur. .

[Example] First, a radar signal processing method of the present invention for suppressing a sea surface reflection signal according to a Weibull distribution included in a radar video signal (hereinafter simply referred to as a radar signal) is obtained by the following processing steps.

■Logarithmically transform the radar signal.

■Calculate the moving average from the logarithmically transformed radar signal.

■ Subtract the moving average obtained in (■) from the logarithmically converted radar signal to obtain a difference signal.

■Antilogarithmically transform the differential signal to make it a linear signal.

■Calculate moment signals from the 1st order to the nth order of the linear signal.

■ Multiply each moment signal by a predetermined coefficient using a predetermined coefficient.

(2) Calculate the threshold signal as the sum of the first to nth moment moment signals multiplied by the coefficient.

(2) Finally, the threshold signal is subtracted from the linear signal obtained in (2) above to obtain an output signal.

The radar signal processing method of the present invention is based on the following principle.

First, the probability density function P of the radar signal X following the Weibull distribution
c(X) is expressed by the following equation (1).

where b is the scale parameter and C is the sharpness (shape)
This is called a parameter and determines the characteristics of the Weibull distribution.

The logarithmic conversion output Y of the radar signal X is expressed as Y=LnX (2),
The moving average value <y> of the logarithmically converted output Y is expressed by the following equation using parameters b and c of equation (1).

<y>= j2nb-(1/c) ・7 (3)
Here γ is Euler's constant.

Next, the Loll-CFAR output V output of the logarithmic conversion output Y is the value obtained by subtracting the moving average <Y> from the logarithm conversion output Y, so V = y - <y> (4), and this Log-CFAR output ■ The output logarithmic conversion output Z is expressed by the following equation.

Z=ev (5) By substituting the above-mentioned equations (2), (3), and (4) into this equation (5), the antilogarithmic transformation output 2 can be transformed as follows.

Z= (X/b)e''' (6) Next, the (6th
) The first moment of the anti-logarithmic transformation output 2 given by the equation is determined by the following equation.

<ZJ>=e “”r (J/c+1) (7
) where r is the gamma function.

On the other hand, the false alarm probability Pfx of the inverse logarithmic transformation output 2 given by the above equation (6) is determined by setting the threshold value (threshold level) to T
Then, it is expressed by the following formula.

Here, if the threshold T is expressed as the linear sum (total sum) of the first-order moments of the anti-logarithmic transformation output 2, then TJa, [<Z'>]''' (9)
becomes. Therefore, the false alarm probability pH of the anti-logarithmic transformation output 2 given by the above equation (8) is: PIs= , -(,E,ar r [(J/C+ 1
)]” ) '(10).

Here, J=
If the value of the coefficient a1 in the above equation (9) is determined as shown in Fig. 4 for the case of 1.2.3 and the case of J = 1.2, 3°4, then in the above equation (10), Given false alarm probability PL! is approximately constant when J = 1 to 3, as shown in Fig. 5, and becomes more constant as shown in Fig. 6 when J = 1 to 4. It has been confirmed that a CFAR output with sufficiently suppressed reflected signals can be obtained.

Note that the coefficient at-a3. of J=1st to 3rd order or J=1st to 4th order shown in FIG. The value of al-a4 is the value obtained by the optimization method, assuming that the number N of logarithmically transformed outputs Y for obtaining the moving average value <y> given by the above equation (3) is N=16, and this coefficient a , depends on the false alarm probability pH and the number N of logarithmically transformed outputs Y for obtaining the moving average <y>. Furthermore, the value of al is not uniquely determined by each parameter,
There are several combinations of optimal solutions.

FIG. 1 is a block diagram showing an embodiment of a radar signal processing device according to the present invention.

In FIG. 1, 100 is a log amplifier, which inputs a radar signal X sampled at a predetermined period, logarithmically amplifies it, and outputs a logarithmically converted signal Y.

Here, the radar signal to be processed by the apparatus shown in FIG. 1 may be an analog signal, or may be a quantized radar signal obtained by converting a sampled analog signal into a digital signal. Naturally, each circuit section adopts a circuit system suitable for analog and digital signals.

The logarithmically transformed signal Y from the log amplifier 100 is given to a moving average circuit 10, which calculates a moving average of a predetermined number N of logarithmically transformed signals Y.

FIG. 2 is a block diagram showing an embodiment of the moving average circuit 10 shown in FIG.

The delay circuit 24 determines the number N1 of moving average values, for example, N=
Sixteen delay elements are connected in series, and the logarithmically converted signal Y input in series is sequentially converted into N parallel outputs.

The adder circuit 26 receives the new logarithmically converted signal Y from the delay circuit 24.
The sum of N=16 parallel outputs obtained for each input is calculated and output to the divider circuit 28. The division circuit 28 divides the output of the addition circuit 26 by the number N for which the average value is to be determined, and outputs a moving average value signal <y>.

FIG. 3 is a block diagram showing another embodiment of the moving average circuit 10 of FIG. 1. In FIG.

In FIG. 3, the moving average circuit 10 is the N/2 delay circuit 3.
0.34, an adder circuit 36, 38, an adder 40, and a divider circuit 28. Furthermore, the output of the N/2 delay circuit 30 is sent to the next N/2 delay circuit 34 via a one-stage delay element 32. is giving to

When N=16, each of the N/2 delay circuits 30.34 has 8 delay elements connected in series, and converts the sequentially input logarithmically converted signal Y into 8 parallel outputs. Output to 38.

The adder circuits 36.38 correspond to the N of each N/2 delay circuit 30.34.
= Find the sum of 8 parallel outputs, give it to the adder 40 and add it to find the sum of N pieces, divide by the number N for which the moving average is to be obtained in the division circuit 28, and output the moving average value signal <y> do.

The one stage delay element 32 provided between the N/2 delay circuits 30 and 34 is used to output the logarithmically converted signal Y to the next stage first difference circuit 12 shown in FIG. There is.

Referring again to FIG. 1, a first difference circuit 12 is provided following the moving average circuit 10, and the first difference circuit 12 converts the logarithmically converted signal Y obtained from the log amplifier 100 into the moving average circuit 10. N moving average value signals calculated by <y
> is subtracted and a difference signal V is output. This difference signal V
becomes the Log-CFAR output given by the above equation (4).

Further, if the first difference circuit 12 is an analog type, it can be realized by a difference circuit using an operational amplifier, whereas if it is a digital type, it can be realized by a digital adder.

The difference signal V from the first difference circuit 12 is given to the antilogarithm calculation circuit 14, and the antilogarithm calculation circuit 14 outputs the antilogarithm conversion signal (linear signal) Z given by the above-mentioned equation (5). The anti-logarithm calculation circuit 14 is realized by a function calculator having exponential function characteristics if it is an analog system, or by a look-up table LUT that stores an anti-logarithm calculated in advance using the difference signal V as an address if it is a digital system. realizable.

The antilogarithm conversion signal 2 from the antilogarithm calculation circuit 14 is
9) is given to the threshold value calculation unit 50 for calculating the threshold value T given by equation 9). In this embodiment, the threshold calculation unit 50 takes first to fourth order moment calculations as an example.

First, the anti-logarithm conversion signal Z is sent to the first moment calculation circuit 1.
6-1. Second-order moment calculation circuit 16-2. Third-order moment calculation circuit 16-3 and fourth-order moment calculation circuit 16
-4, and after raising the inverse logarithmically transformed signal Z to the nth power, an average value is calculated using the same average calculation algorithm as in the moving average circuit 10. That is, for the term on the right side of equation (9) excluding Σa, J = 1.2.3°J
= 1 The moment is calculated for each of 4.

Next, the first-order moment coefficient circuit 18-1. Second-order moment coefficient circuit 18-2. Third-order moment coefficient circuit 18
- third and fourth moment coefficient circuits 18-4 are provided;
Predetermined moment coefficients a1, a2. are applied to each of the first to fourth moments calculated in the previous stage. a3. a
Multiply 4 and match. This 1st to 4th moment coefficient al-
For a4, the values a1 to a4 shown on the right side of FIG. 4 are used.

Each of the first to fourth moments multiplied by a predetermined moment coefficient in this way is finally input to the summation circuit 20, and in the summation circuit 20, J=1 to A threshold T of 4 is calculated. Note that the moment coefficient circuits 18-1 to 18-4 in the threshold calculation unit 50
can be realized using an operational amplifier if it is an analog method, or by a lookup table if it is a digital method.
Similarly, the summation circuit 20 can be realized by a combination of an operational amplifier for an analog system and a look-up table or a digital addition circuit for a digital system.

The antilogarithm conversion signal 2 from the antilogarithm calculation circuit 14 and the threshold signal T calculated by the threshold calculation unit 50 are finally output to the second difference circuit 22, and the second difference circuit 22 converts the antilogarithm conversion signal into an antilogarithm conversion signal. The final CFAR signal is output by subtracting the threshold signal T from (linear signal) 2. The output signal obtained from the second difference circuit 22 uses the threshold value T calculated by the first to fourth order moments according to the moment coefficients shown on the right side of FIG. Thus, even for a sea surface reflection signal that follows a Weibull distribution signal, it is possible to calculate an output signal with a constant false alarm probability Pft regardless of the characteristics of the distribution.

FIG. 7 is an embodiment configuration diagram showing another embodiment of the present invention, in which the second difference circuit 22 shown in FIG. It is characterized in that it has an output signal. That is, in the embodiment shown in FIG. 1, the output signal is obtained by subtracting the threshold signal T from the anti-logarithmically transformed signal (linear signal) Z.
As in the embodiment shown in FIG. 7, the false alarm probability Ptg is constant and even if the Bell threshold signal T itself is displayed as the output signal, it is possible to obtain an output signal in which the sea surface reflection signal is suppressed.

Note that the embodiment shown in FIG. 1.7 takes first- to fourth-order moment calculations as an example, but for example, the moment coefficients a l, a 2 . It may be a third-order moment calculation using a3, and furthermore, the order of the moment calculation can be appropriately determined as necessary.

In addition, although the above embodiment takes as an example the case where the sea surface reflection signal has a Weibull distribution, even when the sea surface reflection signal has a distribution other than the Weibull distribution, the moment calculation of the anti-logarithmically transformed linear signal can be performed in the same way. Results and optimized coefficients 8
1, it is possible to obtain a threshold signal and reliably suppress the sea surface reflection signal.

[Effects of the Invention] As explained above, according to the present invention, it is possible to calculate an output signal that has a constant false alarm probability regardless of the characteristics of the distribution, even for sea surface reflection signals that follow the Weibull distribution. By reliably suppressing sea surface reflection signals and clearly displaying ship images, it is possible to ensure the safety of ship navigation.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the device configuration of the present invention; Fig. 2.3 is a block diagram of an embodiment of the moving average circuit of Fig. 1; Fig. 4 is a block diagram of an embodiment of the moving average circuit of the present invention; An explanatory diagram of the coefficient a1 used; FIG. 5 is an explanatory diagram of the false alarm probability P11 by the third-order moment calculation of the present invention; FIG. 6 is an explanatory diagram of the false alarm probability pH by the fourth-order moment calculation of the present invention; FIG. 7 is an embodiment configuration diagram showing another embodiment of the device configuration of the present invention; FIG. 8 is an explanatory diagram of a conventional system. 10: Moving average circuit 12: First difference circuit 14: Antilogarithm calculation circuit 16-1 to +6-4: 1st to 4th moment calculation circuit 1
8-1 to 18-4: 1st to 4th moment coefficient circuit 20: Summation circuit 22: Second difference circuit 24: Delay circuit 26.36.38.40: Addition circuit 28: Divide circuit 30.34: N/2 delay circuit 32: delay element 50: threshold calculation section 100: log amplifier

Claims (3)

[Claims] (1) Subtract the moving average from the logarithmically converted radar signal to obtain a difference signal, perform antilogarithm conversion on the difference signal to make it a linear signal, and calculate moment signals from the 1st to the nth order of the linear signal. At the same time, each moment signal is multiplied by a predetermined coefficient using a predetermined coefficient, a threshold signal is calculated as the sum of the 1st to nth order moment signals multiplied by the coefficient, and finally the threshold signal is converted to the linear A radar signal processing method characterized by subtracting the signal from the signal to obtain the output signal. (2) a moving average means for calculating a moving average signal of the logarithmically transformed radar signal; a first difference subtracting the moving average signal from the moving average means from the logarithmically transformed radar signal and outputting a difference signal; means; anti-logarithmic conversion means for anti-logarithmically converting the difference signal from the first difference means and outputting a linear signal; moment calculation means that generates in parallel; moment coefficient means that multiplies each of the first to nth order moment signals generated by the moment calculation means by a coefficient using a predetermined coefficient; the moment coefficient means summing means for calculating a threshold signal as the sum of the first to nth order moment signals multiplied by a coefficient; subtracting the threshold signal obtained by the summing means from the linear signal transformed by the anti-logarithmic transformation means to obtain an output signal; Second thing to ask for
A radar signal processing device comprising: a differential means; (3) The radar signal processing method and device according to claim 1 or 2, characterized in that the output signal is a threshold signal obtained as the sum of the first to n-th moment signals.