JPH09231202A - Correlator - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a correlator which can judge its phase advance or delay by itself. SOLUTION: Autocorrelation characteristics obtained when a 1st and a 2nd signal are the same by thinning out information by an information thinning-out means 13 in arbitrary timing as to the multiplication result of the 1st and 2nd signals, inputted to input terminals 11 and 12, which is based upon a function of a parameter τ representing the deviation of one signal from the other are made asymmetrical about the correlation value axis of τ=0. Consequently, the correlator can judge its phase advance or delay by itself.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a correlator that obtains a correlation characteristic that gives a measure of similarity or mutual dependence between two functions.

[0002]

Correlators are circuits that are widely applied in the field of electrical and electronic technology. Depending on the signal handled by the correlator,
Although it may be an analog circuit or a digital circuit, basically it is still the same as determining the cross-correlation characteristic between two signals. Here, if the functions representing two signals are f1 (t) and f2 (t) and the function of the parameter τ representing the deviation of one function from the other, the cross-correlation function R
12 (τ) is expressed by the following equation (1).

[0003]

[Equation 2] Although arbitrary functions can be selected as the functions f1 (t) and f2 (t), the case where the functions f1 (t) and f2 (t) are equal is called an autocorrelation function. In the following description, the autocorrelation function will be mainly described. The reason is that the correlator itself that calculates the correlation function, the signal represented by the function f1 (t) and the function f
This is because it is used for the purpose of determining the similarity or interdependence between signals represented by 2 (t), that is, the periodicity or phase relationship between two signals.

The equation represented by the equation (1) will be described below as a concrete correlator. In the corresponding function, the signal represented by the function f1 (t) is converted into S1 (t). And the signal represented by the function f2 (t) is S2 (t).

As can be seen from the equation (1), the correlator has a multiplier for multiplying the signal S1 (t) and the signal S2 (t) and an integrator for integrating the multiplication result. This is shown in FIG. 1, 2 are input terminals, 3 is a multiplier, 4 is an integrator, 5
Is an output terminal. The signal S1 (t) is input to the input terminal 1, and the signal S2 (t) is input to the input terminal 2. These input signals are multiplied by the multiplier 3 and then integrated by the integrator 4 to output the cross-correlation characteristic of the signal S1 (t) and the signal S2 (t) to the output terminal 5. Where the signal S
If 1 (t) and the signal S2 (t) are the same, the autocorrelation characteristic between the same signals is output to the output terminal 5.

Next, the operation of this correlator will be described after clarifying the signals to be handled. The signals S1 (t) and S2 (t) to be input may be analog signals or digital signals, but in order to simplify understanding, here, a digital signal having periodicity as shown in FIG. 26 is used. I will use it. This signal is a digital signal whose amplitude has two values of -1 and 1, and has a period of T. Further, in the period of the cycle T, a period of 1/7 is defined as Δ. That is, the period of the amplitude 1 is Δ, and Δ = T / 7. Further, the period of amplitude 0 is T-Δ, and T-Δ = 6
It is T / 7.

To obtain the autocorrelation, the signal S1 (t) and the signal S2 (t) are assumed to be the same two signals, and the phase difference τ between the two signals may be swept. Here, as a typical characteristic, the phase of the signal S2 (t) is 2 compared to the phase of the signal S1 (t).
2 shows the input / output signal waveforms of the multiplier 3 at this time by selecting Δ, Δ / 2, 0, −Δ / 2, the leading state.
7 (a) to (d). 27 (a) to (d)
In the figure, the signal S1 (t) is represented by A, the signal S2 (t) is represented by B, and the multiplication result of the signal A and the signal B is represented by C.

First, in the case of (a), the correlation value is obtained from the multiplication result. According to the equation (1), the correlation value is obtained by integrating the multiplication result over one cycle, and thus the period during which the amplitude value 1 is first output is calculated. This is 5Δ when calculated from the figure. Similarly, the period during which the amplitude value 0 is output is calculated. This is 2Δ. Next, the period in which the amplitude value 0 is output is subtracted from the period in which the amplitude value 1 is output, and this is divided by 1 cycle T. Then, (5
Δ-2Δ) / T is obtained. Since T = 7Δ, the calculated correlation value Rss is Rss (τ = 2Δ) = (5Δ-2Δ) / 7Δ = 3/7 (2). Similarly, when the correlation value in the phase relationship of (b) to (d) is obtained, Rss (τ = Δ / 2) = (6Δ-1Δ) / 7Δ = 5/7 (3) Rss (τ = 0) = (7Δ-0Δ) / 7Δ = 7/7 = 1 (4) Rss (τ = -Δ / 2) = (6Δ-1Δ) / 7Δ = 5/7 (5) is obtained.

Since these are typical correlation values at τ, they are discrete numerical values, but in practice, τ is continuously varied to obtain the correlation as shown in FIG. The characteristics are obtained. The vertical axis of FIG. 28 represents the correlation value, and the horizontal axis represents the phase difference τ between two input signals. This correlation characteristic is such that if the phase difference between two input signals is ± Δ or more, it shows a constant value, and if it is within ± Δ, a correlation value that continuously changes according to the phase difference is obtained. .

Consider the application of this characteristic to, for example, a line noise reduction device for television signals. In this case, assuming that one cycle of the signal of A in FIG. 27 is one line of the television signal, the state of amplitude 1 is white,
A state of amplitude -1 is assumed to be black. Since the television signal scans from the upper side to the lower side of the screen, if the correlation between an arbitrary one line and the next one line is the maximum in time, it means that the same signal is scanned for two lines. .
Here, it is assumed that some kind of pulse noise is mixed only in the signal period in which the signal of the first line on the first line represents white. However, since the noise mixing period is shorter than the video signal period, the correlation value as the integration result does not change significantly.

Therefore, consider the following noise reduction system. First, the noise reduction system determines whether to perform noise reduction based on the correlation value obtained from the correlator. If it is determined to be performed, the signal one line before and the current signal are added in time, and then the amplitude is halved and output. If it is determined not to perform, the signal is output without performing any calculation. That is, when the correlation value is high, the above calculation is performed, and when the correlation value is low, the calculation is not performed but the signal is output as it is.

By doing so, even if noise is mixed, if the correlation values of the two lines are high, it is determined that they are the same signal, and the signals are added and halved. By this operation, the signal amplitude does not change, but the noise is mixed in only one line, so it is reduced to ^{1/2 1/2} . On the other hand, when the correlation value is low, for example, when there is a large time difference between the signal periods representing white, if the above calculation is performed, white and black are mixed, and brightness that should not be present appears. I will end up. In order to prevent this, the above calculation is not performed when the correlation value is low.

As described above, the correlator provides a measure of similarity or interdependence between two functions or signals, so that the correlator is applied in many fields. As a familiar application product, a noise reduction circuit of a television receiver or a video tape recorder as described above is typical.

Further, a tracking circuit for signals called Phase Locked Loop is almost always used in consumer equipment. This is a basic circuit for obtaining a signal phase-locked with a reference frequency. It is also used for channel matching by television receivers and the like for radio waves sent from broadcasting stations, and also for rotation control of mechanical systems. Also in these PLLs,
The signal to be tracked and the reference signal are multiplied and integrated. That is, the correlator as described above is used.

As described above, the correlator has been indispensable for a circuit that detects a phase difference between two signals and a circuit that performs processing for matching the phases of two signals. It is a basic circuit and is used in various fields.

However, this correlator has major drawbacks. The phase information obtained from the correlation value is
It is only an absolute value. For example, it is assumed that the correlation value of 5/7 is detected by using the correlator that can obtain the characteristic of FIG. This is shown as P1 and P2 in FIG. That is, from the correlation value of 5/7, τ = + described in P1
Although the phase difference of Δ / 2 and the phase difference of τ = −Δ / 2 described in P2 can be detected, is it + Δ / 2 or −Δ / 2?
It means that it cannot be determined. This is obvious because the correlation characteristic is bilaterally symmetric with respect to the correlation value axis of τ = 0. Therefore, there are many cases where two conventional correlators are prepared, each of which is operated on a different time axis, and the phase is advanced or delayed by combining them.

[0017]

In the above-described conventional correlator, the phase information obtained from the correlation value is only an absolute value, and two correlators are prepared to judge the lead or lag of the phase. , It was necessary to operate them on different time axes and combine them.

The present invention provides a correlator that enables the correlator itself to determine whether the phase is advanced or delayed.

[0019]

In order to solve the above-mentioned problems, the correlator according to the present invention uses a first signal and a second signal as a function of a parameter τ representing the deviation of one signal with respect to the other. From the autocorrelation characteristic obtained by the calculation means for calculating τ = 0 and symmetric with respect to the amplitude axis at τ = 0, information is thinned out using the information thinning means to obtain the self-asymmetry with respect to the amplitude axis at τ = 0. It is characterized by obtaining a correlation characteristic.

With this configuration, information is thinned out at an arbitrary timing from the multiplication result of the first signal and the second signal, which is a function of the parameter τ representing the deviation of one signal with respect to the other. , The autocorrelation characteristics obtained when the first signal and the second signal are the same are asymmetric with respect to the correlation value axis at τ = 0. As a result, the correlator itself can judge whether the phase is advanced or delayed.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram for explaining an embodiment of the present invention. Reference numerals 10 and 11 are input terminals, 12 is a multiplier, 13 is an information thinning means for thinning information, 14 is an integrator, and 15 is an output terminal. It is assumed that two signals having the same signal but different phases are input to the input terminals 10 and 11. After the signals applied to the input terminals 10 and 11 are multiplied by the multiplier 12, the information is decimated by the information decimating means 13 and integrated by the integrator 14 to obtain the correlation characteristic between the both input signals at the output terminal 15. Will be available.

Now, in order to understand this operation, description will be made with reference to FIG. A is the signal S1 (t) input to the input terminal 10.
And B is the signal S2 (t) input to the input terminal 11. Further, it is assumed that the information thinning-out means 13 is controlled by a signal which is in phase with the signal S2 (t) shown in D, for example. As a concrete example of the control method, consider an operation in which the multiplication result of the signals A and B is output only during the period when the signal shown in D is 0 and the multiplication result is always 1 during the other periods. To do. This is equivalent to thinning out the multiplication result in the period of 1.

The phase difference τ between the signals A and B will be shifted to explain what kind of correlation characteristic is obtained. 2A to 2E respectively show the signal phase difference τ of B based on the signal phase of A by Δ, Δ / 2, 0, −Δ / 2, −.
It shows the signal waveform of each part when Δ. In the case of FIG. 2A, since the multiplication result is output during the period of one cycle T excluding the period of Δ / 2 which is phase-locked to B, the output waveform of the information thinning-out means 13 is the signal shown in E. It becomes a waveform. In this way, the same analysis is performed on FIGS. 2B to 2E. Next, the correlation value is obtained from the output terminal 15 by subtracting the period in which 0 is output from the period in which 1 is output in the waveform of E and dividing this by the period T. In the case of FIG. 2A, the correlation value is 4/7, and FIG.
6/7, 1 in FIG. 2 (c), 5/7 in FIG. 2 (d), and 4/7 in FIG. 2 (e).

The above analysis results are obtained by discretely obtaining a typical correlation value at τ. If τ is continuously varied, the correlation characteristics shown in FIG. 3 can be obtained. Similarly to FIG. 28, in FIG. 3, the vertical axis represents the correlation value and the horizontal axis represents the phase difference τ between two input signals. Also, for reference, FIG.
The correlation values obtained from (e) are P1 to P5 in FIG. 3, respectively.
It was noted in. That is, by providing the information thinning means 13, the correlation characteristic can be made asymmetric with respect to the correlation value axis of τ = 0.

In the above description, the information thinning means 13
In addition, a signal whose period of 1 is Δ / 2 is used in phase synchronization with the rising edge of the signal S2 (t). Next, as a signal for thinning out information, a signal delayed by Δ / 2 from the signal shown in FIG. FIG. 4 shows the signal waveform of each part, and FIG. 5 shows the correlation characteristics. As is clear from the difference in characteristics between FIG. 3 and FIG. 5, it is possible to create asymmetry of the correlation value by changing the timing of thinning out the information.

Consider using the correlator to track the phase of the signal B with the signal A described above. In the case of such an application example, first, the correlation characteristic of FIG. 3 is subtracted from the correlation characteristic of FIG. 5 obtained from the above analysis to generate the characteristic shown in FIG. From this characteristic, it can be determined that the correlation value is a positive value in the range of the phase difference of −Δ to 0 and the correlation value is a negative value in the range of the phase difference of 0 to Δ. For example,
The case where the correlation value is 1/14 is described in P1 and P2 of FIG. The phase difference of P1 is -Δ / 4, and the phase difference of P2 is -3Δ.
It is / 4, and it can be determined that the phase difference is negative.
Based on this judgment, if the signal phase of B with respect to the signal of A is advanced in the + direction, it is possible to quickly match the two.

That is, since the conventional correlator has the characteristic of being bilaterally symmetric with respect to the correlation value axis with the phase difference τ = 0, it is impossible to judge the lead or lag of the phase from the correlation value.
However, with the correlator of the present invention, it is possible to create a characteristic that is bilaterally asymmetric with respect to the correlation value axis with a phase difference τ = 0, and it is possible to judge the lead or lag of the phase from this correlation characteristic. In particular, by utilizing these correlation characteristics, it can be an extremely effective means for an application example in which one signal is phase-locked with the other signal.

Next, a specific example of the information thinning means will be described with reference to FIG. That is, the information thinning means 13 is O
It is composed of an R logical operation unit 16, and the output of the multiplier 12 and the timing signal input to the timing signal input input terminal 17 are input to the input of the OR logical operation unit 16,
The output is input to the input of the integrator 14.

1 except for the information thinning means 13 of FIG.
As described above, only the operation of the information thinning means will be described here.

Here, it is assumed that two signals having the same signal but different phases are input to the input terminals 10 and 11. Then, the multiplication result of the signals applied to both terminals becomes the signal waveform of C shown in FIG. The signal of C, which is the result of this multiplication, is input to one input terminal of the OR logic calculator 16. A signal phase-synchronized with the input terminal 11 is input to the other input terminal of the OR logical operation unit 16. Also in this case, the signal D in FIG.
Suppose you type in. Since the OR logic calculator 16 is a calculator that outputs Low to the output only when two input signals are Low, the output waveform of the OR logic calculator 16 is equal to the signal waveform of E in FIG. As a result, the obtained correlation characteristics are the same as those in FIG.

Similarly, if the signal D of FIG. 4 is input to the timing signal input terminal 17, the correlation characteristic equal to that of FIG. 5 is obtained. By advancing such an idea, it is possible to create a similar correlation characteristic by changing the logic of the information thinning means 13. For example, since the OR logical operation has the same logic as the NAND logical operation of negative logic, the information thinning means 13 shown in FIG.
It may be replaced with the ND logic calculator 18. Obviously, similar correlation characteristics can be obtained.

Next, an example in which the present invention is applied to a synchronous tracking circuit of a spread spectrum receiver will be described with reference to FIG. Before describing this embodiment, a spread spectrum communication system will be briefly described, and processing required on the transmitter side and the receiver side will be described.

In the spread spectrum communication system, the information signal to be transmitted is multiplied by a pseudo-noise code to spread and transmit the power spectrum of the information signal, and the receiving side multiplies the received signal again by the transmitting side. This is a communication method in which the spread spectrum is returned to the original state and an information signal is obtained by multiplying by the same pseudo-noise code. The pseudo noise code multiplication process performed on the transmission side is called spreading process,
The pseudo noise code multiplication process performed on the receiving side is called despreading process. The spreading process and the despreading process will be described with reference to FIG.

FIG. 10A shows a spreading process performed on the transmitting side, and FIG. 10B shows a despreading process performed on the receiving side. A is an information signal, B is a pseudo-noise code, and C is a transmission signal obtained as a result of multiplying A and B. Here, for ease of understanding, it is assumed that the information signal A is a digital signal that always has an amplitude value of 1.
Further, as the pseudo noise code, one pattern of the pseudo noise code obtained from the third-order primitive polynomial is assumed. Further, one period of the pseudo-noise code is T, one period is composed of 7 chips, and the time of one chip is defined as Δ. That is, T = 7Δ.

Since the information signal A is 1, the transmission signal C multiplied by the pseudo noise code B becomes a signal equal to the pseudo noise code B, which is transmitted from the transmitter.
The receiver receives this transmission signal, and multiplies the reception signal C by the pseudo noise code D having the same phase as that used for the spreading process on the transmission side, but the E information signal is obtained. As is clear from these, even if the same code is used for the despreading process as the spreading process, if the multiplication timing, that is, the phase is different, the signal E and the signal A are not the same signal, and the restoration of information is not possible. It will not be possible.

Therefore, the receiver must generate and track a pseudo noise code that is always in phase with the pseudo noise code sent from the transmitter side. The synchronous tracking circuit plays this role.

The operation of this synchronous tracking circuit will be described. From FIG. 9, only the synchronous tracking circuit is cut out in FIG. FIG. 11 shows a sliding correlation type synchronous tracking circuit generally used in a spread spectrum receiver.

Numeral 30 is a reception signal input terminal, and 31, 32, 3
3 is a multiplier, 34 is a pseudo noise code generator, 35 and 36 are integrators, 37 is an adder, 38 is a voltage controlled oscillator, and 39 is an output terminal for a despread signal. First, the correlator, which is a part of FIG. 11, will be described. In the following description, the signal used in the spreading process performed on the transmission side is the information signal of the amplitude value 1 which is also used in the above description. And a pseudo-noise code of the third order.

In FIG. 11, the correlator is a multiplier 32 and an integrator 3.
5 or a multiplier 33 and an integrator 36. These correlators are exactly the same as those described in the conventional example, but since the signal input to the correlator is a pseudo noise code in this case, the pseudo noise code is input again here. The operation and correlation characteristics of the correlator at that time will be described.

FIG. 12 shows the multiplication result of two signals and two pseudo noise codes input to the correlator. Let A be the pseudo-noise code given to one input terminal of the correlator, and let B be the other pseudo-noise code, and let τ be the phase difference of B with respect to A.
As a signal waveform of a typical value of τ is shown in (a) to (e). When considering the correspondence with FIG. 11, A in FIG.
The pseudo noise code described in 1 is input to the reception signal input terminal 30, and the pseudo noise code described in B is the pseudo noise code generator 3.
4 may be interpreted. Since C in FIG. 12 represents the multiplication result of the pseudo noise codes A and B, it can be interpreted as the output signal of the multiplier 32 or 33. Also, the signal of C, which is the result of this multiplication, is integrated over one period of the pseudo noise code, and the result is integrated as a correlation value in the integrator 35 or 36.
Output signal.

The signal waveforms shown in FIGS. 12 (a) to 12 (e) are discrete correlation values obtained at a typical τ, but the characteristic obtained by continuously varying τ is shown. FIG.
It was shown to. As described in the conventional example, it can be seen that the correlation characteristic is symmetrical with respect to the correlation value axis of τ = 0.

Since the correlation characteristics of the correlator can be grasped, FIG.
The tracking operation of No. 1 will be described. The pseudo noise spreading code generator 34 of FIG. 11 has three output terminals, and the respective output signals are connected to the multipliers 31 to 33. Here, it is assumed that the pseudo noise codes output from these three output terminals are pseudo noise codes whose phases are shifted by Δ from each other, as shown in FIG. Corresponding these phase relationships to FIG. 11, the pseudo noise code A is input to the multiplier 33, the pseudo noise code B is input to the multiplier 31, and the pseudo noise code C is input to the multiplier 32. Become. As a result, the correlation characteristic obtained through the multiplier 33 and the integrator 36 has a phase relationship obtained by advancing the correlation characteristic obtained through the multiplier 32 and the integrator 35 by 2Δ with respect to time. The results are shown in (a) and (c) of FIG. The characteristic of (a) is the correlation characteristic obtained from A of FIG. 14, and the characteristic of (c) is the correlation characteristic obtained from C. In FIG. 11, by the adder 37, (c)
Since the characteristic of (a) is subtracted from the characteristic of, the S-shaped characteristic as shown in FIG. 16 is obtained.

The vertical axis of FIG. 16 is the correlation value axis, which is a signal that can be taken out as a voltage in an actual circuit. Therefore, the voltage control oscillator 38 is controlled by this voltage signal, and the oscillation frequency output from the voltage control oscillator 38 is controlled. By controlling the code rate output from the pseudo noise code generator 34 in accordance with the above, it is possible to track the phase at the point of τ = 0 in the S-shaped characteristic. Now, for the signal input to the receiver input terminal 30, the pseudo spread code generator 34 to the multiplier 31
It is assumed that the pseudo spread code output to the is advanced by Δ / 2. At this time, the correlation value obtained from the adder 37 is shown in FIG.
6 indicates a circle.

If the characteristic of the voltage controlled oscillator 38 is shown in FIG.
Assuming that the characteristic is as shown in (1), the oscillation frequency output from the voltage controlled oscillator drops. If the pseudo noise code generator 34 generates a pseudo noise code according to the frequency output from the voltage controlled oscillator 38, the frequency of the pseudo noise code output from the pseudo noise code generator 34 will be low. That is, the phase of the pseudo spread code is delayed with respect to the signal input to the receiver input terminal 30. As a result, the phase difference is controlled to be zero.

The above is the description of the tracking operation, but it is the despread output terminal 39 that the information signal is obtained. This can be understood as follows. Looking at the point of τ = 0 on the mutual correlation characteristics, it can be seen that it is located at exactly half of the shift amount of τ. The pseudo noise code shown in FIG. 14 has the signal phase shown in B of FIG. That is, the received signal input to the input terminal 30 of the receiver matches the pseudo noise code shown in B of FIG. 14, and the despread information signal can be extracted at the output terminal 39 of the multiplier 31. Become.

The receiver of the spread spectrum communication device always causes the receiver side to generate the pseudo noise code phase-synchronized with the pseudo noise code on the transmission side by the above synchronization tracking circuit.
The signal is subjected to despreading processing. However,
The signal that is actually despread is output to the output terminal 39 of FIG.
Output from As is clear from the circuit diagram, the tracking loop and the despread output signal have an off-loop relationship in which there is no feedback from the despread output signal to the tracking loop, and it can be assumed that the tracking circuit has driven the phase difference to 0. However, due to the mismatch of the elements forming FIG. 11 and the wiring delay, the phases of the two signals multiplied by the multiplier 31 are not aligned, and a complete information signal cannot be restored.

Therefore, it is considered that everyone detects a phase shift from the despread signal and performs on-loop phase control. However, since the correlation characteristic obtained from the conventional correlator that only integrates the despread signal is bilaterally symmetric with respect to the correlation value axis of τ = 0, whether the phase should be advanced or delayed. It was impossible to control.

FIG. 9 shows the application of the correlator of the present invention to such phase control. FIG. 9 shows the synchronous tracking circuit described with reference to FIG. 11 provided with the correlator of the present invention, and the on-loop control of the phase control of the despread output signal.
In FIG. 9, the same components as those of FIG. 11 are designated by the same reference numerals and the description thereof will be omitted. In FIG. 9, 40 is a phase inverter, 41,
42 is an OR logic operator, 43 and 44 are integrators, 45 is an adder, 46 is a phase controller, 47 is a first correlator, and 48 is a second correlator.

In this application example, two sets of the correlators of the present invention are provided. The first set is a first correlator 47 including a multiplier 31, an OR logic calculator 42, and an integrator 44, and the second set is a multiplier 31, an OR logic calculator 41, and an integrator 43. The second correlator 48 is configured. First, the operation of the first correlator 47 will be described with reference to FIGS. 18 and 19.

In FIGS. 18 and 19, as in the above description, the pseudo noise code input to the correlator is one pattern of the M sequence code obtained from the cubic polynomial of the third order, and one cycle thereof is T and one chip. Is Δ, and the phase difference between the two pseudo noise codes is τ
Is defined and the signal waveform of each part of the correlator is shown.
A is a pseudo noise code given to the receiver input terminal 30, and B is
Is a pseudo-noise code given to the multiplier 31, C is the multiplication result of A and B, D is a clock signal output from the voltage controlled oscillator 38, and E is an OR logical operation output signal of C and D. That is, an OR logic calculator is used as the information amount thinning means, and the output signal from the voltage controlled oscillator 38, which is a drive signal of the pseudo noise code, is used as the signal phase-locked with the pseudo noise code.

18 (a)-(d) and FIG. 19 (e)-
(G) shows that τ is Δ <τ, Δ / 2 <τ <Δ, 0 <
τ <Δ / 2, τ = 0, −Δ / 2 <τ <0, −Δ <τ <−
The signal waveforms are divided into .DELTA. / 2 and .tau. <-. DELTA., But since the operation itself is multiplication and OR logic operation, the explanation is omitted and only the correlation characteristic is mentioned.

FIG. 20A shows the correlation characteristic obtained by the first correlator 47. τ is τ <-Δ /
In the range of 2 and Δ <τ, the correlation value becomes a constant value of 3/7,
-Linear increase in the range of Δ / 2 <τ <0, 0 <τ <Δ / 2
In the range of 1, the constant value is 1, and in the range of Δ / 2 <τ <Δ, there is a characteristic of linear decrease. This property can also be interpreted as follows. Considering that the information thinning means is a driving clock signal of a pseudo noise code, the information when the duty of the clock signal is 50% is reduced to half of one chip. That is, the information in the range of 0 to Δ / 2 is thinned out and the information in the range of Δ / 2 to Δ is output.

Since the correlation characteristic in which information is not thinned is shown in FIG. 13, from this characteristic, the correlation value in the period in which information is obtained remains unchanged, and the correlation value in the period in which information is thinned is considered to be constant in FIG. The characteristics of (a) can be drawn.

The second correlator 48 inverts the phase of the timing signal for thinning out the information, that is, the clock signal shown in D of FIGS.
It is input to the logical operation unit 41. This signal waveform is shown in Figure 1.
8, shown in F of FIG. G is an OR logical operation result of C and F. The correlation characteristic of the second correlator 48 is shown in FIG.
(B). The characteristic of (b) of FIG. 20 is nothing but the characteristic of (a) delayed by Δ / 2.

Therefore, the correlation characteristic obtained from the adder 45 is as shown in FIG. This characteristic is nothing but the S-shaped characteristic on the time axis required for the tracking operation. In other words, the despread output signal can be used to create a correlation characteristic in which the lead or lag of the phase can be determined.
Once this correlation characteristic is created, the phase controller 46 then determines the output voltage of the adder 45 and feedback-controls the voltage-controlled oscillator 38 to enable on-loop control of the despread signal.

For reference, the correlation value of the characteristic shown in FIG. 20A is expressed by a mathematical expression. Here, if the degree of the primitive polynomial of the M-sequence code that is the pseudo-noise code used as the input signal of the correlator is expressed as k, the maximum value of the correlation value is (6)
The equation and the minimum value are represented by the equation (7).

Maximum correlation value: [1 / (2 ^k -1)] (2 ^k -1) (6) Minimum correlation value: [1 / (2 ^k -1)] [(2 ^k / 2) -1] (7) A second application example of the present invention will be described with reference to the block diagram of FIG. FIG. 22 shows a system for detecting the direction and amount of movement of an image, which can be used as a motion detector for moving images. Before explaining FIG. 22, the image and video signals shown in FIG. 23 will be briefly described.

FIG. 23 (a) shows an image displayed on the television. For ease of understanding, the image is an image of a human in white clothes on a black background image. From this video signal, FIG. 23 shows one line of the video signal corresponding to the position P in FIG.
(B). A horizontal sync signal is present on one line of the actual video signal, but in this case, it represents only the video period displayed on the television screen. The brightness of the video signal is expressed by the DC voltage of the video signal (luminance signal), so the black background of the screen is expressed by the black level of the video signal,
White people are represented as white levels. Here, the video period is defined as T, and the width of the person, that is, the white level video period is defined as Δ.

FIG. 23 (c) shows a video signal delayed by one frame from the video signal of FIG. 23 (a). An image in which a human moves to the left as compared with the image of FIG. 23 (a) is shown. Shows. The image represented by the dotted line is the image one frame before, and the image represented by the solid line is the current image. From this video signal, one line of the video signal corresponding to the position P ′ is shown in FIG. 23D, and the video signal period is T.

The operation of FIG. 22 will be described. In FIG. 22, 50 is a video signal input terminal, 51 is a first counter, 52
Is a shift register, 53 is a second counter, 54 is a frame delay line, 55 is a line delay line, 56 is a multiplier, and 57.
Is an exclusive OR calculator, 58 and 59 are timing signal input terminals, 60 and 61 are OR logic calculators, 62 to 64 are integrators, 65 is an adder, 66 is an amplitude value detector, and 67 is an amplitude value detector. Is a direction detector, 68 is a direction detector, 69 is a direction detector output terminal, and 70 is a timing signal generation circuit.

The timing signal generating circuit 70 is a circuit for generating a timing signal for thinning out the information necessary for creating the asymmetry of the correlation characteristic, and the line delay line 55.
And a first counter 51, a shift register 52, a second counter 53 and an exclusive OR calculator 57.

When the video signal is input to the video signal input terminal 50, the video signal is input to the line delay line 55 in the timing signal generating circuit 70 and the first counter 51. Here, the first counter 51, the shift register 52, and the second counter 53 are set as follows. First, the first counter 51 is an up counter and the second counter 53 is a down counter. Further, the same clock signal is input to the first counter 51, the shift register 52, and the second counter 53, and the counting operation of the counter and the shift operation of the shift register are performed by this clock signal. For the clock signal, for example, a subcarrier frequency used for color signal processing of a video signal can be used.

Next, the input signal of the first counter 51 is H level.
When it becomes igh, it starts counting up, and when it becomes Low, it ends counting. In addition, the second counter 5
No. 3 starts counting down when the input signal becomes high, outputs high to the output, and outputs low when the countdown ends, and both counters are reset at the rising edge of the horizontal synchronizing signal existing in the video signal. To be set. Further, the shift register 52 is set to read the data of the first counter 51 at the falling edge of the horizontal synchronizing signal, and the second counter 53 is set to read the data from the shift register 52 at the rising edge of the horizontal synchronizing signal. .

Now, a series of operations of the timing generating circuit 70 will be sequentially described with the passage of time. First, when the video signal applied to the video signal input terminal 50 becomes High, the first counter 51 starts counting up, and when the video signal becomes Low after a lapse of Δ time, the counting ends.
This count data is read into the shift register 52 at the falling edge of the horizontal synchronizing signal, and data is shifted at the next clock. This data-shifted data is read into the second counter 53 at the rising edge of the horizontal synchronizing signal, and the video signal that has passed through the line delay line 55 is H level.
When it becomes igh, at the same time, the second counter 53 starts counting down and outputs High to the output terminal, and at the same time when the counting down ends, the output terminal is set to Low. Since the video signal from the line delay line 55 and the output of the Δ / 2 width second counter 53 synchronized with the rising edge of the video signal are input to the exclusive OR calculator 57, the output is The signal has a Δ / 2 width that rises with a delay of Δ / 2 from the rising edge of the video signal.

That is, the width Δ of the person is measured from the video signal supplied to the video signal input terminal 50, and this value is halved to be read by the second counter 53. On the other hand, the video signal of which the width of the person is measured is delayed by one line by the line delay line 55 and input to the second counter 53. The second counter 53 outputs High when the video signal from the line delay line 55 becomes High, and Δ / 2
Low is output after passing the width period. Exclusive OR calculator 5
From 7, the output from the second counter 53 becomes a signal of Δ / 2 width which rises at the same time when it falls.

A method of thinning out information from the video signal using the timing signal thus obtained will be described.

The signals input to the multiplier 56 are a video signal which has passed through the line delay line 55 and a video signal which is obtained by delaying this video signal by one frame. The video signals having different one frame times are multiplied by the multiplier 56, and the integrator 6
3 and OR logic operation units 60 and 61. Similar to the conventional correlator, it is the output of the integrator 63 that obtains a correlation characteristic which is bilaterally symmetrical with respect to the correlation value axis where the phase difference between the two input signals is 0. On the other hand, the information thinning means for creating the asymmetrical correlation characteristic is ORed.
The timing signal generating circuit 70 corresponds to the logical operation units 60 and 61 and generates a timing signal for thinning out information.
It is.

FIG. 24 shows the timing relationship of these signals. FIG. 24 (c) is similar to FIG. 23 (c) and shows one frame before and the current image. 1
The image before the frame is drawn with a dotted line, and the current image is drawn with a solid line. FIG. 24B shows a video signal period of one line from the video signal of one frame before, which is a video signal output from the frame delay line 54. FIG. 24D shows a video signal period of one line from the current video signal, which is a video signal output from the line delay line 55.

A in FIG. 24 (a) corresponds to FIG. 24 (b) and FIG.
4 (d), which is an output signal of the multiplier 56. B is a timing signal output from the second counter 53, and D is a timing signal output from the exclusive OR calculator 57. C is the result of thinning out the information by the timing signals shown in the multiplied video signals A to B, and is output from the OR logic operation unit 61. Further, E is a result of thinning information by the timing signals shown in the multiplied video signals A to D, and OR
It is output from the logical operation unit 60.

With this configuration, the output signal of the multiplier 56 is input to the integrator 63 without thinning out the information. Therefore, the output of the integrator 63 is the video signal of one frame before and the current video signal. When the deviation amount is 0, the correlation characteristic is symmetrical with respect to the correlation value axis. On the other hand, the signals whose information has been thinned out separately by the OR logical operation unit 61 and the OR logical operation unit 60 are integrated by the integrator 64 or the integrator 62, so that the video signal of one frame before and the current When the amount of deviation between video signals is 0, the correlation characteristic is asymmetric with respect to the correlation value axis.

Then, assuming that the time difference between the video signal of one frame before and the current video signal is δ, the integrators 62-
The correlation value output from 64 will be calculated. First, the correlation value is calculated from the integrator 63 from the signal waveform of A in FIG. Since the correlation value is obtained by subtracting the period of Low from the period of High as a result of multiplication and dividing this by 1 cycle T, when this is expressed by the equation, it becomes equation (8).

A: 1− (4δ / T) (8) Similarly, the correlation value obtained from the integrator 64 is given by the equation (9),
The correlation value obtained from the integrator 62 is shown in equation (10).

C: (1/2) [1- (4δ / T)] (9) E: 1- (4δ / T) (10) In the case of a video signal, each frame is composed of elements constituting an image. It is rare that a moving image pattern is different and continuous correlation characteristics are obtained. However, theoretically, as shown in FIG.
Similar correlation characteristics to those described above can be obtained. That is,
From the output terminal of the integrator 63, the correlation characteristic similar to FIG. 28,
That is, a correlation characteristic is obtained that is bilaterally symmetric with respect to the correlation value axis with a time difference δ of 0. From the integrator 64 and the integrator 62, a correlation characteristic similar to that of FIGS. 3 and 5, that is, a time difference δ of 0 is obtained. A correlation characteristic that is asymmetric with respect to the correlation value axis is obtained. Regarding these asymmetries, the above equation (9) and (1
It is also clear from the fact that the correlation values obtained from the equation (0) are different. Further, the characteristic of the integrator 64 is subtracted from the characteristic of the integrator 62 from the adder 65, and the correlation characteristic is similar to that in FIG.

That is, if there is movement of the image (time difference δ) between the video signal of one frame before and the current video signal,
By detecting the correlation value in the amplitude value detection circuit 66,
The moving distance (time difference δ) of the image is obtained, and if the direction detection circuit 68 determines whether the correlation value is positive or negative, the moving direction of the image is obtained. The moving distance and the moving direction are the output terminal 67 of the amplitude value detecting circuit and the direction detecting circuit 6, respectively.
Since this information is used, it is possible to use this information to perform adaptive image processing from the distance and direction of movement, and also to predict the distance and direction of video that will move in the future. Becomes

When the conventional correlator was used, the moving distance could be judged, but the moving direction could not be judged, so that there is a limit to the adaptive processing of the moving video signal portion. It was However, by using the correlator of the present invention, it is possible to determine the moving direction of the image. As a result, when it is used in a circuit that three-dimensionally separates a luminance signal and a color signal, it is possible to perform image motion adaptive processing.

As described above, the correlator of the present invention makes it possible to create a characteristic which is bilaterally asymmetric with respect to the correlation value axis where the phase difference between the input signals to the correlator is 0. Leading or lagging, or a phase shift direction can be determined. Therefore, when the correlator of the present invention is applied to, for example, a synchronous tracking circuit in a spread spectrum communication receiver, it becomes possible to perform on-loop control of the phase, which has been impossible in the past, and thereby phase distortion of the despread signal is suppressed. It is possible to keep it extremely small.

When applied to a video signal processing circuit or the like, it is possible to detect the moving direction of an image, which has been impossible in the past, and it is possible to predict the movement of the video signal. This can be applied to, for example, a three-dimensional separation process of a luminance signal and a color signal, a moving image tracking system such as a camera movie, or the like.

[0078]

As described above, according to the correlator of the present invention, it is possible to judge whether the phase is advanced or delayed.
In particular, it is extremely effective for applications in which one of the signals is phase-synchronized with the other signal by utilizing these correlation characteristics.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining an embodiment of the present invention.

FIG. 2 is a signal waveform diagram for explaining the operation of FIG.

3 is a characteristic diagram of the output signal of FIG. 1 when the signal waveform of FIG. 1 is set to FIG.

FIG. 4 is a signal waveform diagram for explaining the operation of FIG.

5 is a characteristic diagram of the output signal of FIG. 1 when the signal waveform of FIG. 1 is changed to FIG.

FIG. 6 is a characteristic diagram created by subtracting the correlation characteristic of FIG. 3 from the correlation characteristic of FIG.

7 is a block diagram for explaining a specific example of the information thinning means of FIG.

8 is a block diagram for explaining another specific example of the information thinning means of FIG. 1. FIG.

FIG. 9 is a block diagram for explaining an application example of the present invention.

FIG. 10 is a signal waveform diagram for explaining spread / despread processing of the spread spectrum communication device.

FIG. 11 is a block diagram in which a part of the circuit of FIG. 9 is omitted.

FIG. 12 is a signal waveform diagram of a pseudo noise code and an output signal used in FIG. 11.

FIG. 13 is an explanatory diagram for explaining a correlation characteristic of the spread spectrum receiver.

FIG. 14 is a signal waveform diagram used for explaining the operation of FIG. 11.

FIG. 15 is an explanatory diagram for explaining a correlation characteristic of a spread spectrum receiver.

FIG. 16 is an explanatory diagram for explaining a correlation characteristic of a spread spectrum receiver.

FIG. 17 is a characteristic diagram of the voltage controlled oscillator of FIG. 11.

FIG. 18 is a signal waveform diagram for explaining the operation of the first correlator in FIG.

FIG. 19 is a signal waveform diagram for explaining the operation of the first correlator in FIG.

FIG. 20 is a characteristic diagram obtained from the first correlator of FIG. 9.

21 is a correlation characteristic diagram obtained from FIG.

FIG. 22 is a block diagram for explaining a second application example of the present invention.

23 is an explanatory diagram for explaining the operation of FIG. 22. FIG.

FIG. 24 is a signal waveform diagram for explaining the operation of FIG. 22.

FIG. 25 is a block diagram for explaining a conventional correlator.

FIG. 26 is a signal waveform diagram for explaining the operation of FIG. 25.

FIG. 27 is a signal waveform diagram for explaining the operation of FIG. 25.

FIG. 28 is a characteristic diagram for explaining a characteristic example of the correlator in FIG. 25.

[Explanation of symbols]

10, 11 ... Input terminal, 12 ... Multiplier, 13 ... Information thinning means, 14 ... Integrator, 15 ... Output terminal.

Claims (17)

[Claims] 1. A calculation means for calculating a first signal and a second signal, which is a function of a parameter τ representing a deviation of one signal with respect to the other, and is symmetric with respect to an amplitude axis at τ = 0. A correlator that obtains an autocorrelation characteristic asymmetric with respect to the amplitude axis at τ = 0 by thinning out information using the information thinning means from the autocorrelation characteristic. 2. A first signal f1 (t) and a second signal f2 (t) as a function of a parameter τ representing the deviation of one signal with respect to the other are expressed as follows: From the autocorrelation characteristic which is derived by substituting it into the arithmetic expression of ## EQU1 ## and which is symmetric with respect to the amplitude axis at τ = 0, the information is thinned out by using the information thinning means, and thus the asymmetry with respect to the amplitude axis at τ = 0 A correlator characterized by obtaining an autocorrelation characteristic. 3. A first signal and a second signal, which are functions of a parameter τ representing the deviation of one signal with respect to the other, are input, and information is thinned out from the first and second input signals. By thinning out the first signal and the second signal
A correlator characterized in that the autocorrelation characteristics obtained when the signals of 1 are the same are asymmetrical with respect to the amplitude axis at τ = 0. 4. A first signal and a second signal, each of which is a function of a parameter τ representing the deviation of one signal with respect to the other, are input, the first and second signals are multiplied, and the information thinning means is used. It is characterized in that the autocorrelation characteristic obtained when the first signal and the second signal are the same by integrating after thinning out the information is asymmetric with respect to the amplitude axis at τ = 0. And the correlator. 5. A first signal and a second signal, each of which is a function of a parameter τ representing a deviation of one signal with respect to the other, are input, the first and second signals are multiplied, and the multiplication result is obtained. When the first signal and the second signal are the same, by performing a logical operation with the third signal that is phase-locked with either the first signal or the second signal, and then performing integration. A correlator characterized in that the autocorrelation characteristic obtained in (1) is asymmetric with respect to the amplitude axis at τ = 0. 6. A first signal and a second signal, each of which is a function of a parameter τ representing a deviation of one signal with respect to the other, are input, the first and second signals are multiplied, and the multiplication result is obtained. Obtained when the first signal and the second signal are the same by performing a logical operation with a clock signal that is phase-synchronized with either the first signal or the second signal, and then performing integration. A correlator characterized in that the obtained autocorrelation characteristic is asymmetric with respect to the amplitude axis at τ = 0. 7. A thinning means for multiplying a first signal and a second signal, and thinning out information by a signal which is phase-synchronized with either the first signal or the second signal, A correlator, characterized in that a correlation characteristic of the first and second signals is obtained by integrating a decimated signal of the information. 8. The correlator according to claim 1, wherein the first signal and the second signal are pseudo noise codes. 9. The correlator according to claim 1, wherein the first signal and the second signal are pseudo-noise codes obtained by inputting a clock signal into a feedback register. . 10. The correlator according to any one of claims 1 to 4 and 7 to 9, wherein an OR logic operator is used as the information thinning means. 11. The correlator according to claim 5, 6, 8 or 9, wherein the logical operation is OR logic. 12. The correlator according to any one of claims 1 to 4 and 7 to 9, wherein a NAND logical operation unit is used as the information thinning means. 13. The correlator according to claim 5, 6, 8 or 9, wherein the logical operation is NAND logic. 14. The correlator according to claim 4, wherein the multiplication is an exclusive OR. 15. A signal set by an arbitrary function and an input signal are the same by setting an arbitrary function, performing a multiplication operation with an input signal, and thinning and integrating information from the multiplication result. The autocorrelation characteristic obtained at
A correlator characterized by being asymmetric with respect to the amplitude axis at. 16. A receiver receives a transmission signal, which has been subjected to spread processing of information by a pseudo noise code and is transmitted, and multiplies the reception signal by a pseudo noise code having the same code as the pseudo noise code but having the same phase. In a spread spectrum receiver that obtains information by performing despreading processing, in order to suppress the phase distortion of the information obtained by performing the despreading processing, a pseudo noise code included in the transmission signal and a pseudo noise code used in the despreading processing. By thinning and integrating the information from the multiplication result of, the autocorrelation characteristics of the pseudo noise code included in the transmission signal and the pseudo noise code used in the despreading process are correlated with the correlation value axis of the same phase. A correlator characterized in that the phase advance or delay is judged from this asymmetry and the pseudo noise code used in the despreading process is phase-controlled. 17. A first timing signal in which a video signal is input, a period during which an arbitrary luminance signal level is obtained is measured, and a half period of the measurement period from the beginning of the measurement period becomes High, and the measurement is performed. A second timing signal is generated in which half of the measurement period becomes High after half of the measurement period has elapsed from the beginning of the period, and the second timing signal is generated by the first timing signal and the second timing signal. The autocorrelation characteristics of the video signals obtained by thinning and integrating the information from the multiplication results of the video signals different by 1 frame from each other are shown in FIG. A correlator which has an asymmetric correlation characteristic and judges the direction of a temporally moving image from this asymmetry.