JP3248261B2 - Signal equalization method, signal equalizer, and video tape recorder - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a signal equalizing method, a signal equalizing apparatus, and a video tape recorder suitable for digitally recording and reproducing a video signal on a magnetic tape, for example.

[0002]

2. Description of the Related Art FIG. 3 is a block diagram showing an example of the configuration of a conventional video tape recorder. Video signals are digitally recorded on the magnetic tape 1. The magnetic head 2 generates a signal recorded on the magnetic tape 1, and the reproducing amplifier 3 amplifies the output of the magnetic head 2 to
/ D converter 4. The A / D converter 4 A / D converts the input signal and outputs the signal to a signal equalization circuit (EQ) 5. The signal equalization circuit 5 compensates for the amplitude and phase of the input signal and outputs the signal to the decoder 6. The decoder 6 decodes the input signal to logic 1 or 0 and outputs it to an error detection and correction (ECC) circuit 7. The microcomputer 8 calculates the coefficient of the multiplier of the signal equalization circuit 5 according to the data input from the A / D converter 4 via the I / O bus, and outputs the coefficient to the multiplier. .

The microcomputer 8 is shown in FIG.
It has a program for performing the same processing as that in the circuit including the adaptive controller 11, the FIR transversal filter 12, the decoder 13, and the subtractor 14 (in other words, the circuit substantially shown in FIG. are doing). The FIR transversal filter 12 has the same configuration as the FIR transversal filter 30 shown in FIG.

[0004] As shown in FIG. 4, FIR-type transversal filter 12 of the microcomputer 8, in this example, a delay circuit 21 ₁ to 21 ₄ for sequentially delaying and outputting an input signal by one clock minute, the delay multipliers 22 ₀ to 22 ₄ for multiplying a predetermined coefficient W ₀ through W ₄ for input and output of the circuit 21 ₁ to 21 _4, is constituted by an adder 23 for adding the outputs of the multipliers 22 ₀ to 22 ₄ I have.

[0005] Similarly, as shown in FIG. 5, FIR-type transversal filter 30 of the signal equalizer circuit 5, in this example, the delay circuits 31 ₁ to sequentially delay and outputs an input signal by one clock minute 31 ₄ and a delay circuit 31 ₁ to 31 ₄ of the input and output multipliers 32 ₀ to 32 ₄ for multiplying a predetermined coefficient W ₀ through W _4, the multipliers 32 ₀ to 32 ₄ adder 33 for adding the output of the It is composed of

Next, the operation will be described. The video signal recorded on the magnetic tape 1 is reproduced by the magnetic head 2, amplified by the amplifier 3 and input to the A / D converter 4. The A / D converter 4 converts the input signal into an A / D signal.
The data is D-converted and output to the microcomputer 8.

The microcomputer 8 performs the same processing as the processing performed by the filter 30 having the configuration shown in FIG. 5 on the data input from the A / D converter 4 in a software manner according to the flowchart of FIG. I do.

[0008] In the microcomputer 8, the delay circuit 21 ₁ to 21 ₄ of the filter 12, generates the same data as when it is sequentially delayed by one clock minute.
As a result, an input data string X _(n) represented by the following equation is obtained (step S1). X _(n) = [x _(n) , x _(n-1) , ... x _(n-4) ] ...
(1)

On the other hand, the multipliers 22 ₀ to 22 ₄ tap coefficients W ₀ to be loaded into _{(n), W 1 (n} ), constituted by · · · W 4 _(n), the tap coefficients represented by the following formula Generate a vector W _(n) . W _(n) = [W0 _(n) , W1 _(n) , ... W4 _(n) ] ...
(2)

[0010] Next, obtained when the multiplier 22 ₀ to 22 _4, the input and output of the delay circuits 21 ₁ to 21 _4, were added by the adder 23 the value output by multiplying each of these tap coefficients The obtained output y _(n) is calculated according to the following equation (step S2). y _(n) = W _(n) X ' _(n) (3) Here, X' _(n) represents a transposed matrix of the vector X _(n) of the input data sequence.

Next, in step S2, the adder 23
An error ε _(n) is calculated by the subtractor 14 from the value y _(n) obtained in the above and the target value d _(n) (step S3).

The target value d _(n) is calculated by the decoder 13 as y
The value of _(n) is decoded with reference to a predetermined threshold TH set in advance, and is obtained as follows.

That is, d _(n) is set to 1 when y _(n) > TH, and is set to -1 when y _(n) <-TH, and -TH <y
_(n) It is set to 0 when <TH. Then, in each of these cases, the error ε _(n) is obtained from the following equations. When y _(n) > TH, ε _(n) = d _(n) −y _(n) = 1−y _(n) (4) When y _(n) <− TH, ε _(n) = d _(n) −y _(n) = − 1−y _(n) (5) When −TH <y _(n) <TH ε _(n) = d _(n) −y _(n) = 0− y _(n) = −y _(n)・ ・ ・
(6)

Next, according to the LMS method (least mean square method) as a gradient method, the average 2
When the multiplication error as E [{ε _(n)} ^2], the following coefficients of the multipliers 22 ₀ to 22 ₄ in accordance with the following equation W _{(n + 1)} is calculated by the adaptive controller 11. W _{(n + 1)} = W _(n) − (α / 2 ₎ ∇W _(n) E [｛ε _(n) ｝ ² ] = W _(n) −αE [ε _(n) ∇W _(n) ｛ ε _(n) ｝] = W _(n) −αE [ε _(n) X _(n) ] (7) where α is a step size, and is a magnitude that falls down a gradient (gradient) at a time. A positive number to determine.

However, when the state of the signal input from the A / D converter 4 changes with time, the expected value of the signal E [ε _(n) X _(n) ] also changes. Therefore, the unbiased variance ε _(n) X _(n) is used instead of the expected value. Thus, the above equation (7) can be rewritten as follows. W _{(n + 1)} = W _(n) −αε _(n) X _(n) (8)

According to this equation, the microcomputer 8
Each tap coefficients of the software-multiplier 22 ₀ to 22 ₄ in the inner is updated by adaptive controller 11 (step S4).

Then, the coefficients calculated in this manner are loaded into an actual multiplier of the signal equalizing circuit 5 (step S5).

As described above, the coefficients calculated by the microcomputer 8 are loaded into the respective multipliers of the filter 30 of the signal equalization circuit 5, and the data output from the A / D converter 4 are added to the coefficients. Are multiplied to perform equalization. Then, the equalized output is output to the decoder 6.
The decoder 6 compares the output of the filter 30 with a predetermined threshold value, and determines (decodes) the logic value of 1 or 0. The output of the decoder 6 is input to the ECC circuit 7, subjected to error detection and correction processing, and output to a circuit (not shown).

The above operation is repeated, and the recording / reproducing process of the partial response class IV is performed by the magnetic recording / reproducing system including the signal equalizing circuit 5.

[0020]

Conventionally, as described above, L
The MS method, the filter 30, because you have to update the coefficients of the multipliers 32 ₀ to 32 _4, as parameters for the algorithm, and the step size α for determining the rate of convergence, a predetermined value of time for updating the coefficients Must be set to However, it is relatively difficult to set these parameters to appropriate values, and if a substantially zero level no-signal portion is played back for a relatively long time, the coefficient will be almost zero and then the signal Even if the level recovers,
There is a problem that the coefficient remains 0 (so-called divergent state).

The present invention has been made in view of such circumstances, and suppresses divergence.

[0022]

A signal equalizing method according to claim 1 equalizes an input signal with predetermined characteristics, decodes the equalized signal, and detects an error rate of the decoded data. , The detected error rate with a predetermined time constant
When the input error rate is within a predetermined reference value range, the coefficient for setting the characteristic is fixed to a predetermined value, and the time constant of the input error rate is set to a predetermined value.
When the error rate is outside the range of a predetermined reference value, the coefficient for setting the characteristic is updated according to a predetermined algorithm, and the input error rate is set.
And this <br/> for setting the time constant of the bets to a first value smaller than the second value, characterized.

The signal equalizer according to claim 2 is
An FIR transversal filter 30 as filter means for equalizing an input signal with predetermined characteristics;
A decoder 6 as decoding means for decoding a signal equalized by 0, an error detection and correction circuit 7 as detection means for detecting an error rate of data decoded by the decoder 6, and an error detection and correction circuit 7
Input means for inputting the error rate with a predetermined time constant
The error rate input by the force means is compared with a predetermined reference value, and when the error rate is within the range of the predetermined reference value, the coefficient for setting the characteristic is fixed to a predetermined value, and the error rate is set to the predetermined reference value. Coefficient setting means for updating a coefficient for setting a characteristic according to a predetermined algorithm when the error rate is out of a range of values, and an error rate falling within a range of a predetermined reference value.
At some point, the time constant of the input error rate is
Set to 1 and the error rate is out of the range of the prescribed reference value
, The time constant of the input error rate is set to the first
And a microcomputer 8 as a time constant setting means for setting a second value smaller than the second value .

A video tape recorder according to a third aspect of the present invention includes a magnetic head 2 as a reproducing means for reproducing a signal recorded on a magnetic tape 1, and a signal reproduced by the magnetic head 2 having a predetermined characteristic. FIR transversal filter 30 as filter means for equalizing, decoder 6 as decoding means for decoding a signal equalized by filter 30, and detecting means for detecting an error rate of data decoded by decoder 6 Error detection and correction circuit 7 and error detection and correction circuit 7
Input method for inputting the issued error rate with a predetermined time constant
The error rate input by the input means is compared with a predetermined reference value, and when the error rate is within the range of the predetermined reference value, a coefficient for setting characteristics is fixed to a predetermined value, and the error rate is set to a predetermined value. Coefficient setting means for updating a coefficient for setting characteristics according to a predetermined algorithm when the error rate is out of the range of the reference value, and when the error rate is within the range of the predetermined reference value.
When within the box, specify the time constant of the input error rate.
Set to the first fixed value and the error rate is
Time constant of the input error rate when out of range
Constant setting to set a second value smaller than the first value
And a macro computer 8 as means .

[0025]

According to the present invention, the coefficient for determining the characteristic for equalizing the input signal is changed according to the error rate of the decoded data. Therefore,
Divergence can be reliably suppressed.

[0026]

FIG. 1 is a block diagram showing the configuration of an embodiment of a video tape recorder according to the present invention, and portions corresponding to those in FIG. 3 are denoted by the same reference numerals. In this embodiment, the error rate detected by the error detection and correction circuit 7 is supplied to the microcomputer 8. Other configurations are the same as those in FIG. However, the method of calculating the coefficients in the microcomputer 8 is different from the conventional case.

Next, the operation will be described with reference to the flowchart of FIG. A digital video signal recorded on a magnetic tape 1 is reproduced by a magnetic head 2,
Amplified by the amplifier 3 and input to the A / D converter 4,
A / D conversion is performed. The A / D converter 4 outputs A / D
The D-converted data is multiplied by a predetermined coefficient by an FIR transversal filter 30 as shown in FIG. 5 and then output to the decoder 6 to be decoded into logic 1 or 0 digital data. Then, this digital data is subjected to error detection and correction in the ECC circuit 7 and output to a circuit (not shown). The above operation is the same as in FIG.

[0028] In the present embodiment, the coefficient of the multiplier 32 ₀ to 32 ₄ of the filter 30 constituting the signal equalization circuit 5
The microcomputer 8 according to the flowchart of FIG.
Is calculated and set by

[0029] That is, first in step S11, the microcomputer 8, the multipliers 32 ₀ through 3 of the filter 30 an initial coefficient stored in the internal memory in advance
It is set to 2 to _4. As a result, a signal processed to have a characteristic corresponding to the initial coefficient is supplied to the decoder 6 and decoded. The ECC circuit 7 detects an error in digital data input from the decoder 6.

In step S12, the error rate output from this ECC circuit 7 is
It is taken in. At this time, the time constant of the fetched error rate is set to a relatively large value.
For example, an average value of the error rate for a predetermined time is calculated, and the average value is monitored by the microcomputer 8.

In step S13, the microcomputer 8 compares the fetched error rate with a predetermined reference value (for example, 10 ^-4 ), and determines whether the average error rate is higher than the reference value. Determine whether or not. If the average error rate is smaller than the reference value, it is determined that appropriate signal equalization processing has been performed, and the process returns to step S12 to perform the next error rate detection operation. That is, as long as the error rate does not exceed the reference value, the loop 1 composed of steps S12 and S13 is repeatedly executed.

If it is determined in step S13 that the error rate is larger than the reference value, the process proceeds to step S14, where the processing of the LMS algorithm is executed. That is,
At this time, the microcomputer 8 subtracts the output of the adder 23 from the output of the decoder 13 in the subtractor 14 to generate an error signal, as in the case described above. And
With reference to the error signal and the signal input from the A / D converter 4, a new coefficient is calculated by the LMS method (least mean square method). Then, the coefficient obtained by the calculation is
The process proceeds to step S15, the load (updated) as the coefficient of the multiplier 32 ₀ to 32 _4. That is, thereby, the initial coefficient set up to now in step S11 is changed to the coefficient calculated in step S14.

Thereafter, in step S16, the microcomputer 8 again outputs the error output from the ECC circuit 7.
-Input the rate . At this time, the input error
The time constant of the rate is input in step S12.
The error rate is set to be smaller than the time constant. That is, since the processing of the above-described loop 1 is performed in a stable state, the time constant is set to be large so as to maintain this stable state as much as possible. Since it is performed in an unstable state, a quicker response is required. Therefore, the time constant is set to a small value.

Next, proceeding to step S17, the microcomputer 8 determines whether or not there is a tendency to diverge based on the error taken in step S16. In step S14, the microcomputer 8 stores a history of error rates since the LMS algorithm started operating. If the error rate is continuously deteriorating (increased), the divergence is determined. If the error rate is decreasing, the divergence is determined to be a non-divergent state.

If it is determined in step S17 that there is a tendency of the divergent state, the process proceeds to step S18, and the coefficient updated in step S15 is rewritten to a previously stored initial coefficient (coefficient set in step S11). Can be Then, various coefficients in the algorithm are also reset to predetermined initial values. These initial values are
Since none of them is in a divergent state, loading the initial coefficients in this way makes it possible to quickly converge the coefficients to an appropriate value.

In step S18, after a predetermined initial coefficient is loaded, the process returns to step S14, and the subsequent processing is repeated. That is, a new coefficient is calculated based on the coefficient that does not diverge in accordance with the LMS algorithm.

As described above, when the coefficient is set to an appropriate value, divergence is avoided. If it is determined in step S17 that the divergence has not occurred (when the error has continuously changed to a smaller value), the process proceeds to step S17.
The process shifts from S17 to S13, and the subsequent processes are repeated. That is, in step S13, the error rate is compared with the predetermined reference value again, and when the error rate is within the range of the reference value, the process returns to step S14, and the processing of loop 2 is repeatedly executed. In contrast, when it is out of range error rate reference value, it exits the loop 2, the process returns to the loop 1, the coefficient of the multiplier 32 ₀ to 32 ₄ to the coefficient at that time is fixed.

[0038]

As described above, according to the first to third aspects of the present invention, the equalized signal is decoded, and the coefficient defining the characteristic is controlled in accordance with the error rate of the decoded data. As a result, divergence can be reliably suppressed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a video tape recorder of the present invention.

FIG. 2 is a flowchart illustrating the operation of the embodiment of FIG.

FIG. 3 is a block diagram showing a configuration of an example of a conventional video tape recorder.

FIG. 4 is a block diagram showing functional blocks of the microcomputer 8 of FIG.

FIG. 5 is a block diagram showing a configuration example of a signal equalization circuit 5 of FIG. 3;

FIG. 6 is a flowchart illustrating an operation of the example of FIG. 3;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Magnetic tape 2 Magnetic head 4 A / D converter 5 Signal equalization circuit 6 Decoder 7 Error detection and correction circuit 8 Microcomputer 11 Adaptive controller 12 FIR type transversal filter 13 Decoder 14 Subtractor 21 _{1 to} 21 ₄ Delay circuit 22 ₀ To 22 ₄ multiplier 23 adder 30 FIR type transversal filter 31 _{1 to} 31 ₄ delay circuit 32 _{0 to} 32 ₄ multiplier 33 adder

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification code FI H03H 17/06 635 H03H 17/06 635B H04B 3/06 H04B 3/06 C

Claims (3)

(57) [Claims] An input signal is equalized with a predetermined characteristic, an equalized signal is decoded, an error rate of the decoded data is detected, and the detected error rate is input with a predetermined time constant.
When the input error rate is within a range of a predetermined reference value, the coefficient for setting the characteristic is fixed to a predetermined value .
And the time constant of the input error rate
It was set to a predetermined first value, when the error rate is input is outside the range of the predetermined reference value, and updates the coefficient for setting the characteristics according to a predetermined algorithm, the input Previous
The time constant of the error rate is smaller than the first value.
A signal equalization method characterized by setting to a second value . 2. A filter for equalizing an input signal with a predetermined characteristic, a decoder for decoding a signal equalized by the filter, and a detection for detecting an error rate of data decoded by the decoder. Means for detecting the error rate detected by the detection means.
Input means for inputting a time constant of, the error rate input by the input means with a predetermined reference value, when the error rate is within the range of the predetermined reference value, setting the characteristic fixing the coefficients to a predetermined value when said error rate is out of the range of the predetermined reference value, a coefficient setting means for updating the coefficients for setting the characteristic in accordance with a predetermined algorithm, the error rate is the If it is within the range of the predetermined reference value
The input time constant of the error rate
Set to a first value, wherein the error rate is equal to the predetermined reference
The error rate that is input when the value is out of the range
Is set to a second value smaller than the first value.
And a time constant setting means for setting the time constant . 3. A reproducing means for reproducing a signal recorded on a magnetic tape, a filter means for equalizing a signal reproduced by the reproducing means with a predetermined characteristic, and a signal equalized by the filter means. Decoding means for decoding; detecting means for detecting an error rate of the data decoded by the decoding means; and detecting the error rate detected by the detecting means in a predetermined manner.
Input means for inputting a time constant of, the error rate input by the input means with a predetermined reference value, when the error rate is within the range of the predetermined reference value, setting the characteristic fixing the coefficients to a predetermined value when said error rate is out of the range of the predetermined reference value, a coefficient setting means for updating the coefficients for setting the characteristic in accordance with a predetermined algorithm, the error rate is the If it is within the range of the predetermined reference value
The input time constant of the error rate
Set to a first value, wherein the error rate is equal to the predetermined reference
The error rate that is input when the value is out of the range
Is set to a second value smaller than the first value.
And a time constant setting means for setting the time constant .