JP3178213B2 - Encoding device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding apparatus effective for use in a high-density recording encoding method such as a digital VTR.

[0002]

2. Description of the Related Art At present, digital VTRs have been studied for the purpose of high image quality and long-time recording. A technique called digital pilot tone is one that contributes to high-density recording that is essential for recording a video signal having a large amount of information for a long time. It encodes (modulates) a specific frequency for tracking so that the recording data sequence itself is generated, so that the tracking accuracy is greatly enhanced without deteriorating the error rate in practical use, and at the same time, only tracking information is used. This eliminates the recording area and improves the recording density.

[0003] For example, as shown in FIG.
Modulation that forms a notch at 2 (a frequency component thereof is suppressed) is F0 modulation, pilot is formed at frequency f1, modulation that forms a notch at frequency f2 is F1 modulation, a notch is formed at frequency f1, and a pilot is formed at frequency f2. Modulate F2
The continuous track is referred to as F1 modulation and F0 modulation.
It is assumed that the modulation pattern is repeatedly changed and recorded, such as modulation, F2 modulation, and F0 modulation. When the head scans the F0-modulated track, the frequency f from the adjacent F1-modulated track is read.
This is realized by comparing the crosstalk of one pilot component with the crosstalk of the pilot component of frequency f2 from the adjacent F2 modulated track.

In other words, when the head has to trace the F0-modulated track completely but moves toward the F1-modulated track, the crosstalk of the pilot component of the frequency f1 from the adjacent F1-modulated track occurs. The amount becomes larger than the crosstalk amount of the pilot component of the frequency f2 from the adjacent F2 modulation track. (In the case of magnetic recording, even if azimuth recording is performed, if the frequency is low, this component is not easily affected by azimuth loss.) Therefore, a band-pass filter that extracts the component of frequency f1 and the component of frequency f2 are extracted. The output of the head can be determined by comparing the two outputs to determine which side the head is shifted from the original position.

[0005] Further, when a notch detects crosstalk from an adjacent track, its frequency component is small.
Signal components rarely interfere with crosstalk detection as noise. Using this result, accurate tracking is realized by controlling the voltage applied to the piezoelectric element to adjust the height of the head mounted on the piezoelectric element, or adjusting the tape feed speed.

FIG. 7 shows a configuration diagram of a conventional encoding device.
This encoder adds one control bit to the beginning of 24-bit data as shown in FIG. 8, performs encoding (interleaved NRZI modulation) for every 25-bit data, and selects the control bit polarity as appropriate. By doing so, a DC component is suppressed and pilots and notches are formed at the frequencies f1 and f2.

[0007] In FIG. 7, the encoding unit 11 is a circuit for adding 0,1 to the control bits interleaved NRZI modulation, with respect to the input a _k at time kT the configurations shown in FIG. 8,

[0008]

(Equation 1)

Is a circuit for generating the recording sequence {b _k } given by here

[0010]

[Outside 1]

Is an exclusive OR. Here, since the control bits are added to the first bit, the odd-numbered bits of one word are sequentially encoded by the control bits. On the other hand, the even-numbered bits are sequentially encoded by the last bit of the previous word . The subtractors 12 and 13 are circuits for subtracting a reference waveform having a desired frequency component from the respective encoded data, and the detector 14 is a detection circuit for extracting the frequency component of the encoded data after the subtraction. This circuit extracts the difference (error component) between the frequency component of the reference waveform and the frequency component of the encoded data because the subtraction has been performed. The detector 41 constituting the detection unit 14 is a circuit for extracting an instantaneous frequency component in units of 25 bits. The selection unit 43 selects the frequency component value for which data selection is determined, and the holding unit 44 selects the frequency component value. By temporarily holding the selected value and adding the latest 25-bit detection values by the adder 42, the frequency components of the entire output data sequence including the output data sequence for which data selection has been determined are extracted. It is. Further, it comprises five circuits of the same configuration corresponding to the extraction of the five components of the DC component, the sine component and the cosine component of the frequency f1, and the sine component and the cosine component of the frequency f2, which differ only in the detection waveform. The square summation units 15 and 16 are circuits for obtaining the square values of a plurality of error components extracted by the detection unit 14 from the two encoded data and calculating the sum thereof. This is a circuit that compares the sum of squares of the error component in the case of “0” and the sum of squares of the error component in the case where the control bit is “1”, and outputs the polarity of the smaller control bit. The selection unit 18 is a circuit that selects encoded data to be finally output according to the polarity detection result of the magnitude comparison unit 17.

With respect to the operation of the coding apparatus configured as described above, a pilot is formed at the frequency f1 and the frequency f1
An example in which a notch is formed in 2 will be described.

First, the encoding unit 11 generates two encoded data corresponding to the polarity of the control bit. Next, the two kinds of generated encoded data strings are ± 1.0 as shown in FIG.
3 is subtracted by the subtracters 12 and 13 to obtain a difference waveform. As the reference waveform, a rectangular wave of frequency f1 is used as a signal having no DC component and no component of frequency f2. The detection section 14 extracts the frequency component from the obtained difference waveform. The components to be extracted are a DC component, a sine component and a cosine component of the frequency f1, a sine component and a cosine component of the frequency f2, and the detection waveform is a DC, a sine wave and a cosine wave of the frequency f1, a sine wave and a cosine wave of the frequency f2. Each component is extracted by a method equivalent to the sine transform and cosine transform of the so-called Fourier transform used. Specifically, the detector 41
, The cumulative sum of the difference waveform and the detected waveform is calculated, and the components of the past output data string held in the holding unit 44 are added by the adder 42 to calculate the component value of the latest data string. However, the above five component values are values indicating errors between the components of the encoded data and the components of the reference waveform.

Next, the square summation units 15 and 16 square all five error component values of the encoded data when the control bits are "0" and "1", respectively, and determine the sum. As a result, the sum of squares of the error component obtained for each control bit is compared by the magnitude comparison unit 17 to detect the control bit polarity of the smaller sum of squares.

Finally, according to the output polarity of the magnitude comparing section 17, the selecting section 18 selects the encoded data having the smaller error and outputs the encoded data. Further, the component error value regarding the two encoded data obtained by the detection unit 14 is selected by the selection unit 43 according to the determination result, and the component value for the encoded data whose output has been determined by being held by the holding unit 44 is updated. deep.

The above operation is repeated for each 24-bit input data, and the DC component and the frequency f2 component, which are the frequencies evaluated by successively determining the encoded output, are suppressed, and the reference waveform has a frequency f1. Ingredients appear and form notches and pilots.

As described above, each component of the encoded output is extracted for each word, and the frequency component is uniformly evaluated by sequentially selecting the smaller one of the sum of the squared values. By selecting encoded data with the smallest possible error, it has a function of converging the error of each component to zero.
In the above example of the reference waveform, since the error between the DC component having the frequency f1 component and the waveform having no frequency f2 component is converged to 0, the frequency component of the reference waveform is included in the frequency f1 of the encoded output. Remain to form a pilot at the frequency f1, and at the same time, suppression of the DC component and formation of a notch at the frequency f2 are realized. When a rectangular wave of frequency f2 is input to the reference waveform, the frequency f1
A notch is formed at a frequency f2, and a pilot is formed at a frequency f2.
In both cases, notches are formed.

[0018]

However, in the above-mentioned conventional configuration, a circuit for obtaining a square value and a circuit for obtaining a sum thereof are required in addition to performing a Fourier transform for extracting an error component. Therefore, there is a problem that the circuit scale is very large.

The present invention has been made to solve the above-mentioned conventional problems, and has as its object to provide an encoding apparatus which can obtain an encoded output which is appropriately approximated to a desired frequency component with a simple algorithm and a small circuit scale. Aim.

[0020]

In order to achieve this object, an encoding apparatus according to the present invention performs encoding according to a predetermined rule by adding one control bit to each L-bit data.
Error detection means for detecting each of the error between the plurality of frequency components in the frequency components and a plurality of frequencies having two types of encoded data obtained by the polarity of the control bits, in the error of each frequency component (absolute value) And coded data selecting means for selecting coded data having no maximum error, and having a configuration for selecting and outputting coded data having a smaller error with respect to a desired frequency characteristic. are doing.

[0021]

According to the present invention, coded data having a poor degree of approximation with respect to a desired frequency is reliably eliminated by the above configuration, and the frequency characteristics of a series of coded outputs are made closer to the desired characteristics.

[0022]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an encoding device according to an embodiment of the present invention. This coding apparatus performs coding by applying interleaved NRZI modulation in units of 25 bits with 1-bit control bits added to 24-bit data.

In FIG. 1, an encoding unit 1 adds “0” and “1” to control bits, encodes a bit sequence in each case by performing interleaved NRZI modulation, and encodes two types of encoded data. The output circuit is equivalent to the conventional configuration shown in FIG. The subtractors 2 and 3 are circuits for subtracting a reference waveform having a desired frequency component from each of the encoded data, and the detector 4 is a detection circuit for extracting the frequency component of the encoded data after the subtraction. This circuit extracts the difference (error component) between the frequency component of the reference waveform and the frequency component of the encoded data because the subtraction has been performed. The configuration of the detector 4 is the same as that of the conventional example, and is to extract the frequency components of the entire series of output data including the past output. Maximum value detector 5
Is a circuit for calculating the absolute values of a plurality of error components extracted by the detection unit 4 for the two encoded data and detecting the maximum value. The polarity detection unit 6 detects the maximum error by the maximum value detection unit 5. This is a circuit for detecting the polarity opposite to the polarity of the control bit of the encoded data on the side. The selection unit 7 is a circuit that selects encoded data to be finally output according to the polarity detection result of the polarity detection unit 6.

The operation of the coding apparatus configured as described above will be described. As an example, a case where a DC component is cut and a pilot is generated at frequency f1 and a notch is generated at frequency f2 will be described. The apparatus sequentially selects encoded output using 24-bit input data as one evaluation unit.

First, the coding unit 1 adds one control bit and performs interleaved NRZI modulation, and then generates the two types of coded data strings as shown in FIG.
The difference waveform is obtained by subtracting the reference waveform shown in FIG. This reference waveform is a rectangular wave having a frequency f1 having no DC component and no component of the frequency f2, and this frequency characteristic is a target.

Next, the detection section 4 extracts the frequency component from the obtained difference waveform. The components to be extracted are DC, frequency f1,
This is a component of the frequency f2. FIG. 4 shows a specific configuration of the detector 41. In the present embodiment, a DC, a sine wave and a cosine wave having a frequency f1 and a sine wave and a cosine wave having a frequency f2 are inputted as detection waveforms, and the coded data is multiplied by each detection waveform and integrated.
This is a circuit equivalent to obtaining a sine component, a cosine component, and a DC component of a so-called Fourier transform. Since this is a routine operation having only different detection waveforms, the circuit is shared by time division. Thereby, the DC component and the frequency f1 with respect to the difference waveform
, A sine component of the frequency f2, and a cosine component. However, this value is an instantaneous frequency component in a 25-bit section, and the frequency component value of the past determined coded output held in the holding unit 44 is added to this value by the adder 42 to obtain two values. To obtain the frequency component of the coded output sequence. Further, this value is a value indicating an error between a component of the encoded data and a component of the reference waveform, and a total of ten types of error components are obtained corresponding to two modulation data.

Next, the ten kinds of error component values are input to the maximum value detecting section 5, the absolute value of the error is obtained, and the maximum value is detected. Here, the encoded data for which the maximum error is detected means that at least one component cannot be approximated to the component of the reference waveform. Therefore, the detection of the maximum error is output to the polarity detection unit 6, and the polarity detection unit 6 detects and outputs the polarity opposite to the polarity of the control bit in which the detection was performed.

Finally, according to the output polarity of the polarity detector 6, the selector 7 selects the encoded data having the smaller error and outputs the encoded data. Further, a component error value relating to the two encoded data obtained by the detection unit 4 is selected by the selection unit 43 according to the output polarity of the polarity detection unit 6, held by the holding unit 44, and the component corresponding to the encoded data whose output has been determined. Update the value.

The above operation is repeated for each input data of 24 bits, and the DC component and the frequency f2 component, which are the frequencies evaluated by sequentially determining the encoded output, are suppressed, and the frequency f1 has a reference waveform. Ingredients appear and form notches and pilots. By changing the waveform of the reference in the same manner as in the conventional example, the formation of the pilot and the notch can be controlled.

As described above, according to this embodiment, the maximum value of the difference (error with respect to the target) between the frequency component of the reference waveform and the frequency component of the two coded data is detected, and the code that does not include the maximum error is detected. It is possible to obtain a modulation output approximated by a desired frequency characteristic only by outputting the converted data. Also, in the conventional method of equally evaluating the sum of the squared values of the error, since the error is squared, the one with a large error becomes the dominant term of the sum of the squared values. The evaluation method of the embodiment for eliminating the coded output including the maximum error can achieve almost the same performance as the conventional example.

In this embodiment, two encoded data are generated according to the polarity of the control bit, and the respective frequency components are detected. However, as shown in FIG.
Since the I-modulation is inverted every other bit depending on the polarity of the control bit, the frequency component Podd of the inverted bit group and the frequency component Peven of the bit group independent of the control bit are classified and the frequency components are inverted. Detect (P
Even + Pod) and (Peven-Pod) may be calculated as frequency components of two types of encoded data.

In the interleaved NRZI modulation as shown in FIG. 5, since the polarity of one control bit has a relationship of inverting even-numbered bits of the next word (25 bits), the control as shown in FIG. It is also possible to extract all the bit groups depending on the bits over two words, extract the frequency components, and sequentially determine the bit polarity every other bit.

In this embodiment, the error component is directly calculated by subtracting the reference waveform in advance. However, since the frequency component of the reference waveform becomes a fixed pattern with respect to the phase every 25 bits, the fixed pattern is obtained. May be prepared in advance and appropriately subtracted from the frequency component of the modulation data.

The sine wave and cosine wave input to the detector may be simplified by approximating them with a step-like (discrete value) waveform such as a rectangular wave having a frequency to be extracted.

Further, a part or all of them may be realized by software.

[0036]

As described above, according to the present invention, one control bit is added to each L-bit data, encoding is performed according to a predetermined rule, and two types of encoded data obtained by the polarity of the control bits are obtained. Error detecting means for detecting an error between a frequency component at a plurality of frequencies and a desired frequency component, and coded data having no maximum error among errors (absolute values) of the respective frequency components. By providing the encoded data selection means, encoded data having a large frequency component error can be eliminated, and encoded data having characteristics closer to desired frequency characteristics can be obtained. In addition, only the maximum value and its control bit polarity need be managed, which can be realized by a small-scale circuit, and has good adaptability even when high-speed operation is required. Further, by inputting the reference waveform and the detection waveform from outside, the frequency characteristics of the encoded output can be freely controlled.

As described above, the present invention exerts a great effect when actually used in a high-density recording device such as a digital VTR in which narrow track recording is indispensable.

[Brief description of the drawings]

FIG. 1 is a block diagram of an encoding device according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram showing waveform conversion of a code data sequence.

FIG. 3 is a reference waveform diagram having a desired frequency characteristic.

FIG. 4 is a block diagram showing a configuration example of a detector.

FIG. 5 is a schematic diagram showing characteristics of interleaved NRZI modulation.

FIG. 6 is a schematic diagram of an operation of determining based on a frequency characteristic of a bit string over two words.

FIG. 7 is a block diagram of a conventional encoding device.

FIG. 8 is a schematic diagram showing control bit addition and a modulation scheme.

FIG. 9 is a frequency characteristic diagram based on a modulation pattern.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Encoding part 2 Subtractor 3 Subtractor 4 Detection part 5 Maximum value detection part 6 Polarity detection part 7 Selection part 11 Encoding part 12 Subtractor 13 Subtractor 14 Detection part 15 Square summation part 16 Square summation part 17 Large and small Comparison section 18 Selection section 41 Detector 42 Adder 43 Selection section 44 Holding section

────────────────────────────────────────────────── (5) References JP-A-5-284034 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H03M ^7/ 00-7/50

Claims (7)

(57) [Claims] 1. An encoding device for adding one control bit to each L-bit data to perform encoding according to a predetermined rule, and includes two types of encoded data obtained according to the polarity of the control bits. A coding apparatus that selects and outputs coded data having characteristics closer to a desired frequency characteristic, and outputs the coded data to a plurality of frequency components at a plurality of frequencies.
On the other hand, an error from the frequency component of the encoded data
And error detection means each for detecting said each frequency component
A coded data selecting means for obtaining an absolute value of the error and selecting coded data having no maximum absolute value . Wherein the error detecting means 請 for detecting the frequency component is regarded as a rectangular wave having a predetermined amplitude encoded data string
The encoding device according to claim 1 . 3. The error detecting means regards the encoded data string as a rectangular wave having a predetermined amplitude, and detects a frequency component from a waveform obtained by subtracting a reference waveform having a desired frequency characteristic or a waveform subtracted from the reference waveform. Frequency
The encoding device according to claim 1, wherein an error of the minute is detected . Wherein the error detection means, according to claim 1, wherein the detecting the error of the sine component and the cosine component of the plurality of frequencies including DC
Encoding device. 5. An error detecting means for extracting a frequency component by Fourier detection in which a sine component and a cosine component of a plurality of frequencies are added to a sine waveform of each frequency and a cosine waveform is cumulatively added.
2. The encoding apparatus according to claim 1 , wherein a frequency component is extracted by pseudo Fourier detection in which a sine waveform and a cosine waveform are cumulatively added as an approximate waveform having a discrete amplitude. 6. The code according to claim 5 , wherein only a detection waveform such as a sine wave and a cosine wave required for the Fourier detection or the pseudo Fourier detection is switched to perform a fixed operation, and the error detection means is shared in a time division manner. Device. 7. An error detecting means selects a control bit dependent bit group whose bit polarity is determined only by the polarity of the control bit from the encoded data , detects an error of each frequency component, and selects an encoded data selection. means, according to claim 1, selects the polarity of the control bits dependent bit group sequentially output every consecutive (L + 1) bit coded data string is determined
An encoding device according to claim 1.