KR19980047708A - Digital data restoration device - Google Patents

Abstract

A digital data decompression apparatus comprising a channel equalizer and a data detector for detecting end user data for a signal passed through the channel equalizer, the apparatus comprising: A judging unit for judging whether or not the output signal is within a predetermined range and outputting a control signal according to the judgment; An error detector for detecting an error with respect to a peripheral value of a current output value of the channel equalizer; A data correcting unit for calculating a corrected value of a current output value using the detected error value; And a selector for selectively outputting the corrected value output from the data correction unit and the equalization output value passed through the channel equalization unit according to a control signal of the determination unit to the data detection unit, So that accurate user data can be restored.

Description

Digital data restoration device

The present invention relates to a digital VCR, and more particularly, to a digital data restoration apparatus.

Generally, in a magnetic recording channel, three-level detection or Viterbi detection is performed as a signal detection apparatus, and these two methods have a trade-off relationship between performance and complexity.

That is, although the three-level detection scheme can be implemented simply, the performance is much lower than that of the Viterbi detection scheme.

On the other hand, the Viterbi detection method is superior to the three-level detection method but has a very complicated structure.

Accordingly, it is necessary to develop a detection scheme capable of achieving an appropriate harmonization of these two schemes by using the correlation between the equalized symbols which has been overlooked by the three-level detection scheme.

Hereinafter, an apparatus for restoring data of a digital VCR according to the related art will be described with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an apparatus for recovering data of a digital VCR according to the related art, FIG. 2 is a diagram illustrating a data detection method in the comparison detector shown in FIG. 1, and FIGS. And the ideal output data of the equalized portion.

First, referring to FIG. 1, a configuration of the present invention will be described with reference to FIG. 1. Referring to FIG. 1, a precoder 10 for pre-coding a user data value to be transmitted in advance with an arbitrary code; A data conversion unit (20) for converting a logic signal output from the precoder (10) into an arbitrary symbol value; A channel (30) for transmitting the converted data value; An equalizer 40 for optimally equalizing the data transmitted through the channel 30; And a comparison detector 50 for comparing the equalized data with a preset reference comparison value to detect user data.

The precoder 10 includes delay elements 11 and 12 for arbitrarily encoding the data and delayed data through the delay elements 11 and 12 to perform an exclusive excitation with input user data And an exclusive or gate 13.

The channel 30 may include delay elements 21 and 22 for delaying the symbol values converted by the data conversion unit 20 by a predetermined time period and delay elements 21 and 22 for outputting the delayed data values, And an adder 23 for adding the data value to each other.

The operation of the above configuration will be described with reference to Table 1 below.

Table 1 shows an embodiment of the data flow for the data recovery apparatus of the DVCR according to the prior art.

[Table 1]

1, when a user data value b _k recorded on a magnetic recording medium DVCR is input to the precoder 10 by a reproducing head (not shown), the precoder 10 generates a delay element by 11, 12 and the extreme clutch sheave OR gate (13) b _k a _k-2 model, and outputs the result to the data converter 20.

The binary data value a _k outputted from the precoder 10 is supplied to the data converter 20

-1 and 1, and the output value of x _k is obtained.

That is, if the binary data output from the precoder 10 is 0, the data is converted to -1, and if the binary data is 1, the data is converted to 1.

The converted data value x _k is input to the channel 30 and is modeled in the form of x _k - x _k-2 by the delay elements 21 and 22 and the adder 23 and input to the equalizer 40.

The equalizer 40 optimally equalizes the signal output from the channel, optimally equalizes the signal to a value of y _k , and outputs it to the comparison detector 50.

At this time, in the ideal case in which there is no noise in the optimum equalization, it results in one of +2, 0, and -2.

As shown in FIGS. 3A and 3B, the data string of the y _k value through the equalizer 40 can be separated into an even data string and an odd data string, thereby indicating a redundancy property of signaling .

Here, the redundancy property means that all non-zero symbols in the even-numbered column and the odd-numbered column have polarities different from those of the non-zero symbol immediately before.

That is, if a symbol at a certain moment has a value of +2 or -2, then the symbol value at the next clock moment has a value of 0 or -2 or 0 or +2 and does not have a value of +2 or -2 to be.

The data value y _k applied to the comparison detector 50 is used to detect the user data b _k by comparing and detecting data by a three level detection method as shown in FIG.

That is, as shown in FIG. 2, the 3-level detection scheme is such that when the equalized data in the inter-equalizer 40 is +1 and the signal applied on the basis of -1 2 thresholds is +1 or more, When the data is judged to be less than -1, it is judged to be binary signal 1 and data is detected.

On the other hand, the data between +1 and -1 is 0, and the user data is detected by the prescaler.

The 3-level detection scheme used in the data recovery apparatus of the DVCR according to the related art has an advantage in that it is simple in structure and easy to implement. However, since it can not cope with the non-linearity in which the magnetic recording channel becomes more and more dense, There is a problem that the performance is deteriorated in the channel.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a data recovery apparatus for a DVCR according to the present invention, Level detection method without increasing the complexity of the conventional three-level detection method.

1 is a block diagram showing a digital data restoration apparatus according to the prior art;

2 is a diagram showing a data detection method in the comparison detecting unit shown in Fig. 1

Figs. 3A and 3B are diagrams showing ideal output data of the equalizer shown in Fig. 1

4 is a block diagram illustrating a digital data restoration apparatus according to the present invention.

5 is a detailed block diagram of the equalization error detecting unit shown in FIG.

6 is a diagram showing the detailed configuration of the data correction unit shown in FIG. 4

FIG. 7 is a graph showing the probability density distribution of the equalizer output shown in FIG. 4

Description of the Related Art [0002]

10: precoder 20: data conversion unit

30: channel 40:

50: equalization error detection unit 60: data correction unit

70: Judgment section 80: Switching section

90:

The digital data decompression apparatus according to the present invention is characterized in that the digital data decompression apparatus includes a channel equalizer and a data detector for detecting end user data for a signal passed through the channel equalizer, Determining whether or not the predetermined range is within a predetermined range and outputting a control signal according to the determined range; An error detector for detecting an error with respect to a peripheral value of a current output value of the channel equalizer; A data correcting unit for calculating a corrected value of a current output value using the detected error value; And a selector for selectively outputting the corrected value output from the data correction unit and the equalization output value passed through the channel equalization unit according to the control signal of the determination unit to the data detection unit.

Hereinafter, a digital data restoration apparatus according to the present invention will be described with reference to the accompanying drawings.

FIG. 4 is a block diagram illustrating a digital data restoration apparatus according to the present invention. FIG. 5 is a detailed block diagram of the equalization error detecting unit shown in FIG. 4. FIG. And the same components as those in FIG. 1 will not be described in detail.

Referring to FIG. 4, an equalization error detector 50 for detecting an equalization error of an equalized signal output from the equalizer 40; A data corrector 60 for correcting the equalization error detected by the equalization error detector 50 according to an equalized signal output from the equalizer 40; A determining unit (70) for determining a range of equalized data output from the equalizing unit (40); A switching unit (80) for selectively outputting data corrected by the data correcting unit (60) and data output from the equalizing unit (40) according to the determination result of the determining unit (70); And a comparison detector 9 for comparing the selected output data with a one-basis comparison value to restore the end user data.

5, the equalization error detector 50 compares the data y _{k + 2} output from the equalizer 40 with one reference comparison level value to detect other data y _{k + 2} A three level comparator 51; The data detected by the three-level comparator 51 _{k + 2)} and by the equalization calculating a difference unit 40, data (y _{k + 2)} output from the error (e _{k + 2)} detecting the adder (52 to output to the data correction unit 60) ; And a plurality of delay elements 53, 54, 55, and 56 for delaying the detected error e _{k + 2} by a predetermined time delay to detect another error value e _{k + 2} .

6, the data correcting unit 60 delays the equalized data value y _{k + 2} output from the equalizer 40 shown in FIG. 4 by a predetermined time to generate another data value y _k), the plate delay element (61, 62 to output to the part 70 and the switching unit 80); (E _{k + 2} , e _k-2 ) output from the equalization error detecting unit 50 and a value y _k delayed through the delay element to obtain a data value _k ) to the switching unit (80).

The operation of the above configuration will be described.

FIG. 7 is a diagram showing the probability density distribution of the equalizer output shown in FIG.

Referring to FIG. 7, the probability density distribution of the data output from the equalizer 40 of FIG. 4 is highest at +2, 0, -2.

Therefore, when the threshold values of -1 and +1, which are boundary points, are used, appropriate signal detection can be performed.

This threshold detection method is the three-level detection method described in the prior art.

However, as shown in FIG. 7, the threshold values -1 and +1 are concentrated in the detection error since the existence probabilities of the symbols are mixed.

Thus, in the present invention, when a threshold is set to an appropriate boundary in the vicinity of -1 and +1, and the output of the equalizer 40 falls within the set boundary, a new detection scheme is applied instead of a direct three-level detection scheme.

That is, in general, the output of the equalizer 40 has a correlation between the preceding and following symbols, that is, between the output odd number column of the equalizer 40 and the leading and trailing symbols of each even number column.

Therefore, the equalization error also has a correlation between the before and after symbols. Therefore, if the equalization error of the preceding and following symbols is subtracted from the output of the current equalizer 40, the correlation between the errors can be removed, and more accurate signal detection can be performed.

This is a new data detection scheme proposed by the present invention.

In other words, as shown in FIG. 7, the equalization error of the equalizer 40 is added to the output value of the equalizer 40 in the ambiguous zone near the threshold +1 and -1 in the negative direction The value near -1, +1 is shifted to a value near -2, 0, + 2 that we desire.

That is, as shown in FIG. 4, when the output of the equalizer 40 is within the ambiguous region shown in FIG. 7, the signal corrected by the data corrector 60 _k is output to the comparison detecting section 90 and the output y _k of the equalizing section 40 is outputted to the comparison detecting section 90 as it is.

At this time, whether or not the output of the equalizer 40 is present in an ambiguous zone is determined by the judging unit 70. The presence or absence of the output in the ambiguous zone is expressed as follows.

1 - bound | y _k | 1 + bound

On the other hand, the comparison detection operation of the comparison detector 90 is the same as the conventional comparison detector, that is, the three-level detection method shown in FIG.

The above operation will be described in detail.

First, a look divided in two operation when the output (y _k) of the equalizer 40 is also not located in the case which is located in the zone of the ambiguous 7.

1) If y _k is located in an ambiguous zone

First, y _{k + 2} output from the equalizer 40 is output to the equalization error detector 50 and the data corrector 60.

The y _{k + 2} input to the data correcting unit 60 is delayed by a predetermined time through the delay elements 61 and 62 shown in FIG. 6 to output the delayed y _k values to the switching unit 80, the calculator 63, And outputs it to the determination section 70.

The y _{k + 2} input to the equalization error detector 50 is input to the 3-level comparator 51 and the adder 52 shown in Fig.

The 3-level comparator 51 detects +2 if the inputted y _{k + 2} value is larger than the threshold value +1, detects 2 if it is smaller than the threshold value -2, 0 " is detected.

In other words,

+2, if y _{k + 2} + 1

_{k + 2k + 2}

-2, if y _{k + 2} -1

to be.

Thus detected _{k + 2} is input to the adder and added to the input y _{k + 2} , that is, the difference value is calculated to detect the equalization error e _{k + 2} .

That is, the equalization error e _{k + 2} is

e _{k + 2} = _{k + 2} - y _{k + 2} .

The detected equalization error e _{k + 2} are the data beam with a delay element (53, 54, 55, 56), the equalization error to another equalization detecting the error value e _k-2 through e _{k + 2} shown in Fig. 5 (60).

At this time, the judging unit 70 judges whether the inputted y _k value is a value located in the ambiguous zone shown in FIG. 7, and if the value exists at the position, the data correcting unit 60 corrects the data Corrected output _k

The data correction in the data correcting unit 60 corrects the data by subtracting the equalization errors e _{k + 2} and e _k-2 of the front and rear symbols from the current output value of the equalizer 40, thereby reducing the error correlation.

In other words, = y _k - e _k-2 - e _{k + 2} .

The data thus corrected _k is input to the switching unit 80 and is output to the comparison detecting unit 90 according to the output selection signal of the determining unit 70 to compare and detect the data, .

Here, the comparison detector 90 detects data in the same manner as the conventional three-level detection method described with reference to FIG.

2) If y _k is outside the ambiguous zone

In Figure 6 above y _k is 7, the data correction in y _k is is output to the determination section 70, determining section 70 to be output through a delay element (61, 62) of the section (60) as shown in It is determined whether or not it is present in the shown ambiguous zone.

If it is determined that the value of y _k does not exist in the ambiguous zone, the value of y _k is directly output to the switching unit 80 without data correction.

At this time, the determination section 70 is the y _k value of the y _k values to provide an output select signal to the switching unit 80 to the y _k value is selected, the output since the value does not exist in the ambiguous zone And outputs it to the comparison detector 90.

The comparison detector 90 compares the data with the conventional three-level detection scheme shown in FIG. 2, The value is restored.

The apparatus for recovering data of a digital VCSEL according to the present invention has values of a zone in which detection errors are frequently generated in threshold detection in a conventional three-level detection system, with equalization errors of an equalizer of a front- It is possible to restore the accurate digital data by reducing the probability of error detection even in a tendency that the magnetic recording medium is going to have a high density without increasing the complexity in structure.

Claims (8)

And a data detector for detecting end user data for a signal that has passed through the channel equalizer and the channel equalizer, A determining unit for determining whether a current output value of the channel equalizing unit is within a predetermined constant range and outputting a control signal according to the current output value; An error detector for detecting an error with respect to a peripheral value of a current output value of the channel equalizer; A data correcting unit for calculating a corrected value of a current output value using the detected error value; And a selector for selectively outputting the corrected value output from the data correction unit and the equalization output value passed through the channel equalizer to the data detector according to the control signal of the determination unit. . The method according to claim 1, Wherein the error value obtained by the equalization error detection unit is obtained from a corresponding row before and after the current output value. The method according to claim 1, The equalization error detector is a level comparator 3 for detecting other data (y _{k + 2)} Compare data (y _{k + 2)} output from the conversion unit, such as the channel and one reference level comparison values; The data detected by the three-level comparator ( _{(k + 2} ) and the data (y _{k + 2} ) output from the channel equalizer to detect an error (e _{k + 2} ) and output the error to the data correcting unit; And a plurality of delay elements for delaying the detected error e _{k + 2} by a predetermined time to detect another error value e _k-2 . The method according to claim 1, Wherein the predetermined range is ± 1 ± bound. 5. The method of claim 4, Wherein the bound value is greater than one. The method according to claim 1, And outputs the current output value to the data detector directly without data correction if the current output value of the channel equalizer is not within a preset constant range. The method according to claim 1, The data corrector may include a plurality of delay elements delaying the equalized data value y _{k + 2} output from the channel equalizer by a predetermined time to output another data value y _k to the decision unit and the selection unit 80; (E _{k + 2} , e _k-2 ) output from the equalization error detecting unit and a value (y _k ) delayed through the delay element to obtain a data value _k ) to the selection unit. < Desc / Clms Page number 19 > 8. The method of claim 7, Wherein the calculator subtracts the equalization error value of the front and back symbols from the current output value of the channel equalizer.