JPH06267006A - Magnetic reproducing device - Google Patents

Abstract

(57) [Abstract] [Purpose] A signal is reproduced by using all the signal energy recorded in a track without requiring high accuracy in tracking and without increasing adjacent crosstalk. [Structure] The reproducing heads 3a to 3e are arranged in the track width direction, and the vicinity of the track 2a to be reproduced is scanned in the longitudinal direction. The signal selection circuit 11 selects only the target track reproduction signal reproduced by the reproduction heads 3b to 3d that scans the track 2a. Then, only the switches 10b to 10d corresponding to these are turned on. Therefore, the adder 6 sums and outputs only the signals from the reproducing head that scans the track 2a.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic reproducing apparatus for reproducing a signal recorded on a magnetic recording medium such as a video tape recorder (VTR).

[0002]

2. Description of the Related Art In a conventional magnetic reproducing apparatus, the track width is becoming narrower in order to increase the recording density. In particular, in recent years, in order to record / reproduce high-definition television, the number of channels has been increased, and along with this, the track width has been further narrowed.

When the track width becomes narrow, it becomes difficult for the reproducing head to scan the track accurately. On the other hand, there was a magnetic reproducing device that took countermeasures in the mechanical section by ultra-precision processing, high-precision assembly, etc. (for example, Tomita et al .: Hi-Vision digital VTR HDD-
1000 / HDDP-1000, Broadcasting Technology November Extra Edition, Volume 43, No. 12, Kenrokukan Publishing Co., Ltd., 64
P., 1990).

Further, there is a magnetic reproducing apparatus in which the reproducing head width is determined on the assumption that tracking deviation occurs during reproduction (for example, Umemoto et al .: Examination of recording / reproducing system of 1.2 Gb / s digital VTR, television). John Society Technical Report, VIR 89-1, Vol. 13, No. 31, pp. 1-
Page 6, 1989).

[0005]

However, there is a limit to the measures to be taken in the mechanism section by ultra-precision machining, high-precision assembly, or the like. In particular, it is difficult to always scan the reproducing head without causing tracking deviation with respect to changes in temperature and changes with time.

For this reason, a method of determining the reproducing head width is used on the assumption that tracking deviation occurs during reproduction. However, when this method is used, when the reproducing head width is widened, noise and crosstalk from adjacent tracks are increased. Further, when the reproducing head width is narrowed, not all the signal energy recorded on the track can be reproduced. Therefore, the signal-to-noise ratio is deteriorated even when the tracking deviation does not occur.

In view of the above problems, the present invention uses a plurality of reproducing heads arranged in the track width direction to scan the target track to be reproduced and its surroundings, select only the signals from the target track, and total them. By outputting all the signal energy recorded on the target track without requiring high accuracy in tracking when scanning the reproducing head and without increasing crosstalk from the adjacent track. An object of the present invention is to provide a magnetic reproducing device for reproducing a signal.

Further, by multiplying the reproduction signal from the reproduction head which scans the adjacent track by a coefficient and synthesizing the signal with the signal from the target track, the reproduction signal from the target track to be reproduced is included in the reproduction signal from the adjacent track. It is an object of the present invention to provide a magnetic reproducing device capable of canceling a crosstalk signal.

[0009]

In order to solve the above problems, the present invention is arranged in the width direction of a magnetic recording medium in which signals are recorded on a plurality of tracks, and the widths of the magnetic recording media are equal to or less than the track width. And the total width of the whole is N playback heads wider than the track width (N is an integer of 2 or more)
When the target track to be reproduced and its periphery are scanned in the track longitudinal direction by using the N reproduction heads, N system reproduction signals output from the N reproduction heads are input, and the target track is reproduced. A signal selecting means for selecting only n system target track reproduced signals (n is a positive integer not larger than N) reproduced by the reproducing head which has been scanned upward, and n system targets selected by the signal selecting means. It is provided with an adding means for adding and outputting the track reproduction signals.

Particularly preferably, as the signal selecting means, only the reproducing signal whose amplitude is larger than a predetermined reference value among the reproducing signals of the N systems is selected as the target track reproducing signal of the n system. It is equipped with.

Alternatively, as the signal selecting means, a reproduction signal from a predetermined reproducing head among the reproduction signals of the N systems is used as a reference reproduction signal, and a reproduction signal of the remaining (N-1) systems is used as a reference reproduction signal. A reproduction signal having a value difference from the reproduction signal smaller than a predetermined reference value, and signal selection means for selecting the reference reproduction signal as the n-system target track reproduction signal are provided.

A magnetic recording medium in which digital signals are recorded on a plurality of tracks, N reproducing heads (N is an integer of 2 or more) arranged in the width direction of the tracks,
Of the plurality of tracks, when the target track to be reproduced and its periphery are scanned in the track longitudinal direction by using the N reproduction heads, N reproduction signals output from the N reproduction heads are input. And N coefficients C _j
Reproduced digital data from N coefficient multiplying means for multiplying (coefficients C _{1 to} C _N ) by each other, adding means for adding the outputs of the N coefficient multiplying means, and output of the adding means. Of the N coefficients C _j by the minimum error type automatic equalization algorithm by using the identification means for identifying, the N-system reproduced signal, the output signal of the adding means, and the reproduced digital data. And a coefficient calculation means for controlling the distance.

[0013]

With the above structure, the present invention scans the target track to be reproduced and its surroundings by using a plurality of reproduction heads arranged in the track width direction, and only the signal from the target track is detected by the amplitude value of the reproduction signal. By selecting and outputting the total, the signal recorded on the target track without requiring high tracking accuracy when scanning the read head and without increasing crosstalk from adjacent tracks. Using all the energy,
The signal can be regenerated.

Further, one of the plurality of reproducing heads is used as a reference, and among reproduction signals from other reproducing heads, a signal having a small difference from the reproducing signal from this reference head is reproduced from the same track as the reference head. By selecting the selected signal as the reproduced signal, it is possible to select only the reproduced signal from the target track to be reproduced without recording adjacent tracks at different azimuth angles and without providing a guard band.

Further, by multiplying the reproduction signal from the reproduction head which scans the adjacent track by a coefficient and synthesizing the signal with the signal from the target track, the cross signal from the adjacent track included in the reproduction signal from the target track to be reproduced is obtained. The talk signal can be canceled.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A magnetic reproducing apparatus according to a first embodiment of the present invention will be described below with reference to the drawings. FIG. 1A is a conceptual diagram showing the positional relationship between the tracks and the reproducing head of the magnetic reproducing apparatus in the first embodiment of the present invention, and FIG. 1B is the magnetic reproducing apparatus in the first embodiment of the present invention. It is a block diagram.

In FIG. 1A, the signal to be reproduced is
It is assumed that the data is recorded on the track 2a on the magnetic recording medium 1. In addition, it is assumed that signals different from the track 2a are recorded on the tracks 2b and 2c adjacent to each other at different azimuth angles from the track 2a.

This track 2a and adjacent tracks 2b and 2c on both sides are arranged in the width direction of the track and all five reproducing heads 3a, 3b, 3c and 3d have the same azimuth angle as the track 2a. 3e is used to scan in the track longitudinal direction. In FIG. 1, among the five reproducing heads, three of 3b, 3c, and 3d scan the target track 2a, but two of 3a and 3e scan the track 2a. Absent.

As shown in the block diagram of FIG. 1 (b),
The signals reproduced by the five reproducing heads are respectively
It is amplified by the amplifiers 12a, 12b, 12c, 12d and 12e. The amplified reproduction signal is input to the adder 6 and the signal selection circuit 11 via the switches 10a, 10b, 10c, 10d, and 10e, respectively.

The signal selecting circuit 11 is reproduced by the reproducing heads 3b, 3c and 3d which scan the track 2a.
Only the target track reproduction signals of the three systems are selected. And the switches 10b, 10c, 10d corresponding to these
Only turn on. Therefore, the adder 6 sums only the signals from the reproducing head that scans the track 2a and outputs the summed signal from the output terminal 13.

FIG. 2 shows a signal selection circuit 1 according to this embodiment.
1 shows an example of the configuration. 5 playback heads 3a, 3b, 3
The signals reproduced by c, 3d, and 3e are inputted to the level detection circuits 21a, 21b, 21c, and 21d via the input terminals 20a, 20b, 20c, 20d, and 20e, respectively.
21e is input. Level detection circuits 21a, 21b,
Reference numerals 21c, 21d, and 21e respectively detect the amplitude of the reproduction signal, and use the detected values as comparators 22a, 22b, 22c, and 22.
It outputs to d and 22e. Comparators 22a, 22b, 22
c, 22d and 22e are the amplitude values of the input reproduction signal,
Compared with a predetermined reference level, a signal for turning on the switch when it is larger and for turning off the switch when it is smaller is output terminals 23a, 23b, 23c, 23d,
23e through switches 10a, 10b, 1 respectively.
Output to 0c, 10d, and 10e.

In the present embodiment, three reproducing heads 3b, 3c and 3d among the five reproducing heads are the target tracks 2a.
Is reproduced with the same azimuth angle, the amplitude value of the reproduction signal is large. On the other hand, the two 3a and 3e reproduce the adjacent tracks 2b and 2c on both sides at an azimuth angle different from that at the time of recording, so that the amplitude value of the reproduction signal is small. Therefore, depending on the amplitude value of the reproduction signal, the target track 2
Only the signal reproduced from a can be selected.

Therefore, as shown in FIG. 3, a tracking shift occurs and the reproducing head 3 out of the 5 reproducing heads.
c, 3d, and 3e scan the target track 2a, and two reproducing heads 3a and 3b scan the track 2a.
Even when the upper part is not scanned, the adder 6 outputs only the sum of the signals reproduced from the three reproducing heads 3c, 3d and 3e.

As described above, in the present embodiment, the track 2a and its surroundings are scanned by using the five reproducing heads arranged in the track width direction, and the target track 2a is selected according to the amplitude value of the reproduced signal. By selecting only the signal reproduced from and outputting the total, without requiring high accuracy in tracking when scanning the reproducing head, and without increasing crosstalk from adjacent tracks, All of the signal energy recorded on the target track can be utilized to regenerate the signal.

In this embodiment, the tracks 2b and 2c adjacent to both sides of the target track 2a are
It is assumed that the recording is performed at an azimuth angle different from that of the track 2a while being in contact with the track 2a. However, according to the present invention, the track 2a and the tracks 2b adjacent to both sides are
It is also useful when a guard band in which no signal is recorded is formed between 2c. In this case, the signal can be reproduced by using all the signal energy recorded on the target track without adding the noise reproduced from the guard band to the reproduced signal. In this case, if the width of the guard band is wide enough and the reproducing head does not scan the adjacent track, it is not necessary to record the adjacent tracks at different azimuth angles. May be recorded at.

FIG. 4 shows another configuration example of the signal selection circuit 11 in the first embodiment. 5 playback heads 3
The signals reproduced by a, 3b, 3c, 3d and 3e are
Input terminals 20a, 20b, 20c, 20d, 2 respectively
0e is input. In this configuration example, the signal from the input terminal 20c is input to the subtractors 30a, 30b, 30d and 30e. Further, the input terminals 20a, 20b, 20d, 2
0e from the subtracter 30a, 30b, 30
d and 30e, and the difference from the signal from the input terminal 20c is used as a difference signal to detect the level detection circuits 21a and 21a, respectively.
It outputs to 21b, 21d, and 21e.

Level detection circuits 21a, 21b, 21d,
21e respectively detects the amplitude of the difference signal and outputs the value to the comparators 22a, 22b, 22d and 22e. The comparators 22a, 22b, 22d, and 22e compare the amplitude value of the input difference signal with a predetermined reference level, and when the difference value is small, turn on the switch, and when it is large, turn off the switch. , Output terminals 23a, 23b, 2
The switches 10a, 10 are connected via 3d, 23e, respectively.
b, 10d, and 10e. The output terminal 23c
Outputs a signal that always turns on the switch 10c.

As described above, in this configuration example, one of the five reproducing heads, the reproducing head 3c, is used as a reference,
A signal having a small difference from the reproduction signal from this head is selected as a signal reproduced from the same track 2a as the reproduction head 3c.

According to this configuration example, when signals of adjacent tracks have low correlation, it is not necessary to record adjacent tracks at different azimuth angles as in the configuration example of FIG. There is also no need to provide a guard band. When using this configuration example, the switch 10c is not necessary and may be omitted.

In this embodiment, when the relationship between the reproducing head and the track is as shown in FIG. 1 (a), two of the three reproducing heads 3b and 3d are targeted. Since not only the track 2a but also the tracks 2b and 2c adjacent to both sides are reproduced, the reproduction signal includes the crosstalk component from the tracks 2b and 2c. In order to reduce this, in addition to the method of recording adjacent tracks at different azimuth angles and providing a guard band as described above, the number of heads is increased to narrow the width of each head, 2a and tracks 2 adjacent to both sides
There is a method of finely dividing the boundary between b and 2c. Further, when the signal to be reproduced is a digital signal, the crosstalk component can be canceled by the configuration described in the second embodiment below.

Next, a magnetic reproducing apparatus according to a second embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a block diagram of a magnetic reproducing apparatus according to the second embodiment of the present invention. The relationship between the reproducing head and the track is shown in FIG.
Same as. However, in this embodiment, it is assumed that a digital signal is recorded on each track.

As shown in the block diagram of FIG. 5, the signals reproduced by the five reproducing heads are amplified and equalized by the equalizing circuits 4a, 4b, 4c, 4d, 4e, respectively. The output signals of the equalization circuits 4a, 4b, 4c, 4d, and 4e are coefficient multipliers 5a, 5b, 5c, 5d, and 5 respectively.
It is input to the adder 6 via e and the coefficient calculation circuit 9. The adder 6 adds all the reproduced signals that have been input, and outputs them to the discrimination circuit 7 and also to the coefficient calculation circuit 9. The discriminating circuit 7 discriminates the digital data from the inputted reproduced signal and outputs it to the output terminal 8 and the coefficient calculating circuit 9.

The coefficient calculation circuit 9 includes equalization circuits 4a, 4b,
From the output signals of 4c, 4d, and 4e, the output signal of the adder 6, and the reproduced digital data output from the discrimination circuit 7, a coefficient value to be applied to the reproduced signal is adapted using a minimum error type automatic equalization algorithm. Output to the coefficient multipliers 5a, 5b, 5c, 5d, and 5e.

The automatic equalization algorithm controls the characteristic of the equalizer to the optimum one in accordance with the change of the characteristic of the transmission line in the transmission of the digital signal. inside that,
The minimum error type automatic equalization algorithm is an algorithm for minimizing the m-th power mean error (m is a positive integer) between the output of the equalizer, that is, the input of the discrimination circuit and the digital data which is the source of the input signal. Is.

In this embodiment, assuming that the error rate of data on the transmission line is low, the output of the identification circuit 7 is used instead of the original digital data. The algorithm will be described for the case of m = 2, that is, the case of using a so-called MSE algorithm that minimizes the mean square error between the input and the output of the discrimination circuit 7.

The MSE algorithm uses F at time n.
When the j-th tap coefficient of the IR filter is C _j (n), the tap coefficient is sequentially updated by the following (Equation 1).

[0037]

[Equation 1]

In (Equation 1), X _kj is the input signal of the coefficient multiplier at the j-th tap of the FIR filter. Further, e _k is error information obtained as a difference between the output of the FIR filter and the determination output. Further, α is a coefficient for controlling the convergence speed. For details of this algorithm, see, for example, "Digital Signal Processing" by Miyakawa et al.
(First edition November 10, 1975), published by The Institute of Electronics and Communication Engineers,
It is described in.

Although this algorithm is described as updating the tap coefficient of the FIR filter in the above-mentioned reference, it is intended to minimize the mean square error between the input and output of the discrimination circuit. It is also applicable to the configuration shown.

The equalization circuits 4a, 4b, 4 in FIG.
The output signals of c, 4d, and 4e are respectively assigned to X _{k + 2} , X _{k + 1} , and X
_{Letting k} , X _k-1 , and X _k-2 be the output signal Y _k of the adder 6,
It is represented by the following (Equation 2).

[0041]

[Equation 2]

The reproduced digital data output from the discrimination circuit 7 is _defined as a _k, and when the digital information is "1", a _k =
When 1, "0", it is assumed that a _k = -1. At this time,
The error information e _k can be expressed by the following (Equation 3), and the mean square error D between the input and output of the discrimination circuit can be expressed by the following (Equation 4).

[0043]

[Equation 3]

[0044]

[Equation 4]

Partial differentiation of this (Equation 4) with respect to C _j ,
In other words, by determining the sensitivity of the mean square error of the input and output of the discrimination circuit to the change of each coefficient,
The algorithm shown in (Equation 1) is derived. Therefore,
The algorithm of (Equation 1) can be used to calculate the coefficient that minimizes the mean square error between the input and the output of the discrimination circuit even in the configuration of FIG.

FIG. 6 shows the coefficient calculation circuit 9 in this embodiment.
An example of the configuration will be shown. 5 playback heads 3a, 3b, 3
The signals reproduced by c, 3d, and 3e are input terminals 50a, 50b, 50c, 50d, and 50e, respectively,
It is input to the multipliers 51a, 51b, 51c, 51d, 51e. On the other hand, the output signal of the adder 6 is input through the input terminal 55, and the output signal of the identification circuit 7 is input terminal 56.
Is input to the subtractor 54 via the. Subtractor 5
4 subtracts the output signal of the discriminating circuit 7 from the output signal of the adder 6 to create an error signal, and the multipliers 51a, 51b,
It outputs to 51c, 51d, and 51e.

Multipliers 51a, 51b, 51c, 51d,
51e multiplies each input signal by the error signal from the subtracter 54, and the result is integrated by the integrating circuits 52a, 52b, 52.
It outputs to c, 52d, and 52e. Integrating circuits 52a, 52
b, 52c, 52d, and 52e integrate input multiplication results, and output terminal 53a and 53 as coefficient values, respectively.
Coefficient multiplier 5 via b, 53c, 53d and 53e
a, 5b, 5c, 5d, 5e.

With the above configuration, the coefficient calculation circuit 9 uses the output signals of the equalization circuits 4a, 4b, 4c, 4d, 4e, the output signal of the adder 6 and the reproduced digital data output from the discrimination circuit 7. , (Equation 1) can be used to adaptively control the value of the coefficient by which the reproduction signal is multiplied.

In this embodiment, as described above, the relationship between the reproducing head and the track is the same as that shown in FIG.
At this time, of the five reproducing heads, the reproducing head 3c reproduces only the signal of the target track 2a. Since the two reproducing heads 3b and 3d reproduce not only the target track 2a but also the adjacent tracks 2b and 2c on both sides, the reproduced signal contains the crosstalk component from the tracks 2b and 2c. Two more reproducing heads 3a and 3e
Reproduces only the crosstalk components from the tracks 2b and 2c adjacent to each other.

Therefore, the reproducing heads 3a and 3e are calculated by calculating the coefficients so as to minimize the root mean square error between the signal obtained by multiplying these signals by a coefficient and the synthesized signal and the digital data identifying this signal. By multiplying the crosstalk components reproduced by the above two by a negative coefficient and adding them, it is possible to cancel the crosstalk components from the tracks 2b and 2c included in the two reproduced signals of the reproducing heads 3b and 3d. it can.

As described above, the present embodiment has an excellent feature that the crosstalk signal from the adjacent track can be canceled in addition to the same features as the first embodiment.

In this embodiment, the track 2
For the purpose of only canceling the crosstalk component from b and 2c, as shown in FIG. 7, the inner three of the five reproducing heads, that is, the reproducing heads 3b and 3c.
And 3d may be put together and replaced with a single reproducing head having a width wider than the track width.

Further, in this embodiment, in order to reduce the crosstalk component originally included, it is possible to use a method of recording adjacent tracks at different azimuth angles and providing a guard band.

Further, in the above-mentioned first and second embodiments, the number of heads is 5, but the number of heads can be set arbitrarily.

[0055]

As described above, according to the present invention, a plurality of reproducing heads arranged in the track width direction are used to scan the target track to be reproduced and its surroundings, and the amplitude value of the reproducing signal By selecting only the signal from the target track and outputting the total, without requiring high accuracy in tracking when scanning the reproducing head, and without increasing crosstalk from adjacent tracks, All of the signal energy recorded on the target track can be utilized to regenerate the signal.

Among the plurality of reproducing heads, the reproducing head at the center is used as a reference, and a signal having a small difference from the reproducing signal from the reference head is selected as a signal reproduced from the same track as the reference head. Thus, only the reproduction signal from the target track to be reproduced can be selected without recording adjacent tracks with different azimuth angles and without providing a guard band.

Further, by calculating the coefficients so as to minimize the root mean square error between the signals obtained by multiplying the reproduced signals from a plurality of reproducing heads by the coefficients and the digital data identifying the signals, The crosstalk signal from the adjacent track included in the reproduction signal from the target track to be reproduced can be canceled.

[Brief description of drawings]

FIG. 1A is a diagram showing a positional relationship between tracks and a reproducing head of a magnetic reproducing apparatus according to a first embodiment of the present invention. FIG. 1B is a block diagram of a magnetic reproducing apparatus according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration example of a signal selection circuit.

FIG. 3 is a diagram showing a positional relationship between a track and a reproducing head.

FIG. 4 is a block diagram showing another configuration example of a signal selection circuit.

FIG. 5 is a block diagram of a magnetic reproducing apparatus according to a second embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration example of a coefficient calculation circuit.

FIG. 7 is a diagram showing a positional relationship between a track and a reproducing head in a second embodiment of the present invention.

[Explanation of symbols]

1 magnetic recording medium 2a, 2b, 2c tracks 3a, 3b, 3c, 3d, 3e reproducing head 4a, 4b, 4c, 4d, 4e equalization circuit 5a, 5b, 5c, 5d, 5e coefficient multiplier 6 adder 7 identification Circuit 8 Output terminal 9 Coefficient calculation circuit 10a, 10b, 10c, 10d, 10e Switch 11 Signal selection circuit 12a, 12b, 12c, 12d, 12e Amplifier 13 Output terminal 20a, 20b, 20c, 20d, 20e Input terminal 21a, 21b, 21c, 21d, 21e Level detection circuit 22a, 22b, 22c, 22d, 22e Comparator 23a, 23b, 23c, 23d, 23e Output terminal 30a, 30b, 30d, 30e Subtractor 50a, 50b, 50c, 50d, 50e Input terminal 51a, 51b, 51c, 51d, 51e Multipliers 52a, 52b, 52c, 52d, 52e Integration circuit 53a, 53b, 53c, 53d, 53e Output terminal 54 Subtractor 55 Input terminal 56 Input terminal

Claims (6)

[Claims] 1. A magnetic recording medium in which signals are recorded on a plurality of tracks, and the signals are arranged in the width direction of the tracks, each width is less than or equal to the track width, and the total width is wider than the track width. Number of reproducing heads (N is an integer of 2 or more) and the N reproducing heads, when the target track to be reproduced and the periphery thereof are scanned in the track longitudinal direction among the plurality of tracks, the N reproducing heads are recorded. Of the reproduction signal output from the reproduction head of FIG.
Signal selecting means for selecting only n-system target track reproduction signals (n is a positive integer equal to or less than N) and addition means for adding and outputting n-system target track reproduction signals selected by the signal selecting means. A magnetic reproducing apparatus comprising: 2. The magnetic reproduction according to claim 1, wherein the signal selecting means selects only a reproduction signal having an amplitude larger than a predetermined reference value among the reproduction signals of the N systems as a target track reproduction signal of the n system. apparatus. 3. The magnetic recording medium according to claim 2, wherein when recording signals on a plurality of tracks, adjacent tracks are recorded by using recording heads having different azimuth angles. Playback device. 4. A magnetic recording medium is provided between a plurality of tracks.
The magnetic reproducing apparatus according to claim 2, wherein the magnetic reproducing medium is a magnetic recording medium having a guard band on which no signal is recorded. 5. The signal selecting means uses a reproduction signal from a predetermined reproducing head among the reproduction signals of N systems as a reference reproduction signal, and among the reproduction signals of the remaining (N-1) systems, the reference reproduction signal. 2. The magnetic reproducing apparatus according to claim 1, wherein a reproduction signal having a difference in value between and is smaller than a predetermined reference value and the reference reproduction signal are selected as target track reproduction signals of n systems. 6. A magnetic recording medium in which digital signals are recorded on a plurality of tracks, N reproducing heads (N is an integer of 2 or more) arranged in the width direction of the tracks, and N reproducing heads. Of the plurality of tracks, when the target track to be reproduced and the periphery thereof are scanned in the track longitudinal direction, N system reproduction signals output from the N reproduction heads are input, and N coefficient C _j (coefficient C ₁
.About.C _N ), respectively, and outputs them by multiplying them by N coefficient multiplying means, adding means for adding the outputs of the N coefficient multiplying means, and identifying the reproduced digital data from the output of the adding means. Means, the N-system reproduction signal, the output signal of the adding means, and the reproduction digital data, the value of the N coefficients C _j is adaptively controlled by the minimum error type automatic equalization algorithm. A magnetic reproducing device comprising a coefficient calculating means.