Device and method for magnetic recording/reproducing

The present invention relates to a magnetic recording / reproducing device and a magnetic recording / reproducing method for recording data on a magnetic recording medium with a recording head and reproducing a signal from the plurality of recording tracks with a single unit or a plurality of reproduction heads.

In recent years, in magnetic recording / playback devices, in order to increase the capacity of magnetic recording media, higher density recording is required, and the track width of the recording track is narrowed (hereinafter referred to as "narrowing"). Is required. Improving the accuracy of truck servos is the key to narrowing the width of trucks.

Therefore, a so-called non-tracking method has been proposed and has been put into practical use (for example, Patent Documents 1-5). In this non-tracking method, the magnetic tape recording / playback device records the target track once by dividing the data into blocks for identification on the track that has been double-azimuth-recorded by helical scanning. Data can be reconstructed even if it cannot be reproduced. This non-tracking method makes it possible to obtain more than four times the allowable amount for the position control of the playback head within one recording track required by the conventional track servo.

Further, the possibility of using the non-tracking method in a magnetic tape recording / playback device not only for helical scanning but also for linear recording is being investigated (for example, Patent Documents 6 and 7).

However, when a non-magnetic support having elasticity such as a polyester film is used for the substrate of the tape which is a magnetic recording medium, even if double azimuth recording is performed, the allowable amount of deformation is the track servo. In combination, for example, it was up to about twice the recording track width, and when further deformation occurred, the signal could not be reproduced with a sufficient SN ratio. Further, in the case of recording without double azimuth, the width of the so-called guard band that does not straddle the track is less than the deformation amount of the tape so as not to deteriorate the reliability such as the error rate even when the track servo is used together. I had to hold it down.

Such a problem is caused by the fact that in the signal reproduction method realized so far, the signal quality is remarkably deteriorated by reading a signal from a plurality of recording tracks at the same time by at least one reproduction head.

This is the same even in the non-tracking method, and the playback head appears to reproduce the signal across multiple recording tracks, but when time-divided, the reproduced signal is always one recording track. It was not done to play multiple recording tracks at the same time.

Considering further densification, the limitation of narrowing the track is that only the signal of a single recording track can be reproduced at the same time.

In the background technology of the magnetic head device, in order to improve the recording density, a technique of arranging a plurality of heads in one block and forming the same azimuth block is used to record a plurality of data frames at one time. There are (eg, Patent Document 8 and Patent Document 9).

Since these known techniques require the width of the playback head to be about half the width of the recording track, there is a restriction that the output of the playback signal cannot be increased, which leads to deterioration of the SN ratio and further high density. It was not always suitable for recording.

Further, by narrowing the width of the truck, there is a limit that the azimuth cannot be used. <patcit num="1"><text>Japanese Patent No. 1842057</text></patcit><patcit num="2"><text>Japanese Patent No. 1842058</text></patcit><patcit num="3"><text>Japanese Patent No. 1842059</text></patcit><patcit num="4"><text>Japanese Unexamined Patent Publication No. 04-370580</text></patcit><patcit num="5"><text>Japanese Patent Application Laid-Open No. 05-020788</text></patcit><patcit num="6"><text>Japanese Unexamined Patent Publication No. 10-283620</text></patcit><patcit num="7"><text>Japanese Unexamined Patent Publication No. 2003-132504</text></patcit><patcit num="8"><text>Japanese Patent Application Laid-Open No. 2003-338012</text></patcit><patcit num="9"><text>Japanese Unexamined Patent Publication No. 2004-071014</text></patcit>

<p> As described above, in the conventional magnetic recording / reproduction method, when it is desired to increase the recording density, it is necessary to narrow the recording track width. However, if the track width is narrowed as in the prior art, there is a problem that the track cannot be followed during playback. On the other hand, the playback head can read the signal even if it deviates from the reading track. , The non-tracking method has been proposed. However, in order to obtain an appropriate signal by the non-tracking method, there are restrictions on the setting of the playback head, so when aiming for a narrower track, build a system that takes a large SN ratio of the playback signal. Things were difficult.</p><p> Therefore, the present inventors have been developing a technique capable of reducing the constraint of determining the width of the playback head, narrowing the track width, and increasing the recording density.</p><p> However, in order to realize such a magnetic recording / reproducing device, a technique for separating the reproduction signals from a plurality of recording tracks obtained by the reproduction head into signals for each recording track is required.</p><p> Further, by narrowing the width of the recording track, there is a limit that the azimuth cannot be used, so that it is difficult to use the conventional method.</p><p> In view of such circumstances, the present invention is a magnetic recording / playback device capable of narrowing the recording track width and increasing the recording density, and the playback head width can be obtained by the playback head without depending on the narrowing of the recording track. A magnetic recording / playback device and a magnetic recording / playback method that can separate the playback signals from a plurality of recorded tracks into signals for each recording track and provide track servo accuracy equivalent to that when double azimuth recording is used. Is intended to provide.</p>

<p> In order to solve the above problems, the magnetic recording / reproducing device of the present invention has a recording unit that performs signal processing for recording on a magnetic recording medium and a reproduction capable of reproducing a signal from a recording track of the magnetic recording medium. The head reproduces signals from a plurality of recording tracks, obtains position information of the reproduction head for a plurality of recording tracks, and based on this position information and the reproduction signal from the reproduction head, generates a reproduction signal for each recording track. In order to give the reproduction unit the position information of the reproduction head for a plurality of recording tracks, which is provided in the reproduction unit and the recording unit to be generated, the reproduction unit has a recording wavelength equal to or higher than the minimum recording wavelength, and is between adjacent recording tracks. It is provided with an identification signal adder that adds an identification signal that does not interfere to the recording signal.</p><p> As the identification signal, signals orthogonal to each other on the time axis can be used. Further, as the identification signal, it is also possible to use a signal orthogonal to each other on the frequency axis. Further, the identification signal may be an orthogonal code.</p><p> When the identification signal is an orthogonal signal on the time axis, the identification signal is recorded between adjacent recording tracks so as not to overlap in the traveling direction of the magnetic recording medium.</p><p> Further, by arranging the identification signals in front of the data area and in the data area, the position information of the playhead with respect to the plurality of recording tracks may be recalculated in the middle of the data area. As a result, it is possible to improve the ability to follow changes in the positional relationship between the playback head and each recording track due to deformation of the media or the like.</p><p> A learning signal for automatic gain control and / or bit synchronization detection may be placed in front of the identification signal. Further, the identification signal is arranged after the synchronization signal used for detecting the identification signal start position.</p><p> Further, when a plurality of recording tracks are recorded as one group on the magnetic recording medium, an area called a guard for which recording is prohibited is arranged between the groups. As a result, it is possible to avoid interference with the recording track of the adjacent group.</p>

<p> As described above, according to the present invention, in the magnetic recording / playback apparatus capable of narrowing the recording track width and increasing the recording density, the playback head width does not depend on the narrowing of the recording track, and is determined by the playback head. The reproduced signals from the obtained plurality of recording tracks can be separated into signals for each recording track.</p>

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(First Embodiment)

FIG. 1 is a diagram showing a configuration of a recording unit of a magnetic recording / playback device using a multi-head according to an embodiment of the present invention, and FIG. 2 is a diagram showing a configuration of a playback unit of the magnetic recording / playback device. The number of recording heads is M, and the number of playback heads is N.

First, the configuration of the recording unit 100 of FIG. 1 will be described.

The recording unit 100 includes a data distributor 101, a recording encoder 102, a preamble adder 103, a delay element 104, a recording compensator 105, a recording amplifier 106, and a recording head 107. The number of the recording coder 102, the preamble adder 103, the delay element 104, the recording compensator 105, the recording amplifier 106, and the recording head 107 are provided by M each.

The data distributor 101 distributes the recorded data 1 input from the terminal 108 by the number M of the recording heads 107 according to a certain rule. The recording encoder 102 encodes the recorded data distributed by the data distributor 101. The preamble adder 103 adds a preamble including a training signal, a synchronization signal, and an identification signal to the encoded recorded data. The details of this preamble will be described later. The delay element 104 adjusts the phase of the recording signal for each recording head by giving a desired timing to the recording data to which the preamble is added. The recording compensator 105 performs the recording compensating process of the recorded data given the desired timing. The recording amplifier 106 converts the recorded data after the recording compensation process from voltage to current and sends it to the recording head 107. The recording head 107 records signals on the tape-shaped magnetic recording medium 2 to form M recording tracks.

The reproduction unit 200 includes a reproduction head 201, a reproduction amplifier 202, a variable gain amplifier 203, an LPF (low-pass filter) 204, an A / D converter 205, a preamble synchronous signal detector 206, a signal separator 207, an equalizer 208, and a PLL 209. , Detector 210, Synchronous signal detector 211, Decoder 212, Data combiner 213, Gain regulator 214.

The N minute signals read by the N reproduction heads 201 are amplified by the reproduction amplifier 202. The variable gain amplifier 203 changes the gain so that the amplitude level of the output of the reproduction amplifier 202 becomes a predetermined value. The gain regulator 214 controls the gain of the variable gain amplifier 203 by comparing the reproduced signal after the synchronization signal is detected by the preamble synchronous signal detector 206 with the gain target value. The LPF204 removes unnecessary high frequency components from the output of the variable gain amplifier 203 in order to prevent folding distortion. The A / D converter 205 quantizes the output of the LPF204.

The preamble synchronization signal detector 206 detects the synchronization signal in the briam from the output of the A / D converter 205 and identifies the start position of the identification signal. Further, the reproduced signal after the synchronization signal is detected by the preamble synchronization signal detector 206 is output to the gain adjuster 214. The signal separator 207 separates into signals for each of M recording tracks by using the identification signal in the brilliance.

The equalizer 208 performs waveform shaping on the reproduced signal separated for each of the M recording tracks. The PLL 209 performs bit synchronization with the output signal of the equalizer 208. The detector 210 detects the bit-synchronized signal with the PLL 209. The synchronization signal detector 211 detects the synchronization signal in the data area in order to identify the data start position from the output signal of the detector 210. The decoder 212 decodes the signal from the output signal of the detector 210 and the detection result of the synchronization signal detector 211. The data combiner 213 operates in the reverse manner to the data distributor 101 of FIG. 1, rearranges the arrangement of M data under a certain rule, and restores the recorded data.

Next, the basic recording and reproduction operations of the magnetic recording / reproducing device of this embodiment will be described.

First, the operation of the recording unit will be described. FIG. 3 is a flowchart relating to the basic operation of recording. First, the recording unit 100 distributes the input recording data 1 to the number of recording heads and M data strings by the data distributor 101 (step S101). Each of the M distributed data strings is encoded by the recording encoder 102 (step S102).

Next, a preamble composed of a training signal, a synchronization signal, and an identification signal is added to the beginning of the data string by the preamble adder 103 for each of the M encoded data strings (step S103). A predetermined delay amount corresponding to each recording head is given to the M data strings to which the preamble is added, and the delay element 104 delays the data strings (step S104). Then, the recording compensator 105 performs recording compensation for each of the M data strings, and the data is sent to the recording amplifier 106 (step S105). The recording amplifier 106 converts the sent data string from voltage to current (step S106). The recording head 107 records the sent data string on the magnetic recording medium 2 (step S107).

Next, the basic operation of playback will be described. FIG. 4 is a flowchart relating to the basic operation of reproduction.

In the reproduction unit 200, first, M recording tracks on the magnetic recording medium 2 are reproduced by N reproduction heads 201 (step S201). The minute N reproduced signal trains are amplified by the reproduction amplifier 202 (step S202). The gain of the amplified N reproduced signal trains is adjusted to a predetermined value by the variable gain amplifier 203 (step S203). For the N reproduced signals whose gains have been adjusted, unnecessary high frequency components are removed by LPF204 in order to prevent folding distortion (step S204). After the LPF204 is output, it is quantized by the A / D converter 205 (step S205).

The preamble synchronization signal detector 206 detects the synchronization signal during the preamble to identify the start position of the identification signal from the output of the A / D converter 205 (step S206). Further, the M reproduced signal trains are sent to the gain adjuster 214, an error from the gain target value is obtained, and the gain of the variable gain amplifier 203 is determined (step S210). In the signal separator 207, a channel is obtained from the identification signals of N reproduced signal sequences, and the signals are separated into signal sequences corresponding to the number of reproduced M recording tracks (step S207).

After that, the equalizer 208 performs waveform shaping on the reproduced signals separated for each M track (step S208). Bit synchronization detection is performed for each of the M reproduced signal strings (step S209). The signal of the reproduced signal string synchronized with M bits is detected by the detector 210 (step S211). The synchronization signal detector 211 detects the synchronization signal at the head of each of the M detected signal sequences and identifies the data start position (step S212). The M detected signal sequences are decoded by the decoder 212 (step S213). Then, the M decoded signal strings are returned to the original data arrangement by the data combiner 213 (step S214).

Next, a means for separating the reproduction signal for each recording track from the reproduction signals obtained by the plurality of reproduction heads 201 will be described.

In the following description, a linear tape magnetic recording / playback method will be described as an example.

A group is a group of M data frames recorded. Here, the data frame means the structure of the signal to be recorded. Considering the case of playing back with one playback head, for example, by changing the position little by little and playing back M times or more, all the recording tracks of one group are played back. That is, the recording track and the playback head do not have to have a one-to-one correspondence. For example, assuming that the recording track and the playback head are played back N times, it is sufficient that the N playback signals always include one signal of the M recording track. .. When there are N playback heads, if N &lt;M, the position of the playback heads is shifted and playback is performed M / N times or more. Also, when N&gt; = M, it only needs to be played once.

Hereinafter, the case of M = 4 and N = 4 will be described as an example.

FIG. 5 is a diagram showing a method of recording on the magnetic recording medium 2. From the figure, M data frames 3 are recorded on the magnetic recording medium 2, and these are treated as one group for signal processing. A plurality of these groups are arranged in parallel to each other on the magnetic recording medium 2 with an unrecorded area called a guard 5 in between. This guard 5 is for avoiding the signal being reproduced from the track of the adjacent group at the time of reproduction. In this example, since M = 4, one group is composed of four tracks (1)-(4).

Next, the preamble arranged in the data frame 3 will be described.

FIG. 6 is a diagram showing the configuration of the preamble recorded in the data frame 3 of one group. In this example, as the identification signal 13 for separating the M recorded signals from the N reproduced signals, signals orthogonal to each other on the time axis are used.

As shown in the figure, a training signal 11, a synchronization signal 12, and an identification signal 13 are sequentially recorded as preambles at the beginning of the data frame 3. Data 14 is recorded after this preamble. The training signal 11 is a learning signal used for training of a gain adjuster, a bit synchronization detector, and the like. The synchronization signal 12 is detected by the preamble synchronization signal detector 206 and identifies the start position of the identification signal 13.

The identification signal 13 does not overlap each other in physical positions so that the identification signals 13 (1), 13 (2), 13 (3), 13 (4) do not interfere with each other in the adjacent recording track for each data frame 3. As recorded). That is, the identification signal 13 (1) which is the identification signal 13 of the track (1) is in the T1 section, the identification signal 13 (2) which is the identification signal 13 of the track (2) is in the T2 section, and the identification signal 13 (2) which is the identification signal 13 of the track (2) is in the T3 section. The identification signal 13 (3), which is the identification signal 13 of the track (3), is recorded, and the identification signal 13 (4), which is the identification signal 13 of the track (4), is recorded in the T4 section. A gap Tg for a predetermined time is provided between the identification signals 13 of the adjacent recording tracks in order to avoid interference with the adjacent recording tracks.

Here, the identification signal 13 is a recording signal having a minimum recording wavelength or higher, and may be a repeating or random recording wavelength of the minimum recording wavelength. Preferably, it is desirable to use the same signal as the training signal 11 and the identification signal 13. This is because, for example, if a single frequency repeat signal is recorded in the training signal 11, the gain of that frequency is adjusted by the gain adjuster 214. Therefore, if the same signal is recorded in the identification signal 13, it will be described below. This is because the accuracy of the reproduction signal separation processing is improved.

The recording method of the identification signal 13 in FIG. 6 is an example, and is not particularly limited to this recording method.

Next, the reproduction method will be described.

The width of the playback head 201 is set to be equal to or wider than the width of the recording track. In the example of FIG. 6, the width of the playback head 201 is set to 1.5 times the width of the recording track, and the individual playback heads 201 (1), 201 (2), 201 (3), 201 (4) are set to 1.5 times the width of the recording track, respectively. Play signals from one or more recording tracks. The method of arranging the playback head 201 is not limited to FIG.

FIG. 7 is a timing chart of the signal waveform reproduced from each of the reproduction heads 201 (1), 201 (2), 201 (3), 201 (4), respectively. Here, the recorded data vector is X, and the vector of the reproduction signal reproduced by each reproduction head 201 (1), 201 (2), 201 (3), 201 (4) is Y. At this time, the channel characteristics can be expressed by the matrix H of N rows and M columns, that is, 4 rows and 4 columns, and the relationship between X and Y is given by the following equation.

<maths num="1"><img file="JP4697012B2_D0001.tif" /></maths>

Therefore, if the matrix H is regular, multiply by the general inverse matrix of the matrix H from the left of Eq. (1).

<maths num="2"><img file="JP4697012B2_D0002.tif" /></maths>

Therefore, the data for each recording track can be separated. Here, in order for the matrix H to be regular, the outputs of all the playback heads 201 (1), 201 (2), 201 (3), 201 (4) need be different.

Next, how to obtain the 4 × 4 matrix H will be described. here,

<maths num="3"><img file="JP4697012B2_D0003.tif" /></maths>And.

When the head position of the playback head 201 (1) in FIG. 6 straddles the track (1) and the track (2), the signal reproduced by the playback head 201 (1) is in the T1 section as shown in FIG. The signal of the track (1) appears in, the signal of the track (2) appears in the T2 section, and there is no signal in the T3 period and the T4 section. From this, the first row of the matrix of Eq. (3) can be obtained.

Here, the signal recorded at the kth of the p (p = 1,2,3,4) th track is r._pLet y1 (k) be the reproduction signal of (k) and the reproduction head 201 (1). From T1 section to h₁₁Is sought,

<maths num="4"><img file="JP4697012B2_D0004.tif" /></maths>Is.

As mentioned earlier, the signal recorded in the identification signal 13 may have the minimum recording wavelength, but the signal recorded here has a certain length and extends over the sections T1, T2, T3 and T4. The accuracy can be improved by taking the average of each.

Also, in the same way, from the T2 section to h₁₂Is sought,

<maths num="5"><img file="JP4697012B2_D0005.tif" /></maths>

Also, from the T3 section to h₁₃But from the T4 section to h₁₄Is sought. Here, there is no signal, so

<maths num="6"><img file="JP4697012B2_D0006.tif" /></maths>Is.

Next, consider the reproduction signal of the reproduction head 201 (2). When the playback head 201 (2) straddles the track (2) and the track (3), the timing chart of the playback head 201 (2) shown in FIG. 7 shows that the T1 section has no signal and the T2 section has no signal. The signal of track (2) appears, and the signal of track (3) appears in the T3 section. In addition, there is no signal in the T4 section. From this, the second row of the matrix of Eq. (3) can be obtained.

Let the reproduction signal of the reproduction head 201 (2) be y2 (k). From T1 section to h₂₁Is obtained and there is no signal

<maths num="7"><img file="JP4697012B2_D0007.tif" /></maths>Is. From T2 section to h₂₂Is sought,

<maths num="8"><img file="JP4697012B2_D0008.tif" /></maths>Is. H in the T3 section₂₃Is sought,

<maths num="9"><img file="JP4697012B2_D0009.tif" /></maths>Is. H in the T4 section₂₄ Is obtained and there is no signal

<maths num="10"><img file="JP4697012B2_D0010.tif" /></maths>Is.

Further, consider the reproduction signal of the reproduction head 201 (3). When the playback head 201 (3) straddles the track (3) and the track (4), the timing chart of the playback head 201 (4) shown in FIG. 7 shows that there is no signal during the T1 period and the T2 section, and T3. The signal of track (3) appears in the section, and the signal of track (4) appears in the T4 section. From this, the third row of the matrix of Eq. (3) can be obtained.

Let the reproduction signal of the reproduction head 201 (3) be y3 (k). From T1 section to h₃₁, T2 section to h₃₂Is obtained, and since these are no signals,

<maths num="11"><img file="JP4697012B2_D0011.tif" /></maths>Is. H in the T3 section₃₃Is sought,

<maths num="12"><img file="JP4697012B2_D0012.tif" /></maths>Is. H in the T4 section₃₄Is sought,

<maths num="13"><img file="JP4697012B2_D0013.tif" /></maths>Is.

Finally, consider the reproduction signal of the reproduction head 201 (4). When the playhead 201 (4) straddles the track (4) and the guard area, the timing chart of the playhead 201 (4) shown in FIG. 7 shows that there is no signal during the T1 period, the T2 period, and the T3 section. The signal of track (4) appears in the T4 section. From this, the fourth row of the matrix of Eq. (3) can be obtained.

Let the reproduction signal of the reproduction head 201 (4) be y4 (k). From T1 section to h₄₁, T2 section to h₄₂, T3 section to h₄₃Is obtained, and since these are no signals,

<maths num="14"><img file="JP4697012B2_D0014.tif" /></maths>Is. H in the T4 section₄₄Is sought,

<maths num="15"><img file="JP4697012B2_D0015.tif" /></maths>Is.

In this way, the matrix H is obtained. Therefore, this general inverse matrix can be obtained, and the signal corresponding to the recorded signal can be separated from the reproduced signal from Eq. (2).

Next, a recording / reproducing method when a signal orthogonal to each other on the frequency axis is used as the identification signal 13 for separating the N reproduced signals into the M recorded signals will be described.

First, the recording method will be described with reference to FIG.

In this case, the explanation becomes very complicated when M = 4 and N = 4, so the case of M = 2 and N = 2 will be described as an example.

FIG. 8 is a diagram showing a configuration of a preamble arranged in the data frame 3 when a signal orthogonal to each other on the frequency axis is used as the identification signal.

As shown in the figure, a training signal 11, a synchronization signal 12, and an identification signal 13 are sequentially recorded as preambles at the beginning of the data frame 3. Data 14 is recorded after this preamble. The training signal 11 is a learning signal used for training of a gain adjuster, a bit synchronization detector, and the like. The synchronization signal 12 is detected by the preamble synchronization signal detector 206 and identifies the start position of the identification signal 13. As the identification signal 13, a signal orthogonal to each track on the frequency axis is used.

Signals that are orthogonal on the frequency axis are 2 such as "10", "1100", and "11110000".^kIt is represented by (k = 0,1,2, ...) times the frequency. In FIG. 8, a signal having a frequency of 1 times "1010" is used for the track (1), and a signal having a frequency "1100" twice that frequency is used for the track (2). In this case, the frequency of the orthogonal signal is four times that of the signal to be recorded. Here, "1011" is used as the identification signal. This is not particularly limited to this case. Hereinafter, the orthogonal signal represented by the binary values of "1" and "0", the identification signal "1" is replaced with "-1", and "0" is replaced with "+1". The signal recorded on the track (1) is obtained by multiplying the identification signal "-1 + 1-1-1" by the orthogonal signal "-1 + 1-1 + 1".

<maths num="16"><img file="JP4697012B2_D0016.tif" /></maths>Will be. Similarly, the signal recorded on the track (2) is obtained by multiplying the identification signal "-1 + 1-1-1" by the orthogonal signal "-1-1 + 1 + 1".

<maths num="17"><img file="JP4697012B2_D0017.tif" /></maths>Is.

In the case of playback, multiply (16) by the orthogonal signal "-1 + 1-1 + 1" used for track (1).

<maths num="18"><img file="JP4697012B2_D0018.tif" /></maths>Will be. If this is integrated at the frequency of the recorded signal, that is, every 4 symbols at (18),

<maths num="19"><img file="JP4697012B2_D0019.tif" /></maths>Will be. The original identification signal "-1 + 1-1-1" can be restored by averaging this for threshold detection and every 4 symbols. Also, multiplying (17) by the orthogonal signal "-1 + 1-1 + 1" used for track (1),

<maths num="20"><img file="JP4697012B2_D0020.tif" /></maths>Will be. If you integrate this every 4 symbols,

<maths num="21"><img file="JP4697012B2_D0021.tif" /></maths>And even if (16) contains the signal of track (2), it disappears and disappears.

Next, the reproduction method will be specifically described.

The width of the playback head 201 is set to be equal to or larger than the width of the recording track. In the example of FIG. 8, the width of the playback head 201 is set to 1.5 times the width of the recording track, and the individual playback heads 201 (1) and 201 (2) each transmit signals from one or more recording tracks. Reproduce. The method of arranging the playback head 201 is not limited to FIG.

FIG. 9 is a timing chart of the signal waveforms reproduced from the reproduction heads 201 (1) and 201 (2), respectively.

Here, the matrix H of Eq. (3) is

<maths num="22"><img file="JP4697012B2_D0022.tif" /></maths>It becomes a 2 × 2 matrix of. How to find this matrix H will be described.

From the reproduction signal of the reproduction head 201 (1), the first row of the matrix of the equation (22) can be obtained. In FIG. 8, when the playback head 201 (1) straddles the track (1) and the track (2), the playback head 201 of the signal recorded at the k (k = 1,2, ..., 16) th The playback signal in (1) is y_kAnd.

H in equation (22)₁₁Is the playback signal y_kIt is obtained from the signal of track (1) included in. Therefore, the orthogonal signal "-1 + 1-1 + 1" of the track (1) is reproduced as the reproduction signal y._kHang on.

<maths num="23"><img file="JP4697012B2_D0023.tif" /></maths>Therefore, take the average for each of the four symbols and divide by the corresponding identification signal.

<maths num="24"><img file="JP4697012B2_D0024.tif" /></maths>Then, take the average value of the four calculation results of (24) and h₁₁And.

Next, h in equation (22)₁₂Is the playback signal y_kIt is obtained from the signal of track (2) included in. So, if you multiply the orthogonal signal "-1-1 + 1 + 1",

<maths num="25"><img file="JP4697012B2_D0025.tif" /></maths>Will be. Take the average for each of the 4 symbols in (25) and divide by the corresponding identification signal.

<maths num="26"><img file="JP4697012B2_D0026.tif" /></maths>Then, take the average value of the four calculation results in (26) and h₁₂And.

From the reproduction signal of the reproduction head 201 (2), the second row of the matrix of Eq. (22) can be obtained. As shown in FIG. 8, consider the case where the playhead 201 (2) straddles the track (2) and the guard area. As in the case of the reproduction head 201 (1), the reproduction signal in the reproduction head 201 (2) of the k (k = 1,2, ..., 16) th recorded signal is set to y._kAnd.

H in equation (22)₂₁Is the playback signal y_kDoes not contain the signal of track (1), so

<maths num="27"><img file="JP4697012B2_D0027.tif" /></maths>Is.

Next, h in equation (22)₂₂To ask. Playback signal y_kIs multiplied by the orthogonal signal "-1-1 + 1 + 1" of the track (2).

<maths num="28"><img file="JP4697012B2_D0028.tif" /></maths>Therefore, take the average for each of the four symbols and divide by the corresponding identification signal.

<maths num="29"><img file="JP4697012B2_D0029.tif" /></maths>Will be. Then, take the average value of the four calculation results of (29) and h₂₂And.

Finally, a recording / reproducing method when an orthogonal code is used for the identification signal 13 for separating the N reproduced signals into the M recorded signals will be described.

First, the recording method will be described with reference to FIG. Here, the case of M = 4 and N = 4 will be described as an example.

FIG. 10 is a diagram showing a configuration of a preamble arranged in the data frame 3 when an orthogonal code is used for the identification signal 13.

As shown in the figure, a training signal 11, a synchronization signal 12, and an identification signal 13 are sequentially recorded as preambles at the beginning of the data frame 3. Data 14 is recorded after this preamble. The training signal 11 is a learning signal used for training of a gain adjuster, a bit synchronization detector, and the like. The synchronization signal 12 is detected by the preamble synchronization signal detector 206 and identifies the start position of the identification signal 13. For the identification signal 13, an orthogonal code is used for each recording track.

A case where a known Hadamard matrix is used as the orthogonal code used for the identification signal 13 will be described as an example.

In the recording, since M = 4, the Hadamard matrix of order 4 is used. This matrix is expressed as follows.

<maths num="30"><img file="JP4697012B2_D0030.tif" /></maths> The rows of this matrix are used as the identification signal 13. That is, the first row of the matrix is assigned to the track (1), the second row is assigned to the track (2), the third row is assigned to the track (3), and the fourth row is assigned to the track (4). This is not particularly limited to this case. Further, for the sake of simplicity, the matrix of Eq. (30) is used as it is, but an appropriate sequence may be used instead of the matrix of Eq. (30). Therefore, these signals (eg, "-1" is phase-inverted) are recorded.

Next, the reproduction method will be described.

The width of the playback head 201 is set to be equal to or larger than the width of the recording track. In the example of FIG. 10, the width of the playback head 201 is set to 1.5 times the width of the recording track, and the individual playback heads 201 (1), 201 (2), 201 (3), and 201 (4) are set to 1.5 times the width of the recording track, respectively. , Play signals from one or more recording tracks. The method of arranging the playback head 201 is not limited to FIG.

FIG. 11 is a timing chart of the signal waveform reproduced from each of the reproduction heads 201 (1), 201 (2), 201 (3), 201 (4), respectively.

By Hadamard transforming the reproduced signal, it can be separated into the original recorded signal. Therefore, multiplying both sides of Eq. (1) by the Hadamard matrix of Eq. (30) from the right,

<maths num="31"><img file="JP4697012B2_D0031.tif" /></maths>Is. Than this,

<maths num="32"><img file="JP4697012B2_D0032.tif" /></maths>Is. Therefore,

<maths num="33"><img file="JP4697012B2_D0033.tif" /></maths>Then, the matrix H is obtained.

The identification signal 13 described above is always required in front of the data area, but an identification signal may be provided in the data area in order to obtain followability. As a result, the ability to follow changes in the positional relationship between the playback head 201 and each recording track due to deformation of the magnetic recording medium 2 or the like is improved. (Second embodiment)

Next, as a second embodiment of the present invention, a magnetic recording / reproducing device using a single head will be described with reference to FIGS. 12 to 15.

FIG. 12 is a diagram showing a configuration of a recording unit of the magnetic recording / playback device according to the present embodiment, and FIG. 13 is a diagram showing a configuration of a playback unit of the magnetic recording / playback device.

In the recording unit 300, a storage unit 110 capable of storing one group of recorded data is arranged between the preamble adder 103 and the delay element 104.

In the reproduction unit 400, a storage unit 215 is arranged between the preamble synchronization signal detector 206 and the signal separator 207. In one playback operation, the playback head 201 reproduces one group of recording tracks. A reproduction signal is recorded in the storage unit 215 for each reproduction until all the recording tracks are reproduced at least once.

Next, the basic recording and reproduction operations of this embodiment will be described.

First, the operation of recording will be described. FIG. 14 is a flowchart relating to the basic operation of recording.

In the recording unit 300, first, the input recorded data 1 is distributed to M data strings by the data distributor 101 (step S301). Each of the M distributed data strings is encoded by the recording encoder 102 (step S302). Next, a preamble composed of a training signal, a synchronization signal, and an identification signal is added to the beginning of the data string by the preamble adder 103 for each of the M encoded data strings (step S303). The M data strings generated in this way are recorded in the storage unit 110 (step S304).

Next, the data string for one recording track is read from the storage unit 110 (step S305). A predetermined delay amount is given to the data string by the delay element 104 (step S306). Then, the data string is subjected to recording compensation by the recording compensator 105 and sent to the recording amplifier 106 (step S307). The recording amplifier 106 converts the transmitted data string from voltage to current (step S308). The recording head 107 records the sent data string on the magnetic recording medium 2 (step S309).

After the recording of the data for one recording track is completed, it is determined whether or not the recording of the data for one group is completed (step S310), and if not (No in step S310), the storage unit 110 The next data string is read out and the process for recording is repeated in the same manner. The above operation is repeated until the recording of data for one group is completed.

Next, the basic operation of playback will be described. FIG. 15 is a flowchart relating to the basic operation of reproduction.

In the reproduction unit 400, first, the recording track on the magnetic recording medium 2 is reproduced by the reproduction head (step S401). The minute reproduction signal is amplified by the reproduction amplifier 202 (step S402). The gain of the amplified reproduced signal is adjusted to a predetermined value by the variable gain amplifier 203 (step S403). In the reproduced signal whose gain has been adjusted, unnecessary high frequency components are removed by LPF204 in order to prevent folding distortion (step S404). After the LPF204 is output, it is quantized by the A / D converter 205 (step S405). The synchronization signal in the preamble is detected by the preamble synchronization signal detector 206, and the start position of the identification signal is specified (step S406). Further, the reproduced signal is sent to the gain adjuster 214, an error from the gain target value is obtained, and the gain of the variable gain amplifier 203 is determined (step S409). The output of the preamble synchronous signal detector 206 is recorded in the storage unit 215 (step S407). For each playback of one recording track, it is determined whether or not one group of playback signal strings has been recorded in the recording unit 215 (step S408), and if not completed (No in step S408), the next recording track is played back. .. The above operation is repeated until the reproduction signal for one group is stored in the storage unit 215.

When the reproduction signal for one group is stored in the storage unit 215 (Yes in step S408), the reproduction signal for one group stored in the storage unit 215 is read out (step S410). The signal separator 207 finds a channel using the identification signal and separates it into a signal sequence corresponding to the number of reproduced M recording tracks (step S411).

After that, the equalizer 208 performs waveform shaping on the reproduced signals separated for each of the M recording tracks (step S412). Bit synchronization detection is performed for each of the M reproduced signal strings (step S413). A signal is detected by the detector 210 for the reproduced signal sequence in which M bits are synchronized (step S414). The synchronization signal detector 211 detects the synchronization signal at the beginning of each data area from the M detected signal sequences and identifies the data start position (step S415). The M detected signal sequences are decoded by the decoder 212 (step S416). Then, the M decoded signal strings are returned to the original data arrangement by the data combiner 213 (step S417).

(Third embodiment)

The present invention is not limited to being applied to a non-azimuth type magnetic recording / reproduction method, but can be similarly applied to a so-called double azimuth type magnetic recording / reproducing method having a plurality of azimuth directions.

FIG. 16 is a diagram showing a method of recording on the magnetic recording medium 2 by the double azimuth method.

In this example as well, as in FIG. 5, the case where the number of recording tracks M is M = 4 is taken as an example. Four data frames 3 are recorded on the magnetic recording medium 2, and these are treated as one group in the signal processing of reproduction. Here, the azimuth angle is changed from that of the adjacent group for recording. By recording azimuth on the magnetic recording medium 2, it is possible to prevent the interference of the recording tracks of the adjacent groups. Therefore, the adjacent groups are arranged without the respective groups sandwiching the guard area of FIG. .. As a result, the recording density can be further increased.

(Fourth Embodiment)

Although the magnetic tape recording / playback method of the linear recording method has been described above, the present invention can be similarly applied to the helical scan method.

FIG. 17 is a diagram showing a method of recording on the magnetic recording medium 2 by the helical scan method. In this example as well, as in FIG. 5, the case where the number of recording tracks M is M = 4 is taken as an example. Four data frames 3 are recorded on the magnetic recording medium 2, and these are treated as one group in the signal processing of reproduction. Guard 5 is placed between adjacent groups.

FIG. 18 is a diagram showing a method of recording on the magnetic recording medium 2 by the double azimuth helical scan method. In this example as well, as in FIG. 5, the case where the number of recording tracks M is M = 4 is taken as an example. Four data frames 3 are recorded on the magnetic recording medium 2, and these are treated as one group in the signal processing of reproduction. Here, the azimuth angle is changed from that of the adjacent group for recording. By recording azimuth on the magnetic recording medium 2, interference of recording tracks of adjacent groups can be prevented, so that adjacent groups are arranged without this group sandwiching the guard area of FIG. As a result, the recording density can be further increased.

<figref num="1">It is a figure which shows the structure of the recording part in the magnetic recording / reproduction apparatus using the multi-head which is one Embodiment of this invention.</figref><figref num="2">It is a figure which shows the structure of the reproduction part of the magnetic recording / reproduction apparatus of FIG.</figref><figref num="3">It is a flowchart about the basic operation of the recording part of FIG.</figref><figref num="4">It is a flowchart about the basic operation of the reproduction part of FIG.</figref><figref num="5">It is a figure which shows the method of recording on a magnetic recording medium.</figref><figref num="6">It is a figure which shows the structure of the preamble arranged in the data frame when the signal which is orthogonal on the time axis is used as the identification signal.</figref><figref num="7">It is a timing chart of the signal waveform reproduced from the reproduction head of FIG.</figref><figref num="8">It is a figure which shows the structure of the preamble arranged in the data frame when the signal orthogonal on the frequency axis is used as the identification signal.</figref><figref num="9">It is a timing chart of the signal waveform reproduced from the reproduction head of FIG.</figref><figref num="10">It is a figure which shows the structure of the preamble arranged in the data frame when the orthogonal code is used for the identification signal.</figref><figref num="11">It is a timing chart of the signal waveform reproduced from the reproduction head of FIG.</figref><figref num="12">It is a figure which shows the structure of the recording part of the magnetic recording / reproduction apparatus using a single head which is 2nd Embodiment of this invention.</figref><figref num="13">It is a figure which shows the structure of the reproduction part of the magnetic recording / reproduction apparatus of FIG.</figref><figref num="14">It is a flowchart about the basic operation of the recording part of FIG.</figref><figref num="15">It is a flowchart about the basic operation of the reproduction part of FIG.</figref><figref num="16">It is a figure which shows the method of recording on a magnetic recording medium by a double azimuth method.</figref><figref num="17">It is a figure which shows the method of recording on a magnetic recording medium by a helical scan method.</figref><figref num="18">It is a figure which shows the method of recording on a magnetic recording medium when double azimuth is used for a helical scan method.</figref>

Code description

1 Recorded data 2 Magnetic recording media 3 data frames 5 guard 11 Training signal 12 Sync signal 13 Identification signal 14 data 100 recording section 101 data distributor 102 Recording code 103 preamble adder 104 Delay element 105 Record Compensator 106 Recording amplifier 107 Recording head 110 storage 200 playback section 201 Playhead 202 playback amplifier 203 Variable gain amplifier 204 LPF 205 A / D converter 206 Preamble Sync Signal Detector 207 Signal separator 208 equalizer 209 PLL 210 detector 211 Synchronous signal detector 212 Decoder 213 Data combiner 214 Gain controller