JPH02295227A - Data modulating method - Google Patents

Abstract

PURPOSE:To improve the recording density by limiting the continuous same codes excluding O within a prescribed number, the continuous codes of O within a prescribed number less than that of the continuous same codes, and the length where the adverse codes are alternately repetitive within L bits respectively. CONSTITUTION:In a Fig. A showing the modulation codes, two patterns A and B are available out of 8 patterns with L (bit number) and M (unit number) equal to 6 and 8 respectively. The record signals are shown in a waveform having a ternary logic levels as shown in a Fig. B against the modulation codes. Under such conditions, the continuous same codes excluding O are limited within a prescribed number together with the continuous codes of O limited within a prescribed number less than that of the continuous same codes, and the length where the adverse codes are alternately repetitive limited within L bits respectively. The prescribed m-bit data is converted into the prescribed L bits. Thus it is possible to reduce the loss and the non-linearity inherent to the magnetic recording and to improve the recording density.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention relates to a data modulation method, particularly a multivalued data modulation method for recording and reproducing a large amount of information on a magnetic medium (including a magneto-optical medium). This invention relates to a modulation code generation method.

In this case, the number of consecutive identical codes other than 0 is within a predetermined number, the number of consecutive 0 codes is within a predetermined number smaller than the above-mentioned predetermined number, and the length of alternately repeated opposite codes is within L bits, This is to improve recording density.

[Conventional technology]

Conventionally used modulation codes are based on the principle that 1.0 of a digital signal corresponds to the magnetization or the presence or absence of magnetization reversal in a magnetic medium. On top of that, M F M (
Modified Frequency Modulat
ion), 8-10 conversion, partial response, etc. to control the band, DC7! J
We are trying to unify them and match them with the system.

In other words, it is a modulation code that is based on the premise that it can be detected bit by bit.

[Summary of the invention]
[Problem to be solved by the invention] This invention solves the following two problems.
However, the magnetic particles in the magnetic medium have a continuous succession in the modulation method of the data converted into predetermined L pits of the magnetic special value. Therefore, when the amount of recorded information is increased, that is, when the recording density is increased using a binary modulation code, the maximum repetition frequency after modulation increases, signal deterioration occurs as shown in FIG. 8, and false detection occurs frequently.

Or influences other than the head magnetic field that act on the determination of residual magnetization (
(diamagnetic field, interaction of the magnetic particles themselves, etc.) causes nonlinear sign interference called peak shift, making detection impossible.

The losses during recording include, for example, recording demagnetization, separation loss (loss due to the distance between the tape and head), thickness loss (tape thickness loss), and self-demagnetization (self-loss that occurs within the tape). The losses during reproduction include gap loss, separation loss, and azimuth loss.

Even if we try to increase the recording density by using binary modulation codes in this way, various losses occur during recording and playback, so we have no choice but to rely on improving the performance of magnetic media and magnetic heads, and we are not satisfied. It has been difficult to measure the improvement in recording density over time.

The present invention has been made in view of these points, and provides a data modulation method that can improve recording density.

[Means to solve the problem]

This invention provides a data modulation method for converting predetermined binary m-bit data into multi-value predetermined L bits, in which consecutive identical codes other than 0 are within a predetermined number, and consecutive 0 codes are within the predetermined number. The number of bits is within a predetermined number smaller than the number of bits, and the length in which opposite signs can be alternately repeated is L bits or less.

[Effect]

The number of consecutive identical codes other than 0, ie, 1 or -1, is within a predetermined number, for example, 3. The number of consecutive zero codes is within a predetermined number less than the above predetermined number, for example, within two. The length of the opposite sign, that is, 1 and -1 that can be repeated alternately, is within L bits. By performing code conversion that satisfies at least these three conditions, it is possible to reduce loss specific to magnetic recording and nonlinear (intersymbol interference) and improve recording density.

〔Example〕

Hereinafter, one embodiment of the present invention will be described in detail based on FIGS. 1 to 7.

In this embodiment, first, the data string is divided into a plurality of pits, and if the length of the modulation code to be recorded is L bits, this becomes the basic unit during demodulation. The explanation assumes that the modulation code length is L bits, the number of patterns is M, and the length on the magnetic tape corresponding to L bits is lμm, and the pattern distance is Di.
j, and the distance between the closest patterns is pmin.

FIG. 1 shows an example of the relationship between the modulation code, recording signal, and reproduction waveform used in this embodiment.

FIG. 1A shows the modulation code, in the case of L=6 and M=8,
Among the eight patterns, two patterns, A and B, are shown. The recording signal for such a modulation code is
It is represented by a waveform having three logic levels as shown in FIG. 1B.

FIG. 1C shows the divided waveforms obtained when reproducing from a magnetic medium, where the broken lines are each isolated waveform, and the solid line is the reproduced output obtained by adding these isolated waveforms. Here, the following can be said about the isolated waveform. In other words, if we consider a case where a magnetization vector pointing in a certain direction exists independently in a region where the magnetization direction is random, the winding of the head will change when the head gap passes through the leading and trailing ends of the magnetization vector. A current corresponding to the change in magnetization flows through the wire, and the presence and polarity of the magnetization vector can be detected. The reproduced waveform detected at this time is called an isolated waveform. This can be said to be the response waveform of a single magnetization with a certain direction. For details of this isolated waveform, please refer to Japanese Patent Application No. 63-322699.

FIG. 2 shows the distance between pattern A and pattern B in FIG. 1. Waveform W1 is the reproduced output waveform of pattern A, and waveform W is the reproduced output waveform of pattern B. The distance between both patterns at each sampling position is determined as <d, ~d6 as shown in Figure 2.
Then, the distance D between pattern A and pattern B is D = a,
+ d2+ tL+ a4+ ds+da. Let M be the number of patterns recorded in length β μm on the magnetic tape corresponding to L bits on the magnetic medium (M-1
) There is an inter-pattern distance of M/2.

Figure 3 shows the pattern arrangement when M = 8 and L = 3. If the magnetic medium maintains complete linearity, that is, there is no uncertainty factor, each pattern will be arranged from P1 to P1 in Figure 3. Pe
It is assumed that it is placed at the position of In reality, there is a certain spread due to uncertain factors such as noise generated by the magnetic medium, nonlinearity, system noise, and jitter. In Figure 3, this is represented by an ellipse. In Figure 3,
The X+y and z axes represent the reproduction output level at the sampling position, and when L=3, there are three sampling positions.

The probability of making an error during playback depends on the pattern to be detected and
It is determined by the amount of overlap between the area of the uncertainty factor and the pattern at the closest distance (1) min). For example, in FIG. 3, if the pattern to be detected is P2, the pattern closest to this pattern P2 is P3. When these areas of uncertainty factors are respectively represented by broken lines as shown in FIG. 4, the areas where the areas of uncertainty overlap, that is, the areas indicated by diagonal lines, are the areas where errors occur. And when there is no overlap, no errors occur. In other words, in order to prevent such overlap from occurring, it is necessary that Dmin>a+b, where a and b are the lower limits. In addition, the magnitude of the uncertainty factor is
It is non-uniform because the nonlinear factors of the magnetic medium differ depending on the pattern. In the modulation code of this embodiment, one of the important points in creating the modulation code is to maximize ΣDIJ and equalize Dij, that is, to maximize Dmin.

Here, in this embodiment, the performance required of the modulation code is summarized as follows.

1) The distance Dmin between the closest patterns should be maximized, 2) The output obtained during playback should be large, 3)
4) No error propagation; 5) DC-free code; 6) Intersymbol interference occurs linearly (among uncertain factors, this is due to the magnetic medium). ).

In other words, ■, ■, ■ are the conditions based on this invention obtained by improving the conventional one, and ■, ■. ■ is a condition generally used when creating codes. Note that ■ and ■ may not be necessary depending on the system.

Here, as a specific example, the modulation code is ternary, L=3, M=8
Let's take the case of , for example, and consider it in relation to the performance required of the above-mentioned modulation code. Here, regarding the required performance (2) where the code 1 indicates magnetization M having a certain direction and -1 indicates the opposite direction, there are 27 modulation codes from 111 to -1-1-1, of which 8 are selected. The maximum distance on the sign is 111-(-
1 -1 -1) = 6, and in order to maximize Dmin, it is sufficient to select every 3 to 4 at equal intervals.

Regarding the required performance (2), the repetition of 1 and -1 should not be repeated more than L times, that is, not more than 3 times. Incidentally, magnetization reversal M
．． If M is repeated, it will cause a decrease in output as shown in FIG. This is due to an increase in loss due to gap loss, recording demagnetization, and self-demagnetization. In FIG. 9, the broken line represents the isolated waveform, and the solid line represents the reproduced output obtained by adding each isolated waveform, and the detected amplitude s. =s, =s. =s. There is a relationship between

Regarding the required performance ■, there are similarities with the required performance ■.
If magnetization reversal is not performed appropriately, the erasure rate will decrease and nonlinearity due to the interaction of the demagnetizing field and magnetic particles in the magnetic medium will increase.

Regarding the required performance (2), in the case of three values, the DC component can be simply looked at the sum of the signs. For example, the sign is 1, 0, 1, 0, 1
, 0. -1. 0, 1. When it is 1, the sum of its signs is 2, so in this case, there is DCt on the 10 side. Therefore, either the sum within a block of codes is 0, or the sum within 2L is 0 depending on the combination of modulation codes.
It is enough. However, in order to satisfy the required performance (2), it is better not to set a convention between successive modulation codes when recording excess data.

From the above, the following can be concluded in this example.

■ There must be no more than 3 M (or M) in a row, ■ 0
The number of consecutive times is within 2, ■ The number of alternating repetitions of M and M is within L, ■ The DC component is O in 2L (that is, the number of M and M within 2L is be the same).

Example of conversion code that satisfies the above ■～■ (M=8, L=3)
is shown in Figure 5. Nα1 marked * in Figure 5
The sum of the modulation codes of and Nα8 becomes 0 within 2L (that is, 6 bits), indicating that the signal becomes DC free.

Therefore, at this time, both codes are alternately switched for each IL. All other modulation codes are IL (3 bits)
The sum becomes 0 within, and DC is free.

An example of the conversion code in the case of M=8 and L=6 is shown in FIG.

In this case, the sum is 0 within the IL (6 bits), making it DC free.

FIG. 7 shows the relationship between recording density and error rate according to this embodiment in comparison with the conventional one, where curve a is conventional, curve b is
5 according to this embodiment. Incidentally, in this example, a vapor-deposited tape is used as the magnetic medium, and in the conventional case, 8/10 conversion is performed with binary bit detection, and in this example, M = 8, L = 6.
This is an example of ternary block detection. For example, when the pit error rate is lO-5, the conventional recording density is about 85 KBPI, but in this embodiment, the recording density is about 200 KBPI, which shows that the recording density is greatly improved.

In this way, in this embodiment, the modulation method is based on demodulation in units of multiple bits (L) rather than detection on a bit-by-bit basis, and the pattern that can be created by continuous magnetization (L pieces) corresponds to the code. Since a multilevel code is used to increase the distance between the patterns, the loss and nonlinearity (in intersymbol interference) peculiar to magnetic recording can be reduced, and the recording linear density can be improved.

In addition, although the above-mentioned embodiment has been described with reference to the case of three values as the multivalue, the present invention is not limited to this, and can be similarly applied to cases of other multivalues exceeding three values.

〔Effect of the invention〕

As described above, according to the present invention, the number of successive identical codes other than 0 is within a predetermined number, the number of consecutive O codes is within a predetermined number less than the predetermined number, and the length in which opposite codes are alternately repeated is L. Since the predetermined m-bit data is converted into multi-value predetermined L bits, the loss and nonlinearity peculiar to magnetic recording can be reduced and the recording density can be improved.

In addition, since this method can ultimately be applied entirely to LSI, it is possible to achieve higher system performance, higher reliability, lower cost, etc. Furthermore, the effects of improved performance of magnetic media and devices can be reflected to a greater extent.

[Brief explanation of drawings]

1 to 7 are diagrams for explaining the present invention, and FIGS. 8 and 9 are diagrams for explaining a conventional example. Agent: Hide Matsukuma Mori Pataso City (M・8, L = 3) Fig. 3 Q7niy+)α+b Fig. 4

Claims (1)

[Claims] In a data modulation method for converting predetermined binary m-bit data into multi-value predetermined L bits, the number of consecutive identical codes other than 0 is within a predetermined number, and the number of consecutive 0 codes is less than the predetermined number. A method for modulating data, characterized in that the number of times the opposite signs are repeated alternately is within the above L bits.