JPS59146414A - Digital modulation method - Google Patents

Abstract

PURPOSE:To attain the high density of magnetic recording by realizing self- clocking and at the same time reducing greatly the maximum magnetization inverting interval while maintaining the performance of 3PM. CONSTITUTION:An FEM-4 converts the data of 4 bits into a code word of 8 bits and then modulates the series of the code word by an NRZI. For the 1st rule of modulation of the FEM-4, the 8th bit P8 of a code word which is old in terms of time series is inverted to bit ''1'' in case the next two continuous binary data of 4 bits are combined. Then the 3rd bit P3 of the following code word is inverted to bit ''1''. For the 3rd rule of modultion of the FEM-4, the 2nd code word P1P2P3P4P5P6P7P8 and 00001000 are allowed to correspond to the 2nd data ''O'' which is set at the center in terms of time series in the case a combination of the next three continuous binary data. Here a conversion table of modulation method (FEM-5) is shown for this invention. The 1st rule of modulation of the FEM-5 is equal to the 1st rule of modulation of the FEM-4, and the 2nd rule of modulation of the FEM-5 is equal to the FEM-4. Then the 3rd rule of modulation of the FEM-5 corresponds to the 3rd modulation rule of the FEM-4.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention particularly relates to a 4-8 conversion digital modulation method for converting 4 bits of data bits into 8 bits of a code word.

Conventional configurations and their problems In general, digital magnetic recording has the characteristics of being able to economically record a large amount of information and to store it stably over a long period of time. An example of a modulation method for information transmission for this purpose is RZ (Return.

Zero), RB (Return to Bi)
as), NRZ (Non-Return to Z
ero), NRZI (Non-Return to
ZeroI), FM (FrequeueyMod
ulatiern), PE (Phase Enc
oding), MFM (Modified Fr.
EQUITY MODULATION), M FM
Various methods such as (Modified MFM) have been proposed. However, recently, various new modulation methods other than these have been proposed. For example, / MNRZI
, 3PM
ern), ZM (Zer.

Moclulot engineering), etc. In the digital magnetic recording, the detection window width Tw is large and the maximum /JS
The magnetization reversal interval 'l' pin is large, the maximum magnetization reversal 1 interval T BOX is small, and the ratio OR (Density Rat
io) The crab is large and the magnetization reversal interval is T! It is said that a modulation system capable of self-clocking with a large product force of lI+I&detection window width Tw is desirable. Recently, MFM and 3PM are popular modulation methods. 1st
The figure shows a comparison of various modulation methods.

In Figure 1, C0DE RATE = - T, MIN knee (d-+-1) ・T T, Max = -GC+1) ・T Tw = -, T n CLOCK PATE MA - Tw d: bit 61 and bit The minimum number of bits 60〃 between \\1//, l: Binoto\\1 〃 and BiJ 1. 'NY1 u]・\\o Maximum number of rr As is clear from Figure 1, if the data period is T, then M
F M pJ, T min is T, Tmax is 2T,
OR is 1, Tw is 0.5, 'l'l1iI+×T
w =: 0.6 T' and overall balance and'! It can be seen that this is a different modulation method.

However, when achieving further l/(-higher density, 'f
A modulation method with a large pin is desirable. Zokote proposal i-
The other thing is 3PM. This way (I: As shown in Figure 1, Tvy=O, EiT, DR=
1.6. Tm1qXTW2O,715T and MFM improved by 2! 1 from T of LMFM to T min
, 5T, which indicates that there is a trend toward high-density Hfl. But Tm? Lx is MFM 2
Compared to T, this is 6T, which is a big step back. This is because when reproducing a recorded iris number, the center line of the waveform, that is, the baseline fluctuation appears greatly, so when converting the reproduced waveform to ・S, /1S with a gomparator, a special one large If this is not done properly, the reproduction margin will become small and demodulation error will occur. A method of integrating the reproduced waveform after equalizing it'
cd1, into the bulk type, the rad shows no differential characteristics, and its four DC components are lost, and it is impossible to reproduce the DC even if it is integrated (-7). The center line of the reproduced waveform is detected from the envelope t'l - below -1- of the cow waveform [7, and the so-called OCC reprecipitation method is often used, which uses it as a direct current.With this integration method, j4 Although it is strong against high-frequency Xf sound due to the integral type mechanism and low-pass-high-frequency reduction characteristic, on the other hand, as mentioned above, since DC reproduction is not required, low-frequency waste "?", for example・・・
)・It has a drawback that it is very weak. In addition, in the differential method in which the half waveform is differentiated (~, its peak value is detected and pulsed), if the maximum magnetization reversal interval Tmax is large, due to the differential! property, in the middle of the 6T interval (There is a drawback that false peaks due to sound occur. Therefore, in order to prevent false peaks, we masked the false peaks in the middle of the 6T interval and detected only the original peak peaks. A masking gate generation circuit is used.This method uses a method of pulsing the waveform before differentiation at an appropriate level of the amplitude of the main waveform.However, this method depends on the amplitude of the endogenous waveform. lr-me, there is a change of heart in which the level of the reproduced waveform is constant.Therefore, AGC, that is,
A method to always keep the level of the main waveform constant is essential.If the maximum magnetization reversal interval Tmax is large, the margin for jitter becomes small.In this way, if the maximum magnetization reversal interval Tmax is large ( The shortcomings of

In addition, the 3PM force type 1d3bi・1・butabisoto is 6bi・
The method of converting to ``G'', the modulation and demodulation of the gods, and the ``pu'' that creates the modulation/demodulation Kuro-ri, - are a little simplistic 1, /3 frequency division 2:
In addition, when configuring a circuit using commercially available general-purpose Ron ri, there are many circuits with a 4-bit configuration, and in digital audio etc. One zumbling data is 16 bits, not Va times 3,
This means that if the modulated signal is demodulated, demodulation errors due to dropouts and peak shifts will occur, but the error propagation in that case will be in units of 4 bits. The 4-bit high-order conversion method is more advantageous in that it can be suppressed within one data segment.

For these reasons, the inventor proposed 3 as another modulation method.
FEM-1 (Fouv t) has performance comparable to PMf1.
o Eight Modulation-1), the maximum magnetization reversal interval Tmax has been improved from 6T to 5.5TF
E M-2, further improved Tmax to 5T and 2 de hF
EM-3. Furthermore, Tmax was improved to 4.6T (7 FE
I have already proposed M-a.

- First, before explaining the modulation method of the present invention, 3PM and FE
Let me explain about M-4.

FIG. 2 is a conversion table for the SPM system, which converts 3-bit data into a 6-bit code word and modulates the sequence of the code word with NRZI. The code word is p1
．． (At least two bits Xt 1 tt, if a sequence of code words occurs in which the first bit P1 of the following code word is SS 1 //, bit 61 〃 and the next bit 8 \1
Since the condition that the minimum number of consecutive bits N○〃 that are inserted between The pit P1 is inverted to bit SS □ //, and the 6th pit P6 of the chronologically older code word is changed to bit SS 1
There is a special rule for reversing //. Furthermore, if the 6th bit P5 of a code word is SS1// and the 1st bit P1 of the next code word is S11//, the minimum magnetization reversal interval will be T, so in this case, the above 6th bit P5. Invert both the 1st pit P1 to st □ tt and change the 6th bit i-th P6 which is always SS □ // to \\
1, and the reversal interval is maintained at 1.6T. Furthermore, in this aPM system, the maximum number of 7XSO1/s that can be inserted between bits XX 1 and XX 1 of the codeword sequence is limited to 11. Therefore, the code words are selected so that the maximum inversion interval is 6τ.

Next, the maximum magnetization reversal interval was improved from 6T to 4.6T,
FEM-4, which maintains the performance of 3PM except for the above, will be explained. FIG. 3 is a conversion table for FEM-4.

This method converts 4-bit data into 8-bit code words and modulates the code word sequence with NRZI. The code word is bit % 1 // and the next bit St
At least 2 bits between 1//\\o
〃 is included, and one restriction is added so that the number is at most 8 or less. The first modulation rule of FEM-4 is to invert the 8th bit P8 of the chronologically older code word to bit \\1 for the following two consecutive combinations of the 4-bit binary data in Figure 3. Furthermore, the third bit P3 of the code word following this is inverted to bit \\1. As this continuous combinational pinoli data, XX2 // (N[L3) -10〃(N(L
1)XX3. tt, (4) →60//
(Nα1) 8\4// (N(L5) →XXO
// (Nα1)\\8 // (Na 9)
→XXc)' // (Na 1)"19〃(N
ulo) →SSO//(N11)ss 15//
(N(116), XXOtl (N[Ll
)62〃(Nα3) →XX1 // (Nα
2) XX3 // (Nα4) →ss 1
tt (Na2)XX4 // (N[L5)
→1\1〃(Nα2)XX9 // (N[L 1
0) →XX1 // (N[L2).

The second modulation rule of FEM-4 is that in the case of the following two consecutive combinations of the 4-bit binary data in FIG. Py P801
001001. d as this continuous combination binary data.

XX4 // (H(15) →SS 2. /
/ (Nu 3) SS 4 // (Nα5) →
' 5 // (Nl16).

The third modulation rule of FEM-4 is that in the case of the following three consecutive combinations of the 4-bit binary data in FIG. the second code word PI P2 Ps P4 Ps P6 P7 P2O0
001000. This continuous binary data is XX1// (N(L2) −+SS O// (N
lil) →S\5 // (N17)-+XX9
(N (L 10) →1
) →XX6 II (Nα7) →XXg l/ (
Nα1Q) →XX10//(Nα11) →XX14 // (Nα16) →XX16〃(Nα16) XX13 tl (Nu 14) → SS □ // (
N (L 1 ) →XX6 II (Nα7) →\\
9〃(Nα1Q) →(S 1Q // (Nα11) -→614// (N[Ll 5) →λ\16〃(Nα16) \\14〃 (T1α16) -→\\Q 〃 (
NIll) →＼＼ 6 // (Nα7)−
→ts 9 // (Nα1Q) →XX10Il (Nα11) →614〃(Nα15) →XX16〃(Nα16).

The fourth modulation rule of FEM-4 is based on the code word sequence pattern in the concatenated part of the code words in Figure 3, which is based on the γ-hyphen P7 and the 8th bit P8 of the Kasuso language, which is older in time series. The first bit P3 of the code word following is S 1o1 ttj
In this case, invert the 8th bit P8 to bit ＼＼1, and reset the 7th bit I, P7 and the 1st bit P1.
'XO//. For example, the binary data in Figure 3 "51/ (lt116) is 1\6 (
This is the case when Nα7) continuously overlaps. These data other than the modulation rules are made to correspond to the first code word in FIG. Changes in code words in the application examples of the first to fourth modulation rules of FEM-4 are as shown in FIG.

The maximum magnetization reversal interval Tmax of FEM-4 occurs when the following binary data in FIG. 3 are consecutive.

XVl // (Nα2) →XX□ // (
Nα1) SS 3 // (Nα4) →s<
2 // (t's)1\4〃(Nα6) →(
% 31/ (Nα4)λ\7〃(Nα8) →
6Ol/ (Nα1)λ\13//(Nα14)
→\\0〃(Nα1)XX14 // (Nα15)
→XXO// (Nα1)SS 16 //
(Nα16) →\\1〃(Nα2)\\8〃(
Nα9) →ss 1 // (Nil2) However, this is limited to the case where the first to fourth modulation rules of FfCM-4 are applied in this order in time series.

FIG. 6 shows an example in which the maximum magnetization reversal interval Tmax=-4.6T occurs.

In addition to this, 4-8 conversion algorithms and conversion tables with magnetization reversal intervals of 1.6T to 4.5T are possible, but in such cases, when demodulated, the original binary data cannot be reproduced. If there is no data, or if it corresponds to two binary data, it will be 1-.

Now, we will briefly describe demodulation algorithms for FEM- and * that are currently being considered. First, a code word to which the first modulation W5 rule is applied is demodulated. 1) By using some means, for example, detecting a synchronization signal used in a digital audio code or the like directly from a modulated data codeword sequence before demodulation, and thereby detecting an 8-bit code from a continuous codeword sequence. Word boundaries are detected, each codeword is divided into 8-bit units, and the 8th bit P8 of the fortune-telling codeword and the 3rd bit P5 of the next codeword are both bi... If ＼＼１〃, then the 8th bit P8 of the fortune-telling code word and the 3rd pin of the next code word in the chronological order
Invert both p3 to SS Q //.

Next, the code word applied with the second modulation rule of F fi: M-4 is demodulated. This is done by pattern matching out of the continuous modulated code Rtj sequence.
Detects the second code word of Nα5 in the figure, and further follows <
n';3'δt (is Nα3 or N[L6 in Fig. 3)
In this case, for the second code word of JI2 Nil 5 before UL, the binary data XX4 (0
100) and demodulates it. Similarly, Nα in Figure 3
After the second codeword of 5, the next 8-bit codeword PI
P2 Ps P4 Ps P6 P7 Pa○ 001
0001 continues, the second code word of Nα6 is associated with the binary data 641/(0100) of 14α6 in FIG. 3 and demodulated. This is because the binary data of N[L6 in FIG. 3 is followed by 11α7 and N[L 10 , Nα11 . N
If the sharp data of α15°Nα16 continues, FEM-4
The 7th bit P of the code word of Nα6 according to the fourth modulation rule of
This is because 7 is inverted to a bit \\0, and the eighth digit P8 is inverted to a bit \\1.

Next, the code word to which the third modulation rule of FEM-4 is applied is demodulated. This simply detects the second code word of Nα1 in FIG.
Demodulate the binary data ss □ // (OCOO) of α1.

Next, the code word to which the fourth modulation rule of FEM-a is applied is demodulated. This means that out of a continuous codeword sequence,
The 8th bit □ of the code word P8 is a bit)\1 Code word valley detection -: The 7th bit or 7 of this code word and the 1st bit P1 of the following code word □ are bit SS 1 //'
After f inversion, the 8th bit P8 of the chronologically old code word is inverted to bit XX'(>7/'. The code word obtained as a result is the first code word in FIG. 3. Therefore, binary 1)'' data is demodulated corresponding to each first code word.

The demodulation order of the FgMff4th third modulation rule and the FgMff4th modulation rule may be performed in any order. Finally, the first to fourth modulation rules described above are applied and the successful code word is demodulated corresponding to the first code word in FIG.

Objectives of the invention: '□1 The present invention provides an FEM-4 that maintains the performance of Mi-3'PM'.
The purpose of the present invention is to provide a digital modulation method that can improve the TmfLX.

Structure of the invention □ 1 Digital modulation method of the present invention is capable of self-clocking, and Tl1rn is 1; s'I', Tw is 0,
BT, T#Irn'X'Tw='O,”
Maintaining the performance of '76 T'' and 3PM, the above-mentioned 3
By greatly improving Tmax = 6T, which is a drawback of PM, to 4T, we have achieved higher density. In other words, this invention increases the Tmax of FKM-4 to 4T.
It is an improved type and is called FE, M-6.
DESCRIPTION OF THE EMBODIMENTS FIG. 6 is a conversion table of the modulation method (FEM-5) of the present invention. This method also converts 4 pino l- data into 8 bits)
The code/word that is converted into the code/word and the code word sequence is modulated by the NR2 machine. The first modulation rule of FEM-6 is FEM-4. is the same as the first modulation rule. In this case, the combination of 4-bit bicooler data is S< 2 /l (N[L3) →N\0 /l
(N[Ll)"s< 3.n (NI14')
→5sOt/' (N(L 1) ・Ant 4
〃(Nl16) □→＼＼0〃(Nα1) □
\(g /i (14α10) →lX□ //
(Nα1) SS 1'5 // (Nα16)
→＼几〃τN11L1)λ＼2〃(N[L3)′
→\\1〃(Nα2)ss 3 // (Nα4)
→SS 1//” (N(12)SJ” 4
// (N(L '5) →\\1〃(N
[L2)XX' 9” // (Nα1g) →
XX1' // '(Nα2)'St 15 // (
Nα16) →SS 1// '(Nα2), and the answer 9 of □FEM-4 is removed and transmitted. □ 68〃(Nl19) −〇′/i(N(11
) The second modulation rule of FEM-15 is also the same as F'EM-4. In this case, the combination Buy) - Redata and so is XX4' // (N(L'S) □ → Senru 7
/ '('Nα3) Sl 4 // '(Nα6)
→N6//hiring α6) 1N' 4// (N1
15) ” →s＜”7'
>i ′('N[l'8n□ \\4 // (N[L
6) →N11〃(NIl'12)XXB
// (,Nα9) →λ\0〃(NIL'1
)'\\8 /'(Na9)'-+XX1'/
/ jN[L2] Look, it's transparent 4 more times than FEM 24. For the following three consecutive combinations of 4-bit Nobuinaride-Yu in Figure 6 for Female, this combination has the same code word number correspondence as the '2nd modulation rule: Tojite. is' 1\\4〃
Data other than (Nα6) → &'3〃4)-\2〃Kan3) □ Next, the third modulation rule of FfCM-es corresponds to the modulation rule of FEM-4A:3 In the case of the following consecutive M iii (7') combinations, the code of the third figure and the second row is used for the following consecutive M iii (7') combinations: Word PI P2 P3P4 P5 PλP7 P8'000
01000 - In this case, the combination is 1〃6α2
) →\10iα1) io /;' Other than (/4a1□) Nata XX7”// f4(IB) ” -) SS
'Of' (N(Ll)-\o ti liα1)
Data of [data] 13 //l(L 1""4)-d
s Q”// Ill 1) -JS □ l) I
Data other than (L 1 ) Q14//Tsuri J1 et al.) -+
X'0"//%α1) 40 data other than 1)%
'O' // (N(L"1") '-+XX' o
//'(N(11) →ss 6〃(Nα7)→6 9
11 (N[L'10) →\\1o//(Nα11 n one generation 14//(Nα16) →\\151/ (Na 16) XX5// (Nα6) →\\O//(N[Ll) →
\\ 6 // (Nα7) → Is g /7 (N
α10) →\\10//, (Nα11) →\X141/ (Na16) One generation 16/I (N[l 16) \\6〃(Na7) →\\○// (Nα1)→
(S 6 // (N[L7)-, +XX9 //
(N[l 10)→\\10//(Nα11) →\\141/ (N[l 15) →\\15//(Nα16) XX10〃 (j4I111)→\\o // (NI
L 1 ) →\X 6 // (Na7) →\\g
// (Nα10) →A1011 (Nl111) →\\ 14//(Na 16) →\\15//(Nα16) Ss 11 u (Na 12) →\\Ol/ (N
u 1 ) −+\\ 6〃(Nα7) →\\ 9 /
/ (Nα1Q) →\\to l/ (Nα11) →\\14 // (Nα15) -door\li // (Nα16) \\12 // (Nu 1 3) →\\O
〃 (N[11) −+\X (, n (Na '
7)→\\Q n (N[l 1o) -i10 It (14α11) →\\l/L 7/ (N[l 16)→\\I”;
// (NO, 1e). In this case as well, the combination is largely different from the case of FEM-4.

Next, the fourth modulation rule of FEM-5 is as follows, so for the combination of the first word, the second binary data 60〃(00
0o), the third code word P+ P2 P3 P4 Ps P6 Py Psoo in FIG.
Make oo 1 00 1 correspond. In this case, the combination is 'It
1 1/ (N[L2) −+X1 0 /
/ (NOll) −+'h O〃 (N(L
l ) XX1 // (NO, 2) -IO//
(Na1)→＼＼1〃(Nα2)XX7〃(NIL8)
→\\0 // (N(11)-+\\o // (N
ul)\\7 〃 (Nα8), -t\\ ○ //
(No, 1) →\\ 1 It (Na
2) VNl 3 // (Nnl 4) ->” O〃
(Na1) −)\' O// (Nl11)XX
13 // (1suα14)→\\0〃(Nα1)- and XX1// (+4α2)XX14 ” (Na 1t
s ) →\\o 〃(Na1) -+\\o /'
(N(Ll)XX14//(1・4α15)→6O/
/ (Nα1)→＼＼ 1〃(Nα2). This is a modulation rule that does not exist in i7iFEM-4.

Next, for the 6th modulation rule shift 1 of FEM-5: the following Bai-Kari-j-Y combination, the second sign in Fig. 6 is applied to the Pa-Kali data (○011) of Nα4. C3
nrtenPIP2PsP4P5PaPyPg 10001001'17-[Response group. In this case, the combination is
X3 tt (Na4) → XX2 // (Sound 3)
XX3 // (Nα4) →\\6〃(Nα6). These are modulation rules that do not exist in FEM-4.

Then, the sixth modulation rule of FEM-15 is FEM
-4 is the same as the fourth modulation rule.

Other data are transformed in accordance with the first transformer rugotazism shown in FIG.

The code-redundant data of 8 hits modulated in this way is demodulated according to the first to sixth modulation rules, and
It has a one-to-one correspondence with binary data in units of humans. Changes in code words in application examples of these modulation rules are as shown in FIG.

Figure 1 shows a comparison between FEM-5 and other various modulation methods. When demodulating the recording waveform while maintaining the performance of the 3PM method, the maximum magnetization reversal interval Tmax, which is likely to have the most adverse effect as described above, is increased from 3PM's 6T to 4T'!!: 1.
1] is an innovative modulation method that has been improved. The software configuration is also FieldProgr? Lmable L
By combining logic array and microprogramming techniques, it is possible to realize TiJ function more easily than 3PM.

FIG. 8 shows recording current waveforms for various modulation methods.

Effects of the Invention As detailed above, according to the present invention, self-clocking is possible, TsiR is 1.6T, Tw is Q, 6
When maintaining the 3PM performance of T, Tmn X TW = 0.75 T and demodulating the recording waveform -
The maximum magnetization reversal interval Tmax that is likely to have the most adverse effect on
It has the advantage of being able to improve rlJ up to 4T.

[Brief explanation of drawings]

Figure 1 is a comparison diagram of various modulation methods, and Part 2 is a comparison diagram of various modulation methods.
An example conversion table diagram of the SPM method, Figure 3 is FfC
M-4 conversion table diagram, Figure 4 is a diagram showing application examples of each modulation rule of Fli:M-4, and Figure 6 is the recording current waveform of the maximum magnetization reversal interval Tmax type modulation method in Ffi:M-4. FIG. Name of agent: Patent attorney Toshio Nakao and 1 other person
1 Figure 2 Figure 3 □□□□-” fds 4 figure
oooo f. “-mouth-C1〒-two-one-one, six-cho-one-one-j-
u00000 100000000 rO-4-0'+
O□ Bu oo Bu 0-1-1-y1----,
-, y ding-ul, 5 ding-January snoring present 11mAl'JJΔ
1 draw oooooo on the bill, 2π 琶BO-, 儒f0 連 oooooo4o-1-Eiko')'00o010゜-□--35 cho-----÷←--3 cho-1111 Fifth
Figure 00jf00f0 0010000000QjOOO0
-1--=-4y-d-□---- oiooooo f zu 0 1 0000000
0 1 000001=-=--4,51----
11+1+00QQO70000f 00000000
10 rays----1-4sT----11'' et al.
tsu . −()(ノ

Claims (2)

[Claims] (1) The data bits are divided into 4-bit units, and the following 8-bit code words (P1P2P3P
4P5P6P7 P8) ■00000010 ■00000100 ■00010000 (■00100000 ■01000000 ■00010010 ■10010010 ■00100100 ■01001000 010010QO0 ◎10000010 ・0010 0010 [Phase] 01000010 [Phase] 01000100 @hoooo 1o. [Phase] 10001000 are made to correspond, and the code word ■ , ■, ■, ·, o is followed by code word ■, or code word 0.■, ■, O
If any of the 1 [phases] is followed by the wa code word ■, the 8th bit of the fortune-telling code word is chronologically picked. sj 1
//, and further reverses the chronologically new 3rd bit P5 of the new code word to bit 61, and the code word ■ is followed by any of the code words ■, ■, ■, or O, or When codeword ■ is followed by either codeword ■ or ■, or when codeword ■ is followed by codeword ■ and further followed by a codeword other than codeword ■, codeword 01001 is used instead of codeword ■. 001 is assigned, and one of codewords ■, ■, 0.0 is followed by codeword ■ and is further followed by a codeword other than codeword ■, or codeword ■ is followed by codeword ■ and further codeword In the order of ■, ■, O, if either O follows, the code word ■ is followed by the code word ■, and then the code word ■, ・
, ■, ○, [phase] If any one of the following is followed, the code word 0 is followed by the code word ■, and then the code word ■, 0, 0 .
□, [phase], or code word [
phase] is followed by the code word ■, and then the code word ■, ・, @ ,
@, [phase] is followed by code word [phase], or code word [phase] is followed by code word ■, [phase],
If either Q, @), or [stock] follows, the code word ■
If codeword 00001 000 is assigned instead of , and codeword ■ is followed by codeword ■ and further followed by either codeword ■ or ■, or if codeword ■ is followed by codeword ■ and then codeword ■, If one of ■ follows, the code word [stock] is followed by the code word ■, and then either of the code words ■ or ■, or the code word [phase] is followed by the code word ■. Furthermore, the code word ■, ■noi1″
If either one follows, the second code word ■σ) ii (instead assigns the code word 00001001 and assigns the code word ■ to the code word ■, ■q). Word 10001 0
01 is assigned, and the 7th hit P7 and the 8th bit of the chronologically older codeword in the codeword sequence pattern in the codeword-to-codeword concatenation part, and the 1st pinpoint P1 of the codeword following this are \X101. //If it is, select the 8th bit P8 as 1. SS 'l /l, and the 7th bit P
7 and 1st bit P1 are inverted to bit N○〃, and then N
A digital modulation method characterized by RZI modulation. (2) Bit XX1 and bit SS1 of code sequence
Digital modulation method O according to claim 1, characterized in that modulation is performed such that at least two or more bits XXO and no more than seven bits are included between // and