JPH0239145B2 - NICHIJOHOHENCHOHOSHIKI - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a modulation method for binary information. When transmitting information via a transmission path, various studies have been conducted in order to use the transmission path more efficiently and at the same time to make the information less susceptible to deterioration. In recent years, supported by the rapid development of semiconductor and LSI technology, methods of transmitting and processing information digitally are often adopted. However, although digitization makes it possible to make information less susceptible to deterioration, it results in an increase in the frequency band during transmission. As described above, particularly in the transmission of digital information, how to reduce the frequency band required for transmission is an important consideration. First, an example of a conventional digital information transmission device will be explained with reference to the block diagram shown in FIG. In Figure 1, 1 is an input terminal, 2 is a modulator,
3 is a transmission path, 4 is an equalizer, 5 is a detector, 6 is a demodulator, and 7 is an output terminal. The information to be transmitted is applied via input terminal 1 to modulator 2 . The modulator 2 transforms the input digital information in a predetermined manner and guides it to the transmission path 3. The information that has passed through the transmission line 3 is subjected to various waveform corrections in the equalizer 4 and sent to the detector 5.
After being detected by the demodulator 6, the information is demodulated by the demodulator 6, and the information is extracted via the output terminal 7. The transmission line 3 is, for example, a recording/reproducing medium such as a cable, an optical fiber waveguide, or a magnetic tape, and the equalizer 4 reversely corrects the waveform deterioration received when passing through the transmission line 3. At detector 5, equalizer 4
The output signal from the modulator 2 is shaped in a predetermined manner to reproduce the output waveform of the modulator 2. The signal obtained in this way is applied to the demodulator 6, and the information applied to the input terminal 1 is transferred to the output terminal 7 by applying the signal obtained in this way to the demodulator 6 and performing the operation opposite to that of the modulator 2.
is retrieved via. When transmitting digital information, it goes without saying that the modulator 2 plays the role of an interface between the transmission line 3 and the information source applied to the input terminal 1, and the modulator 2 plays the role of an interface between the transmission line 3 and the information source applied to the input terminal 1. This is extremely important for efficiently utilizing the transmission line 3. Now, some of the conventionally used modulation methods will be explained with reference to the waveform diagram shown in FIG. In Figure 2, 9 is the data string to be transmitted, 1
0 to 12 are waveforms after data string 9 is modulated using a modulation method called Tenten PM, FM, MFM, 8
is the bit cell length. Each modulation method shown here is a well-known method, so a description thereof will be omitted. By the way, parameters for evaluating the characteristics of a modulation method include a minimum inversion interval (hereinafter referred to as "Tmin"), a window width (hereinafter referred to as "Tw"), and the like. Tmin is the shortest value of the inversion interval of the waveform after modulation, and is strongly related to the high range (upper limit) of the occupied frequency bandwidth of the signal after modulation. Generally, the smaller Tmin, the higher the upper limit frequency component included in the modulated signal. Therefore, it is desirable that Tmin be as large as possible. Furthermore, Tw is what is called the "window width" as already mentioned. In the process of detection and demodulation, it is normal to determine whether the signal is at a high level or a low level based on a timing signal (clock signal) created on the receiving side. Therefore, if the time error between the timing of the timing signal and the received signal is within a predetermined range, demodulation will be performed without error, and the allowable error range will take on different values depending on the modulation method. In normal cases, the apparent zero-crossing position of the received signal changes due to waveform deterioration and various noises in the transmission path, leading to error occurrence. Therefore, it goes without saying that a modulation system as wide as possible is strongly desired for Tw. However, for each modulation method in Figure 2, Tmin is
1.0 (MFM), 0.5 (FM, PM), Tw is 0.5 each
There is only one. Each of the above modulation methods is only a small part of conventional methods, but in general, in digital modulation methods, if you try to increase Tmin in order to reduce the occupied bandwidth, you will have to decrease Tw as a result. . Therefore, in order to pass as much information as possible within the passband of the transmission path, Tmin must be increased, and in order to increase the margin against noise and various types of deterioration.
It is important to increase Tw. Therefore, in the present invention, Tmin is 1.33 (=4/3, a value normalized by the bit cell length, of course), and Tw is 0.67 (=2/3, a value normalized by the bit cell length, of course), and Tmin is large and Tw It also provides a relatively wide range of modulation methods. Of course, various m/n conversions have been announced, but one with a large Tmin of 1.33 and a relatively large Tw of 0.67 has not been announced so far. The principle of the present invention will be explained below. The present invention is a type of m/n conversion block modulation method, in which a data string to be transmitted is divided into m blocks and each block is converted into a predetermined n separate data strings for transmission. I can say it. The present invention sets Tmin to 1.33 (4/3) and Tw=
This makes it 0.672/3. Therefore, first, in order to achieve Tw=0.67 (2/3), the values of m/n are set to m=2 and n=3. The bit cell length of the conversion is 2/2 of the bit cell length of the original information.
3, so in 3/2 conversion, Tw=0.67(2/
3). Next, another feature of the modulation method of the present invention is
The purpose is to set Tmin=1.33 (4/3). There are data patterns that cannot be used to achieve this property. In other words, the data string after converting into three pieces of data, that is, the “converted data” is 101 or
010 itself has an inversion interval of 0.67 (2/3). Therefore, 101 and 010 cannot be used. In this way, the data pattern of the conversion data is 000,
001, 011, 100, 110, 111 are by themselves
Tmin=1.33 (4/3) is not violated. However, when the converted data continues one after another,
There are some combinations that cannot be used in terms of Tmin. Therefore, if the conversion data is Mn (m _1o , m _2o , m _3o ), the pattern that can be used as Mn (m _1o , m _2o , m _3o ) is the immediately preceding data sequence m _o-1 (m _{1(o-1 )} ), m _2(o-1) ,
m _3(o-1)) , and those that violate Tmin = 4/3 cannot be used. Hereinafter, to simplify the explanation, a block of data array to be transmitted divided into two pieces will be referred to as an "input data block", and a block of data string consisting of three data pieces after conversion will be referred to as a "converted data block". ”. Then Dn the former
(d _1o , d _2o ) The latter is denoted by Mn (m _1o , M _2o , m _3o ).
The subscript n here means the nth block. Therefore, Mn (m _1o , m _2o , m _3o ) and M _o-1 (m _1(o-1) ,
m _2(o-1) , m _3(o-1) ) is shown in Table 1. In Table 1, the (n-1)th conversion data block m _o-1 (m _1(o-1) , m _2(o-1) , m _3(o-1) ) and the nth Possible conversion data block mn (m _1o , m _2o ,
m _3o ), for example, if "000" is used for the (n-1)th, the nth is 000,001,
Five patterns, 011, 110, and 111, can be used, and even if the nth pattern is used consecutively after the (n-1)th pattern, Tmin=1.33 (4/3) will not be violated. M _o-1 (m _1(o-1) , m _2(o-1) , m _3(o-) ) is 001, 011,
The same idea applies to the cases of 100, 110, and 110, and the usable Mn (m _1o , m _2o , m _3o ) is shown in Table 1, respectively.

[Table] Incidentally, since the input data block is composed of 2 bits, there are ⁴ types of data patterns for the input data block. Therefore, it is necessary to prepare four copies of the conversion data. However, as can be seen from Table 1, M _o-1 (m _1(o-1) ,
m _2(o-1) , m _3(o-1) ) followed by available Mn (m _1o ,
m _2o , m _3o ) do not necessarily exist in four ways.
001 and 110 as M _o-1 (m _1(o-1) , m _2(o-1) , m _3(o-1) )
When using , only three ways can be used as Mn (m _1o , m _2o , m _3o ). However, if 001 and 110 are not used, the number of patterns that can be used after 000, 011, 100, and 110 will be reduced to three. When this happens, information cannot be transmitted. Therefore, in Table 1, 000, 011, 100 and 111
By mainly using 001 and 110, Tmin=1.33 (4/3) and Tw=0.67
(2/3) can be satisfied. For example, the n-1st conversion data M _o-1 (m _1(o-1) ,
m _2(o-1) , m _3(o-1) ), if 000 is used, n
The th conversion data Mn is 000, 011, and 111.
Use either 001 or 110, a total of 4 pieces. As an example, the nth input data block Dn(d _1o ,
d _2o ), 000 for 00, 011 for 01,
In the case of 10, 111 is used as Mn (m _1o , m _2o , m _3o ), respectively. 001 and 110 are selectively used as conversion data for the remaining pattern 11. For example, the (n+1)th input data block is 00 or
For 01, enter 001

【table】

[Table] When it is 10 or 11, use 110 as conversion data. Hereinafter, as M _o-1 (m _1(o-1) , m _2(o-1) , m _3(o-1) )
In the case of 001, 011, 100, 110, and 111, conversion data is also allocated according to Table 2. Of course, by following Table 2, Tw=0.67 (2/
3) and Tmin=1.33 (4/3). The above is the principle of the modulation method according to the present invention. Figure 3 shows the waveforms obtained as a result of _the modulation method obtained _according to _this principle.
The trends in m _2o and m _3o are shown in Table 3.

[Table] In Table 3, the input data block obtained by dividing the input data into two pieces is written as “Dn”.
n means the nth block. The case where the input data is 000110111011000110001110... is shown. If this is used as an input data block, it will be 00, 01, 10, 11, etc. 1st Dn
Assuming that Mn (that is, Mo) immediately before is 000, from Table 2, the converted data M ₁ becomes 000 for D ₁ =00 when n=1. (In order to simplify the description, Mn (m _1o , m _2o , m _3o ) is replaced by Mn, Dn (d _1o ,
d _2o ) is written as Dn) Next, when n = 2, the previous conversion data M ₁ was 000, so for D ₂ = 01,
From Table 2, M ₂ will adopt 011. Similarly, convert data blocks according to Table 2.
Mn is created. What you need to be careful about here is when n = 4, 6, and 11. For example, when n = 4, D ₄ = 11, the next D ₅ is 10, and M ₃ is 111.
As M ₄ , adopt 110 from Table 2. Conversion data block Mn obtained in this way
000011111110000001100011111000110…
...becomes... Table 3 is shown in FIG. 3 together with a waveform diagram. In FIG. 3, the input data, block number, input data block, conversion data block, conversion data block, etc. are the same as in Table 3, so their explanation will be omitted. 13 is the bit cell length of the input data, 14 is the block length, 19 is a waveform diagram of the input data 15, and 20 is a waveform diagram of the converted data block 18. 22 is the minimum inversion interval Tmin of the input data 15. On the other hand, 21 indicates the Tmin of the present modulation method, and it goes without saying that the Tmin of the present invention is equal to the bit cell length 1.
1.33 (4/3) times larger than 3. Further, regarding Tw, it is obvious that it is 0.67 (2/3) since 2 bits are converted to 3 bits, so the explanation will be omitted. By the way, the relationship between the input data block and the converted data block of the present invention is shown in Table 2, but if this is rewritten, it becomes Table 4.

[Table] In Table 4, for example, if 00 is input as Dn, when M _o-1 = 001, Mn = 100, M _o-1 ≠
When Dn is 001, Mn=000; When Dn is 01, according to M _o-1 , when M _o-1 = 000 or 100, Mn = 011,
When M _o-1 = 011 or 111, Mn = 100, M _o-1 = 001
When Dn is 10, Mn=000 when M _o-1 = 110,
When M _o-1 ≠ 110, Mn = 111; when Dn is 11,
When D _o+1 = 00 or 01, Mn = 001, D _o+1 = 10 or
11, Mn=110, M _o-1 = 110, Mn=110
becomes. Needless to say, this is the same relationship as in Table 2. Next, a configuration example for implementing the modulation method of the present invention will be shown below. FIG. 4 is a block diagram showing one configuration example of the present invention. In FIG. 4, 23 is an input terminal, 24 is a block generator, 25 is a first delay device, 26 is an address signal generator, 27 is a memory,
28 is a second delay device, 29 is a parallel-to-serial converter 30
is the output terminal. The data string to be transmitted is applied to a blocker 24 via an input terminal 23. The data string to be transmitted is divided into two blocks by a block generator 24 to form input data blocks. Blockizer 24
The output data of the address signal generator 26 and the first
is applied to the delay device 25 of. On the other hand, the output of the second delay device 28 is also applied to the address signal generator 26. An address signal generator 26 generates an address signal according to the output data of the blocking device 24, the first delay device 25, and the second delay device 28, and controls the memory 27. The memory 27 sends a predetermined converted data block to the second delay device 28 and the parallel-serial converter 29 according to the address signal applied from the address signal generator 26. The parallel-serial converter 29 converts the converted data output from the memory 27. Converts a data block (parallel 3-bit configuration) into a serial data string,
A modulated signal is sent out via the output terminal 30. As already explained, in the case of the present invention, the data string to be transmitted is divided into two blocks to form an input data block, so the block generator 24 separates the data string applied via the input terminal 23 into two blocks. Constructs a delimited input data block (parallel 2-bit configuration). The first delay device 25 is for delaying the input data block created by the block generator 24 by one block period, and can be constructed of, for example, a latch. First delay device 2 at present
5 is the nth input data block, that is, Dn (d _1o , d _2o ), the block generator 24
The output becomes the (n+1)th input data block, ie, D _o+1 (d _{1 (o+1)} , d _{2 (o+1)} ). On the other hand, the second
The output of the delay device 28 is the (n-1)th converted data block M _(o-1) (m _{1 (o-1)} , m _{2 (o-1)} ). This point will be explained again later. These Dn (d _1o , d _2o ), D _o+1 (d _1(o+1) , d _2(o+1) )
and M _(o-1) (m _{1 (o-1)} , m _{2 (o-1)} ), an address signal is generated in the address generator 26. A predetermined converted data block is obtained from the memory 27 in accordance with this address signal. Next, the relationship between the addresses of the memory 27 and the stored data will be explained with reference to Table 5. In Table 5, "address" indicates the address of the memory 27, and "stored data" indicates the data stored in the memory 27. For example,

[Table] 5 5 5

Claims (1)

[Claims] 1 Create an input data block by sequentially dividing a data string consisting of a binary signal to be transmitted into two pieces of data each, and divide the input data block into three pieces corresponding to each of the four types of data patterns of the input data block. When converting into a conversion data block consisting of a data string of binary signals, for the first data pattern among the four types of data patterns of the input data block, if the immediately preceding conversion data block is 001, the input data block is 100. , 000 if the previous converted data block is other than 001, and 000 for the second data pattern, the previous converted data block is
If it is 000 or 100, use 011. If the previous converted data block is 011 or 111, use 100. If the previous converted data block is 001, use 111. For the third data pattern, use the previous conversion. 000 when the data block is 110, 111 when the previous converted data block is other than 110, and for the fourth data pattern, the next input data pattern is the first data pattern or the second data pattern. 001 when the data pattern is the above data pattern, 110 when the next input data pattern is the third data pattern or the fourth data pattern, and 110 when the previous data block is the third data pattern.
A binary information modulation method characterized in that when the data is 110, 011 is used as the converted data block.