JP3515526B2 - Coding method of control information in a communication system - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control information coding method in a communication system, and more particularly, in inter-node communication in which an 8B / 10B code is interposed, control information error detection is performed when a bit error occurs during communication. The present invention relates to a control information encoding method in a communication system that enables the above.

[0002]

2. Description of the Related Art A communication system for communicating control information for a variable-length data frame and its communication processing will be briefly described below as an example. The transmitting node, for the data frame divided into a plurality of 8-bit length blocks,
Each block is encoded into a communication data code according to the 8B / 10B encoding conversion rule and transmitted. Also, regarding the control information divided into M (M is a natural number) 8-bit length blocks during the period in which no data frame is transmitted, each block is encoded according to the 8B / 10B encoding conversion rule. A control inertial information transmission process of transmitting as a control information set is performed.

The receiving node decodes the received communication data code according to the 8B / 10B coding conversion rule, and restores the data frame divided into a plurality of 8-bit length blocks. In addition, a control information reception process of decoding M pieces of control information from the control information set according to the 8B / 10B encoding conversion rule is also performed. Next, the 8B / 10B encoding conversion rule will be described. For details, see JP-A-59-10056.
It is described in the official gazette. This conversion rule considers the polarity (disparity) of the codeword when converting an 8-bit length block into a 10-bit codeword, and adjusts so that the cumulative number of 1s and 0s is always almost equal.

Here, the polarity of the code word is the difference between the numbers of 1 and 0 in 10 bits. When the number of 1's and 0's is equal, the polarity is neutral, and when the number of 1's and 0's is different, the polarity is non-neutral. Further, in the latter case, a code word having two more 1's is called "polarity is positive (+2)" and a code word having more two 0's is called "polarity is negative (-2)". Of 256 blocks of 8 bits, 134 patterns are converted into codewords with a neutral polarity. For the remaining 122 ways, two codeword pairs with positive and negative polarities are assigned to each block, and if the cumulative polarity up to the point of encoding the block is positive, the codeword is converted into a negative codeword. If the cumulative polarity is negative, it is converted to a code word with positive polarity. Cumulative polarity has an initial value of -1 (negative)
Is a simple addition of the polarity of the code word (any one of 0, +2, and -2) for each conversion. The cumulative polarity does not change upon conversion to a codeword whose polarity is neutral, and the polarity is inverted upon conversion to a codeword whose polarity is non-neutral.

A 10-bit special code word K that does not correspond to any of the above 8-bit 256 blocks is defined and used to identify that it is control information. Finally, a conventional control information encoding method will be described.
For details, see Japanese Patent Laid-Open No. 3-243042 and textbooks on fiber channel standards (for example, Alan F. Benner, "Fib
re Channel ", McGraw-Hill, 1996, Chapter 6, Ordered Se
ts) etc. The control information divided into M (M is a natural number) 8-bit length blocks is ordered by (1 + M) codewords during a period (inter frame gap) in which a communication data code is not transmitted. Communication is performed using a set of code words (hereinafter referred to as a control information set). The control information set starts with the special codeword K and carries the control information with the remaining M codewords.

For example, in the Fiber Channel standard,
The control information is transferred using a 40-bit fixed-length control information set consisting of (1 + 3) code words. This is called a primitive signal. Table 1 shows a list of control information sets used by the primitive signals.

[Table 1] The code word notation in Table 1 is as described above in Japanese Patent Laid-Open No. 59-100.
Based on No. 56 publication. The first control information set is a special code word K28.5 having a non-neutral polarity and is used to identify the control information set. The remaining 3 code words carry 3 pieces of control information. When the combination of the remaining three codewords does not correspond to any of Table 1, the control information reception processing unit ignores this. Therefore, the conventional device does not malfunction even if a new control information set is added.

The existing information set shown in Table 1 is designed so that the cumulative polarity per control information set is neutral. That is, the polarity of the second codeword is non-neutral, the polarities of the third and fourth codewords are both neutral, or
Both are non-neutral. Therefore, the control information set can be controlled to start only from the codeword having the positive polarity, and the configuration of the control information reception processing unit is simplified. Some of the control information sets shown in Table 1 are fixedly assigned to the remaining three codewords and used for special status display and notification. For example, IDLE indicates an idle state in which there is no data frame to be sent or control information, and R_RDY notifies the transmitting side that the receiving side has an empty buffer.

In another control information set, the second code word is fixedly assigned to identify the type of notification, and the control information divided into two 8-bit length blocks is assigned to the third and fourth code words. Encode. Of the 8-bit length block 256 types, only 134 types that are converted into neutral codewords are used so that the polarities of the third and fourth codewords are neutral after encoding. Therefore, 128 bits of 7-bit control information are added to the above 1
7B / one-to-one correspondence to 34 different codewords
The 10B coding conversion rule is defined as a code table.

[0009]

However, the conventional coding method has a serious drawback that when a new control information set is added, it is likely to be erroneously received as an existing control information set due to a bit error generated during communication. It was Less than,
This drawback will be detailed using the case of a Fiber Channel primitive signal. Consider a case where a new control information set is added to Table 1. A code word whose polarity is non-neutral, for example, D28.7 is fixedly assigned to the second code word. 7th and 12th bits for 3rd and 4th
Eight types of control information are encoded into codewords having a neutral polarity by the conventional 7B / 10B encoding conversion rule.

At this time, only the second codeword D28.7 is definitely different from the existing control information set.
The 3rd and 4th codewords may take exactly the same codeword as the existing control information set depending on the control information to be encoded. Therefore, for example, if bit errors concentrate in the second codeword during communication and D28.7 is erroneously received as D21.7, the new control information set is confused with the existing control information set VC_RDY, which is a serious problem. Cause problems.

The bit error resistance in the control information set is measured by the sum of Hamming distances between corresponding code words. The Hamming distance corresponds to the number of bits in which 1 and 0 are inverted at the same bit position in two bit strings. Codeword with negative polarity + D28.7 (001110 0
001) and a code word with a negative polarity + D21.7 (101010 0001)
Has different values at the 1st and 4th bit positions, and the Hamming distance is 2. Here, a code word having a negative polarity is selected when the cumulative polarity up to that time is positive, and thus "+" is prefixed.

That is, if a double bit error occurs during communication of a new control information set, + D28.7 becomes + D2.
It may be received as 1.7 and may cause a serious problem by confusing it with the existing control information set. In order to solve the above drawbacks, it is an object of the present invention to provide an encoding method that facilitates detection of a bit error occurring during communication when encoding a control information set in a communication system.

[0013]

In order to achieve the above-mentioned object, the invention according to claim 1 uses communication data of 8B / 10B.
In addition to generating and transmitting a codeword in consideration of polarity according to the encoding conversion rule, M (M is a natural number of 2 or more) added for transmitting and processing the communication data.
The control information of each is converted into a code word, and the special code word is
In the coding method of the control information in a communication system for transmitting the control information set comprising a head (1 + M) codewords, at least one control information set, multiple sets of control Chi sac M pieces of control information For encoding information
In the other control information set, at least one of the plurality of control information is applied in the other control information set by applying a coding conversion rule that generates only one of the codewords whose polarity is neutral or non-neutral. non-neutral encoding conversion rule is applied, at least another one Oite the encoding of control information polar neutral code Oite to the encoding control information whose polarity is generated only non-neutral code word The feature is that the neutral coding conversion rule for generating only words is applied.

The invention described in claim 2 is the same as claim 1
In the control information encoding method in the communication system according to, in the codeword group consisting of all the codewords generated by the non-neutral encoding conversion rule and the neutral encoding conversion rule, one codeword and another The Hamming distances to all codewords are at least N + 1 (N is a natural number).

The invention described in claim 3 is the same as that of claim 2
In the method for encoding control information in the communication system according to the above item, in the codeword group, 8B / 10B blocks before encoding 8B / 10B corresponding to one codeword and 8B corresponding to all other codewords. The Hamming distance to the 8-bit block before / 10B encoding is at least N + 1 (N
Is a natural number).

The invention according to claim 4 is the same as claim 1
In the control information encoding method in the communication system according to any one of 1 to 3, the neutral encoding conversion rule is applied to the second codeword in the control information set with M = 3.

The invention according to claim 5 is the same as claim 2
In the control information encoding method in the communication system according to any one of 3 or 3, the codeword generated by the neutral encoding conversion rule has a Hamming distance of at least N with a predetermined codeword group. Number).

The invention described in claim 6 is the same as claim 5
In the method for encoding control information in the communication system according to claim 8, the 8B / 10B block before 8B / 10B encoding corresponding to the codeword generated by the neutral encoding conversion rule belongs to a predetermined codeword group. It is characterized in that the Hamming distance to the 8-bit long block before 8B / 10B encoding corresponding to all codewords is at least N (N is a natural number).

The invention described in claim 7 is the same as claim 5
Or the control information encoding method in the communication system according to any one of 6 to 6, wherein the predetermined codeword group is
(D20.4, D5.4, D10.4, D31.2, D
17.4, D21.4, D21.7, D31.3, D3
1,5, D31.6), and N = 4.

The invention according to claim 8 is the same as claim 5
Or the control information encoding method in the communication system according to any one of 6 to 6, wherein the predetermined codeword group is
(D20.4, D5.4, D10.4, D31.2, D
17.4, D21.4, D21.7, D31.3, D3
1.5, D31.6), the codeword group generated by the neutral coding conversion rule is (D6.6, D14.1, D25.
6, D25.1, D19.2, D13.5, D13.
2, D4.0) , the code word group generated by the non-neutral coding conversion rule is (D18.7, D1.3, D0.5,
D11.7, D3.4, D26.0, D28.7, D2
It is characterized in that it is 3.1) .

The invention described in claim 9 is the same as claim 1.
In the control information encoding method in the communication system according to any one of 1 to 8, in the other control information set, at least one of the M pieces of control information has a neutral and non-neutral codeword. Applying the compatible encoding conversion rule, the other control information, when the polarity of the code word that encodes the one control information is neutral, the non-neutral encoding conversion rule is applied, and the non-neutral In the case of, the neutral coding conversion rule is applied.

Further, in the invention described in claim 10, in the control information encoding method in the communication system according to claim 9, the one control information is encoded before the other control information. Is characterized by.

According to the present invention, by taking the above-mentioned means, the position of the code word having the neutral polarity and the position of the code word having the non-neutral polarity are completely matched at the plurality of code word positions of the different control information sets. It provides a coding method that does not include the Hamming distance at a specific codeword position and can add a Hamming distance at another codeword position, which facilitates detection of transmission bit errors and makes it possible to erroneously decode different control information sets. The possibility of doing so can be reduced.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings and tables. <First Embodiment> A control information coding method in a communication system according to a first embodiment of the present invention will be described. A coding method for adding a new control information set NEW to which D28.7 is fixedly assigned as the second code word in the communication system using the control information set shown in Table 1 will be described in detail. Table 2 shows the control information set NEW to be added.

[0025]

[Table 2] The control information set NEW carries two pieces of control information using the third and fourth codewords. Codeword ± D28.7 and the codeword group (± D20.
4, ± D5.4, ± D10.4, ± D31.2, ± D1
7.4, ± D21.4, ± D21.7, ± D31.3,
The minimum Hamming distance with respect to ± D3 1,5, ± D31.6) is 2. That is, + D28.7 having a negative polarity
(001110 0001) and -D28.7 (0 with positive polarity
01110 1110) differs from the second codeword in the existing information set by 0 and 1 in at least two bit positions.

FIG. 1 shows the contents of the control information set NEW encoding method according to the first embodiment of the present invention. In the figure, first, a special code word + K2 indicating that it is a control information set depending on whether the cumulative polarity at the start of encoding is positive or negative.
8.5 or -K28.5 is output (step 1
1). Since K28.5 is a code word whose polarity is non-neutral, the cumulative polarity is inverted. Then, the second code word + D28.7 or -D28.7 indicating the control information set NEW is output according to whether the cumulative polarity is positive or negative (step 12). The cumulative polarity is reversed again. Further, the first control information is encoded into a code word having a neutral polarity (third code word) and output as in the conventional method (step 1).
3).

For example, control information of 128 bits of 7 bits is converted into a code word having a neutral polarity by the conventional 7B / 10B encoding conversion rule. The cumulative polarity does not change. Further, the second control information is encoded into a code word having a non-neutral polarity (fourth code word) and output (step 14). As the control information, of the 8-bit length block 256 types, only 122 types in which a non-neutral codeword is assigned can be used. For example, 5B, in which 32 types of control information of 5 bits are simply assigned to a 10-bit codeword whose polarity is non-neutral 5B
/ 10B encoding conversion rule is used. The details of this encoding conversion rule are described in Japanese Patent Application No. 2000-040653. Since the fourth codeword has a non-neutral polarity, the cumulative polarity is reversed.

On the other hand, in the existing control information set shown in Table 1, the polarities of the 3rd and 4th codewords are all neutral, or all of them are non-neutral. Therefore 3,
At least one of the code words is always different from the control information set NEW in which the polarity of the fourth code word is neutral or non-neutral. That is, the Hamming distance between the control information set NEW and the existing information set can be at least 3 or more by adding 1 in the 3rd and 4th codewords. For example, while transmitting the control information set NEW, 2
Double bit error occurs + 2nd codeword + D28.7
Even if it is erroneously received as D21.7, the polarity of the fourth code word is non-neutral, so there is no fear of confusing it with VC_RDY.

That is, according to the control information coding method in the communication system according to the first embodiment of the present invention, when the control information is coded, 8B / 10 having polarity is used.
In a communication system using the B coding conversion rule, a plurality of control information is added to a control information set that is coded by a single coding conversion rule that generates only one of a code word having a neutral or non-neutral polarity. Then, the coding conversion wax rule that generates only codewords with neutral polarity is applied to one control information, and the coding conversion rule that generates only codewords with non-neutral polarity is applied to other control information. By adding the control information set to be added, the Hamming distance increases based on the polarity of the codewords to be encoded, in addition to the Hamming distance between the fixed allocation codewords in the related art, and the transmission bit error can be easily detected.

<Second Embodiment> Next, a control information coding method in a communication system according to a second embodiment of the present invention will be described. In this embodiment, the control information set N
Conventional 7B / 1 to generate the third codeword of EW
Instead of the 0B coding conversion rule, for example, a coding method for allocating control information of 8 bits of 3 bits to 8 code words of which polarity is neutral as shown in Table 3 is used. These eight codewords are
In each case, a Hamming distance of 3 or more with respect to the other 7 code words was selected.

[Table 3] Also, instead of the 5B / 10B coding conversion rule, for example, a coding method for assigning control information of 8 bits of 3 bits to 8 code words of non-neutral polarity shown in Table 4 is used for the fourth code word generation. To use. The eight codewords were selected so that the Hamming distances to the other seven codewords were 3 or more regardless of their polarities.

[Table 4]

Therefore, just as in the first embodiment, even if a double bit error occurs in the second codeword, the error can be detected without being erroneously received with the existing control information set and the double bit error can be detected. Even if a bit error occurs in the third or fourth codeword, the error can be detected without erroneously decoding the two pieces of control information carried by the control information set NEW, so that highly reliable exchange of control information is possible. It will be possible.
In the 5B / 10B coding conversion rule used in the first embodiment and the conventional 7B / 10B coding conversion rule, the minimum Hamming distance between codewords after conversion is 1,
If there is a bit error in the third or fourth code word during communication, the error could not be detected and erroneous control information could be restored. According to the control information coding method in the communication system according to the second exemplary embodiment of the present invention, it is possible to easily detect the transmission bit error not only between the control information set and the existing control information set but also inside the control information set NEW. Become.

<Third Embodiment> In the first and second embodiments, a communication system such as a fiber channel for sequentially transmitting 10-bit codewords converted based on the 8B / 10B coding conversion rule is provided. set to target. However, in the 10-gigabit version of Fiber Channel, which is being standardized, in addition to the conventional communication section using 10-bit codewords,
A section for transmitting an 8-bit length block before encoding is also allowed. In this section, the control information divided into M (M is a natural number) 8-bit long blocks is communicated as M 8-bit long blocks. In the present embodiment, such 8B /
(EN) A coding method capable of easily detecting an error in control information even in a section where the 10B coding conversion rule is not applied.

In the third embodiment, an encoding method is used for generating the third codeword of the control information set NEW, for example, 8 bits of control information of 3 bits are assigned to 8 8-bit length blocks shown in Table 5. . Each of the 8-bit length blocks has a Hamming distance of 3 or more with the other 7 blocks, and the 8B / 10B code-converted 8 codewords have the neutral polarity, and the other 7 blocks. We chose a Hamming distance of 3 or more from the codeword.

[Table 5] Similarly, for the fourth codeword generation, a coding method is used in which, for example, 3 bits and 8 types of control information are assigned to 8 blocks of 8 bit length shown in Table 6.

All of the 8-bit length blocks have a Hamming distance of 3 or more with the other 7 blocks, and the 8B / 10B code-converted 8 codewords have non-neutral polarities. The Hamming distances to the other seven codewords were selected to be 3 or more regardless of their polarities.

[Table 6] Here, the 8-bit length block before conversion of the second code word D28.7 in the control information set NEW is Oxfc.
(111 11100).

On the other hand, the code words D20.4, D5.4, D10.
4, D31.2, D17.4, D21.4, D21.
7, D31.3, D3 1.5, D31.6,
Table 7 shows the 8-bit length block before the conversion.

[Table 7] As is clear from Table 7, the 8-bit length block before conversion of the code word used second in the existing control information set is Hamming with the 8-bit length block before conversion of D28.7. The distance is 2 or more.

Therefore, according to the control information coding method in the communication system of the third embodiment of the present invention,
The same effect as in the first embodiment or the second embodiment is obtained in the transmission section using the 8B / 10B encoding conversion rule, and further, in the transmission section not using the codeword, the double effect is obtained. Even if a bit error occurs in the second codeword, since at least one of the third and fourth information sets is different from the existing information set, an 8-bit length block,
The error can be detected without being accidentally restored, and
Even if a double-bit error occurs in the third or fourth 8-bit long block, communication is performed with the control information set NEW 2
An error can be detected without accidentally restoring individual control information.

<Fourth Embodiment> Next, a control information coding method in a communication system according to a fourth embodiment of the present invention will be described. The first to third embodiments of the present invention are 2
A new control information set NEW in which the non-neutral D28.7 is assigned to the third codeword is added, and a method of carrying two control information by using the third and fourth codewords is illustrated. The fourth embodiment of the present invention provides an encoding method for assigning the second neutral codeword. Further, the second codeword is also assigned the control information similarly to the third and fourth codewords, and an encoding method for carrying a total of three pieces of control information is also provided.

Encoding that assigns the first control information of 8 bits of 3 bits to the second code word of the control information set NEW, for example, to eight neutral code words shown in Table 5 described in the third embodiment. Use the method. The 3rd and 4th codewords are the 3rd
Just as in the embodiment, each is a neutral or non-neutral code word, and carries 3 bits and 8 ways of second and third control information, respectively. As described in the third embodiment, the eight codewords shown in Table 5 have Hamming distances of 3 or more with respect to the remaining seven codewords and 8-bit length blocks before encoding. . Further, these eight codewords are actually two or more Hamming distances for both the codeword secondly used in the existing control information set shown in Table 7 and the 8-bit length block before the codeword. I chose to be.

Table 8 shows the contents of the control information set NEW based on the fourth embodiment of the present invention, and the encoding method will be described with reference to FIG.

[Table 8] The cumulative polarity immediately before sending the control information set is always controlled to be negative. In FIG. 2, first, a special code word -K28.5 having a positive polarity indicating that it is a control information set is output (step 21). The cumulative polarity reverses to positive.

Next, one of the code words having the neutral polarity shown in Table 5 by encoding the first control information (second code word)
Is output (step 22). The cumulative polarity does not change.
Further, the second control information is encoded and one of the code words having the neutral polarity shown in Table 5 (the third code word) is output (step 23). The cumulative polarity does not change. Next, the third control information is encoded and the code word having the negative polarity (fourth code word) shown in Table 6 is output (step 24). The cumulative polarity reverses to negative.

According to the fourth embodiment of the present invention, in addition to all the effects described in the first to third embodiments, the second neutral codeword of the control information set NEW to be added is used. The cumulative polarity per control information set becomes neutral, and the control information reception processing unit can be simplified as in the case of using only the existing control information set. That is, since the first special code word K28.5 of the control information set can always be limited to start with the same polarity (for example, -K28.5 having a positive polarity), + K28 is included in the control information set detection circuit. The detection of .5 becomes unnecessary.

Further, according to the encoding method of the fourth embodiment of the present invention, eight kinds of control information can be sent by the second code word of the control information set NEW. At this time, just as in the third embodiment, even if double bit errors occur concentrated in any of the second, third, and fourth codewords or 8-bit length blocks in all transmission sections, the error occurs. Can be detected without being erroneously restored from the existing control information set, and an error in the control information transmitted by the control information set NEW can also be detected.

<Fifth Embodiment> Next, a control information coding method in a communication system according to a fifth embodiment of the present invention will be described. In the second to fourth embodiments,
An encoding method is provided in which a Hamming distance between a control information set NEW to be added and an existing control information set and a Hamming distance between control information transmitted in the control information set NEW are 3 or more. The fifth embodiment of the present invention provides an encoding method in which the Hamming distance is 4 or more.

In the fifth embodiment of the present invention, an encoding method for assigning, for example, five neutral codewords shown in Table 9 is used for generating the second and third codewords in the control information set NEW. Table 1 for the 4th codeword
The non-neutral codeword D12.0 shown in 0 is fixedly assigned.

[Table 9]

[Table 10] The total of 6 codewords shown in Tables 9 and 10 have a Hamming distance of 4 or more with the remaining 5 in any of the codewords and the 8-bit length block before encoding, and the existing Fiber Channel standard. The Hamming distance was selected to be 3 or more for both the second code word used in the control information set and the 8-bit long block before the code word.

Also in the fifth embodiment of the present invention, just as in the fourth embodiment, the control information set NEW and the existing control information set have the polarity of either the third code word or the fourth code word. different. Therefore, together with the Hamming distance of 3 in the second codeword, a Hamming distance of 4 or more is secured. In this way, even if triple bit errors are concentrated in any of the second, third, and fourth codewords in all transmission sections, the error can be detected and is erroneously restored from the existing control information set. Without
Further, it is possible to detect an error in the control information carried by the control information set NEW.

In general, a coding conversion rule for generating only code words having a neutral polarity is applied to one control information, and a code conversion for generating only code words having a non-neutral polarity is applied to other control information. In a coding method for adding a control information set to which a rule is applied, as a codeword group generated by each coding conversion rule, a Hamming distance from a codeword used in an existing control information set is N or more (N is A natural number of 2 or more),
Moreover, it is only necessary to select a codeword group in which the Hamming distance between codewords belonging to the codeword group is (N + 1) or more.

At this time, the Hamming distance from the existing control information set is also increased by 1 based on the polarity of the codeword (N + 1).
As described above, the error can be detected regardless of which code word in the control information set is concentrated in N-bit error, and the error can be detected as the existing control information set, and the added control information set NEW can be used. There is no concern that the transmitted control information will be accidentally restored. Further, as for the code word group, similarly for an 8-bit long block before encoding, a block having a Hamming distance of N or more with an 8-bit long block used in the existing control information set and belonging to the block group By selecting a codeword group such that the mutual Hamming distance is (N + 1) or more, N double bit errors can be detected not only in the transmission section using the 8B / 10B coding conversion rule but also in all the transmission sections.

We perform a full search of 256 code words in the conventional 8B / 10B coding conversion rule, and for the code words shown in Table 7 used in the existing control information set,
It has been found that the maximum value of N in which a codeword group satisfying the above-described condition exists is 3. At this time, the maximum value of the number of code words that can form the code word group is 6, and there are 77 combinations. Table 9 and Table 10 show only one of them. Similarly, when N is 2, the maximum value of the number of code words that can form a code word group is 16, and it was found that there is only one combination. A total of 16 codewords shown in Tables 5 and 6 form the complete codeword group.

<Sixth Embodiment> Next, a control information coding method in a communication system according to a sixth embodiment of the present invention will be described. In each of the first to fifth embodiments, each of the M pieces of control information carried in the control information set can carry only 7 bits and 128 ways. The sixth embodiment of the present invention provides an encoding method in which at least one piece of control information can carry 256 8-bit information.

As a sixth embodiment of the present invention, a case of adding a new control information set NEW to the existing control information set of the fiber channel shown in Table 1 will be described with reference to FIG. Control information set N
The EW carries three pieces of control information using the second, third and fourth codewords. In FIG. 3, first, a special code word K28.5 having a non-neutral polarity indicating that it is a control information set is output (step 31). In this case, the cumulative polarity is reversed. Next, 8 bits of the first 3 bits of control information are encoded and output to a codeword group (second codeword) whose polarity is neutral as shown in Table 5 (step 32). In this case, the cumulative polarity does not change.

Further, a code word (third code word) obtained by coding the second set of 8 bits of the control information 256 according to the conventional 8B / 10B coding conversion rule is output (step 33). This third code word may be neutral (134 types / 256 types) or non-neutral (122 types / 256 types) in polarity. In the former case, the cumulative polarity does not change, and in the latter case, In this case, the cumulative polarity is reversed. Further, the third control information of 3 bits and 8 patterns is encoded and output to a codeword group (fourth codeword) having non-neutral polarity shown in Table 6 (step 34). In this case, the cumulative polarity is reversed. Table 11 shows the control information set added in the sixth embodiment.

[Table 11] One of the code words shown in Table 5 is assigned to the second code word in the control information set NEW, and as described in the fourth embodiment, this code word and its 8-bit length block before encoding are In each case, the Hamming distance is 2 or more with respect to the second codeword in the existing control information set and the 8-bit length block before the encoding.

Further, 4 in the control information set NEW
At least a Hamming distance of 1 can be expected in the th codeword. The reason is that the fourth non-neutral codeword that can exist in the existing control information set shown in Table 1 is D.
22.4, and this codeword is not included in Table 6. Thus, according to the control information encoding method in the communication system of the sixth embodiment of the present invention,
With the same Hamming distance 3 as in the first to fourth embodiments, it is possible to use the third codeword and communicate the second set of 8 bits of control information 256 bits. The second or fourth control information set NEW to be added
Second, even if double-bit errors concentrate, errors can be detected,
The first or third control information will not be erroneously restored or will not be erroneously restored as another existing control information set.

In the SONET optical transmission system used for the backbone optical transmission line, control information is communicated in 1-byte 8-bit units. Therefore, it is possible to communicate 1-byte unit control information according to the sixth embodiment of the present invention. Very beneficial from a compatibility point of view.

<Seventh Embodiment> Next, a control information coding method in a communication system according to a seventh embodiment of the present invention will be described. In the sixth embodiment, the cumulative polarity per control information set is not necessarily neutral. The seventh embodiment of the present invention provides an encoding method in which at least one of the control information can carry 256 types of 8-bit information, and the cumulative polarity per control information set is neutral.

In this embodiment, the control information set NEW
Shall carry three pieces of control information using the second, third and fourth codewords. FIG. 4 shows a control information set encoding method according to the present embodiment. The cumulative polarity immediately before sending the control information set may be always controlled to be negative. In FIG. 4, at the time of encoding the control information set, first, a special code word -K28.5 having a positive polarity and indicating the control information set is output (step 41). In this case, the cumulative polarity is positively inverted. Next, the first 8 bits of 3 bits of control information are encoded and output to a codeword group (second codeword) whose polarity is neutral as shown in Table 5 (step 42). In this case, the cumulative polarity does not change.

Further, a code word (third code word) obtained by coding the second set of 8 bits of the control information 256 according to the conventional 8B / 10B coding conversion rule is output (step 43). Here, (134 types / 256 types) does not change the cumulative polarity, and when it is non-neutral (122 types / 256 types), the cumulative polarity is reversed. Next, if the polarity of the third code word is neutral, the third control information of 3 bits and 8 ways is encoded into a code word group of which the polarity is non-neutral as shown in Table 6 (steps 44 and 45), 3 If the polarity of the th code word is non-neutral, it is encoded into a code word group having the neutral polarity shown in Table 5 (steps 44 and 46).

Here, when the 8 bits of the third control information of 3 bits are encoded into the code word group whose polarity is not shown in Table 6, the cumulative polarity is inverted and the polarity shown in Table 5 is neutral. When encoded into a group of code words, the cumulative polarity does not change.
Table 12 shows the added control information set.

[Table 12] The greatest feature of the seventh embodiment of the present invention is that, when the third control information is encoded into the fourth codeword, the third codeword has a polarity that is neutral or non-neutral. The point is to encode so that the polar characteristics (neutral or non-neutral) do not match. Therefore, the Hamming distance 1 can be additionally ensured with respect to the existing control information set shown in Table 1 in which the polarity characteristics of the third and fourth codewords in the control information set are both neutral or both non-neutral. That is, together with the Hamming distance of 2 or more in the second codeword, the Hamming distance of 3 or more with the existing control information set can be set.

Furthermore, the 16 code words in Tables 5 and 6 which can be the fourth code word in the control information set are:
It is selected so that the Hamming distances to the remaining 15 bits are all 3 or more in both the code word and the 8-bit length block before encoding. In this way, according to the seventh embodiment of the present invention, while the same effect as that of the sixth embodiment is obtained, the cumulative polarity per control information set can be made the same neutral as that of the existing control information set. The reception processing unit can be simplified.

The sixth embodiment or the seventh embodiment of the present invention
In the embodiment, the same effect can be obtained even if the third and fourth codewords are exchanged. However, prior to the encoding of the third code word, the second control information encoded into the fourth code word is encoded into a code word having a neutral polarity or a code having a non-neutral polarity. It needs to be checked if it is encoded in a word. Therefore, it is easier to configure the circuit by first encoding the 256-bit control information of 8 bits and then applying the conversion rule used for encoding according to the polarity of the code word.

In general, in addition to a control information set that applies a single coding conversion rule that generates only codewords with either neutral or non-neutral polarity, one control information has a code with an arbitrary polarity. Apply a coding conversion rule for generating a word, and select a coding conversion rule for generating a code word having a polarity characteristic different from the polarity characteristic of the code word obtained by encoding the one control information as the other control information. By encoding with, the Hamming distance increases based on the polarity of the codeword to be encoded, in addition to the Hamming distance between codewords at a specific codeword position,
It becomes easy to detect communication bit errors.

[0061]

As described above, according to the control information coding method in the communication system according to the present invention, a code word having a polarity neutrality and a polarity are provided at a plurality of code word positions of different control information sets. Since the control information is encoded so that the positions of the non-neutral codewords do not completely match, the Hamming distance at another codeword position is added to the Hamming distance at a specific codeword position,
The transmission bit error can be easily detected, and the possibility of erroneous decoding with different control information sets can be reduced.

Furthermore, not only the codeword but also 8 before encoding
By securing the Hamming distance also in the bit length block, even when there is a communication section that does not use the 8B / 10B coding conversion rule, similar transmission bit errors can be easily detected in all the communication sections. Furthermore, by providing a coding conversion rule that generates a code word whose polarity is neutral and a coding conversion rule that generates a code word whose polarity is non-neutral, the code words generated according to the polarities of other code words The polarity can be manipulated, and 256 bytes in 1-byte 8-bit unit
Various control information can be encoded into one codeword for communication. In this way, when a control information set is newly added to the communication system in which the control information set already exists, a large number of highly reliable control information sets can be provided.

[Brief description of drawings]

FIG. 1 is a flowchart showing the content of a control information encoding method in a communication system according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing the content of a control information encoding method in a communication system according to a fourth embodiment of the present invention.

FIG. 3 is a flowchart showing the content of a control information encoding method in a communication system according to a sixth embodiment of the present invention.

FIG. 4 is a flowchart showing the contents of a control information encoding method in a communication system according to a seventh embodiment of the present invention.

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-3-243042 (JP, A) JP-A-56-65314 (JP, A) JP-A-59-10056 (JP, A) JP-A-2001-230678 (JP, A) (58) Fields surveyed (Int.Cl. ⁷ , DB name) H03M 3/00-11/00 G06F 11/10 330

Claims (10)

(57) [Claims] 1. The communication data is generated and transmitted in accordance with a 8B / 10B encoding conversion rule to generate a code word, and at the same time, M (M is 2) added for transmission and processing of the communication data. its control information of a natural number greater than or equal to)
Convert each to a codeword and start with the special codeword (1+
M) A control information encoding method in a communication system for transmitting as a control information set consisting of M codewords, wherein at least one control information set includes M control
Oite for encoding the plurality pieces of control information Chi caries information, the polarity is applied a coding conversion rule to generate only either one neutral or non-neutral code word, in the other control information sets, the plurality Control information
At least a Oite for encoding the control information applies the non-neutral encoding conversion rule polarity produces only non-neutral code words, placed on the coding of other at least one control information among
In particular, a control information encoding method in a communication system is characterized by applying a neutral encoding conversion rule for generating only codewords having a neutral polarity. 2. In a codeword group including all codewords generated by the non-neutral coding conversion rule and the neutral coding conversion rule, a Hamming distance between one codeword and all other codewords is The method for encoding control information in a communication system according to claim 1, wherein at least N + 1 (N is a natural number). 3. An 8-bit length block before 8B / 10B encoding corresponding to one codeword in the codeword group,
The Hamming distance to the 8-bit long block before 8B / 10B coding corresponding to all other codewords is at least N +.
3. The control information encoding method in the communication system according to claim 2, wherein the control information is 1 (N is a natural number). 4. The control in a communication system according to claim 1, wherein the neutral coding conversion rule is applied to a second code word in the control information set with M = 3. Information encoding method. 5. The code word generated according to the neutral coding conversion rule has a Hamming distance of at least N (N is a natural number) from a predetermined code word group. A method for encoding control information in the communication system according to any one of 1. 6. An 8B / 10-bit block before 8B / 10B coding corresponding to a codeword generated by the neutral coding conversion rule and 8B / corresponding to all codewords belonging to a predetermined codeword group. The method for encoding control information in a communication system according to claim 5, wherein the Hamming distance from the 8-bit length block before 10B encoding is at least N (N is a natural number). 7. The predetermined codeword group is (D20.4,
D5.4, D10.4, D31.2, D17.4, D2
1.4, D21.7, D31.3, D3 1, 5, D3
The method for encoding control information in a communication system according to claim 5, wherein N = 4 in 1.6). 8. The predetermined codeword group is (D20.4, D
5.4, D10.4, D31.2, D17.4, D2
1.4, D21.7, D31.3, D31.5, D3
1.6), the code word group generated by the neutral coding conversion rule is (D6.6, D14.1, D25.6, D25.
1, D19.2, D13.5, D13.2, D4.0)
Then, the codeword group generated by the non-neutral coding conversion rule is
(D18.7, D1.3, D0.5, D11.7, D
3.4, D26.0, D28.7, D23.1) , The control information coding method in the communication system according to claim 5, wherein 9. In the other control information set, M
At least one of the control information is applied with the compatible coding conversion rule that generates both a neutral and a non-neutral codeword, and the other control information is a code obtained by encoding the one control information. 9. The control information in the communication system according to claim 1, wherein the non-neutral coding conversion rule is applied when the word polarity is neutral, and the neutral coding conversion rule is applied when the word polarity is non-neutral. Encoding method. 10. The method for encoding control information in a communication system according to claim 9, wherein the one control information is encoded earlier than the other control information.