Synchronous data transmission method and device implementing same

• US 4,682,334 A
• Filed: 05/20/1985
• Issued: 07/21/1987
• Est. Priority Date: 05/23/1984
• Status: Expired due to Term

• Current Assignee: Compagnie industrielle des télécommunications CIT-ALCATEL
• Original Assignee: COMPAGNIE INDUSTRIELLE DES TELECOMMUNICATIONS CIT-ALCATEL
• Inventors: Le Bescont, Herve, Le Mouel, Bernard, Le Meur, Jean-Paul
• Primary Examiner(s): Smith, Jerry
• Assistant Examiner(s): Beausoliel, Jr., Robert W.
• Application Number:
  US06/736,038
• Time in Patent Office: 792 Days
• Field of Search: 371/47, 371/49, 371/55, 371/56, 178/69 D, 375/19
• US Class Current: 714/809
• CPC Class Codes: H04L 25/4906 using binary codes

Abstract

A synchronous data transmission method uses an MB 1C 1F type binary-binary code in which the binary data to be transmitted is subdivided into successive blocks of M bits complemented when they feature a disparity (marks minus spaces) of the same sign as the data already encoded, to which is added a complement bit indicating if complementing has been applied and a frame bit consisting of a parity bit enabling the subdivision into blocks carried out at the encoding stage to be recovered at the decoding stage. The encoder has at the input a demultiplexer carrying out the subdivision into blocks followed by circuits for computing the word digital sum and parity and the running digital sum, together with an inverter circuit which processes the blocks from the multiplexer and their parity bit under the control of a complementing decision circuit and a multiplexer transforming the encoded binary signal from the inverter circuit into an isochronous sequence to which a frame bit is added.

Claims

1. A synchronous data transmission method employing an MB 1C 1F type code for coding each M bits of binary data to be transmitted into an M+2 bit code word, said method steps of:

dividing the binary data to be transmitted into successive blocks of M bits each (where M is an even integer);

adding to each block a frame bit and a complement bit to form an intermediate word, said frame bit having a value determined in accordance with a parity of each block of M bits;

determining a word digital sum (WDS) in accordance with the difference between the number of marks (logic

     1) and spaces (logic

     0) of said intermediate word and complement bit;

determining a running digital sum (RDS) in accordance with the difference between the number of ones and zeros in the data already encoded;

complementing said intermediate word in accordance with a comparison of the signs of WDS and RDS, while changing the value of said complement bit, if necessary, to indicate whether said intermediate word has been complemented or not; and

combining said complement bit and said complemented intermediate word to form an output code word to be transmitted having M+2 bits.

2. Method according to claim 1, wherein the value of M is 12, the complement bit C is at logic 1 to indicate a non-complemented block or at logic 0 to indicate a complemented block, and the word digital sum WDS computed at the encoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by two units.

3. Method according to claim 1, wherein the value of M is 12, the complement bit C is at logic 0 to indicate a non-complemented block or at logic 1 to indicate a complemented block, and the word digital sum WDS computed at the encoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by one unit.

4. Method according to claim 1, wherein the frame bit is a parity bit transforming the associated block of M bits with its complemented bit C into an odd parity word of (M+2) bits.

5. Method according to claim 4, wherein the value of M is 12, the complement bit C is at logic 1 to indicate a non-complemented block and at logic 0 to indicate a complemented block, the running digital sum RDS computed at the endcoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by algebraic addition of the word digital sums WDS produced for each block by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by two units and with the sign changed in the event that the block is complemented, the least significant bit of each word digital sum WDS being ignored.

6. Method according to claim 4, wherein the value of M is 12, the complement bit C is at logic 0 to indicate a non-complemented block and at logic 1 to indicate a complemented block, the running digital sum RDS computed at the encoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by algebraic addition of the word digital sums WDS produced for each block by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by one unit and with the sign changed in the event that the block is complemented, the least significant bit of each word digital sum WDS being ignored.

7. Method according to claim 1, wherein the frame bit is a parity bit transforming the associated block of M bits with its complement bit C into an even parity word of (M+2) bits.

8. Method according to claim 7, wherein the value of M is 12, the complement bit C is at logic 1 to indicate a non-complemented block and at logic 0 to indicate a complemented block, the running digital sum RDS computed at the encoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by algebraic addition of the word digital sums WDS produced for each block by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by two units and with the sign changed in the event that the block is complemented.

9. Method according to claim 7, wherein the value of M is 12, the complemented bit C is at logic 0 to indicate a non-complemented block and at logic 1 to indicate a complemented block, the running digital sum RDS computed at the encoding stage to decide whether or not to complement the blocks is produced in two'"'"'s complement binary arithmetic by algebraic addition of the word digital sums WDS produced for each block by natural binary addition of the logic ones of the block concerned and its frame bit, with the sum increased by one unit and with the sign changed in the event that the block is complemented.

10. A method according to claim 1, further comprising the step of reconstituting said blocks, said reconstituting step including the step of monitoring a parity error rate, wherein the division into blocks effected at the encoding stage is reconstituted at the decoding stage by monitoring the parity error rate.

11. An encoder for implementing a synchronous data transmission method employing an MB 1C 1F type code in which the binary data to be transmitted is divided into successive blocks each of M bits (where M is an integer) each coded by a word comprising (M+2) bits made up of the M bits of the block, complemented or not according to whether the word which would be obtained without complementing would have a word digital sum WDS equal to the difference between the number of marks (logic 1) and spaces (logic 0) of the word of the same or opposite sign as a running digital sum RDS equal to the difference between the numbers of ones and zeros in data already encoded, a complement bit C indicating whether complementing is applied or not and a frame bit complemented or not with the M bits of the block, whereby on decoding the blocks formed at the encoding stage may be reconstituted, in which method M is even and the value of the frame bit is determined according to a parity of each block of M bits, the encoder comprising:

a demultiplexer for transforming the data to be transmitted into successive blocks of M bits,

a circuit for computing the word digital sum WDS and the parity of each block connected to the output of said demultiplexer,

a complementing decision circuit for comparing the signs of the word digital sum WDS and the running digital sum RDS,

a circuit for computing the running digital sum RDS of the binary data already encoded on the basis of the output signal from said circuit computing the digital word sum and the output signal from said complementing decision circuit,

a complementing circuit controlled by said complementing decision circuit, said complementing circuit processing the block of M bits available at the output of said demultiplexer and its parity bit delivered by said circuit computing the word digital sum and parity, and

a multiplexer for transforming into an isochronous binary data bit stream the words delivered by said complementing circuit to which has been added a complement bit delivered by said complementing decision circuit.

12. Encoder according to claim 11, wherein the value of M is 12 and said circuit computing the word digital sum WDS and parity comprises:

a first group of four full binary adders adapted to add three binary numbers each of one bit receiving on their individual inputs 12 parallel bits of each block delivered by said demultiplexer,

a second group of two fully binary adders one of which is adapted to add two two-bit binary numbers with a first carry input and another of which is adapted to add two two-bit binary numbers with a second carry input, each of said one and another adders adding pairs of binary numbers delivered by said first group of adders,

a full binary adder adapted to add two binary words each of four bits with a third carry input adding binary numbers delivered by said second group of adders and delivering the word digital sum WDS, and

a parity computation circuit the output signal of which is fed to one of said first, second and third carry inputs, another of said carry inputs being held at logic 1 and a last of said carry inputs being held at a logic level corresponding for the complement bit to the indication of a non-complemented block.