Pulse or digital communications – Systems using alternating or pulsating current – Plural channels for transmission of a single pulse train
Reexamination Certificate
1997-12-19
2003-04-01
Corrielus, Jean (Department: 2631)
Pulse or digital communications
Systems using alternating or pulsating current
Plural channels for transmission of a single pulse train
C375S298000
Reexamination Certificate
active
06542553
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a device and a method for coding data, a device and a method for decoding data and communication devices using them.
This invention is of use in all areas of coding, storage and transmission of digital data, and in particular those using an alphabet in which the number of symbols is different from 256.
In particular, the present invention applies to transmission of sextuples, modulated by an amplitude modulation in accordance with two carriers in quadrature (hereinafter called “QAM”) with 64 states (hereinafter called “QAM-64”).
2. Description of the Prior Art
There are many coding methods allowing error correction of digital data. Among the best known codes used at present, the Reed-Solomon codes may be mentioned. These constitute a powerful means of correcting data transmission errors. They may be constructed on any alphabet containing a number of symbols which is equal to a power, p
m
, of a prime number, p.
Very often a value of m equal to 8 and a value of p equal to 2 is chosen. The consequence of this large predominance of codes on alphabets with 2
8
(=256) symbols is that the majority of Reed-Solomon coders and decoders which are found on the market work on this alphabet. Their low relative cost and their high efficiency means that they are used in many areas notably in the transmission or storage of digital data on tape or disc. This is because a Reed-Solomon coder or decoder constructed to work on 2
8
symbols can also work on an alphabet containing 2
4
(=16), 2
2
(=4) or 2
1
(=2) symbols. The corresponding codes are commonly known under the name “BCH codes” on respectively GF(2
4
), GF(2
2
) or GF(2) (where GF means “Galois Field”). Nevertheless, an alphabet with 64 symbols cannot be treated in this way because the Galois field GF(2
6
) is not a sub-field of GF(2
8
).
Therefore, when the natural alphabet of an application contains 64 symbols, as in a system using a QAM-64 modulation, these symbols cannot be considered as words of a code on GF(2
8
).
Consequently, in the case of transmission of data modulated with a QAM-64 modulation, a person skilled in the art of transmission wishing to use inexpensive Reed-Solomon coding components uses them in a non-optimal way: he considers a sequence of binary data as a flow of octets which he codes with a Reed-Solomon coder. The code words produced are considered with no particular care as a sequence of 6-uples; each 6-uple is finally modulated in the form of a QAM-64 symbol.
On receipt, each symbol received is interpreted as a binary 6-uple. The resulting sequence of binary data is considered as a sequence of octets specifying one GF(2
8
) element. This sequence of GF(2
8
) elements, entering a Reed-Solomon decoder corresponding to the coder used at transmission, will be decoded in an ordinary manner. This manner of formatting QAM-64 symbols in octets has a significant drawback. As in any transmission system, transmission errors occur on QAM-64 symbols. However, the 6 bits of the same QAM-64 symbol may have been coded over two consecutive octets. As the Reed-Solomon decoder works on octets, it is possible that an error on a single QAM-64 symbol could produce an error on two consecutive octets, which amounts to doubling the error affecting the data transmitted in this manner. This reduces the correction capability of a Reed-Solomon coder expressed as a number of correctable QAM-64 symbols.
In order to solve the problem disclosed previously, a code specified on GF(2
6
) could be chosen. Two other problems then arise:
on the one hand, in this case, a component of this type is not readily found today on general sale, and
on the other hand, if it is wished to use words of length greater than or equal to 64 binary 6-uples, no Reed-Solomon code of this length is known on GF(2
6
).
Consequently, the redundancy of the codes is used less efficiently; for a given power of correction, a greater redundancy is required (in other words, the efficiency of the code is lower).
SUMMARY OF THE INVENTION
The present invention intends to remedy these drawbacks. It aims, above all, to allow the transmission of symbols forming part of an alphabet having a number of symbols equal to the alphabet used by a transmission means, while coding the said symbols with a code adapted to another alphabet.
To this end, the invention relates, according to a first aspect, to a device for transmitting digital data to be transmitted, representing a physical quantity and belonging to a first alphabet having P symbols, characterized in that it has:
a labelling means adapted to map each symbol of the first alphabet to secondary digital data belonging to a second alphabet having Q symbols, with Q strictly greater than P, P symbols of the second alphabet each representing exclusively one and only one symbol of the first alphabet,
a coding means adapted to determine redundant data belonging to the second alphabet, using predetermined coding rules taking into account the secondary digital data, and
a transmission means adapted to modulate at least one physical quantity into a series of signals each capable of taking a number P of different values, and successively representing, according to predetermined transmission rules, on the one hand, the digital data to be transmitted and, on the other hand, the said redundant data.
Correlatively, the invention relates, according to a second aspect, to a device for receiving signals representing so-called “to be transmitted” digital data belonging to a first alphabet having P symbols, characterised in that it has:
an identification means adapted to receive received digital data belonging to an alphabet having P symbols, and to map received digital data to so-called “to be decoded” symbols belonging to an alphabet having Q symbols, with Q strictly greater than P,
a decoding means adapted to correct errors affecting the symbols to be decoded, using predetermined decoding rules taking into account so-called “redundant” symbols to be decoded, and to supply so-called “corrected” symbols, and
a translation means adapted to supply so-called “transmitted” digital data capable of taking a number P of different values, and successively representing, according to predetermined so-called “translation” rules, corrected symbols.
Correlatively, the invention relates, according to a third aspect, to a method for transmitting digital data to be transmitted, representing a physical quantity and belonging to a first alphabet having P symbols, characterized in that it has:
a labelling step during which each symbol of the first alphabet is mapped to secondary digital data belonging to a second alphabet having Q symbols, with Q strictly greater than P, P symbols of the second alphabet each representing exclusively one and only one symbol of the first alphabet,
a coding step during which redundant data belonging to the second alphabet are determined, using predetermined coding rules taking into account the secondary digital data, and
a transmission step during which at least one physical quantity is modulated into a series of signals each capable of taking a number P of different values, and successively representing, according to predetermined transmission rules, on the one hand, digital data to be transmitted and, on the other hand, the said redundant data.
Correlatively, the invention relates, according to a third aspect, to a method for receiving signals representing so-called “to be transmitted” digital data belonging to a first alphabet having P symbols, characterized in that it has:
an identification step during which digital data belonging to an alphabet having P symbols are received, and received digital data are mapped to so-called “to be decoded” symbols belonging to an alphabet having Q symbols, with Q strictly greater than P,
a decoding step during which errors affecting the symbols to be decoded are corrected, using predetermined decoding rules taking into account so-called “redundant” symbols to be decoded, and to supply so
Le Dantec Claude
Piret Philippe
Canon Kabushiki Kaisha
Corrielus Jean
Fitzpatrick ,Cella, Harper & Scinto
LandOfFree
Device and method for transmitting digital data, device and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Device and method for transmitting digital data, device and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Device and method for transmitting digital data, device and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3098883