초록 ▼

The present invention concerns a method for obtaining an error correcting code of a given first size (N), including a systematic information part of a given second size (K) and a redundancy part. A block turbo-code is first obtained from said systematic information part and from predetermined inform

The present invention concerns a method for obtaining an error correcting code of a given first size (N), including a systematic information part of a given second size (K) and a redundancy part. A block turbo-code is first obtained from said systematic information part and from predetermined information elements provided at known locations, said block turbo-code consisting in a product of elementary systematic block codes, at least said predetermined information elements being removed from the obtained block turbo-code to produce a shortened block turbo-code. An additional error correcting code is derived from at least a portion of said shortened block turbo-code and said shortened block turbo-code is stuffed with stuffing elements so as to attain said given first size, said stuffing elements being information elements of said additional error correcting code.

대표청구항 ▼

What is claimed is: 1. A method for obtaining an error correcting code of a given first size (N), said error correcting code including a systematic information part of a given second size (K) and a redundancy part, according to which a block turbo-code is first obtained from said systematic informa

What is claimed is: 1. A method for obtaining an error correcting code of a given first size (N), said error correcting code including a systematic information part of a given second size (K) and a redundancy part, according to which a block turbo-code is first obtained from said systematic information part and from predetermined information elements provided at known locations; said block turbo-code being a product of elementary systematic block codes; at least said predetermined information elements being removed from said block turbo-code to produce a shortened block turbo-code, said method comprising: deriving an additional error correcting code from at least a portion of said shortened block turbo-code; and stuffing said shortened block turbo-code with stuffing elements to attain said given first size, said stuffing elements being information elements of said additional error correcting code. 2. The method for obtaining an error correcting code according to claim 1, wherein: said stuffing elements are information elements of said redundancy part of said additional error correcting code. 3. The method for obtaining an error correcting code according to claim 2, wherein: said error correcting code is a replication code. 4. The method for obtaining an error correcting code according to claim 1, wherein: said stuffing elements are inserted in said block turbo-code at locations where said predetermined information elements have been removed. 5. The method for obtaining an error correcting code according to claim 4, wherein: said information elements of said error correcting code are inserted according to a pseudo-random pattern. 6. The method for obtaining an error correcting code according to claim 4, wherein: said information elements of said error correcting code are inserted according to a uniform regular pattern. 7. The method for decoding an error correcting code obtained by a method according to claim 1, wherein: said error correcting code is provided by a demodulation step as a first block of soft values; a soft pre-decoding step using soft values corresponding to said stuffing elements is performed to produce a second block of soft values of higher reliability; a removing step where the soft values corresponding to said stuffing elements are removed from said second block is performed; said predetermined information elements are then inserted in said second block at said known locations as soft values of highest reliability to produce a third block of soft values; said third block of soft values is turbo-decoded; and said predetermined information elements are removed from said turbo-decoded block to produce a systematic information part having said given second size. 8. A coder for coding a block of information elements having a given second size (K) into an error correcting code having a given first size (N), said coder comprising; means for inserting at known locations predetermined information elements; means for turbo-coding a resulting block of the means for inserting by applying a product of elementary systematic block codes; means for removing at least said predetermined information elements from the turbo-coded block to produce a shortened block turbo-code; additional error correction coding means for coding at least a portion of said shortened block turbo-code; and stuffing means for stuffing said shortened block turbo-code with information elements of said additional error correcting code. 9. A decoder for decoding a first block of soft values corresponding to an error correcting code generated by a coder according to claim 8, said decoder comprising: means for performing a soft pre-decoding step using the soft values corresponding to the stuffing elements and for producing a second block of soft values of higher reliability; means for removing front said second block the soft values corresponding to said stuffing elements; means for inserting in said second block predetermined information elements with the highest reliability level at said known locations to produce a third block of soft values; means for turbo-decoding said third block of soft values; and means for removing from said turbo-decoded block said predetermined information elements to produce an information block having said second size. 10. The coder according to claim 8, wherein: said stuffing means stuffs information elements of a redundancy code part obtained from said additional error correction coding means. 11. The coder according to claim 10, wherein: said error correcting code is a replication code. 12. The coder according to claim 8, wherein: said error correcting code is a replication code. 13. The coder according to claim 8, wherein: said stuffing means stuffs information elements in said shortened block turbo-code at locations where said predetermined information elements have been removed. 14. The coder according to claim 13, wherein: said information elements of said error correcting code are inserted according to a pseudo-random pattern. 15. The coder according to claim 13, wherein: said information elements of said error correcting code are inserted according to a uniform regular pattern. 16. The coder according to claim 8, wherein: said error correcting code is provided as a first block of soft values by demodulation; a second block of soft values of higher reliability is produced using soft values corresponding to said information elements stuffed by said stuffing means; said soft values corresponding to said information elements stuffed by said stuffing means are removed from said second block; a third block of soft values is produced by inserting said predetermined information elements in said second block at said known locations using said means for inserting; said third block of soft values is turbo-decoded; and said predetermined information elements are removed from said turbo-decoded block using said means for removing to produce a systematic information part having said given second size. 17. A computer program product storing instructions for execution on a computer system, which when executed by the computer system, causes the computer system to obtain an error correcting code of a given first size (N), said error correcting code including a systematic information part of a given second size (K) and a redundancy part, according to which a block turbo-code is first obtained from said systematic information part and from predetermined information elements provided at known locations, said block turbo-code is a product of elementary systematic block codes, at least said predetermined information elements are removed from said block turbo-code to produce a shortened block turbo-code, by performing the steps of: deriving an additional error correcting code from at least a portion of said shortened block turbo-code; and stuffing said shortened block turbo-code with stuffing elements to attain said given first size, said stuffing elements being information elements of said additional error correcting code. 18. The computer program product according to claim 17, wherein: said stuffing elements are information elements of said redundancy part of said additional error correcting code. 19. The computer program product according to claim 18, wherein: said error correcting code is a replication code. 20. The computer program product according to claim 17, wherein: said stuffing elements are inserted in said block turbo-code at locations where said predetermined information elements have been removed.