Pipelined architecture to decode parallel and serial concatenated codes

• US 6,304,995 B1
• Filed: 01/26/1999
• Issued: 10/16/2001
• Est. Priority Date: 01/26/1999
• Status: Expired due to Term

• Current Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Original Assignee: TRW Limited (Zeppelin - Stiftung der Stadt Friedrichshafen)
• Inventors: Smith, Ronald P., Saunders, Oliver W.
• Primary Examiner(s): Decady, Albert
• Assistant Examiner(s): Torres, Joseph D.
• Application Number:
  US09/237,714
• Time in Patent Office: 994 Days
• Field of Search: 714/786, 714/794, 714/796, 714/792, 375/259, 375/265, 375/340
• US Class Current: 714/786
• CPC Class Codes: H03M 13/2957 Turbo codes and decoding
• CPC Class Codes: H03M 13/2987 using more component decoders than component codes, e.g. pipelined turbo iterations

Abstract

The present invention provides a method and apparatus (400) for iteratively decoding data which has been encoded with contatenated codes. The apparatus (400) includes pipelined and cascaded decoder processors (406, 430 and 436) connected to a multiple block memory device (402), through a multiplexing and data control block (404). A data decision element (437) is provided for generating decoded output data. The method includes receiving encoded data (802) while data already received is processed iteratively by decoder processors in a pipelined fashion. Decoder processors are designated to perform particular iterations (810) of an iterative decoding process which are performed simultaneously. As a decoder processor completes processing its designated iteration on a block of data, the decoder processor outputs decoding information (808) to the decoding processor designated to perform the subsequent iteration. Upon completion of all iterations for a block of data, the method includes generating output (814) consisting of the decoded data block. The method provides that once processing is complete on a data block, the memory block is made available (816) for the storing of new encoded input data.

Claims

1. A cascaded processor decoder for decoding encoded data, the decoder comprising:

a plurality of decoder processors, including a first cascaded decoder processor and a second cascaded decoder processor;

a metric output of said first cascaded decoder connected to a metric input of said second cascaded decoder processor; and

a multiple-block memory connected to said decoder processors for storing encoded input data arranged in multiple data blocks.

2. The cascaded processor decoder of claim 1, further comprising a plurality of multiplexers connected between said decoder processors and said memory, said multiplexers allowing any decoder processor to selectively access any data block in said memory.

3. The cascaded processor decoder of claim 1, further comprising a data decision element for generating decoded output data, said data decision element connected to said second cascaded decoder processor.

4. The cascaded processor decoder of claim 1, further comprising:

a plurality of multiplexers connected between said decoder processors and said memory, said multiplexers allowing any decoder processor to access any data block in said memory; and

a data decision element for generating decoded output data, said data decision element connected to said second cascaded decoder processor.

5. The cascaded processor decoder of claim 1, further comprising:

at least one multiplexer randomly accessing a data block of said multiple-block memory.

6. A cascaded processor decoder for iteratively decoding data, the decoder comprising:

a plurality of decoder processors equal in number to a predetermined number of decoding iterations, said decoder processors including at least a first cascaded decoder processor and a second cascaded decoder processor;

a metric output of said first cascaded decoder processor connected to a metric input of said second cascaded decoder processor;

a multiple-block memory comprising data blocks at least equal in number to said decoder processors;

a plurality of multiplexers connected between said data blocks and said decoder processors, said multiplexers allowing said decoder processors to selectively access any of said data blocks; and

a data decision element for generating decoded output data, said data decision element connected to said second cascaded decoder processor.

7. The cascaded processor decoder of claim 6 wherein each of said data blocks comprise:

a systematic bits section;

a first parity bits section; and

a second parity bits section.

8. The cascaded processor decoder of claim 7 wherein each of said decoder processors comprise:

a first decoder connected to a first data block including an associated systematic bits section and an associated first parity bits section;

a permuter connected to a metric output of said first decoder;

a second decoder connected to said first data block including an associated second parity bits section, said second decoder connected to an output of said permuter; and

an inverse permuter connected to a metric output of said second decoder.

9. The cascaded processor decoder of claim 8, wherein said data decision element comprises:

a delay element connected to a metric output of said first decoder of said second cascaded decoder processor;

an adder connected to an output of said delay element, said adder connected to an output of said inverse permuter of said second cascaded decoder processor;

a data decision processor connected to an output of said adder, said data decision processor including a decoded data bit output.

10. A method for iteratively decoding encoded data using a predetermined total number of decoding iterations, the method comprising:

receiving encoded data into a data block in a multiple-block memory;

accessing the encoded data in the data block by a first decoder processor through a plurality of multiplexers; and

iteratively decoding said encoded data a predetermined total number of times with said cascaded decoder processors.

11. The method for iteratively decoding encoded data of claim 10, further comprising:

generating an output bit stream of decoded encoded data with a data decision element.

12. The method for iteratively decoding encoded data of claim 10, wherein said step of iteratively decoding comprises iteratively decoding among at least two interconnected cascaded decoder processors.

13. The method for iteratively decoding encoded data of claim 10, wherein said step of iteratively decoding comprises decoding at most one iteration on each of said cascaded decoder processors.

14. The method for iteratively decoding encoded data of claim 10, wherein said step of iteratively decoding comprises:

performing a first number of decoding iterations on said encoded data with a first decoder processor, and generating first resultant metrics;

passing said first resultant metrics to a second decoder processor;

accessing the encoded data in the data block by the second decoder processor through a plurality of multiplexers;

performing a second number of decoding iterations on said encoded data and first resultant metrics with a second decoder processor and generating second resultant metrics; and

passing said second resultant metrics to a third decoder processor.

15. The method for iteratively decoding encoded data of claim 14, further comprising:

performing a last decoding iteration with a last decoder processor, and generating last resultant metrics;

passing said last resultant metrics from said last decoder processor to a data decision element; and

performing data decisions with said data decision element.

16. The method for iteratively decoding encoded data of claim 15 wherein said step of performing data decisions comprises generating decoded data based on a polarity of said last resultant metrics.

17. The method for iteratively decoding encoded data of claim 15, wherein said step of generating last resultant metrics comprises:

generating a first subset;

generating a second subset; and

generating said last resultant metrics by adding said first subset to said second subset.

18. The method for iteratively decoding encoded data of claim 17 wherein said step of performing data decisions comprises generating decoded data based on a polarity of said last resultant metrics.