High speed iterative decoder

• US 7,853,855 B1
• Filed: 08/11/2009
• Issued: 12/14/2010
• Est. Priority Date: 08/25/2005
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: Marvell International Limited (Marvell Technology Group Limited)
• Inventors: He, Runsheng
• Primary Examiner(s): Lamarre; Guy J
• Application Number:
  US12/538,953
• Time in Patent Office: 490 Days
• Field of Search: 714/755, 714/786, 714/774
• US Class Current: 714/755
• CPC Class Codes: H03M 13/2957 Turbo codes and decoding
• CPC Class Codes: H03M 13/2975 Judging correct decoding, e.g. iteration stopping criteria
• CPC Class Codes: H03M 13/6544 IEEE 802.16 (WIMAX and broadband wireless access)

Abstract

A system includes a clock having a clock frequency. A first buffer configured to buffer a plurality of encoded data frames. An iterative decoder configured to iteratively decode, in accordance with the clock frequency. A first encoded data frame of the plurality of encoded data frames buffered in the first buffer. The iterative decoder generates a confidence result. The iterative decoder compares the confidence result to a predetermined value. The iterative decoder stops iteratively decoding the first encoded data frame based on the comparison of the confidence result to the predetermined value or based on a total number of decoding iterations. The clock frequency of the clock is based on a total number of encoded data frames buffered in the first buffer.

Claims

1. A system comprising:

a clock having a clock frequency;

a first buffer configured to buffer a plurality of encoded data frames; and

an iterative decoder configured to

iteratively decode, in accordance with the clock frequency, a first encoded data frame of the plurality of encoded data frames buffered in the first buffer, wherein for each decoding iteration, the iterative decoder

generates a confidence result; and

performs a comparison of the confidence result to a predetermined value; and

stop iteratively decoding the first encoded data frame (i) based on the comparison of the confidence result to the predetermined value or (ii) based on a total number of decoding iterations,

wherein the clock frequency of the clock is based on a total number of encoded data frames buffered in the first buffer.

2. The system of claim 1, wherein the clock frequency of the clock is further based on an average number of decoding iterations needed for the confidence result to achieve the predetermined value.

3. The system of claim 1, further comprising a second buffer configured to receive decoded data frames from the iterative decoder.

4. The system of claim 3, wherein the second buffer has a same memory space as the first buffer.

5. The system of claim 1, further comprising a physical layer module in communication with the first buffer.

6. The system of claim 5, wherein the physical layer module is compatible with at least one standard from a group consisting of a Bluetooth standard;

and Institute of Electrical and Electronics Engineers (IEEE) standards 802.3, 802.3an, 802.11, 802.11a, 802.11b, 802.11g, 802.11h, 802.11n, 802.16, and 802.20.

7. A transceiver comprising:

the system of claim 5; and

a forward error correction encoder configured to employ a low-density parity check (LDPC) code or a concatenated code.

8. A method comprising:

buffering, in a first buffer, a plurality of encoded data frames; and

iteratively decoding, in accordance with a clock frequency, first encoded data frame of the plurality of encoded data frames buffered in the first buffer, wherein each decoding iteration comprises

generating a confidence result; and

performing a comparison of the confidence result to a predetermined value; and

stopping the iterative decoding of the first encoded data frame (i) based on the comparison of the confidence result to the predetermined value or (ii) based on a total number of decoding iterations,

wherein the clock frequency is based on a total number of encoded data frames buffered in the first buffer.

9. The method of claim 8, wherein the clock frequency is further based on an average number of decoding iterations needed for the confidence result to achieve the predetermined value.

10. The method of claim 8, further comprising clocking the iterative decoding step at a clock frequency based on an average number of iterations needed for the confidence result to achieve the predetermined value.

11. The method of claim 8, further comprising further comprising buffering, in a second buffer, decoded data frames.

12. The method of claim 11, wherein the second buffer has a same memory space as the first buffer.