Symbol by symbol map detection for signals corrupted by colored and/or signal dependent noise

• US 20070226599A1
• Filed: 05/22/2006
• Published: 09/27/2007
• Est. Priority Date: 03/23/2006
• Status: Active Grant

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Motwani, Ravi
• Granted Patent: US 8,091,009 B2
• Time in Patent Office: Days
• US Class Current: 714/792
• CPC Class Codes: G11B 20/10009 Improvement or modification of read or write signals
• CPC Class Codes: G11B 20/18 Error detection or correction; Testing , e.g. of drop-outs
• CPC Class Codes: H03M 13/1102 Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes
• CPC Class Codes: H03M 13/2957 Turbo codes and decoding

Abstract

Symbol by symbol MAP detection for signals corrupted by colored and/or signal dependent noise. A novel means is presented for recursive calculation of forward metrics (α), backward metrics (β), and corresponding soft information (e.g., which can be provided as LLRs (log likelihood ratios)) within communication systems in which a trellis can be employed to perform demodulation of a received signal sequence. For signals that have been corrupted by colored and/or signal dependent noise, this means provides for the ability to perform novel soft information calculation for subsequent use in iterative decoding processing. Many types of communication channels can benefit from this novel means of detection including communication channels within hard disk drives (HDDs).

Claims

1. An apparatus, comprising:

a processing module; and

a memory, coupled to the processing module, that is operable to store operational instructions that enable the processing module to;

employ a trellis to demodulate a plurality of symbols having colored noise;

for each symbol of the plurality of symbols, recursively calculate a corresponding forward metric with respect to the trellis using a first branch metric term that is based on at least one other symbol of the plurality of symbols;

for each symbol of the plurality of symbols, recursively calculate a corresponding backward metric with respect to the trellis using a second branch metric term that is based on at least one other symbol of the plurality of symbols; and

for each symbol of the plurality of symbols, calculate a corresponding LLR (log likelihood ratio) using at least one forward metric corresponding to a first time and at least one backward metric corresponding to a second time.

2. The apparatus of claim 1, further comprising:

an iterative decoder that is operable to;

receive the plurality of LLRs corresponding to the plurality of symbols; and

perform iterative decoding processing thereby making a best estimate for each symbol of the plurality of symbols.

3. The apparatus of claim 2, wherein:

the iterative decoder performs a plurality of local decoding iterations to generate the plurality of best estimates such that each best estimate corresponds to one symbol of the plurality of symbols;

the iterative decoder feeds back the plurality of best estimates to the processing module as “

a priori”

information for use during a global decoding iteration;

the memory, coupled to the processing module, is operable to store operational instructions that enable the processing module to calculate at least one additional corresponding LLR using at least one additional forward metric corresponding to the first time and at least one additional backward metric corresponding to the second time for each symbol of the plurality of symbols; and

the iterative decoder is operable to;

receive the at least one additional plurality of LLRs corresponding to the plurality of symbols; and

perform iterative decoding processing thereby making at least one additional best estimate for each symbol of the plurality of symbols.

4. The apparatus of claim 2, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the signal that is received from the communication channel is an LDPC (Low Density Parity Check) coded signal; and

the iterative decoder is an LDPC code decoder.

5. The apparatus of claim 2, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the signal that is received from the communication channel is a turbo coded signal; and

the iterative decoder is a turbo code decoder.

6. The apparatus of claim 1, wherein:

at least one symbol of the plurality of symbols includes only one bit.

7. The apparatus of claim 1, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the communication channel is a read channel that is coupled to a storage media of a hard disk drive (HDD).

8. The apparatus of claim 1, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the communication channel is a wired or a wireless communication channel.

9. An apparatus, comprising:

a detector that is operable to;

employ a trellis to demodulate a plurality of symbols having colored noise;

for each symbol of the plurality of symbols, recursively calculate a corresponding forward metric with respect to the trellis;

for each symbol of the plurality of symbols, recursively calculate a corresponding backward metric with respect to the trellis; and

for each symbol of the plurality of symbols, calculate a corresponding LLR (log likelihood ratio) using at least one forward metric corresponding to a first time and at least one backward metric corresponding to a second time; and

an iterative decoder that is operable to;

receive the plurality of LLRs corresponding to the plurality of symbols; and

perform iterative decoding processing thereby making a best estimate for each symbol of the plurality of symbols.

10. The apparatus of claim 9, wherein:

the detector is operable to;

for each symbol of the plurality of symbols, recursively calculate the corresponding forward metric with respect to the trellis using a first branch metric term that is based on at least one other symbol of the plurality of symbols; and

for each symbol of the plurality of symbols, recursively calculate the corresponding backward metric with respect to the trellis using a second branch metric term that is based on at least one other symbol of the plurality of symbols.

11. The apparatus of claim 9, further comprising:

an AFE(analog front-end) analog that is operable to receive a continuous-time signal from a communication channel;

an ADC (analog to digital converter) that is operable to sample the continuous-time signal thereby generating a discrete-time signal; and

a digital equalizer that is operable to perform equalization on the discrete-time signal thereby generating the plurality of symbols having colored noise.

12. The apparatus of claim 9, wherein:

the iterative decoder performs a plurality of local decoding iterations to generate the plurality of best estimates such that each best estimate corresponds to one symbol of the plurality of symbols;

the iterative decoder feeds back the plurality of best estimates to the detector as “

a priori”

information for use during a global decoding iteration;

for each symbol of the plurality of symbols, the detector is operable to calculate at least one additional corresponding LLR using at least one additional forward metric corresponding to the first time and at least one additional backward metric corresponding to the second time; and

the iterative decoder is operable to;

receive the at least one additional plurality of LLRs corresponding to the plurality of symbols; and

perform iterative decoding processing thereby making at least one additional best estimate for each symbol of the plurality of symbols.

13. The apparatus of claim 9, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the signal that is received from the communication channel is an LDPC (Low Density Parity Check) coded signal; and

the iterative decoder is an LDPC code decoder.

14. The apparatus of claim 9, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel;

the signal that is received from the communication channel is a turbo coded signal; and

the iterative decoder is a turbo code decoder.

15. The apparatus of claim 9, wherein:

at least one symbol of the plurality of symbols includes only one bit.

16. The apparatus of claim 9, wherein:

the plurality of symbols having colored noise is generated by processing a signal that is received from a communication channel; and

the communication channel is a read channel that is coupled to a storage media of a hard disk drive (HDD).

17. A method, comprising:

employing a trellis to demodulate a plurality of symbols having colored noise;

for each symbol of the plurality of symbols, recursively calculating a corresponding forward metric with respect to the trellis using a first branch metric term that is based on at least one other symbol of the plurality of symbols;

for each symbol of the plurality of symbols, recursively calculating a corresponding backward metric with respect to the trellis using a second branch metric term that is based on at least one other symbol of the plurality of symbols; and

for each symbol of the plurality of symbols, calculating a corresponding LLR (log likelihood ratio) using at least one forward metric corresponding to a first time and at least one backward metric corresponding to a second time.

18. The method of claim 17, further comprising:

performing iterative decoding processing, using the plurality of LLRs corresponding to the plurality of symbols, thereby making a best estimate for each symbol of the plurality of symbols.

19. The method of claim 17, further comprising:

performing a first plurality of local decoding iterations, such that at least one local decoding iteration thereof uses the plurality of LLRs corresponding to the plurality of symbols, thereby making a best estimate for each symbol of the plurality of symbols;

feeding back the plurality of best estimates, such that each best estimate corresponds to one symbol of the plurality of symbols, as “

a priori”

information for use during a global decoding iteration;

for each symbol of the plurality of symbols, calculating at least one additional corresponding LLR using at least one additional forward metric corresponding to the first time and at least one additional backward metric corresponding to the second time; and

performing a second plurality of local decoding iterations, such that at least one local decoding iteration thereof uses the at least one additional plurality of LLRs corresponding to the plurality of symbols, thereby making at least one additional best estimate for each symbol of the plurality of symbols.

20. The method of claim 17, further comprising:

processing a signal that is received from a communication channel to generate the plurality of symbols having colored noise, wherein the communication channel is a read channel that is coupled to a storage media of a hard disk drive (HDD).