Efficient, programmable and scalable low density parity check decoder

1. A low density parity check decoder comprising:

a plurality of message processing blocks arranged in a logical array; a plurality of info bit processors, one for each column of said message processing blocks in said logical array;

a plurality of lookup table memories, one for each column of said message processing blocks in said logical array; and

a plurality of check message processors, one for each row of said message processing blocks in said logical array,

wherein said decoder iteratively operates to decode a data signal, and

wherein the lookup table memories are loaded with:

information as to which bits of a corresponding compressed parity matrix are set to 1, said corresponding compressed parity matrix having an equal number of columns as the plurality of lookup table memories.

2. The decoder of claim 1 implemented in an ASIC.

3. The decoder of claim 1 implemented in a FPGA.

4. The decoder of claim 1 , wherein the lookup table memories are also loaded with bit-node message memory access pointers required for computing bit messages on a column basis.

5. A low density parity check decoder optimized for a parity check matrix in eIRA format comprising:

a plurality of message processing blocks arranged in a logical array;

a plurality of info bit processors, one for all but two columns of said message processing blocks in said logical array;

a plurality of lookup table memories, one for all but two columns of said message processing blocks in said logical array;

a plurality of check message processors, one for each row of said message processing blocks in said logical array;

wherein said decoder iteratively operates to decode an original data signal,

and

wherein the lookup table memories are loaded with information as to which bits of a corresponding compressed parity matrix are set to 1, said corresponding compressed parity matrix having an equal number of columns as the plurality of lookup table memories.

6. The decoder of claim 5 , implemented in an ASIC.

7. The decoder of claim 5 , implemented in a FPGA.

8. The decoder of claim 5 , wherein the lookup table memories are also loaded with bit-node message memory access pointers required for computing bit messages on a column basis.

9. A method of increasing processing efficiency in an LDPC decoder, comprising:

compressing an H parity matrix so as to arrange bit-node message “columns” to facilitate mapping to Message Processing Block MPB “columns” of an LDPC decoder and corresponding access via LUT pointer tables to minimize processing cycles,

wherein said compressing includes either:

(i) combining individual columns of the H parity matrix to create combined columns of the compressed matrix, where no row of the combined column has more than a single 1 bit,

or

(ii) for an H parity matrix in eIRA format:

mapping diagonal lines of a staircase portion of the H parity matrix to a first set of columns of the compressed matrix, and

combining individual columns of the random portion of the H parity matrix to create combined columns of the compressed matrix, where no row of the combined column has more than a single 1 bit.

10. The method of claim 9 , wherein said minimization of processing cycles minimizes address conflicts within the same MPB that will take multiple access cycles to resolve.

11. The method of claim 9 , wherein said minimization of processing cycles minimizes the splitting of bit-node messages across MPB “columns” that will take multiple access cycles to resolve.

12. The method of claim 9 , wherein said minimization of processing cycles balances the bit-node computations across all the MPB/LUT “columns” so that they will complete their computations at nearly the same time.