Staggered transmission and reception for reducing latency and memory
View Patent ↗A demodulator processes a continuous-time signal to generate at a plurality of encoded bits. An inner decoder processes a first subset of bits within the plurality of encoded bits to correct selected ones of the first subset of bits to form a corrected first subset of bits and to generate partially corrected data from the plurality of encoded bits based on the corrected first subset of bits. An outer decoder processes the partially decoded data, to correct selected ones of a second subset of the plurality of encoded bits to form a corrected second subset of bits. A bit combiner generates data estimates by combining the corrected first subset of bits and the corrected second subset of bits.
1. An apparatus, comprising:
a demodulator to process a continuous-time signal to generate a plurality of encoded bits, the plurality of encoded bits are generated from a plurality of symbols, the plurality of symbols having n bits per symbol;
an inner decoder to process a first subset of bits within the plurality of encoded bits to correct selected ones of the first subset of bits to form a corrected first subset of bits and to generate partially corrected data from the plurality of encoded bits based on the corrected first subset of bits, wherein the first subset of bits correspond to a plurality of least significant bits of the n bits per symbol;
an outer decoder to process the partially corrected data, to correct selected ones of a second subset of the plurality of encoded bits to form a corrected second subset of bits; and
a bit combiner to generate data estimates by combining the corrected first subset of bits and the corrected second subset of bits.
2. The apparatus of claim 1 , wherein:
the second subset of the plurality of encoded bits correspond to a plurality of most significant bits of the n bits per symbol.
3. The apparatus of claim 1 , wherein:
the continuous-time signal is a set partition coded modulation (SPCM) signal.
4. The apparatus of claim 1 , wherein:
the plurality of encoded bits are generated from one of a plurality of different data channels; and
the inner decoder includes a low density parity check (LDPC) decoder that generates the corrected first subset of bits for the plurality of different data channels.
5. The apparatus of claim 1 , wherein the inner decoder includes:
a low density parity check (LDPC) decoder that generates the corrected first subset of bits.
6. The apparatus of claim 5 , wherein the inner decoder further includes:
a low density parity check (LDPC) encoder that re-encodes the corrected first subset of bits to form a re-encoded first subset of bits;
an error detector, coupled to the low density parity check (LDPC) encoder, that generates an error detection signal by comparing the re-encoded first subset of bits to the first subset of bits; and
a bit corrector, coupled to the error detector, that generates the partially corrected data based on the error detection signal.
7. The apparatus of claim 1 , wherein the outer decoder includes:
a Reed-Solomon (RS) decoder that generates the corrected second subset of bits.
8. The apparatus of claim 1 , wherein the outer decoder includes:
a deinterleaver, that de-interleaves the partially corrected data to generate a plurality of deinterleaved data; and
a Reed-Solomon (RS) decoder that generates the corrected second subset of bits, based on the plurality of deinterleaved data.
9. The apparatus of claim 1 , wherein the outer decoder includes:
a symbol converter that processes the partially corrected data to generate a plurality of Reed-Solomon encoded data;
a deinterleaver, that de-interleaves the plurality of Reed-Solomon encoded data to generate a plurality of deinterleaved data; and
a Reed-Solomon (RS) decoder that generates the corrected second subset of bits, based on the plurality of deinterleaved data.
10. A method comprising:
inner decoding to process a first subset of bits within a plurality of encoded bits to correct selected ones of the first subset of bits to form a corrected first subset of bits and to generate partially corrected data from the plurality of encoded bits based on the corrected first subset of bits, wherein the plurality of encoded bits are generated from a plurality of symbols, the plurality of symbols having n bits per symbol and wherein the first subset of bits correspond to a plurality of most significant bits of the n bits per symbol;
outer decoding to process the partially corrected data, to correct selected ones of a second subset of the plurality of encoded bits to form a corrected second subset of bits, wherein the second subset of the plurality of encoded bits correspond to a plurality of most significant bits of the n bits per symbol; and
generating data estimates by combining the corrected first subset of bits and the corrected second subset of bits.
11. The method of claim 10 , wherein:
the plurality of encoded bits are demodulated from a set partition coded modulation (SPCM) signal.
12. The method of claim 10 , wherein the inner decoding includes:
generating the corrected first subset of bits for a plurality of different data channels via a single low density parity check (LDPC) decoder.
13. The method of claim 10 , wherein the inner decoding includes:
generating the corrected first subset of bits via a low density parity check (LDPC) decoding.
14. The method of claim 13 , wherein the inner decoding further includes:
re-encoding the corrected first subset of bits to form a re-encoded first subset of bits via a low density parity check (LDPC) encoder;
generating an error detection signal by comparing the re-encoded first subset of bits to the first subset of bits; and
generating the partially corrected data based on the error detection signal.
15. The method of claim 10 , wherein the outer decoding includes:
a Reed-Solomon (RS) decoder that generates the corrected second subset of bits.
16. The method of claim 10 , wherein the outer decoding includes:
de-interleaving the partially corrected data to generate a plurality of deinterleaved data; and
a Reed-Solomon (RS) decoding that generates the corrected second subset of bits, based on the plurality of deinterleaved data.
17. The method of claim 10 , wherein the outer decoding includes:
processing the partially corrected data to generate a plurality of Reed-Solomon encoded data;
de-interleaving the plurality of Reed-Solomon encoded data to generate a plurality of deinterleaved data; and
generating the corrected second subset of bits, by decoding the plurality of deinterleaved data.
18. An apparatus, comprising:
a demodulator to process a continuous-time signal from a plurality of downstream cable channels to generate a plurality of encoded bits, the plurality of encoded bits are generated from a plurality of symbols, the plurality of symbols having n bits per symbol;
an inner decoder to process a first subset of bits within the plurality of encoded bits to correct selected ones of the first subset of bits to form a corrected first subset of bits and to generate partially corrected data from the plurality of encoded bits based on the corrected first subset of bits, wherein the first subset of bits correspond to a plurality of least significant bits of the n bits per symbol;
an outer decoder to process the partially corrected data, to correct selected ones of a second subset of the plurality of encoded bits to form a corrected second subset of bits; and
a bit combiner to generate data estimates by combining the corrected first subset of bits and the corrected second subset of bits.
19. The apparatus of claim 18 , wherein:
the continuous-time signal is a set partition coded modulation (SPCM) signal.
20. The apparatus of claim 18 , wherein:
the plurality of encoded bits are generated from one of a plurality of different data channels; and
the inner decoder includes a low density parity check (LDPC) decoder that generates the corrected first subset of bits for the plurality of different data channels.