Probabilistic amplitude shaping
This disclosure provides methods, devices and systems for encoding data in wireless communications. Some implementations more specifically relate to performing a first encoding operation on data bits of a code block to shape the amplitudes of the resultant symbols such that the amplitudes have a non-uniform distribution. In some aspects, the probabilities associated with the respective amplitudes generally increase with decreasing amplitude. For example, the non-uniform distribution of the amplitudes of the symbols may be approximately Gaussian. In some aspects, the first encoding operation is or includes a prefix encoding operation having an effective coding rate greater than 0.94 but less than 1. The first encoding operation is followed by a second encoding operation that also adds redundancy but does not alter the data bits themselves. In some aspects, the second encoding operation is or includes a low-density parity-check (LDPC) encoding operation associated with a coding rate greater than 5/6.
1. A method for wireless communication by a wireless communication device, the method comprising:
performing a first encoding operation on a plurality of information bits that produces a plurality of amplitude-shaped bits;
arranging the plurality of amplitude-shaped bits into a plurality of code blocks, each code block having a number (M) of bits that includes one or more of the plurality of amplitude-shaped bits;
performing a second encoding operation on the plurality of code blocks that produces a plurality of respective codewords, each codeword having a number (L) of bits that includes the one or more amplitude-shaped bits of the respective code block and one or more parity bits resulting from the second encoding operation, where M/L>5/6;
arranging the one or more amplitude-shaped bits and the one or more parity bits of each of the plurality of codewords into a plurality of symbols, each symbol having an amplitude based on the respective amplitude-shaped bits arranged in the symbol, the first encoding operation producing the plurality of amplitude-shaped bits so that the amplitudes of the plurality of symbols have a non-uniform distribution; and
transmitting a wireless packet, including the plurality of symbols, to at least one receiving device.
2. The method of claim 1 , wherein M/L=7/8.
3. The method of claim 1 , wherein the first encoding operation has an effective coding rate greater than or equal to 0.5 and less than 1.
4. The method of claim 1 , wherein the performing of the first encoding operation comprises:
selecting, from a look-up table (LUT), a pattern of bit values that matches a subset of the information bits, the LUT storing a plurality of patterns of bit values corresponding to a respective plurality of patterns of amplitude-shaped bits, the plurality of amplitude-shaped bits including the pattern of amplitude-shaped bits corresponding to the selected pattern of bit values.
5. The method of claim 4 , wherein each of the plurality of patterns of amplitude-shaped bits represents a respective amplitude level having a probability of occurrence based on a probability mass function (PMF).
6. The method of claim 4 , wherein each of the plurality of patterns of amplitude-shaped bits represents two or more amplitude levels having a probability of occurrence based on a PMF.
7. The method of claim 1 , wherein the first encoding operation is not performed on a plurality of unshaped information bits, the M bits of each code block further including one or more of the plurality of unshaped information bits.
8. The method of claim 1 , wherein the second encoding operation is based on a low-density parity-check (LDPC) code.
9. The method of claim 8 , wherein the LDPC code has a coding rate equal to 7/8.
10. The method of claim 8 , wherein the LDPC code has a coding rate equal to 5/6.
11. The method of claim 8 , wherein the performing of the second encoding operation comprises:
encoding the M bits of each code block as a respective number (N) of codeword bits based on the LDPC code; and
puncturing a number (K) of the codeword bits associated with each code block so that L=N−K.
12. The method of claim 11 , wherein the puncturing is not performed on the plurality of amplitude-shaped bits.
13. The method of claim 11 , wherein M=1620, N=1944, and K≥90.
14. The method of claim 11 , wherein M=1701, N=2106, and K=162.
15. A wireless communication device comprising:
at least one modem;
at least one processor communicatively coupled with the at least one modem; and
at least one memory communicatively coupled with the at least one processor and storing processor-readable code that, when executed by the at least one processor in conjunction with the at least one modem, is configured to:
perform a first encoding operation on a plurality of information bits that produces a plurality of amplitude-shaped bits;
arrange the plurality of amplitude-shaped bits into a plurality of code blocks, each code block having a number (M) of bits that includes one or more of the plurality of amplitude-shaped bits;
perform a second encoding operation on the plurality of code blocks that produces a plurality of respective codewords, each codeword having a number (L) of bits that includes the one or more amplitude-shaped bits of the respective code block and one or more parity bits resulting from the second encoding operation, where M/L>5/6;
arrange the one or more amplitude-shaped bits and the one or more parity bits of each of the plurality of codewords into a plurality of symbols, each symbol having an amplitude based on the respective amplitude-shaped bits arranged in the symbol, the first encoding operation producing the plurality of amplitude-shaped bits so that the amplitudes of the plurality of symbols have a non-uniform distribution; and
transmit a wireless packet, including the plurality of symbols, to at least one receiving device.
16. The wireless communication device of claim 15 , wherein M/L=7/8.
17. The wireless communication device of claim 15 , wherein the first encoding operation has an effective coding rate greater than or equal to 0.5 and less than 1.
18. The wireless communication device of claim 15 , wherein the performing of the first encoding operation comprises:
selecting, from a look-up table (LUT), a pattern of bit values that matches a subset of the information bits, the LUT storing a plurality of patterns of bit values corresponding to a respective plurality of patterns of amplitude-shaped bits, the plurality of amplitude-shaped bits including the pattern of amplitude-shaped bits corresponding to the selected pattern of bit values.
19. The wireless communication device of claim 15 , wherein the performing of the second encoding operation comprises:
encoding the M bits of each code block as a respective number (N) of codeword bits based on a low-density parity-check (LDPC) code; and
puncturing a number (K) of the codeword bits associated with each code block so that L=N−K.
20. The wireless communication device of claim 19 , wherein the puncturing is not performed on the plurality of amplitude-shaped bits.
21. A method for wireless communication by a wireless communication device, the method comprising:
receiving a wireless packet including a plurality of symbols having a plurality of amplitudes, the plurality of symbols representing a plurality of codeword bits, the plurality of amplitudes having a non-uniform distribution;
arranging the plurality of codeword bits into a plurality of code blocks, each code block including a number (L) of the codeword bits;
performing a first decoding operation on the plurality of code blocks that produces a plurality of respective codewords, each codeword having a number (M) of bits that includes a plurality of amplitude-shaped bits and a plurality of parity bits, where M/L>5/6, the plurality of amplitude-shaped bits of each codeword indicating the amplitude of a respective symbol of the plurality of symbols; and
performing a second decoding operation on the plurality of amplitude-shaped bits of each codeword that produces a plurality of respective de-shaped bits for each of the plurality of codewords.
22. The method of claim 21 , wherein M/L=7/8.
23. The method of claim 21 , wherein the first decoding operation is based on a low-density parity-check (LDPC) code.
24. The method of claim 23 , wherein the LDPC code has a coding rate equal to 7/8.
25. The method of claim 23 , wherein the LDPC code has a coding rate equal to 5/6.
26. The method of claim 21 , wherein the second decoding operation reverses a prefix encoding operation having an effective coding rate greater than or equal to 0.5 and less than 1.
27. The method of claim 21 , wherein the performing of the second decoding operation comprises:
selecting, from a look-up table (LUT), a pattern of de-shaped bits that matches the plurality of amplitude-shaped bits of a respective codeword of the plurality of codewords, the LUT storing a plurality of patterns of de-shaped bits corresponding to a respective plurality of patterns of amplitude-shaped bits, the plurality of de-shaped bits including the selected pattern of de-shaped bits.
28. The method of claim 27 , wherein each of the plurality of patterns of amplitude-shaped bits represents a respective amplitude of the plurality of amplitudes having a probability of occurrence based on a probability mass function (PMF).
29. The method of claim 27 , wherein each of the plurality of patterns of amplitude-shaped bits represents two or more amplitudes of the plurality of amplitudes having a probability of occurrence based on a PMF.
30. A wireless communication device comprising:
at least one modem;
at least one processor communicatively coupled with the at least one modem; and
at least one memory communicatively coupled with the at least one processor and storing processor-readable code that, when executed by the at least one processor in conjunction with the at least one modem, is configured to:
receive a wireless packet including a plurality of symbols having a plurality of amplitudes, the plurality of symbols representing a plurality of codeword bits, the plurality of amplitudes having a non-uniform distribution;
arrange the plurality of codeword bits into a plurality of code blocks, each code block including a number (L) of the codeword bits;
perform a first decoding operation on the plurality of code blocks that produces a plurality of respective codewords, each codeword having a number (M) of bits that includes a plurality of amplitude-shaped bits and a plurality of parity bits, where M/L>5/6, the plurality of amplitude-shaped bits of each codeword indicating the amplitude of a respective symbol of the plurality of symbols; and
perform a second decoding operation on the plurality of amplitude-shaped bits of each codeword that produces a plurality of respective de-shaped bits for each of the plurality of codewords.
31. The wireless communication device of claim 30 , wherein M/L=7/8.
32. The wireless communication device of claim 30 , wherein the first decoding operation is based on a low-density parity-check (LDPC) code.
33. The wireless communication device of claim 30 , wherein the second decoding operation reverses a prefix encoding operation having an effective coding rate greater than or equal to 0.5 and less than 1.
34. The wireless communication device of claim 30 , wherein the performing of the second decoding operation comprises:
selecting, from a look-up table (LUT), a pattern of de-shaped bits that matches the plurality of amplitude-shaped bits of a respective codeword of the plurality of codewords, the LUT storing a plurality of patterns of de-shaped bits corresponding to a respective plurality of patterns of amplitude-shaped bits, the plurality of de-shaped bits including the selected pattern of de-shaped bits.