Successive interference cancellation and multi-user minimum mean square channel estimation based on soft decoding information

1. A method comprising:

receiving a first signal that is a superposition of symbols transmitted concurrently by a set of users in shared resources of an aft interface; and

performing successive interference cancellation for the set of users, wherein an iteration of successive interference cancellation for a current user in the set of users comprises:

canceling, for the current user, interference in the first signal produced by the symbols transmitted by previous users in the set of users, wherein canceling the interference is performed based on soft estimates of symbols transmitted by the previous users, wherein the soft estimates of the symbols transmitted by the previous users were reconstructed in previous iterations of the successive interference cancellation based on previous log likelihood ratios (LLRs) generated by decoding of the symbols transmitted by the previous users;

decoding a current symbol transmitted by the current user and received in the first signal after canceling the interference based on the soft estimates of the symbols transmitted by the previous users;

reconstructing a current soft estimate of the current symbol transmitted by the current user based on a current LLR that represents a likelihood that the decoded current symbol is the same as the current symbol transmitted by the current user; and

feeding back the current soft estimate of the current symbol transmitted by the current user to cancel interference in at least one subsequent iteration of the successive interference cancellation for the set of users.

2. The method of claim 1 , further comprising:

performing multi-user minimum mean square estimation (MMSE) to determine estimated channels based on pilot symbols received from the set of users.

3. The method of claim 2 , wherein canceling, for the current user, the interference in the first signal produced by the symbols transmitted by the previous users in the set of users further comprises generating an estimated interference by combining the estimated channels with the reconstructed soft estimates of the symbols transmitted by the previous users and subtracting the estimated interference from the first signal.

4. The method of claim 3 , wherein the iteration of successive interference cancellation for the current user in the set of users further comprises:

generating an error variance matrix based on the current LLR.

5. The method of claim 4 , wherein generating the error variance matrix comprises generating diagonal elements of the error variance matrix using a quadratic function of the current LLR.

6. The method of claim 4 , wherein the iteration of successive interference cancellation for the current user in the set of users further comprises:

applying MMSE equalization to the first signal after subtracting the estimated interference from the first signal to generate an equalized signal, wherein the MMSE is performed for the current user based on the error variance matrix and the estimated channels.

7. The method of claim 6 , wherein decoding the symbols current symbol transmitted by the current user comprises decoding the equalized signal.

8. The method of claim 7 , wherein the iteration of successive interference cancellation for the current user in the set of users further comprises:

updating a value of the current LLR and the current soft estimate in response to decoding the equalized signal; and

providing the updated value of the current LLR and the current soft estimate for use in a subsequent iteration of successive interference cancellation.

9. The method of claim 1 , wherein performing the successive interference cancellation comprises performing the iterations of successive interference cancellation until detecting all of the set of users or reaching a maximum number of iterations.

10. The method of claim 1 , further comprising:

determining a first number of the set of users that are to transmit the first signal including the superposition of symbols in the shared resources of the air interface; and

providing, to the first number of the set of users, information indicating modulation and coding schemes, power control settings, and pilot sequences to be used for transmitting the symbols.

11. An apparatus comprising:

a transceiver configured to receive a first signal that is a superposition of symbols transmitted concurrently by a set of users in shared resources of an air interface; and

a processor configured to perform successive interference cancellation for the set of users, wherein an iteration of successive interference cancellation for a current user in the set of users comprises:

canceling, for the current user, interference produced in the first signal by the symbols transmitted by previous users in the set of users based on soft estimates of symbols transmitted by the previous users, wherein the soft estimates of the symbols transmitted by the previous users were reconstructed in previous iterations of the successive interference cancellation based on previous log likelihood ratios (LLRs) generated by decoding of the symbols transmitted by the previous users;

decoding a current symbol transmitted by the current user and received in the first signal after canceling the interference based on the soft estimates of the symbols transmitted by the previous users;

reconstructing a current soft estimate of the current symbol transmitted by the current user based on a current LLR that represents a likelihood that the decoded current symbol is the same as the current symbol transmitted by the current user; and

feeding back the current soft estimate of the current symbol transmitted by the current user to cancel interference in at least one subsequent iteration of the successive interference cancellation.

12. The apparatus of claim 11 , wherein the processor is further configured to estimate channels using multi-user minimum mean square estimation (MMSE) based on pilot symbols received from the set of users.

13. The apparatus of claim 12 , wherein the processor is further configured to generate an estimated interference in the iteration of successive interference cancellation for the current user by combining the estimated channels with the reconstructed soft estimates of the symbols transmitted by the previous users, and wherein the processor is configured to subtract the estimated interference from the first signal in the iteration of successive interference cancellation for the current user.

14. The apparatus of claim 13 , wherein the processor is further configured to generate an error variance matrix based on the current LLR in the iteration of successive interference cancellation for the current user.

15. The apparatus of claim 14 , wherein the processor is further configured to generate diagonal elements of the error variance matrix using a quadratic function of the current LLR in the iteration of successive interference cancellation for the current user.

16. The apparatus of claim 14 , wherein the processor is further configured to apply MMSE equalization to the first signal in the iteration of successive interference cancellation for the current user after subtracting the estimated interference from the first signal to generate an equalized signal, wherein the MMSE is performed on the basis of the error variance matrix and the estimated channels.

17. The apparatus of claim 16 , wherein the processor is further configured to decode the equalized signal in the iteration of successive interference cancellation for the current user.

18. The apparatus of claim 17 , wherein the processor is further configured to:

update a value of the current LLR and the current soft estimate in response to decoding the equalized signal in the iteration of successive interference cancellation for the current user; and

provide the updated value of the current LLR and the current soft estimate in the iteration of successive interference cancellation for the current user for use in a subsequent iteration of successive interference cancellation.

19. The apparatus of claim 11 , wherein the processor is configured to perform additional iterations of the successive interference cancellation until detecting all of the set of users or reaching a maximum number of iterations.

20. The apparatus of claim 11 , wherein the apparatus is a base station, and wherein:

the processor is further configured to determine a first number of the set of users that are to concurrently transmit symbols in the shared resources of the air interface; and

the transceiver is further configured to provide, to the first number of the set of users, information indicating modulation and coding schemes, power control settings, and pilot sequences to be used for transmitting the symbols.

21. An apparatus comprising:

a receiver configured to receive a first signal that is a superposition of data streams transmitted concurrently in shared resources of an air interface; and

a processor configured to perform successive interference cancellation (SIC) on the first signal, wherein an iteration of the SIC comprises:

canceling, for a first data stream in the superposition of data streams, interference produced in the first signal based on soft estimates of symbols transmitted in at least one second data stream in the superposition of data streams, wherein the soft estimates were reconstructed in previous iterations of the SIC based on previous log likelihood ratios (LLRs) generated by decoding of the symbols transmitted in the at least one second data stream;

decoding a first symbol transmitted in the first data stream and received in the first signal after canceling the interference based on the soft estimates of the symbols transmitted in the at least one second data stream;

reconstructing a first soft estimate of the first symbol in the first data stream based on a first LLR that represents a first likelihood that the decoded first symbol is the same as the first symbol transmitted in the first data stream; and

feeding back the first estimate of the first symbol transmitted in the first data stream to cancel interference in at least one subsequent iteration of the SIC.