Claims
- 1. A maximum likelihood sequence estimator (MLSE) equalizer device being included in an MLSE sub-receiver comprising:
a feedforward circuit responsive to input data for processing the same to generate feedforward circuit output, said input data being generated from transmitted data being transmitted by wireless transmission; a feedback circuit responsive to said input data for processing the same to generate feedback circuit output; and an equalizer training module responsive to said feedforward circuit output and said feedback circuit output for training said MLSE equalizer device by minimizing the difference between said feedforward circuit output and said feedback circuit output, said MLSE equalizer device being trained to generate equalized data, said MLSE sub-receiver for decoding said equalized data to generate decoded transmitted data by mitigating the effects of multi-path communication channel due to wireless transmission of said transmitted data.
- 2. An MLSE equalizer device as recited in claim 1 wherein said feedforward circuit is a feedforward fractionally spaced filter (FFF) circuit and said feedback circuit is a decision feedback equalizer (DFE) circuit.
- 3. An MLSE equalizer device as recited in claim 2 wherein said FFF circuit is a finite-duration impulse response (FIR) filter circuit performing channel shortening on said input data.
- 4. An MLSE equalizer device as recited in claim 1 wherein said input data including a preamble portion, a header portion and a main data portion.
- 5. An MLSE equalizer device as recited in claim 4 wherein said feedback circuit for processing said preamble portion to generate feedback circuit output using feedback filter coefficients bk, said feedforward circuit responsive to incoming signals for processing the same to generate feedforward circuit output using feedforward filter coefficients fk.
- 6. An MLSE equalizer device as recited in claim 5 wherein said equalizer training module for training said MLSE equalizer device by adapting said feedback filter coefficients bk and said feedforward filter coefficients fk to minimize the difference between said feedforward circuit output and said feedback circuit output.
- 7. An MLSE equalizer device as recited in claim 6 wherein said equalizer training module adapts said feedback filter coefficients bk and said feedforward filter coefficients fk by updating a coefficient accumulator.
- 8. An MLSE equalizer device as recited in claim 7 wherein said equalizer training module adapts said feedback filter coefficients bk and said feedforward filter coefficients fk using one or more adaptation constants.
- 9. An MLSE equalizer device as recited in claim 6 wherein said equalizer training module for establishing a reference sequence by identifying sequence bits in said preamble portion.
- 10. An MLSE equalizer device as recited in claim 9 wherein said reference bits being used to locate a reference location in said preamble portion, said reference location for indicating the location of said reference sequence in said preamble portion, said reference sequence for initializing timing controls and frequency offset loops in an MLSE receiver system.
- 11. An MLSE equalizer device as recited in claim 10 wherein said equalizer training module includes a channel estimation module for calculating energy of each channel of data in said preamble portion using said reference sequence, said energy being used in adapting said feedback filter coefficients bk and said feedforward filter coefficients fk.
- 12. An MLSE equalizer device as recited in claim 11 wherein output of said channel estimation module includes a severe channel, a mild channel and an ideal channel, said severe channel having long filter lengths for said feedforward circuit and said feedback circuit, said mild channel having short filter lengths for said feedforward circuit and said feedback circuit, said ideal channel having substantially short filter lengths for said feedforward circuit and said feedback circuit.
- 13. An MLSE equalizer device as recited in claim 11 wherein said equalizer training module for minimizing the difference between said feedforward circuit output and said feedback circuit output to remove residual modeling uncertainty caused by said channel estimation module.
- 14. An MLSE equalizer device as recited in claim 6 for using a set of complementary code keying (CCK) codewords to generate a set of MLSE codewords, said header portion being used to determine said set of CCK codewords, said MLSE sub-receiver using said set of MLSE codewords to decode said equalized data.
- 15. An MLSE equalizer device as recited in claim 14 wherein said equalizer training module for training said MLSE equalizer device on-the-fly, said MLSE equalizer device for generating said set of MLSE codewords on-the-fly.
- 16. An MLSE equalizer device as recited in claim 1 wherein said feedforward circuit being located internally to said equalizer training module.
- 17. An MLSE equalizer device as recited in claim 6 wherein the leading tap of said feedback filter coefficients bk is set to a fixed value to avoid the trivial solution.
- 18. A method for generating decoded transmitted data comprising:
receiving input data and processing the same to generate feedforward circuit output; processing the input data to generate feedback circuit output; performing training by minimizing the difference between the feedforward circuit output and the feedback circuit output; generating equalized data; and decoding the generalized data to generate the decoded transmitted data.
REFERENCE TO PRIOR APPLICATION
[0001] This application claims the benefit of a previously filed U.S. Provisional Application No. 60/389,874 filed on Jun. 18, 2002, and entitled “SYMBOL-BASED DECISION FEEDBACK EQUALIZER (DFE) OPTIMAL EQUALIZATION METHOD AND APPARATUS WITH MAXIMUM LIKELIHOOD SEQUENCE ESTIMATION FOR WIRELESS RECEIVERS UNDER MULTIPATH CHANNELS”.
Provisional Applications (1)
|
Number |
Date |
Country |
|
60389874 |
Jun 2002 |
US |