Low-complexity, highly-spectrally-efficient communications
A system may comprise circuitry that includes a sequence estimation circuit and a non-linearity modeling circuit. The circuitry may be operable to receive a single-carrier signal that was generated by passage of symbols through a partial response filter and through a non-linear circuit. The circuitry may be operable to generate estimated values of the symbols using the sequence estimation circuit and using the non-linearity modeling circuit. An output of the non-linearity modeling circuit may be equal to a corresponding input of the non-linearity modeling circuit modified according to a non-linear model that approximates the non-linearity of the non-linear circuit through which the received signal passed.
1. A method comprising:
in a receiver comprising a sequence estimation circuit:
receiving a first sample of a first signal;
generating, by said sequence estimation circuit, a first symbol vector based on said first sample of said first signal;
generating a partial response feedback signal based on said first symbol vector;
receiving a second sample of said first signal;
non-linearly distorting said partial response feedback signal; and
processing said second sample of said first signal based on said non-linearly-distorted partial response feedback signal.
2. The method of claim 1 , wherein:
said first signal is non-linearly-distorted as a result of non-linearity of a signal path;
said receiver comprises a non-linearity modeling circuit; and
said method comprises determining, via said non-linearity modeling circuit, a model of said non-linearity of said signal path.
3. The method of claim 2 , wherein said model of said non-linearity of said signal path is represented as a polynomial or exponential expression.
4. The method of claim 1 , wherein symbols of said first symbol vector are N-QAM symbols and N is an integer.
5. The method of claim 1 , wherein said generating said partial response feedback signal comprises performing the following:
convolving said first symbol vector with a plurality of tap coefficients.
6. The method of claim 5 , wherein said generating said first symbol vector comprises performing the following:
generating a plurality of metrics, each one of said plurality of metrics corresponding to one of a plurality of candidate symbol vectors;
selecting one of said plurality of candidate symbol vectors based on said plurality of metrics; and
outputting said selected one of said plurality of candidate symbol vectors as said first symbol vector.
7. The method of claim 6 , wherein each of said plurality of metrics is based on a Euclidean distance.
8. The method of claim 1 , comprising non-linearly distorting said partial response feedback signal by applying said model of said non-linearity of said signal path to said partial response feedback signal.
9. The method of claim 1 , wherein said processing said second sample comprises one or both of:
equalizing; and
recovering a carrier frequency.
10. A system comprising:
a receiver comprising a sequence estimation circuit, wherein said receiver is operable to:
receive a first sample of a first signal;
generate, by said sequence estimation circuit, a first symbol vector based on said first sample of said first signal;
generate a partial response feedback signal based on said first symbol vector;
receive a second sample of said first signal;
non-linearly distort said partial response feedback signal; and
process said second sample of said first signal based on said non-linearly-distorted partial response feedback signal.
11. The system of claim 10 , wherein:
said first signal is non-linearly-distorted as a result of non-linearity of a signal path;
said receiver comprises a non-linearity modeling circuit; and
said receiver is operable to determine, via said non-linearity modeling circuit, a model of said non-linearity of said signal path.
12. The system of claim 11 , wherein said model of said non-linearity of said signal path is represented as a polynomial or exponential expression.
13. The system of claim 10 , wherein symbols of said first symbol vector are N-QAM symbols and N is an integer.
14. The system of claim 10 , wherein said generation of said partial response feedback signal comprises convolution of said first symbol vector with a plurality of tap coefficients.
15. The system of claim 14 , wherein said generation of said first symbol vector comprises:
generation of a plurality of metrics, each one of said plurality of metrics corresponding to one of a plurality of candidate symbol vectors;
selection of one of said plurality of candidate symbol vectors based on said plurality of metrics; and
output of said selected one of said plurality of candidate symbol vectors as said symbol vector.
16. The system of claim 15 , wherein each of said plurality of metrics is based on a Euclidean distance.
17. The system of claim 1 , wherein said receiver is operable to non-linearly distort said partial response feedback signal by application of said model of said non-linearity of said signal path to said partial response feedback signal.
18. The system of claim 10 , wherein said processing of said second sample comprises one or both of:
equalization; and
carrier frequency recovery.