Interference cancellation system
View Patent ↗An adaptive interference cancellation system is described. In one example the system operates by receiving a data signal using a DSL (Digital Subscriber Line) and receiving a reference signal, the reference signal corresponding, in part, to noise on the data signal. The reference signal is classified and a noise cancellation signal is applied to the data signal based on the classification.
1. A method comprising:
receiving a data signal including noise using a DSL (Digital Subscriber Line);
receiving a reference signal, the reference signal corresponding, in part, to noise on the data signal;
classifying the noise of the reference signal by estimating properties of the reference signal and comparing the estimated properties to a known set of properties;
selecting a noise cancellation signal based on the classifying; and
applying the selected noise cancellation signal to the data signal.
2. The method of claim 1 , wherein receiving a data signal comprises receiving the data signal on a telephone line within a binder and wherein receiving a reference signal comprises receiving a signal on a second telephone line that is within the same binder as the first telephone line.
3. The method of claim 2 , wherein the second telephone line carries analog voice and wherein receiving the reference signal comprises filtering out the analog voice.
4. The method of claim 2 , wherein the second telephone line carries DSL data using multiple tones and wherein receiving the reference signal comprises receiving the reference signal on tones that are not used to carry DSL data.
5. The method of claim 1 , wherein receiving a data signal comprises receiving the data signal on a telephone line within a binder and wherein receiving a reference signal comprises receiving a signal on a second telephone line that is within the same binder as the first telephone line.
6. The method of claim 1 further comprising measuring the strength of the reference signal; and
if the strength is low, then not applying the noise cancellation to the data signal.
7. The method of claim 1 wherein classifying the reference signal comprises estimating statistical properties of the reference signal and comparing the statistical properties to a set of known properties.
8. The method of claim 7 , wherein selecting a noise cancellation signal comprises selecting an entry in a classification table based on the estimated statistical properties and using a corresponding coefficient from the classification table as the noise cancellation signal.
9. The method of claim 8 , wherein applying the selected noise cancellation signal comprises subtracting the noise cancellation signal from the data signal.
10. The method of claim 1 wherein selecting a noise cancellation signal comprises selecting a coefficient from a plurality of candidate coefficients and producing a noise cancellation signal using the coefficient.
11. The method of claim 10 , further comprising determining an error value for the data signal after applying the noise cancellation signal and updating the corresponding coefficient by using the determined error value.
12. The method of claim 1 wherein applying the noise cancellation signal comprises applying frequency domain noise cancellation signal to a data signal that is in the frequency domain.
13. A non-transitory machine-readable medium storing thereon instructions that when executed by the machine cause the machine to perform operations comprising:
receiving a data signal including noise using a DSL (Digital Subscriber Line);
receiving a reference signal, the reference signal corresponding, in part, to noise on the data signal;
classifying the noise of the reference signal by estimating properties of the reference signal and comparing the estimated properties to a known set of properties;
selecting a noise cancellation signal based on the classifying classification; and
applying the selected noise cancellation signal to the data signal.
14. An apparatus comprising:
a data receive chain to receive a data signal including noise through a DSL (Digital Subscriber Line);
a reference signal receive chain to receive a reference signal corresponding, in part, to noise on the data signal;
a noise classification module to characterize the noise of the reference signal by estimating properties of the reference signal and comparing the estimated properties to a known set of properties and to select a noise cancellation signal based on the classification; and
a noise filter to apply the selected a noise cancellation signal to the data signal.
15. The apparatus of claim 14 , wherein the noise classification module selects a subtraction coefficient based on the classification of the noise signal and combines the coefficient with the noise signal and wherein the noise filter comprises a subtractor to apply the combined coefficient and noise signal to the data signal.
16. The method of claim 6 , wherein measuring the strength comprises comparing the power of the reference signal to a threshold.
17. The method of claim 7 , wherein estimating statistical properties comprises estimating the power spectral density.
18. The method of claim 11 , wherein updating the corresponding coefficient comprises applying the determined error value to a least means square estimation.
19. The method of claim 10 wherein producing the noise cancellation signal comprises multiplying the selected coefficient by the reference signal.