Adaptive signal processing

1. An apparatus for adaptive signal processing, the apparatus comprising:

an input terminal for receiving an input signal;

an output terminal;

an adaptive filter connected to receive the input signal from the input terminal and configured to provide to said output terminal an adaptive filter output with a filter delay D A ;

a reference filter corresponding to the adaptive filter, connected to receive the input signal from the input terminal and configured to provide a reference filter output with a filter delay D R which is greater than D A ;

a delay unit for delaying the adaptive filter output by a delay corresponding to D R -D A ;

and an adaption unit configured to measure the error between the delayed adaptive filter output and the reference filter output and to adapt the adaptive filter in response to the error, wherein the adaptive filter and the reference filter are each linear filters, at least instantaneously.

2. The apparatus according to claim 1 in which both the adaptive filter and the reference filter comprise low pass filters.

3. The apparatus according to claim 1 in which the error is a root mean square error.

4. The apparatus according to claim 1 in which the adaptation is performed using an LMS algorithm.

5. The apparatus according to claim 1 in which the reference filter comprises a fixed symmetric filter.

6. A method of signal processing to transform an input signal in a signal processor to a processed signal, the method comprising: conducting an adaptive process on an input signal to form the processed signal, said adaptive process having controllable parameters; conducting in parallel a corresponding reference process on said input signal; measuring an error between a delayed version of the adaptive process output and the reference process output; and adapting the adaptive process by controlling said parameters to reduce said error, wherein the adaptive process and the reference process are each linear, at least instantaneously.

7. The method according to claim 6 , wherein the delay of the adaptive process output serves to bring the adaptive process output into time correspondence with the reference process output.

8. The method according to claim 6 , wherein said step of adapting the adaptive process serves to minimize said error over a time period not less than said delay.

9. The method according to claim 6 in which the input signal is a sampled signal.

10. The method according to claim 6 in which the adaptive process is a filtering process.

11. The method according to claim 6 in which the input signal describes the position or size of an adaptive window within a video frame.

12. The method according to claim 6 in which the adaptive and the reference processes are the output of the same optimisation procedure operating on the same input conditions, save for respectively different latency constraints.

13. A method of low pass filtering an input signal, the method comprising:

conducting an adaptive low pass filtering process on an input, said process having controllable filter parameters and a filter delay D A ; conducting in parallel a reference low pass filtering process on said input signal, said reference process having a filter delay D R which is greater than D A ; measuring an error between a delayed version of the adaptive process output and the reference process output; and adapting the adaptive process by controlling said filter parameters to reduce said error.

14. The method according to claim 13 in which the error measure is a root mean square error and in which the adaptation is performed using an LMS algorithm.

15. The method according to claim 13 in which the reference process is a fixed filter.

16. The method according to claim 15 in which the fixed filter is symmetric.

17. The method according to claim 13 in which the input signal describes the position or size of an adaptive window within a video frame.