Hello, In my elektor sdr software I am experimenting a little with adaptive noise cancellers. What I do is based on the series of articles of doug smith "signals, samples and stuff: a dsp tutorial", dated back in 1998. He describes a simple and easy to implement adaptive predictor, based on the work of widrow and hoff.
The current use is in voice and music, i.e. am, usb and lsb. To me it seems that it works better in case there is a carrier (i.e. am), and less when there is no such carrier (observations based on listening). Anyone indications of whether this can be true, and what good approaches are for noise/interference cancellation in listening to ssb signals Thanks in advance jan
