The reassignment method warrants its own page, particularly when it contains this much information. However, I'd like to see more references to the early material and a description of why this method sucks (no formal description of the result of reassignment, no amplitude representation, plagued by numerical error) and why anyone would want to reassign a TFR explicitly rather than use a phase-based method of signal-component extraction (e.g. Carmona's). Fitz, can you delete all but one of your references? I don't think we need that many but if anyone then claims this is some obscure technique I'll back you up. Jon Harrop 12:01, 16 April 2007 (UTC)