Talk:Wigner distribution function

autocorrelation

If it is indeed an autocorrelation that is meant, then it is not a time-reversed copy. Auto correlation uses the cross-correlation (shifted), not the convolution (time-reversed, shifted). The current article text essentially suggests one operation, but explains the other. I am not so experienced with integral transforms, but it appears to indeed be a time-reversed, shifted copy based on the definition given. That means that is more like an autoconvolution. This term exists, but doesn't have a Wikipedia page. Any other mathematicians care to weigh in?--75.80.43.80 (talk) 23:39, 18 April 2011 (UTC)[reply]

I do not see how it can be that "The WDF is essentially the Fourier transform of the input signal’s autocorrelation function". The link to autocorrelation shows that the autocorrelaiton itself is an integral. None of it appears in the definition of the WDF. What about removing this sentence? — Preceding unsigned comment added by 130.246.132.178 (talk) 14:34, 12 September 2013 (UTC)[reply]

I fixed this part of the article. Lots of stuff could be added to this article, like references to Ville and sections on negativity and the uncertainty principle. But I don't agree with merging this with the QM article, because the wave function and a time series are two very different things with very different interpretations... even though there will necessarily be many redundant/analogous relations. CHF (talk) 18:45, 19 September 2013 (UTC)[reply]