Hacker News
new
past
comments
ask
show
jobs
points
by
slow_typist
12 hours ago
|
comments
by
uecker
12 hours ago
|
next
[-]
The simplest way to show this is that the adjoint (complex transpose) DFT matrix is the inverse, so U^H U = I.
(For DFT there are different scaling conventions which may be taken into account.)
reply
by
slow_typist
9 hours ago
|
parent
|
[-]
Maybe not the simplest way but constructive proofs are always nice.
reply
by
eliben
6 hours ago
|
prev
|
[-]
Yes! It's discussed in the "Inverse DFT: reconstructing the signal" section, which relies on the earlier post
https://eli.thegreenplace.net/2026/notes-on-the-complex-root...
reply