Lots of possible approaches. If you have a good approx. of the amplitude, you should be able to clear a linear least squares param est
-
-
Replying to @EmilyGorcenski @zip
iow, if you have y = A*sin(kx+b), and a good estimate for A, then perform LLS on kx_i+b = asin(y_i/A) for some N samples.
1 reply 0 retweets 0 likes -
Replying to @EmilyGorcenski @zip
Basically solve the system M [k b]^T = Y using standard LLS algorithm, [k b]^T = inv(M^T M) y, but can do better than computing inv.
1 reply 0 retweets 0 likes -
Replying to @EmilyGorcenski @zip
That solution will essentially be optimal if the noise characteristic is Gaussian and if the algorithm is fast enough.
1 reply 0 retweets 0 likes -
Replying to @zip
QAM in practice needs to be faster, so those algorithms generally operate directly in freq domain, less familiar with those
1 reply 0 retweets 0 likes -
Replying to @zip
Yeah, signal processing generally throws a lot of prerequisite crap at you, even though ultimately the implementation math is simple
1 reply 0 retweets 0 likes
No good way that I know of offhand
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.