I am working on EEG data compression.When Reconstructing the eeg signal
using wavelet coeficient i am using psnr(peak signal to noise ratio) to
calculate the reconstruction error.
Is there any other metric which can be used(like PDR).?
--
Thanks and Regards
Simeranjit Singh Brar
Regarding wavelets
Started by ●November 25, 2008
Reply by ●November 25, 20082008-11-25
Howdy Simeranjit,
That sounds like a really interesting project! Is there any way you can
grab some raw data and compare the full real data stream to the
reconstructed stream? A straight difference and root mean square would
give you a great base measure. Then try all the different models you have
and compare them to the base measure. Just pick the one that comes
closest to what you need.
Patience, persistence, truth,
Dr. mike
On Mon, 24 Nov 2008, simeranjit brar wrote:
> I am working on EEG data compression.When Reconstructing the eeg signal
> using wavelet coeficient i am using psnr(peak signal to noise ratio) to
> calculate the reconstruction error.
> Is there any other metric which can be used(like PDR).?
> --
> Thanks and Regards
> Simeranjit Singh Brar
>
That sounds like a really interesting project! Is there any way you can
grab some raw data and compare the full real data stream to the
reconstructed stream? A straight difference and root mean square would
give you a great base measure. Then try all the different models you have
and compare them to the base measure. Just pick the one that comes
closest to what you need.
Patience, persistence, truth,
Dr. mike
On Mon, 24 Nov 2008, simeranjit brar wrote:
> I am working on EEG data compression.When Reconstructing the eeg signal
> using wavelet coeficient i am using psnr(peak signal to noise ratio) to
> calculate the reconstruction error.
> Is there any other metric which can be used(like PDR).?
> --
> Thanks and Regards
> Simeranjit Singh Brar
>
