Prev: PNET UDP QUESTION
Next: calculating 95-percentile for a non-Gaussian distribution
From: Mark Shore on 5 May 2010 11:08
The Cesaro sums noted in the first reference are more commonly referred to as Fejer sums. The Lanczos sigma factors seem to be more effective, and relatively easily implemented.

The FEX submission from Howard Wilson in my first post will probably be a good starting point for 1D Fourier series as it is appears both well-written and adequately commented.
From: Merciadri Luca on 5 May 2010 17:11
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Mark Shore" <mshore(a)magmageosciences.ca> writes:

> The Cesaro sums noted in the first reference are more commonly
> referred to as Fejer sums.
That gives more results than Cesaro sums, at least on Google.

> The Lanczos sigma factors seem to be more effective, and relatively
> easily implemented.
As you say.

> The FEX submission from Howard Wilson in my first post will probably
> be a good starting point for 1D Fourier series as it is appears both
> well-written and adequately commented.
Nice to hear it. :-)

- --
Merciadri Luca
See http://www.student.montefiore.ulg.ac.be/~merciadri/
- --

Our greatest glory is not in never falling, but in rising every time
we fall. (Conficius)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iEYEARECAAYFAkvh3ukACgkQM0LLzLt8MhxnRQCfR6a2FN2hg1rYbzGgJCbPKS26
bzwAoJnvqOeMtf0mzMbMcWxDEhR0QnhV
=9EvR
-----END PGP SIGNATURE-----
 | 
Pages: 1
Prev: PNET UDP QUESTION
Next: calculating 95-percentile for a non-Gaussian distribution