Implementing the Particle Mesh Ewald method

[cyberdash]

I'm currently attempting to implement the "Particle mesh ewald" method for a molecular dynamics simulation. It has the advantage of having an algorithmic complexity of O(n log(n)), compared to a normal O(n^2) complexity for a regular approach.

I've been reading through the original papers for this ( http://dx.doi.org/10.1063/1.464397 ) But I am finding it to be quite dense, and it's difficult to translate the mathematics into actual code.

I'm wondering if anybody knows of a simple implementation that I could have a look at, even pseudocode would be helpful. If that doesn't exist, any resource explaning the actual algorithm clearly, or an explanation from yourself if you understand it, would be greatly appreciated.

Thanks!

 

[cyberdash]

Or maybe point me to the right place / person to ask?