Calculating entropy in many-body-simulation

Hi all,

as already implied by the title, I'd like to compute the entropy of an ensemble of particles in a many-body-simulation of an ideal gas. At each time step I get new values for the position, speed and direction of every particle.

Can anyone recommend a method for getting the instantaneous entropy from that data?

I thought about dividing the 6 dimensional phase space into small boxes and counting the number of those boxes containing particles, such that there's no double counting if two or more particles can be found in the same box. This should somehow be proportional to the phase space volume, but I don't know if there's a better method or if this is right. So if there's all particles at the same place in real space but at totally random positions in momentum space for instance, the particles are distributed randomly across the phase space on a hyperplane and I'd get an entropy value similar to one for particles that are also randomly distributed in real space and that should not be...

Thanks in advance,
Kami