MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

We have 2 different sequences of real numbers, say s1, s2 obtained in the following fashion. We tie two unit point masses with a spring. Next we let the entire sping-mass system into one dimesional free space (zero gravity). However the space has spatial particles which can collide with either of the point masses and imapart their to them.

Sequence s1 is obtained by sampling the the location of the first point mass and s2 by the second.

Is it possible to distinguish arbitrary sequences from sequences which have been obtained in this fashion?

share|cite|improve this question
don't conservation of momentum and energy impose some constraints (such as boundedness for a start)? – Yemon Choi Dec 27 '09 at 15:29
If I'm correctly imagining the system, s1 will always be on the same side of s2 and the length of the spring will impose a maximum distance between s1 and s2. Speeds won't reach the speed of light, so that's a rough upper bound on (change in position)/(change in time) for each of s1 and s2. I have no idea how many particles will collide with s1 and s2 and with what energies between 2 samplings, so beyond these restriction it seems anything could happen. – Jonas Meyer Dec 27 '09 at 19:15

First off, your question is impossible to answer without more conditions on the initial distribution of momenta of the other particles (for instance they could be carefully arranged so that you can get any sequence of measurements that you wish for). A reasonable one is that the smaller particles are in equilibrium at some temperature T. With this assumption you can now apply the well-developed principles of statistical mechanics.

You can easily write down a Langevin equation for your ball-spring system. This can then be solved with the Fokker-Planck formalism to your system to figure out the probability distribution of the positions as a function of time as the ball-spring system equilibrates with the bath of particles. Of course here you'll have to put in initial conditions (if precisely well-defined, the initial distributions will be delta-function-like) for your ball and spring as well, as that will affect the approach to equilibrium.

Qualitatively, at very long times, you will get diffusion of the center of mass following the Einstein relation and from this, the positions of the two balls can be worked out as well.

I can try to add more details if you have specific questions. Or you can look for books on Fokker-Planck or "approach to equilibrium" in general (I learned the stuff from courses a while back, so I don't have a good reference at hand, sorry).

I can't really answer your question directly unless I know more about what "arbitrary sequences" you have in mind. Also, I've framed this answer from the point of view that you can look at a lot of such sequences and average over them to look at their distribution. If you're just given one such sequence, as far as I know you can't say very much.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.