My apologies, I didn't quite understand your initial post. After reading that paper it makes sense now.
However, information is still being transferred using your definition of speed. You say that:
Then this would be the equivalent of marking a point on the coin at time t = 0 when it is at contact with point P (as from the article). You then measure the time interval taken for the marked point to return to point P, and determine the speed of rotation from that. Is this what you mean, or am I getting the wrong end of the stick?
If that is the case, then clearly this speed cannot exceed c. I can measure that point's position as a function of time, its momentum etc. All of this is information and would violate SR if it was travelling faster than c.
The paper you quote uses non-relativistic mechanics, and rightly so. There might be a singularity but as they say various resistive forces prevent it ever becoming a relativistic problem. If you were able to magically create a system without the resistive forces in question then you would still not be able to exceed c, unfortunately I do not have the time to mathematically prove it to you.