Department Gathering

A conversation on Gaussian noise featuring Shannon (with a guest appearance by Nash).

Abstract

One of the key reason's for assuming a gaussian distribution for the noise in communication is due to the celebrated central limit theorem. Another, slightly less well-known reason outside of the information theory community, is that Gaussian noise can be considered as the worst noise for some coding strategies (mathematically, there is a Nash equilibrium in play here) and this was established by Shannon in his seminal work. In this talk, Prof. Chandra Nair will go over three ways to arrive at this result.

Back