Estimation of the signal-to-noise ratio (SNR) in communication simulation

The authors discuss the estimation of signal-to-noise ratio (SNR), with a view to its implementation in simulation. They extend previous results (see W.H. Tranter and M.D. Turner, 1977; and M.D. Turner et al., 1977) for the distribution of the SNR for an idealized channel to the linear channel. They also place the problem within a Hilbert space context, which allows a decomposition of the output into pleasing definitions of signal, noise, and distortion.<<ETX>>