the strong converse states that the error goes to 1. Thus, is a sharp threshold between perfectly reliable and completely unreliable communication.

Finite-blocklength refinements

Finite-blocklength refinements of the noisy-channel coding theorem study the best coding rate at a fixed block length, rather than only the limiting capacity. Masahito Hayashi used the information-spectrum method to derive second-order coding-rate formulas for channel coding, including stationary memoryless channels and additive Markovian channels.

1.Previous
3.Next