A linear programming based channel coding strong converse for the BSC and BEC
Sharu Theresa Jose, Ankur A. Kulkarni · 2017
This paper illustrates the application of the linear programming (LP) based framework proposed by the authors previously [1] in deriving improved converses for finite blocklength channel coding of a discrete memoryless binary symmetric channel (BSC) and binary erasure channel (BEC). Employing elementary concepts of optimization, finite blocklength channel coding converses for BSC and BEC are derived from first principles. The converses thus derived do not rely on information theoretic constructs like tilted information or information densities. Moreover, it is shown that the converses obtained imply the strong converse for BSC and BEC, thereby introducing a new approach for explaining the strong converse phenomenon.