What is the information rate penalty for perfect transmission (i.e., without wrong symbol transmission)?
Considering the Markov process of transmission for the ARQ (Automatic Repeat reQuest) system example we compute T; the expected number of times one of the n–ordered symbols can be transmitted. This gives us the percentage increase/decrease in the number of transmission.
The product of T and Information rates is the effective information rate. This answers the question–How often, on the average must a n–ordered symbol be transmitted (to successfully deliver the information)? This is the information rate penalty.
In our example for transmitting , . The inequality or extra information is due to noise.
Though noisy information is regarded useless by a communication engineer they are useful for a cryptographer. Some kind of coding is necessary for the inequality
Description of a basic cryptography system.
Error rate is the measure for characterizing modal nature of information transmission.