Answer the question

In order to leave comments, you need to log in

U

U

uvelichitel2019-07-09 02:53:42

Are there general methods to distinguish an encoded message from noise?

An important indicator of the quality of coding and hashing algorithms is the uniformity of the probabilistic distribution of the output sequence. Suppose we have received data, possibly a message encoded by an arbitrary code unknown to us. Are there any reliable methods to decide whether data is meaningful or white noise? What to read about it?

Answer the question

In order to leave comments, you need to log in

2 answer(s)

@uvelichitel

@usdglander

L

longclaps, 2019-07-09 @uvelichitel

No way. Read right here.

One of the oldest encryption methods is the one-time cipher pad .

Let's take an array of random bits as it, the sender will parse his message to them and send it, the recipient will decrypt it.

L

Lander, 2019-07-09 @usdglander

1. You are confusing encryption with encryption.

2. Judging by the question, we are talking about encryption, so I will talk about it. In the case of encryption, everything depends on the algorithm itself. There are bad algorithms for which the distribution of bits (or their groups) in the ciphertext is not statistically uniform. One of the first tests for the "reliability of a cipher" (which, by the way, cannot be proved, but can only be refuted) is an attempt to compress a large ciphertext with an archiver such as Zip or Rar. If the resulting file is less than or equal in size to the ciphertext, then the cipher can immediately be discarded. This is such a simple test at home.

3. The alignment of the statistical characteristics of a cipher or hash is mainly done using the so-called avalanche effect .. The main criterion for a good avalanche effect is that changing one bit in the original message changes half the bits in the encrypted message or hash.

Similar questions

T

W

T

M

P

Didn't find what you were looking for?

Ask your questionAsk a Question

731 491 924 answers to any question