Justification is requested. please help me resolve this problem. Phd and Prof
-
It is really needed me to proof my little cryptographic scheme because of i have used an own created cryptographic scheme to make secure for the mobile data transaction in a database system, and our lectures, who are professional on this cryptography, said "If it is used any created cryptographic algorithm to make secure for something then it would be done for proper cryptanalysis to prove that the algorithm is not hackable and to proof strength of algorithm first then it is possible to say as secure model. Otherwise it is not considered as secure. I have heard about that cryptographic scheme is built after the mathematic justification is done but it is already done for programing implementation before the mathematic justification on the scheme/algorithm is done because of that time i had no idea about that all. If it is needed to rebuilt the scheme/algorithm again, it really hard for me to restart new one. Cryptographic algorithm metric literature (1997), I've found, suggested about a few important metrics such as type, functions, key size, rounds, complexity, attack and strength but If it is new modern metric came out ready, i have not found it yet. It is really needed justification on the proper cryptanalysis to prove that the scheme/algorithm is not hackable and to proof strength of algorithm as well General detail of the algorithm follows as listed below. Type: Symmetric Functions: Confidentiality and authentication Key size: 256 bit Round: 3 Complexity: encryption, decryption and key setup include the expansion, bit operations, modular multiplications and modular exponentiations. Attack: have not started to test it yet. Strength: not started. Additional info encryption and decryption schemes work on numbers only because of data transaction in the database system uses numbers (not text and any other symbols) Encryption Input: 32 bits Output: 128 bits what need to do is better? If Any body, who are Phd, Prof and experts on the cryptography and were doing this before, please advice me? Thank you
-
It is really needed me to proof my little cryptographic scheme because of i have used an own created cryptographic scheme to make secure for the mobile data transaction in a database system, and our lectures, who are professional on this cryptography, said "If it is used any created cryptographic algorithm to make secure for something then it would be done for proper cryptanalysis to prove that the algorithm is not hackable and to proof strength of algorithm first then it is possible to say as secure model. Otherwise it is not considered as secure. I have heard about that cryptographic scheme is built after the mathematic justification is done but it is already done for programing implementation before the mathematic justification on the scheme/algorithm is done because of that time i had no idea about that all. If it is needed to rebuilt the scheme/algorithm again, it really hard for me to restart new one. Cryptographic algorithm metric literature (1997), I've found, suggested about a few important metrics such as type, functions, key size, rounds, complexity, attack and strength but If it is new modern metric came out ready, i have not found it yet. It is really needed justification on the proper cryptanalysis to prove that the scheme/algorithm is not hackable and to proof strength of algorithm as well General detail of the algorithm follows as listed below. Type: Symmetric Functions: Confidentiality and authentication Key size: 256 bit Round: 3 Complexity: encryption, decryption and key setup include the expansion, bit operations, modular multiplications and modular exponentiations. Attack: have not started to test it yet. Strength: not started. Additional info encryption and decryption schemes work on numbers only because of data transaction in the database system uses numbers (not text and any other symbols) Encryption Input: 32 bits Output: 128 bits what need to do is better? If Any body, who are Phd, Prof and experts on the cryptography and were doing this before, please advice me? Thank you
Probably easier to prove its broken. Encrypt some plaintext with your scheme and then feed the results into the NIST random number test suite. If your results aren't random then the scheme is probably weak.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding. -
Probably easier to prove its broken. Encrypt some plaintext with your scheme and then feed the results into the NIST random number test suite. If your results aren't random then the scheme is probably weak.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding.A little bit clarify about NIST random test suite please? If the scheme used it for generator as a random generator function in class of java programming language then what need to do? It is already proven random generator?
-
A little bit clarify about NIST random test suite please? If the scheme used it for generator as a random generator function in class of java programming language then what need to do? It is already proven random generator?
Take some text, like an ebook or something, and encrypt it with your scheme. If the encryption is strong, then the encrypted text should look random and come close to passing the NIST test suite. If your encryption scheme is weak, it will leave entropy in the output, which the NIST test may pick up on.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding. -
Probably easier to prove its broken. Encrypt some plaintext with your scheme and then feed the results into the NIST random number test suite. If your results aren't random then the scheme is probably weak.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding.Probably easier to prove its broken. Encrypt some plaintext with your scheme and then feed the results into the NIST random number test suite. If your results aren't random then the scheme is probably weak. A consistent failure to pass randomness tests would imply cryptographic weaknesses, which would likely be exploitable. On the other hand, it's possible for a scheme to pass any randomness test one might care to throw at it but nonetheless be cryptographically very weak. Given the availability of strong cryptographic methods, I don't see much point to using weak ones unless one is operating under severe hardware constraints. Of course, it may well be that the application requirements don't quite match any existing security protocol. If that is the case, I would suggest building a protocol using existing encryption algorithms. Under that scenario, you may be able to prove that the only way to crack the protocol is to crack one of the underlying algorithms. Developing a secure protocol will often be tricky, even given secure algorithms, but with care you may be able to produce a protocol that can be shown to be as secure as the algorithms at its core.
-
Take some text, like an ebook or something, and encrypt it with your scheme. If the encryption is strong, then the encrypted text should look random and come close to passing the NIST test suite. If your encryption scheme is weak, it will leave entropy in the output, which the NIST test may pick up on.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding.If the encryption scheme works on only numbers, especially limited inside 32 bits input, then how to test it properly?
-
If the encryption scheme works on only numbers, especially limited inside 32 bits input, then how to test it properly?
Then you encode the data. Pack 4 8bit ascii characters into each word. Normally you'd compress data before encryption too, but in this case its important that you don't. If your algorithm takes non-random data, encrypts it, and the result doesn't look random (it has things like high autocorrelation), then the algorithm is most likely seriously broken.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding. -
Then you encode the data. Pack 4 8bit ascii characters into each word. Normally you'd compress data before encryption too, but in this case its important that you don't. If your algorithm takes non-random data, encrypts it, and the result doesn't look random (it has things like high autocorrelation), then the algorithm is most likely seriously broken.
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding.What you said over can be satisfied as proper test?
-
What you said over can be satisfied as proper test?
No, not at all. If this test fails, then its a very strong sign that your algorithm is weak. If this test passes then it definitely doesnt prove the algorithm in any way...
Mark Churchill Director, Dunn & Churchill Pty Ltd Free Download: Diamond Binding: The simple, powerful, reliable, and effective data layer toolkit for Visual Studio.
Entanglar: .Net game engine featuring automatic networking and powerful HLSL gfx binding.