Hi every one....
I wanted some information on encryption ... Even some links will help..
WHat is one time pad.. ANd which is the most secure form of encrytpion... It will be better if you can explain in detail...
Hi every one....
I wanted some information on encryption ... Even some links will help..
WHat is one time pad.. ANd which is the most secure form of encrytpion... It will be better if you can explain in detail...
the most secure form of encryption is that which hasn't been published yet
Yea i know that.. but i wanted to know what one time pad is.. And how to implement it... An algorithm would do... And i would also like to know wheather it would be good to double encrypt a cipher text using the XOR and some other encryption.. Will it make it more secure.. or will it be a loop hole in the encryption.. but most of all i am interested on how to implement the one time pad.....
a one time pad is basically a vigenere cipher which has as a key as long as its message. So its basically unbreakable. XOR encryption can be risky because of the tricky property of XORing that allows data to be recovered after it has XORed many times
I go to encounter for the millionth time the reality of experience and to forge in the smithy of my soul the uncreated conscience of my race.
Windows XP consists of 32 bit extensions and a graphical shell for a 16 bit patch to an 8 bit operating system originally coded for a 4 bit microprocessor, written by a 2 bit company, that can't stand 1 bit of competition.
But i cannot expect the user to enter a key as long as the message... The amx a user can input is a key of 8 char long.. So how do i implement it...
>>The amx a user can input is a key of 8 char long..
are you by any chance using the getpass()...
-
you really cannot implement a one time pad without the key being as long as the message. Also, the key MUST be random! I mean truly random. You cannot use anything like time() or the PID or whatever. Good sources of randomness are cosmic background radiation, or just plain noise(taken on a small scale) and then run through a strong 1way function like MD5
Ok.. But that would not be practical for me to implement..... So it is like i cannot implement ont time pad without the user entering the key as long as the message right... Ok then what is the bect encryption algorithm i can use if the message is very long and the if the key is around 4 to 10 chars long.... In detail please..
Thanx
Or Radioactive decay! http://www.fourmilab.ch/hotbits/cosmic background radiation
Handbook of Applied Cryptography - should have all the information you're looking for.
http://www.cacr.math.uwaterloo.ca/hac/
Also go to the bookstore and check out Applied Cryptography - Schneier. It's a little bit "lighter" of a read but still very good.
Plus, using a one time pad, the user doesn't input the key; it's generated randomly. Then you destroy your copy and give a copy to the reciever.
If you're interested in NBS / DES I've got some links / info.
Visit entropysink.com - It's what your PC is made for!
>So it is like i cannot implement ont time pad without the user
>entering the key as long as the message right...
Officially, yes. You can use XOR-coding with a user-key. Which is in fact one-time-pad with a too-small key.
You also could use noise. A very easy way to get noise is opening your soundcard for reading and read data from it for a while. You will get background noise then.
Or you could design a very advanced algorithm to calculate a key. Use the user's password, retrieve the serial numbers from the devices on your machine and calculate a key. Perhaps by factoring the sum in primes. Something like that.
Ok fine... Consider that i am generatinf random keys as long as the message ..with the help of the user inputed key..Then do i use the vigenere algorithm to encrypt the messgae.. Or something else....
How about using the CPU cycle or the system clock to generate rendom values...
I wouldn't use a Vigenere Algorithm unless the key you used was as long as the data to be encrypted.
Short key Vigenere Encryption is easy to break, especially when the hacker knows what kind of data is encrypted. For example, in English text (as in any other language) certain letters appear at statistically known frequencies. This makes simple Ceaser shift algorithms easy to crack, and without being an expert on encryption I can envisage ways on how to extend this to a Vigenere encryption with a repeating key.
Also, in an earlier post you suggested using a 'double' encryption. I would guess this wouldn't help you, as your encryption would only be as strong as the strongest encryption algorithm (at best). Also the more complicated you make you algorithm, but more scope you create for hidden weakness.
It would seem that we are both looking for a strong symmetric encryption technique. Please have a look at my recent post. But my guess is that we're both better off implementing a well tried (and patent free) existing algorithm. Any ideas?
Davros