This is my bare bones impmementation of the AES-128 (aka Rijndael) encryption algorithm. Insert your usual "don't use in production environment" warning here, as this project always was, from the beginning, made from a recreational and educational point of view.
This Project was built using gcc version 10.1.0 on my Linux system. Although any C compiler with C11 support may be able to compile this project
make clean
make run
gcc cipher.c -c cipher.o
gcc main.c -c main.o
gcc cipher.o main.o -o aes.exe
./aes.exe (on Unix based machines)
OR aes.exe(On Windows machines)
This is a very bare bones implementation of the AES-128. I say that because there are many things that are missing from this utility:
- Variable size password (not key; key will always be 16 bytes long)
- Variable size data (currently supports the standard single 16 byte block data)
- Seperate Encipher and Decipher modes (currently just enciphers, outputs, deciphers and outputs)
- bit optimizations are heavily missing from the basic operations of AES
1 may be achieved by using a hashing functions that creates a 128 bit digest (like MD5 or something)
2 may be achieved by calculating the number of blocks of data, encrypting individual blocks and appending them to a buffer, and then outputting the buffer in the end
3 may be achieved by making a seperate decipher function inside cipher.c and putting all the decipher functions inside it
4 is not really clear to me, but I feel that there may be many places where I have slacked very much, and optimizations may be possible :)
- https://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.197.pdf (official AES-128 specification from FIPS)
- https://web.stanford.edu/class/ee392d/Chap7.pdf (to read about finite fields)
- https://crypto.stackexchange.com/questions/12873/aes-mixcolumn-stage?noredirect=1&lq=1 (in order to get some insigts about MixColumns())
- https://crypto.stackexchange.com/questions/63158/regular-and-efficient-doubling-in-gf2n?noredirect=1&lq=1 (some more insigts on FF multiplication)