# AES (Advanced Encryption Standard) Overview Part I

The **AES (Advanced Encryption Standard) algorithm** [1], known as Rijndael, was so-named by its creators Joan Daemen and Vincent Rijmen, it is the current **international standard for communications encryption since october 2000**. This algorithm is characterized by a **symmetric block cipher with variable key length**, the default key length is \(128\) bits but can also be set to \(192\) or \(256\) bits.

The operation of the algorithm can be split into **two parts or different processes, being the encryption process, the first one and the second corresponds to the sub-key generation process**, Figure #1 shows the interaction of both processes.

The block to be encrypt has a length of \(128\) bits, while the key can vary from \(128\), \(192\) or \(256\) bits, depending on the amount of standard rounds that apply to text \(10\), \(12\) and \(14\) respectively [2]. Generally, the AES encryption is described by four basic features or **so-called transformations**, these are:

**ByteSub****ShiftRow****MixColumns****AddRoundKey**

The rounds, in which the text is subjected, can be divided into three categories: **initial round, standard rounds and final round**. These rounds consist in one or various combinations of the above transformations (see Figure #1). The different types of rounds differ by the combination of the transformations that are applied to the block to be encrypted.

AES interprets the input block of 128 bits, as a \(4x4\) matrix with input of bytes [3], if the block is \(192\) bits \(2\) columns are added, if four columns \(256\) are added, theses matrices are called **state matrices** and their shape is shown below:

## S-Box

Essentially, **an S-Box (substitution box) is a matrix of substitution values used by algorithms symmetric-key cryptography**. The S-Boxes are carefully selected to be resistant to cryptanalysis. For AES, S-Box is the result of applying two functions to the state matrix, \([a_{ij}]\), firstly its multiplicative inverse \([a_{ij}]\) to \(a_{ij}^{-1} \in GF(2^8)\) and then a linear transformation:

- Any byte can be seen as an element of the finite field \(GF(2^8)\), as every element has a multiplicative inverse, multiplicative inverse is associated to \(GF(2^8)\), i.e. \([a_{ij}]\) to \(a_{ij}^{-1} \in GF(2^8)\), the zero element is associated with the same zero.
- The linear transformation is applied bit by bit with the following rule: linear transformation in \(GF(2^8) \rightarrow GF(2^8)\).

in bits is:

\[\begin{matrix} b'_0 = b_0 \oplus b_4 \oplus b_5 \oplus b_6 \oplus b_7 \oplus c_0 \\ b'_1 = b_1 \oplus b_5 \oplus b_6 \oplus b_7 \oplus b_0 \oplus c_1 \\ b'_2 = b_2 \oplus b_6 \oplus b_7 \oplus b_0 \oplus b_1 \oplus c_2 \\ b'_3 = b_3 \oplus b_7 \oplus b_0 \oplus b_1 \oplus b_2 \oplus c_3 \\ b'_4 = b_4 \oplus b_0 \oplus b_1 \oplus b_2 \oplus b_3 \oplus c_4 \\ b'_5 = b_5 \oplus b_1 \oplus b_2 \oplus b_3 \oplus b_4 \oplus c_5 \\ b'_6 = b_6 \oplus b_2 \oplus b_3 \oplus b_4 \oplus b_5 \oplus c_6 \\ b'_7 = b_7 \oplus b_3 \oplus b_4 \oplus b_5 \oplus b_6 \oplus c_7 \\ \end{matrix}\]**This is the most expensive operation in terms of time for the AES algorithm**, therefore, this operation is precalculated. Finally, this process can be summarized in the following table, known as S-boxes [3] which can be used for any \(xy\) byte.

Furthermore, for the decryption process an inverse table to the one used in the encryption process is calculated, said table is:

## References

- Joan Daemen VR. AES Proposal: Rijndael. NIST AES Proposal (1998)
- J A. AES - Advanced Encryption Standard. (2005) Versión 2005
- A M. Seguridad Europea para EEUU Algoritmo criptográfico Rijndael. Madrid (2004)
- http://www.formaestudio.com/rijndaelinspector/
- http://www.cryptosystem.net/aes/
- http://www.criptored.upm.es/
- http://www.kriptopolis.es/

## Enjoy Reading This Article?

Here are some more articles you might like to read next: