Side channel cryptanalysis attempts to exploit physical leakages of cipher implementations in order to extract secret key information. The dissertation at hand provides a detailed insight into new and refined side channel attacks and corresponding countermeasures. The attacks proposed in this thesis embrace internal collision attacks against the Data Encryption Standard (DES), the Advanced Encryption Standard (AES) and the block ciphers Serpent and Kasumi.
Moreover, an attack which combines the multivariate classification of side channel signals and classic Differential Power Analysis (DPA) is presented.

The blinding of intermediate, key-dependent variables in a cipher with randomized masks is the most common countermeasure against side channel attacks. In this thesis various variants of the masking countermeasure are investigated and performance figures are given. Furthermore, security aspects are examined. For example, it is shown that masked hardware implementations may not prevent side channel attacks due to glitching activities in masked circuits.