Exploiting Silence for Ciphertext Only Cryptanalysis of Stream Ciphered Digitized Voice

摘要

Speech is digitized, encrypted and sent between two parties in many situations. We present techniques to exploit the characteristics of compressed as well as uncompressed digitized speech for ciphertext only cryptanalysis of LFSR based stream ciphers, both for single channel as well as time division multiplexed communication systems. This technique is unique in the sense that it works even if the plain text is not statistically biased, in which all other ciphertext only attacks fail. Criteria for resisting such types of attacks are presented separately for single and multiple channel communication systems.