Chaotic encryption of speech signals in transform domains

A speech encryption system based on substitution and permutation is described. The stream of speech segments is divided and reshaped into two fixed size blocks and the elements are permutated by chaotic map technique, then substituted to different values by discrete cosine transform (DCT) and then permutated another time to eliminate attacks with known plaintext attack. Finally, chaotic speech signals that completely different from original signal is generated. The proposed low complexity system will avoid the delay resulting from the use of simple substitution and fast permutation process. the target of the proposed system is to change the characteristics of speech signal and eliminate the intelligibility of the encrypted signal. The security aspects of the proposed system will depend on different parameters including secret blocks size, permutation process and variable number of rounds.