caa a22 4500 605447071 CHVBK 20210128100130.0 cr unu---uuuuu 210128e20151101xx s 000 0 eng 10.1007/s11042-014-2147-3 doi (NATIONALLICENCE)springer-10.1007/s11042-014-2147-3 Security in the speech cryptosystem based on blind sources separation [Elektronische Daten] [Ali Sadr, Raziyeh Okhovat] In this paper, an appropriate selection for the key is investigated to enhance the security level of the speech cryptosystem based on blind sources separation. In fact, if an appropriate key is not selected, the cryptosystem may be attacked and therefore each confidential signal can be recovered only from the encrypted ones. Two important conditions which are studied in this paper are the number of quantized levels for a digital key and also non-sparsity in time-frequency domain. In the case of the first condition, simulation results show that with smaller coefficient for the confidential signal in comparison with the key, the number of quantized levels for the key should be more to guarantee the security. In the case of the second condition, an algorithm is proposed to recover the confidential signal only from the encrypted signal when the key is sparse in time-frequency domain. Springer Science+Business Media New York, 2014 Blind sources separation (BSS) nationallicence Cryptography nationallicence Cryptanalysis nationallicence Quantized levels nationallicence Sparsity nationallicence Time-frequency domain nationallicence Sadr Ali Iran University of Science and Technology, Tehran, Iran aut Okhovat Raziyeh Iran University of Science and Technology, Tehran, Iran aut Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/21(2015-11-01), 9715-9728 1380-7501 74:21<9715 2015 74 11042 https://doi.org/10.1007/s11042-014-2147-3 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s11042-014-2147-3 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Sadr Ali Iran University of Science and Technology, Tehran, Iran aut NATIONALLICENCE 700 1- Okhovat Raziyeh Iran University of Science and Technology, Tehran, Iran aut NATIONALLICENCE 773 0- Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/21(2015-11-01), 9715-9728 1380-7501 74:21<9715 2015 74 11042