In audio watermarking, the robustness against pitch-scaling attack, is one of the most challenging problems. In this paper, we propose an algorithm, based on traditional time-spread(TS) echo hiding based audio watermarking to solve this problem. In TS echo hiding based watermarking, pitch-scaling attack shifts the location of pseudonoise (PN) sequence which appears in the cepstrum domain. Thus, position of the peak, which occurs after correlating with PN-sequence changes by an un-known amount and that causes the error. In the proposed scheme, we replace PN-sequence with unit-sample sequence and modify the decoding algorithm in such a way it will not depend on a particular point in cepstrum domain for extraction of watermark. Moreover proposed algorithm is applied to stereo audio signals to further improve the robustness. Experimental results illustrate the effectiveness of the proposed algorithm against pitch-scaling attacks compared to existing methods. In addition to that proposed algorithm also gives better robustness against other conventional signal processing attacks.
History
Event
Security of Information and Networks. Conference (4th : 2011 : Sydney, N. S. W.)
Series
ACM International Conference Proceeding Series
Pagination
37 - 42
Publisher
ACM
Location
Sydney, N. S. W.
Place of publication
New York, N. Y.
Start date
2011-11-14
End date
2011-11-19
ISBN-13
9781450310208
Language
eng
Publication classification
E1 Full written paper - refereed
Copyright notice
2011, ACM
Title of proceedings
SIN 2011 : Proceedings of the 4th international conference on Security of information and networks