Scientific article
OA Policy
English

Robustness Improvement of Known-Host-State Watermarking Using Host Statistics

Published inSignal processing, vol. 86, no. 5, p. 1076-1088
Publication date2006
Abstract

In this paper we considered the problem of performance improvement of known-host-state (quantization-based) watermarking methods undergo additive white Gaussian noise (AWGN) and uniform noise attacks. We question the optimality of uniform high rate quantizer based design of dither modulation and distortion compensate dither modulation methods from their robustness to these attacks point of view in terms of bit error rate probability. Motivated by superior performance of uniform deadzone quantizer over the uniform one in lossy source coding, we propose to replace the latter one by a former one designed according to the statistics of the host data. Based on suggested modifications we obtained analytical expressions for bit error rate probability analysis of quantization-based watermarking methods in AWGN and uniform noise channels. Experimental results of computer simulations demonstrated performance enhancement of known-host-state watermarking techniques in comparison to the classically elaborated schemes at negative watermark-to-noise ratios.

Keywords
  • Watermarking
  • Information hiding
  • Known-host-state watermarking
  • Quantization-based watermarking
  • High rate quantization
  • Uniform quantizer
  • Uniform deadzone quantizer
  • Bit error rate probability
Citation (ISO format)
KOVAL, Oleksiy et al. Robustness Improvement of Known-Host-State Watermarking Using Host Statistics. In: Signal processing, 2006, vol. 86, n° 5, p. 1076–1088. doi: 10.1016/j.sigpro.2005.07.024
Main files (1)
Article (Published version)
accessLevelPublic
Identifiers
Journal ISSN0165-1684
539views
594downloads

Technical informations

Creation03/03/2015 4:36:13 PM
First validation03/03/2015 4:36:13 PM
Update time03/14/2023 10:56:59 PM
Status update03/14/2023 10:56:59 PM
Last indexation10/30/2024 11:11:56 PM
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack