Real-time feedback watermarking authentication scheme for streaming media

Abstract

Recently, with the improvement of attack technology and computational capability, how to meet the security application of streaming media data has become a significant issue, especially for systems with a high standard of security. This paper presents a real-time authentication scheme for P2P streaming media transmission system, which can at least meet the application of identity authentication, tamper location, and error retransmission. We first tailor and employ off-the-shelf algorithms to propose two types of watermarking algorithms (the Feature Watermarker and the Identity Watermarker). Then, the sliding window is employed to group and provision the above two types of watermarks in the overall authentication scheme. Several metrics are employed to indicate our watermarking meets the application of invisibility and has a very low false alarm rate and missing alarm rate after computing and choosing viable thresholds. Moreover, we sketch an “Attack List” aiming at detecting and deal with malicious behaviors effectively, including two types of attacks which is irresistible for the common watermarking algorithms (Re-embedded attack and Destruction of watermark). The delay in transmission turns out to be acceptable by reasonably tailoring crucial parameters.