A New Blind IoT-Based MP3 Audio Watermarking Scheme for Content Integrity Checking and Copyright Protection

Rapid development in the area of information and embedded technology, mobile communication networks, IoT multimedia applications, compression, and distribution over Internet has led to a significant need to answer the question: "How to protect, secure, and authenticate multimedia documents?" One of the proposed responses to this challenge is digital watermarking, which hides an inaudible watermark in digital multimedia content. Despite the large number of proposed competent watermarking algorithms in the literature, few are suitable for the compression domain, mainly the authentication application. In this context, this paper depicts a new MP3 audio watermarking scheme for copyright protection and content integrity checking operating directly in the compression domain using Huffman data and side information features. This scheme overcomes the problem of computational time as it operates directly on the compressed bitstream. In addition, it provides enough embedding space by using Huffman data features. A strong advantage of our scheme is that it has been used successfully for both authentication and copyright protection applications. Experimental results have revealed that the proposed watermarking schema has achieved very competitive results compared to others from the literature in terms of inaudibility, robustness, and especially capacity ratio. Our approach offers also good values of ODG and NC even after double recompression-StirMark attacks.