Analisis Performansi Teknik Gabungan Discrete Wavelet Transform, Discrete Sine Transform, Dan Qr Decomposition Pada Audio Watermarking Stereo Dengan Compressive Sampling

Abstract

ABSTRAK Teknik watermarking perlu dikembangkan seiring kemajuan teknologi dan informasi. Audio watermarking memiliki kriteria utama meliputi imperceptibility dan robustness. Oleh karena itu, sangat penting mengetahui metode watermarking yang terbaik yang tahan terhadap berbagai serangan pemrosesan sinyal diiringi dengan kualitas audio yang baik ketika diperdengarkan kepada manusia. Penelitian ini merancang audio watermarking dengan gabungan metode Discrete Wavelet Transformation, Discrete Sine Transformation, dan QR Decomposition. Compressive Sampling digunakan untuk mengkompresi data watermark sehingga saat disisipkan kapasitas pada sinyal host meningkat. Teknik penyisipan yang digunakan adalah Quantization Index Modulation. Sistem audio watermarking yang dilakukan dalam penelitian ini menghasilkan kualitas audio dengan rata-rata SNR sebesar 26.96 dB, rata-rata BER 0.18, rata-rata ODG -1.1, dan MOS 4.2 terhadap seluruh genre audio yang diujikan. Kata Kunci: Discrete Wavelet Transformation, Discrete Sine Transformation, QR Decomposition, Compressive Sampling, Audio Watermarking ABSTRACT Watermarking techniques need to be developed along with advances in technology and information. Audio watermarking has the main criteria of imperceptibility and robustness. Therefore, it is important to know the best watermarking method that is robust to various signal processing attacks accompanied by good audio quality when played to humans.This research designs audio watermarking with a combination of Discrete Wavelet Transformation, Discrete Sine Transformation, and QR Decomposition methods. Compressive sampling is used to compress the watermark data so that during the embedding process, the capacity of the host signal increases. The insertion technique used is Quantization Index Modulation.The audio watermarking system performed in this study produces audio quality with an average SNR of 26.96 dB, average BER 0.18, average ODG -1.13, and MOS 4.2 against all audio genres tested. Keywords: Discrete Wavelet Transform, Discrete Sine Transform, QR Decomposition, Compressive Sampling, Audio Watermarking