A Method of Cover Audio Selection for Embedding Based on Various Criteria

¹ Faculty of Science and Technology, Universiti Sains Islam Malaysia, 71800 Nilai, Malaysia
² Cyber Security and Systems Research Unit, Faculty of Science and Technology, Universiti Sains Islam Malaysia, 71800 Nilai, Malaysia
³ Faculty of Engineering and Built Environment, Universiti Sains Islam Malaysia, Malaysia

^* Corresponding author: farida@usim.edu.my

Abstract

The main goal of an audio steganography method is to improve the capacity, im-perceptibility, and robustness. Several methods of audio steganography have been proposed to enhance its capabilities. To further optimize the efficiency of the audio steganography method, this paper proposes to select the appropriate audio cover for concealing the message. The selection of an appropriate cover can significantly improve the quality of the output and facilitate the development of innovative audio steganography techniques. This paper proposed a new audio cover selection method, which can further enhance the characteristics of the resulting output. This method ranked each cover according to the qualities that need to be boosted using various measurements. These measurements are the Maximum Available Space (MAS), Peak Signal to Noise Ratio (PSNR) and Bit Error Rate (BER) for capacity, imperceptibility, and robustness respectively. Based on the experiment conducted using five cover audios stored in the self-created database, each audio is ranked differently based on the measurement used to determine its characteristic level. In conclusion, the proposed cover selection method can be used to select the most proper cover, hence improving the characteristics of audio steganography.