| dc.description.abstract | Information hiding was born in the early 20th century which was an emerging cross-discipline in many technical fields, such as signal processing, information security, communication, and human auditory system theory. It is mainly divided into two major branches: digital watermarking and steganography. Different from the general traditional encrypted information security method, the information hidden type realizes secure communication transmission by hiding the existence of secret messages. As the time when large amounts of information are widely available to many people, largely through computer technology, information hiding acts as one of the means to protect intellectual property rights, secret communications has been increasingly studied and ap-plied in many fields.
This paper proposes a method of audio hiding by hiding information in the frequency domain via short-time Fourier Transform (SFTF). The carrier audio was using MATLAB applied STFT convert into the frequency domain. In the frequency terms which were inaudible, the signal is modulated according to the embedded secret message, and then generated by inverse short-time Fourier Transform (ISFTF) converted into secret audio. After the secret audio is broadcasted by the playback device, it is recorded by the radio device and then short-time Fourier transform can be used to reversely analyze the hidden secret message.
The subsequent papers applied to the SFTF’s window function coefficients of the audio hiding system and verified that the secret message embedding amount of the system was up to 17.24bit/s. Experiments have also confirmed that the system can interpret secret messages normally at a forward radio distance of 90cm, and has excellent robustness when the signal-noise ratio (SNR) is higher than 5dB. In the end, this audio message hiding system is integrated into a graphical user interface (GUI) application with MATLAB′s App Designer function. | zh_TW |