Universiti Teknologi Malaysia Institutional Repository

MP3 audio steganography technique using extended least significant bit

Mohammed Atoum, Mohammed Salem (2014) MP3 audio steganography technique using extended least significant bit. PhD thesis, Universiti Teknologi Malaysia, Faculty of Computing.

[img]
Preview
PDF
514kB

Official URL: http://dms.library.utm.my:8080/vital/access/manage...

Abstract

Audio Steganography is the process of concealing secret messages into audio file. The goal for using audio steganography is to avoid drawing suspicion to the transmission of the secret message. Prior research studies have indicated that the main properties in steganography technique are imperceptibility, robustness and capacity. MP3 file is a popular audio media, which provides different compression rate and performing steganography in MP3 format after compression is the most desirable one. To date, there is not much research work that embeds messages after compression. An audio steganographic technique that utilizes Standard Least Significant Bits (SLSB) of the audio stream to embed secret message has gained popularity over the years. Unfortunately the technique suffers from imperceptibility, security and capacity. This research offers an extended Least Significant Bit (XLSB) technique in order to circumvent the weakness. The secret message is scrambled before embedding. Scrambling technique is introduced in two steps; partitioning the secret message (speech) into blocks followed by block permutation, in order to confuse the contents of the secret message. To enhance difficulty for attackers to retrieve the secret message, the message is not embedded in every byte of the audio file. Instead the first position of embedding bit is chosen randomly and the rest of the bits are embedded only in even value of bytes of the audio file. For extracting the secret message, the permutation code book is used to reorder the message blocks into its original form. Md5sum and SHA-256 are used to verify whether the secret message is altered or not during transmission. Experimental results measured by peak signal to noise ratio, bit error rate, Pearson Correlation and chi-square show that the XLSB performs better than SLSB. Moreover, XLSB can embed a maximum of 750KB into MP3 file with 30db average result. This research contributes to the information security community by providing more secure steganography technique which provides message confidentiality and integrity.

Item Type:Thesis (PhD)
Additional Information:Thesis (Ph.D (Sains Komputer)) - Universiti Teknologi Malaysia, 2014; Supervisors : Prof. Dr. Subariah Ibrahim, Prof. Dr. Ghazali Sulong
Uncontrolled Keywords:Audio Steganography, robustness
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computing
ID Code:78214
Deposited By: Widya Wahid
Deposited On:30 Jul 2018 08:39
Last Modified:30 Jul 2018 08:39

Repository Staff Only: item control page