Modern information technology/4. Information Security

Morozova A.N.

State higher education institution

"National Mining University"

Steganography methods for transmission of messages in audiosignal

Methods of steganography got active development with the advent of computer technology, when it became possible to use multimedia files as a container for information transfer. Ones of the most commonly used containers are multimedia files. Investing in multimedia containers based on the fact that the human senses have certain selectivity, therefore, to provide a good perception, digital multimedia content has the property of redundancy. This makes it possible to use the least significant components of the necessary information required in order to conceal it.

Human hearing system can be represented as a frequency spectrum analyzer, which can detect and identify signals in the range of 20 - 20,000 Hz. Human hearing system can be modelled as 26 band pass filters, which bandwidth increases with increasing frequency. In connection with this particular development methods have digital steganography audiosrede. Hiding data in the sound (audio) signals is particularly promising. Today there are many software products that have different capabilities, operating principles, scope, format of data used and the preferred software platform.

Essential influence on reliability of stegosystem and possibility of detection of the fact of transfer suppressed messages renders a container choice. For realization of introduction of additional information it is best of all to use audio the file in the WAVE format. When we use WAV format it is necessary to make changes to a sound wave that in very small degree influences sounding, it very much increases probability to leave the hidden message unnoticed.

There are various algorithms for data hiding in audio systems. Among them are the phase coding as one of the most effective ways. The system of the human distinguishes changes hearing of a signal phase more weakly than amplitude or frequency changes, such feature of human perception is the reason for the effectiveness of these algorithms.

Implementation of phase modification audio information is a method in which the phase of the initial segment of the audio signal is modified depending on the implemented data. Phase subsequent segments agreed with it to save the phase difference. It is necessary because the phase difference to the human ear is more sensitive. Phase encoding is one of the most effective methods of steganography by the signal noise is perceived, which allows more seamlessly embed additional information. Further two algorithms implements coding with using of modifications of the phase of the original signal.

Method based on a phase difference.

Procedure of phase encoding consists in the following:

1. Sound sequence converted into a series of N short segments (blocks) .

2. The N-th segment of a signal is applied К-point discrete Fourier transform, where , and creates arrays phases and amplitudes for .

3. The difference of phases between each next segments is remembered for :

; .

4. The binary sequence of data is represented as or displaying respectively "1" or "0". .

5. Taking into account a difference of phases the new massif of phases is recreated for .

6. The matrix of phases is anew created, using the kept difference of phases/

7. Using the new array of phases and the array of amplitudes, components of a complex range that is necessary for performance of the return transformation of Fourier are calculated.

8. Restoration of a sound signal is carried out by application of operations of the return DPF to an initial matrix of amplitudes and the modified matrix of phases.

The relative phase difference between each adjacent segments remains. It is this relative difference in the phase of the human ear is most sensitive. The drawback scheme which is based on the method of phase encoding is a low throughput.

For an assessment of this method it is necessary to find an assessment of information message and to define its error. The idea of the proposed method is based on the determination of the phase characteristics of the signal received by the Hilbert transform emissions. Since the analytical solution of the problem is difficult, the proposed solution is justified by computer simulation. Method for solution of the problem involves the following operations:

1. Formation of the signal container containing an information message .

2. Definition Hilbert image of a signal container, where H – the operator of Hilbert transform:

3. Definition of fractional part of the phase characteristic of a signal container:

4. Expansion of the phase characteristic of a signal of u (t) on an interval of its supervision for the purpose of receiving an assessment of the developed phase characteristic, where – the step function determined by jumps:

5. Assessment of an information message as differences of the phase characteristic of a signal and a signal container phase without an information message:

6. Definition of an error of an assessment :

The offered method of the hidden information transfer is based on use of phase characteristics of signals. Increase of reserve is reached for the account modulations of parameters of a signal container on small time intervals, comparable with its period. Increase of accuracy of measurement of phase characteristics is reached by a way additional weight processing of a signal container.

For increase of efficiency of this algorithm it is recommended to use cryptography methods for preliminary encoding of the text of the message. Joint application of algorithms of a steganography and cryptography increases reliability of transfer of the message.

Analysis of trends in computer steganography shows that in the next few years, interest in the development of its methods will grow more and more particular. It is well known that the problem of information security is constantly growing and encourages the search for new methods of information protection. This article has discussed algorithms for hiding information in audio signals based on phase modification of the original signal. Selection of these methods is caused by their simplicity and that they allows to hide in rather small files rather large volumes of information without significant changes in file sounding for perception the person.

References:

1. Basics of computer steganography/ VA Khoroshko, A.D. Azarov, J.E. Rustle and others, Vinnitsa, Vinnitsa state technical University, 2003. - 143 p.

2. Bendat, J., A. Peirsol Applied analysis of random data: Per. from English. - Academic Press, 1989.-540 p.

3. W. Bender, D. Gruhl, N. Morimoto, A. Lu, Techniques for Data Hiding. IBM Systems Journal, 35(3&4): pp.313-336, 1996.