New information technology: “steganographic fingerprinting”.

New information technology: “steganographic fingerprinting”..

NEW INFORMATION TECHNOLOGY: “STEGANOGRAPHIC FINGERPRINTING”  

The article is a continuation of a series of publications on the problems of steganographic protection [1-3]. Based on the results of a comparative analysis of software products offered by the market and distributed on the Internet, the features, modern capabilities and development trends of the promising information technology of steganographic fingerprinting are considered.”

It is often said that “new is well forgotten old”. Today this proverb has received a continuation: “implemented using new technology”. We are especially convinced of this almost daily, using the achievements of modern information technologies (IT).

One ​​of the most pressing problems of IT is the task of highly reliable protection of information, in particular, protection from unauthorized access to it. The development of such network technology areas as e-commerce, electronic banking and many others depends on its solution today. Everything rests on the task identification of a person. And the best biometric identifier has long been considered a “fingerprint”, especially since it can be put even by a person who cannot sign.

“Fingerprint” is the best signature for intellectuals!

The achievements of modern information technologies have breathed new life and new content into the science of dactyloscopy. Dactyloscopy from Greek – the study of fingerprints or literally: “relating to the observation of fingers”. The integration of digital dactyloscopy and computer steganography technology has made it possible to create an amazing tool for protecting information, digital documents and multimedia products (text, graphic, video and audio files) steganographic dactyloscopy”, the main purpose of which is to create identifiers — hidden digital markers (LDM) or, as they are also conventionally called, digital fingerprints”. The main tasks solved by LDM are shown in Fig. 1.

Fig. 1. Main tasks of hidden digital markers

Based on the tasks to be solved, the following main requirements are imposed on the SCM:

• secrecy (absence of unmasking factors) [1];
• noise immunity [4];
• protection from destructive influences of third parties [3].

Recently, SCMs have been actively used in the following applications:

• blocking unauthorized users' access to audio information in networks and on media;
• control of advertising in radio and television broadcasting;
• identification of the speaker;
• encryption of audio recordings, etc.

So where are these “fingers”?

How are hidden digital marker technologies implemented? How are digital “fingerprints” formed? Where are they located? The answer to these questions is briefly presented in Fig. 2.

Fig. 2. Basic technologies of hidden digital markers

The analysis shows that all modern digital marking methods use computer steganography methods, broadband signals and elements of noise theory. However, not all existing computer steganography methods can be used to form the SCM. For example, the method using the least significant bits (LSB method), although it allows hiding information, does not meet the requirements for noise immunity, since when the marked image is distorted or compressed using lossy schemes, the hidden information (label) is lost [3].

Therefore, modern steganographic systems use the principle of hiding a mark, which is a narrow-band signal, in a wide frequency range of the marked image. This principle is implemented using two different algorithms and their possible modifications. In the first case, information is hidden by phase modulation of the information signal (carrier) with a pseudo-random sequence of numbers. In the second, the available frequency range is divided into several channels and transmission is performed between these channels (Fig. 2).

It should be noted that the mark is some additional noise relative to the original image, but since noise is always present in the signal, its slight increase due to the introduction of the mark does not produce any visible distortions. In addition, the mark is scattered throughout the original image, as a result of which it becomes more resistant to cutting.

When using ICE technology(Fig. 2) digital identification signals are periodically embedded throughout the entire length of the recording in narrow “cuts” made in the spectrum of the audio signal being processed. As a result, it is impossible to extract even a small unmarked fragment from the recording. Attempts to remove the coding signals simply result in the destruction of the recording. This system is designed to work with both analog and digital signals, and the ICE identification codes are preserved both during signal digitization and during its transmission in compressed form.

VEC Technologyis based on a similar principle of embedding identification codes, but only in the field of digital images. This task turned out to be more difficult. It is important that the magnitude of the embedded signals be below the threshold of perception, and the signals themselves could be repeated many times within the image, any fragment of which must be marked. Codes embedded in the image (usually 16-bit) are preserved during compression/decompression operations, during data transmission using a modem, and also during image conversion from digital to analog and back. VEC codes are preserved even when the image is compressed using the JPEG method with a degree of 10:1.

Similar to JPEG and MPEG technologies, the VEC algorithm systematically modifies pixel values ​​by dividing the image into blocks and then applying a Fourier transform. In this case, the information used to identify the material is masked by image details that function as “noise”, but is restored by the inverse Fourier transform. The more details an image contains, the higher the density of identification codes embedded in it can be.

During the process, the image may undergo various transformations, both intentional and unintentional. In particular, it may be compressed, including using lossy compression algorithms. Naturally, the immutability of the mark under such transformations is only achievable if it is placed in large areas across the entire image.

What's better?

At present, a fairly large number of software products have been developed that provide for the embedding of hidden digital markers into multimedia files. The main comparative characteristics of marking programs are presented in Table 1.

Table 1. Comparative characteristics of marking programs

For those readers who would like to become more familiar with the results of the practical implementation of the new information technology of steganographic fingerprinting (SD technology) and specific characteristics of the programs, Table 2 provides information sources of materials on the Internet.

Table 2. Information sources of materials on SD technology

Analysis of the materials presented in Tables 1, 2 shows that the most typical digital fingerprint program is the “SureSign” system, developed by SIGNUM TECHNOLOGIES. Its main area of ​​application is copyright protection, verification of the reliability and authenticity of materials in such applications as secret documentation and electronic commerce. This system includes several service programs based on the patented FBI (FINGERPRINTING BINARY IMAGES) technology and allowing to embed, define and read “fingerprint” regardless of the content of the transmitted object.

The system consists of two independent parts:

• SureSign Writer – for forming a label;
• SureSign Detector – for detecting a label.

The SCM must include the user (author) identification number and the document number. If necessary, a visible logo with adjustable transparency can be additionally embedded. And to prevent “fingerprints” from turning into typos, the following restrictions have been introduced into the SureSign system:

• the image size must be more than 100 KB;
• the image size in JPEG format must be more than 50 KB;
• the compression must be no more than 10 times.

What's over the horizon?

Information technologies continue to develop rapidly, so the trends of further development are visible even to the naked eye. For example, the analysis of the characteristics and features of the SureSign system leads to the possibility of its integrationwith a biometric identification device, in particular, with the most advanced fingerprint identification device on the market. Moreover, modern computers are already equipped with such identification devices (Tables 3, 4). In this case, the new integrated system would be able to provide highly reliable user (author) authorization and protect the object from falsification. In this case, the “steganographic fingerprinting technology” used could literally be written without quotation marks.

Table 3. Modern biometric fingerprint identification tools”

Table 4. Main technical characteristics of biometric identification means by “fingerprint”

Currently, both in Russia and abroad, work on creating new biometric security tools and searching for new biometric technologies is developing intensively. Good promising results have been achieved in the field of contactless (remote) biometric identification. Work is also underway to create technical identification tools that use such individual characteristics as DNA code, cardiogram characteristics, smell, gait, etc. However, these tools still have a number of disadvantages (instability of results, low reliability, complexity of implementation, etc.) and are not yet ready for widespread implementation, but intensive work continues.

The analysis shows that the modern capabilities of biometric technologies already today provide the necessary requirements for identification reliability, ease of use and low cost of equipment. The implementation of biometric attachments to computers at prices of about 100 dollars and below provides good prerequisites for significant activation of new electronic technologies, including electronic commerce. Thus, the analysis has clearly shown that integration of biometric technology, cryptography and steganography already today allows us to implement the most reliable methods of information protection and is one of the most promising trends for the coming years.

Another relevant direction of development of the considered technology of digital marking and biometric identification is use as stegokey of compressed biometric fingerprint information. Analysis of the operation of technical identification tools listed in Table 3 shows that almost all of the tools presented use an operating algorithm based on scanning the finger pattern, digitizing the results, compressing the data to several dozen bytes, and then encrypting the results. The resulting file does not allow an intruder to restore the true fingerprint” and can be used as a digital user identifier and steganographic key.

In connection with the widespread use of digital photography and color printing, one of the promising areas for protecting the copyrights of producers of color images and their prints is the method of steganographic protection based on print bitmap encoding technology[5]. This protection can be implemented on the basis of a licensed supply of an electronic copy of the product. The copy verification algorithm is based on the registration of fingerprint bitmaps, visual comparison of the original fingerprint and the suspicious copy, confirmation of authorship using the hidden message extracted from the bitmap and making a decision on the license violation.

It is known that to date, steganographic methods have not yet provided a mathematically proven guaranteed protection of information, so it is quite natural that another direction for the possible improvement of the information technology of steganographic fingerprinting arises”: integrationdocument marking systems and cryptographic protection systems, which will allow achieving guaranteed protection of hidden information. Therefore, this technology is of great interest not only for copyright protection, but also for the protection of information and documents, especially when transmitted over public communication channels. But that's another story.

Literature

1. Ukov V.S., Romantsov A.P. Computer steganography yesterday, today, tomorrow.//Special equipment, 1998, 5.
2. Ukov V.S., Romantsov A.P. Assessment of the secrecy level of multimedia steganographic channels for storing and transmitting information.//Special equipment, 1999, No. 6.
3. Ukov V.S. Steganographic technologies for protecting documents, copyrights and information.//Special technology, 2000, No. 6.
4. Steganography. Thematic issue of the journal.//Information protection. Confidential, 2000, No. 3.
5. Arkhipov O.P. et al. Steganography in PRN files.//Information protection. Confidential, 2002, No. 2.