Some Applications of Information Hiding
Unobtrusive communications are required by military and intelligence agencies: even if the content is encrypted, the detection of a signal on a modern battlefield may lead rapidly to an attack on the signaler. For this reason, military communications use techniques such as spread spectrum modulation or meteor scatter transmission to make signals hard for the enemy to detect or jam. Basics of spread spectrum modulation are reviewed in Section 6.4.1 and meteor-burst communications are studied by Schilling et al. . Criminals also place great value on unobtrusive communications and their preferred technologies include prepaid mobile phones and hacked corporate switchboards through which calls can be rerouted (e.g., ). As a side effect, law enforcement and counterintelligence agencies are interested in understanding these technologies and their weaknesses, so as to detect and trace hidden messages.
Information hiding techniques also underlie many attacks on “multilevel secure” systems used by military organizations. A virus or other malicious code propagates itself from “low security” to ‘‘high security" levels and then signals data downwards using a covert channel in the operating system or by hiding information directly in data that may be declassified (see also Section 2.7.2).
Information hiding techniques can also be used in situations where plausible deniability is required. “The obvious motivation for plausible deniability is when the two communicating parties are engaged in an activity which is somehow illicit, and they wish to avoid being caught”  but more legitimate motives include fair voting, personal privacy, or limitation of liability. One possible mechanism providing such property is the steganographic file system, presented by Anderson, Needham, and Shamir: if a user knows a file’s name, he can retrieve it; but if he does not, he cannot even obtain evidence that the file exists .
Anonymous communications, including anonymous remailers and Web proxies , are required by legitimate users to vote privately in online elections, make political claims, consume sexual material, preserve online free speech, or to use digital cash. But the same techniques can be abused for defamation, blackmail, or unsolicited commercial mailing. The ethical positions of the players in the information hiding game are not very clear so the design of techniques providing such facilities requires careful thought about the possible abuses, which might be nonobvious.
The healthcare industry and especially medical imaging systems may benefit from information hiding techniques. They use standards such as DICOM (digital imaging and communications in medicine) which separates image data from the caption, such as the name of the patient, the date, and the physician. Sometimes the link between image and patient is lost, thus, embedding the name of the patient in the image could be a useful safety measure [45, 46]. It is still an open question whether such marking would have any effect on the accuracy of the diagnosis but recent studies by Cosman et al.  revealing that lossy compression has little effect, let us believe that this might be feasible. Another emerging technique related to the healthcare industry is hiding messages in DNA sequences . This could be used to protect intellectual property in medicine, molecular biology or genetics.
A number of other applications of information hiding have been proposed in the context of multimedia applications. In many cases they can use techniques already developed for copyright marking directly; in others, they can use adapted schemes or shed interesting light on technical issues. They include the following:
Automatic monitoring of copyrighted material on the Web: A robot searches the Web for marked material and hence identifies potential illegal usage. An alternative technique downloads images from the Internet, computes a digest of them, and compares this digest with digests registered in its database [35, 49]. We will revisit these tools later in Section 7.3.2, and show that the actual benefits are not as good as the advertised ones.
Automatic audit of radio transmissions: A computer can listen to a radio station and look for marks, which indicate that a particular piece of music, or advertisement, has been broadcast [50, 51].
Data augmentation: Information is added for the benefit of the public. This can be details about the work, annotations, other channels , or purchasing information (nearest shop, price, producer, etc.) so that someone listening to the radio in a car could simply press a button to order the compact disc. This can also be hidden information used to index pictures or music tracks in order to provide more efficient retrieval from databases (e.g., [45, 53]).
Tamper proofing: The information hidden in a digital object may be a signed “summary” of it, which can be used to prevent or to detect unauthorized modifications (e.g., [54, 55]).
Some of these applications and techniques will be detailed in the next chapters. We tried to keep chapters simple enough such that any computer science graduate student can understand them without much problem. Note however that steganography and digital watermarking require some background in various disciplines including cryptography, image processing, information theory, and statistics. It is outside the scope of this book to detail all the basic techniques on which information hiding techniques are built. If more background is required, we refer the reader to Menezes  for cryptography, Jain  for image processing, and Cover  for information theory.<< Copyright Enforcement || Wisdom from Cryptography >>