Abstract
Ideally, content protection should incorporate four complementary types of protection. Physical security should control access to the asset. Encryption should protect the asset itself. Forensics technology should trace the asset. Scouting should limit the illegal use of the asset. This chapter introduces basic technologies to fulfill these protections: cryptography, digital watermarking, hardware and software tamper resistance, as well as compliance and robustness regimes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Initially, the US National Security Agency required decreasing the key size from 112 bits to its current size in order to decrease its strength. Many national regulations treat cryptographic algorithms as war weapons. Later in this book, we will see how the limitations imposed by such regulations have dramatically impacted some deployed video content protection systems.
- 2.
In fact, PGP supports two formats of certificates: X509 and its own proprietary format.
- 3.
We use video in this section. Nevertheless, the presented concepts are (most of the time) equally applicable to audio.
- 4.
Another interest in the secret key appears in legal cases. A court may demand disclosure of the algorithm used by the watermark. It will be part of the minutes of the process. Nevertheless, changing the key will allow us to safely continue using the watermark technology [82].
- 5.
Having the original content at hand is a constraint and a risk. You have to either completely trust the operators (which is possible if the operators are members of the content owner’s company) or protect the original content. Thus, content owners prefer blind and semi-blind solutions.
- 6.
Forensic marking is sometimes referred to as fingerprinting, following the seminal work of Neal Wagner [89]. In 1983, Wagner proposed slightly modifying each copy of a digital content in an imperceptible way to discriminate between each instance of the same piece of content. Nevertheless, in this book, we reserve the term fingerprint for techniques of content identification; see Sect. 3.4, Fingerprinting. Other terminology is also sometimes used, such as traitor tracing, transactional watermark, content serialization, and user forensics.
- 7.
It is difficult to assess the real impact of this leakage to the movie’s failure. However, the bad mouthing may have deterred many people from watching it in theaters.
- 8.
Indeed, depending on the type of devices, there are other possible positions for the embedder than prior to the rendering process. If the watermark technology operates in the compressed domain (or at least partly), in the case of PVR, the piece of content would be watermarked prior to being recorded on the internal hard drive. Similarly, in the case of a home network media gateway, the piece of content would be watermarked prior to being ingested by the home network [100].
- 9.
Even if they use different keys, the trust model would be weakened. .
- 10.
Most proposed solutions did not erase the “copy once” mark. Therefore, a “copy no more” content holds both “copy once” and “copy no more” marks.
- 11.
Some implementations proposed using a different watermarking technology to embed the “copy no more” mark. This solution does not enhance the trust model because the detector needs still to handle both watermarking technologies.
- 12.
In the operating method, video and audio fingerprints are similar to the human fingerprints used by police. The fingerprints must be in the central database so that the police can potentially identify a suspect. As in video, although the submitted fingerprints may be altered (partially in the case of human fingerprints), the police may still be able to identify the suspect.
- 13.
Low consumption constraints did not initially come from the fact that smart cards were removable. Indeed, for the first foreseen applications, the smart card reader could have provided enough power supply. In fact, the reduced consumption constraint derives from the small form factor and the associated problems of thermal dissipation. The more power consumed, the more thermal energy produced. Too much heat would destroy the component. Later, with its utilization by mobile devices such as mobile phones, or for wireless cards, the power consumption became a crucial issue.
- 14.
Evaluating a security system in white box means that the evaluator has access to the documentation of the internal behavior of the system. Evaluating in black box means that the evaluator has only access to public information. He may only guess the behavior from what he publicly learns. Therefore, white box evaluation is more challenging for the designers than black box. In white box, the evaluator acts like an insider.
- 15.
The newest techniques attack the chip from its backside with a FIB microscope, thus bypassing the sensor mesh.
- 16.
ContentGuard is co-owned by Xerox, Microsoft, Time Warner, and Technicolor.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Diehl, E. (2012). A Tool Box. In: Securing Digital Video. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-17345-5_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-17345-5_3
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-17344-8
Online ISBN: 978-3-642-17345-5
eBook Packages: Computer ScienceComputer Science (R0)