To be secure, tamper resistant cryptographic devices must be protected against DPA and related attacks. Independent testing processes are essential for validating the presence and effectiveness of these countermeasures. Testing methodologies for power analysis vulnerabilities can yield varying degrees of assurance as to the security of the device under test. While insecurity can be demonstrated conclusively, evidence of security is more open-ended. Confidence in a security evaluation depends on many factors including the comprehensiveness of the evaluation, the skill of the evaluator, the nature of the device’s design, and the difficulty of exploiting any identified vulnerabilities. This paper reviews testing strategies for power analysis and related attacks, including black box and clear box methods. The paper also examines how appropriate design architectures and evaluation approaches can be combined to yield the strongest evidence of a device’s security.
Randomness is required for a variety of computational, statistical, and security-related applications. In particular, random numbers and the processes used to generate them are a critical component of secure protocols and cryptographic key generation. Security processes that lack adequate sources of randomness will have poor security. Cryptography Research has evaluated the C3 Nehemiah random number generator, which is an on-chip component of the VIA Technologies Nehemiah processor core. When properly used, the generator was found to be a consistent, high-rate source of entropy which we believe is suitable for use in cryptographic and high-assurance applications.
This report analyzes the Nehemiah RNG design, provides an entropy analysis of the source, and provides developer recommendations for proper use of the Nehemiah RNG. Cryptography Research provided no Nehemiah design assistance to VIA Technologies or Centaur Technology.
Despite the high public profile of piracy as a threat to intellectual property owners, surprisingly little useful research has been done to understand the range of technical solutions that are feasible. This paper presents results from a study sponsored by Cryptography Research, Inc. to determine how cryptographic systems can provide the most effective long-term deterrent to the piracy of digital video and other content distributed on optical media.
Although numerous products and technologies have been advertised as solutions to the problem of piracy, most commercial security systems fail catastrophically once an implementation is compromised. These designs can work in limited deployments, but any technology deployed as part of a major standard will inevitably attract extremely determined attacks – and some implementations will get broken. The long lifespan of media formats, diversity of player implementations, complexity of security/usage models, and constantly-changing risk scenarios provide attackers with numerous avenues of attack and the time and resources to explore them. As a result, effective content protection systems must be able to survive compromises and adapt to new threats.
Good cryptography requires good random numbers. This paper evaluates the hardware-based Intel Random Number Generator (RNG) for use in cryptographic applications.
Almost all cryptographic protocols require the generation and use of secret values that must be unknown to attackers. For example, random number generators are required to generate public/private keypairs for asymmetric (public key) algorithms including RSA, DSA, and Diffie-Hellman. Keys for symmetric and hybrid cryptosystems are also generated randomly. RNGs are also used to create challenges, nonces (salts), padding bytes, and blinding values. The one time pad – the only provably-secure encryption system – uses as much key material as ciphertext and requires that the keystream be generated from a truly random process.
Because security protocols rely on the unpredictability of the keys they use, random number generators for cryptographic applications must meet stringent requirements. The most important is that attackers, including those who know the RNG design, must not be able to make any useful predictions about the RNG outputs. In particular, the apparent entropy of the RNG output should be as close as possible to the bit length.
Abstract: Cryptosystem designers frequently assume that secrets will be manipulated in closed, reliable computing environments. Unfortunately, actual computers and microchips leak information about the operations they process. This paper examines specfic methods for analyzing power consumption measurements to and secret keys from tamper resistant devices. We also discuss approaches for building cryptosystems that can operate securely in existing hardware that leaks information.