Abstract
Abstract Recently, the emergence and increasingly widespread of wireless networks has generated considerable interest in the information-theoretical approach to ensuring secure communications. The basic principle of information-theoretic security requires a combination of cryptographic methods with channel coding techniques that use the randomness of communication channels to ensure that messages sent cannot be intercepted or decrypted by a third party maliciously eavesdropping on the wireless medium. This paper discusses the information-entropy method for assessing security. We show that information-theoretical security means that any algorithm has a negligible probability of violating the security property. This is the same as unconditional security: it does not rely on any computational assumptions and is not limited to probabilistic violators.