The algorithm has been entered in the Competition for Authenticated Encryption: Security, Applicability, and Robustness (CAESAR) project, based on which the algorithm is expected to be deployed for increasingly secure and reliable information technology.
The new algorithmís major advantage is its resistance to multiple misuse in authenticated encryption operations that provide simultaneous confidentiality and integrity.
One problem of misuse is an attacker making a fake message if plaintexts are released before their integrity is verified. Once a conventional system outputs decrypted plaintext from tampered data without authentication, the attacker can show tampered data as being non-tampered. Whereas this occurs with many conventional systems, the new algorithm fixes the problem, thereby enabling relatively low-memory devices to handle large-volume data.
Another typical problem is the reuse of nonce. In the case of a common authentication algorithm called Advanced Encryption Standard with Galois Counter Mode (AES-GCM), a non-repeatable special parameter, or nonce, is required to achieve security. However, the algorithm is largely bleached if the nonce is reused, so the new algorithm fixes this problem to maintain security even after multiple reuse.
The new algorithm accepts messages longer than the 64-gigabyte limit of AES-GCM, and it works faster than AES-GCM on many platforms.
CAESAR is a competition organized to thoroughly evaluate authenticated encryption algorithms by testing their resistance to multiple third-party cryptanalyzing attacks to prove their security, applicability and robustness. Algorithms that receive third-party cryptanalysis through CAESAR are expected to gain wide acceptance, which is why this new algorithm has been submitted to the competition. Candidate algorithms will be screened annually and the first results will be announced on January 15, 2015, with the final results to be announced on December 15, 2017.
Based on the results of the CAESAR competition, NTT and Mitsubishi Electric intend to research and develop services and products for machine-to-machine (M2M) applications incorporating their new algorithm, thereby contributing to increased security and reliability in information technology.