In the past, memoranda of understanding were designed either for no authentication or for a full agreement on public keys. Since this prior agreement must be based on techniques outside the system (e.g. B trusted servers that never go down), it`s a good idea to consider lower levels of key distribution that require as few additional assumptions as possible. Pease M, Shostak R, Lamport L (1980) Agreement in case of error. J ACM 27 (2):228-234 The Byzantine agreement includes a system of n processes, some of which may be defective. The problem is that good processes agree on a binary value emitted by a string that can itself be one of the n processes. If the issuer sends the same value to each process, all correct processes must agree on that value, but in any case, they must agree on a value. An explicit solution that does not use authentication for processes n = 3t + 1 is indicated by using message bits 2t + 3 and message bits O (t3 log t). This solution can be easily extended to the general case of n ⩾ 3t + 1 to obtain a solution with 2t + 3 turns and O (nt + t3 log t). Dolev D, Fischer MJ, Fowler RJ, Lynch NA, Strong HR (1982) An efficient algorithm for Byzantine agreements without authentication.
Inf Control 52 (3):256-274 In this article, we investigate the possible margin of error for Byzantine correspondence, assuming different types of key distribution. It turned out that message authentication is useful for getting a Byzantine match with any number of arbitrarily faulty nodes. Unfortunately, this approach is the additional problem of key allocation. The distribution of keys can be seen as a pre-agreement on the public keys of the participants. Hadzilacos, V. Connectivity requirements for bizantine agreements in case of limited failures. Distrib Comput 2, 95-103 (1987). doi.org/10.1007/BF01667081 Dolev D, Dwork C, Stockmeyer L (1983) On the minimum synchronism needed for distributed consensus. Proc 24th Annual Symp on Foundations of Comput Sci, IEEE Computer Society, Tucson, AZ (November 1983) University of Toronto, 10 King`s College Road, M5S 1A4, Toronto, Ontario, Canada Reischuk R (1982) A new solution for the Byzantine generals problem. IBM Res Rep RJ 3673 (November 1982) Lamport L, Shostak R, Pease M (1982) The Byzantine generals problem. ACM Trans Programme Lang Syst 4 (3):382-401 Instant online access to all editions from 2019. The subscription is automatically renewed every year.
Hadzilacos V (1984) error tolerance problems while calculating. Dissertation, Harvard University Distributed Computing Volume 2, pages95-103(1987)Cite this article We study the problem of making Byzantine agreements on any network where processors and communications are subject to omissions or errors of shutdown. . . .