On the Design and Analysis of Consensus Protocols for Automotive Electronic Systems

Abstract: This thesis deals with the fundamental problem of reaching agreement on a value in a distributedcomputing system in the presence of faults. We address this problem in the contextof safety-critical distributed automotive applications, such as virtual traffic lights. In suchsystems, it is essential that different computers can make coordinated and mutually consistentdecisions. To this end, they must reach agreement on the data they use as a basisfor their decisions. The problem of ensuring that a group of computers agrees on a value,or a set of values, is known as the consensus problem. We consider this problem for systemsthat use unreliable communication channels, e.g., wireless channels, where an arbitrarynumber of messages can be lost during the execution of an agreement algorithm. Previousresearch has shown that it is impossible to construct an algorithm that guarantees consensusunder the assumption that the number of communication failures is unbounded. Our aimis therefore to design consensus algorithms that minimize the probability of disagreementunder this failure model. To this end, we propose and investigate three variants of an agreementalgorithm. These variants are distinguished by their decision criterion, which decideswhether a computer should decide on a value or abort. We conduct our analysis for bothsymmetric and asymmetric communication failures. Our analysis show that the probabilityof disagreement depends strongly on the number of computers in the system, the number ofrounds of message exchange, and the probability of message loss. We show that a moderatelypessimistic decision criterion achieves a lower maximum probability of disagreementcompared to both an optimistic and a pessimistic decision criterion.