Practicals & Theses

List of practicals and theses at the Embedded Computing Systems Group (E191/2)

Details for IoT Distributed Consensus Algorithm Implementation

The goal of this Master thesis is to implement and experimentally analyze the performance of a novel distributed consensus algorithm for dynamic networks.


During the last years, we have developed novel distributed agreement algorithms for directed dynamic networks controlled by a message adversary. Such algorithms operate in synchronous rounds, where all nodes broadcast some information, collect the broadcasts of the other nodes, and finally compute something based on the current local state and the messages received.

The goal of this project is to efficiently implement a provably optimal algorithm developed recently, and to evaluate its performance in some IoT network. There is no preference for a particular IoT platform, the ESP8266 used e.g. in Ezio Bartocci's IoT class would be just fine.

The size of this project is that of a typical Bachelor thesis, but has the potential to be published at some scientific conference (via a paper that could be worked out in a subsequent scientific Master project, for example). Since the existing algorithm can be improved both w.r.t. efficiency and even w.r.t. correctness proofs, however, the topic would also make up a very nice Master thesis.

Required Skills

IoT node programming skills, interest in networked distributed systems


Dipl.-Ing. Dr.techn. Ulrich SCHMID (main responsibility)