Weak models of wireless distributed computing Comparison between radio networks and population protocols

Abstract

This thesis compares weak distributed computing models that are suit- able for extremely limited wireless networks. The comparison is mainly between multiple variations of radio networks and population protocols. The analysis is based on model features, computability and algorithmic complexity. The thesis analyses essential and optional model features, and organizes the models accordingly. It discusses the applicability of results from stronger models to radio network models, including impossibility results, algorithms and their runtime. It analyzes different radio network algorithms for the classical problems in terms of their features, and it discusses their applicability to other radio network models. It reviews the fundamental differences between population protocols and radio networks. Lastly, the comparative analysis summarizes fundamental differences and separating features.