Architecture for responsive emergency communications networks

Patrick Lieser, Flor Alvarez, Paul Gardner-Stephen, Matthias Hollick, Doreen Boehnstedt

Research output: Contribution to conferencePaperpeer-review

46 Citations (Scopus)


Self-organizing Mobile Ad-hoc Networks (MANETs) based on Delay Tolerant Networking (DTN), are powerful tools for maintaining or reestablishing telecommunications following disasters and other infrastructure disrupting events. However, such networks typically have very limited bandwidth compared with infrastructure-based networks, with the practical effect that they cannot satisfy every demand placed upon them. Thus, if the most critical traffic is to be delivered, and in a timely manner, some form of filtering or prioritization is needed. This paper sets out an architecture for solving this problem, and presents supporting simulation and field results. The architecture is built using the input of several emergency and disaster response organizations, to ensure that the key services required by citizens post-disaster were incorporated. Reflecting the dynamic nature of post-disaster communications needs, as identified in the survey, the architecture provides a framework in which arbitrary prioritization policies can be defined, and redefined, so that the humanitarian utility of a network can be maximized according to the prevailing situation and requirements. A proof-of-concept implementation is presented, yielding orders of magnitude reduction in message delivery latency in both simulation and in a field trial of an existing disaster communications system.

Original languageEnglish
Number of pages9
Publication statusPublished - 22 Dec 2017
Event2017 IEEE Global Humanitarian Technology Conference (GHTC) - San Jose, United States
Duration: 19 Oct 201722 Oct 2017


Conference2017 IEEE Global Humanitarian Technology Conference (GHTC)
Abbreviated titleGHTC 2017
Country/TerritoryUnited States
CitySan Jose


Dive into the research topics of 'Architecture for responsive emergency communications networks'. Together they form a unique fingerprint.

Cite this