On the Cost of Geographic Forwarding for Information-Centric Things

Marcel Enguehard, Ralph E. Droms, Dario Rossi

Abstract

Recent research has identified information-centric networking (ICN) as a good fit for Internet of Things (IoT) deployments. However, most studies have focused on ICN as an application enabler, disregarding the behavior from a network viewpoint. In this paper, we address this by studying the most important properties of an ICN-IoT deployment and contrast the operational costs between geographic-based forwarding and name-based forwarding schemes. We aim to understand if, and under which IoT deployment characteristics, geographic forwarding constitutes an advantage over name-based schemes, in terms of feasibility (i.e., memory footprint and computational capability of the devices) and performance (which we analyze as the overall energy cost of operating an ICN-IoT network under either forwarding paradigm). To achieve this goal, we employ a mixture of: 1) modeling; 2) simulative; and 3) experimental methodologies, which are useful to, respectively, 1) state the problem in a principled way; 2) gather information about topological properties that are instrumental to the model; and 3) gather physical properties of the devices to feed the model with realistic data. In a nutshell, our results show that geographic forwarding: 1) halves the memory footprint on our reference deployments and 2) yields significant energy savings, especially for dynamic topologies.
BibTeX
@article{enguehard2018cost,
  author = {Marcel Enguehard and Ralph E. Droms and Dario Rossi},
  title = {{On the Cost of Geographic Forwarding for Information-Centric Things}},
  journal = {IEEE Trans. Green Commun. Netw.},
  year = {2018},
  doi = {10.1109/tgcn.2018.2867267},
}

Loading related content...

Figure