Engineering Papers⌕ Search
Home/Archive/NASA NTRS

NASA NTRS · 20170010412

A Machine Learning Concept for DTN Routing

Dudukovich, Rachel·Hylton, Alan·Papachristou, Christos
View original source ↗Export metadata ↓

Abstract

This paper discusses the concept and architecture of a machine learning based router for delay tolerant space networks. The techniques of reinforcement learning and Bayesian learning are used to supplement the routing decisions of the popular Contact Graph Routing algorithm. An introduction to the concepts of Contact Graph Routing, Q-routing and Naive Bayes classification are given. The development of an architecture for a cross-layer feedback framework for DTN (Delay-Tolerant Networking) protocols is discussed. Finally, initial simulation setup and results are given.

Explore related subjects

Delay Tolerant Networks

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Dudukovich, Rachel, Hylton, Alan, Papachristou, Christos. 2017-10-10. A Machine Learning Concept for DTN Routing. https://ntrs.nasa.gov/citations/20170010412

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

NASA NTRS2018

Delay Tolerant Network Routing as a Machine Learning Classification Problem

This paper discusses a machine learning-based approach to routing for delay tolerant networks (DTNs) [1]. DTNs are networks which experience frequent disconnections between nodes, uncertainty of an end-to-end path, long one-way trip times, and may have high error rates and asymmetric links. Such networks exist in deep space satellite networks, very rural environments, disaster areas and underwater environments. In this work, we use machine learning classifiers to predict a set of neighboring nodes which are the most likely to deliver a message to a desired location based on message history delivery information.We use the Common Open Research Emulator (CORE) [2] to emulate the DTN environment based on real-world location traces and collect network traffic statistics from the Bundle Protocol implementation IBR-DTN [3]. The software architecture for classification-based routing, analysis and preparation of the network history data and prediction results are discussed.

Delay Tolerant Networks↗