Performance Comparison between Network Coding in Space and Routing in Space

TL;DR

This paper quantitatively compares network coding in space and routing in space, analyzing their performance differences and the conditions under which network coding offers advantages in two-dimensional Euclidean space.

Contribution

It introduces a quantitative analysis of cost advantages between network coding in space and routing in space with arbitrary terminal node movement.

Findings

Upper-bound of cost advantage identified

Region where network coding outperforms routing determined

Properties of Space Information Flow analyzed

Abstract

Network coding in geometric space, a new research direction also known as Space Information Flow, is a promising research field which shows the superiority of network coding in space over routing in space. Present literatures proved that given six terminal nodes, network coding in space is strictly superior to routing in space in terms of single-source multicast in regular (5+1) model, in which five terminal nodes forms a regular pentagon centered at a terminal node. In order to compare the performance between network coding in space and routing in space, this paper quantitatively studies two classes of network coding in space and optimal routing in space when any terminal node moves arbitrarily in two-dimensional Euclidean space, and cost advantage is used as the metric. Furthermore, the upper-bound of cost advantage is figured out as well as the region where network coding in space is superior to routing in space. Several properties of Space Information Flow are also presented.