Graph theoretic methods for radio equipment selection

Flood, Ian 2013. Graph theoretic methods for radio equipment selection. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (1MB) | Preview
	PDF - Supplemental Material Restricted to Repository staff only Download (69kB)

Abstract

In the 1970s and 1980s, a small group of American engineers recognised the importance of the graph-colouring ideas studied by mathematicians and the potential for these ideas to be used in practical radio frequency assignment procedures. Some groundbreaking work led to a long period of study in academia where many variants on the Frequency Assignment Problem have been considered and some advanced algorithms developed. This thesis has investigated the Frequency Assignment Problem for microwave fixed links and, taking account of the constraints experienced in professional practice, extended this to include the problem of Equipment Selection. For a particular data-rate, standard radio equipment using relatively lower-or higher order modulation schemes can be deployed by the fixed link operator. While the higher-order options use less bandwidth, they radiate at higher powers and require more protection in the radio interference environment. That is, they are more potent interferers and present a greater challenge to distant interferers. Therefore, when the assigner’s objective is to minimise the span of frequencies used by a network, the higher-order modulation radio is not always the most spectrally efficient. The thesis has hypothesised that by doubling the bandwidth requirement on selected links, the assigner can actually reduce the overall span of frequencies used to support a frequency assignment for the entire network. With a minimum span objective, fixed link deployment scenarios have been exposed to a standard IP Solver that gives exact solutions. Using graph-theoretic methods, equipment selection heuristics have been developed and tested in offline and online environments. This work has gathered significant evidence in support of the hypothesis.

Item Type:	Thesis (PhD)
Status:	Unpublished
Schools:	Computer Science & Informatics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Date of First Compliant Deposit:	30 March 2016
Last Modified:	19 Mar 2016 23:25
URI:	https://orca.cardiff.ac.uk/id/eprint/50910

Actions (repository staff only)

Edit Item