A Kirchhoff Scattering Model for Millimetre Wavelength Wireless Links

Lawrence Sayer, Alberto Loaiza Freire, Evangelos Mellios, Andrew Nix

Research output: Contribution to conferenceConference Paperpeer-review

80 Downloads (Pure)

Abstract

Modelling tools that include diffuse scatter are essential for planning 5G networks that include millimetre wavelength wireless links. In this paper, a Kirchhoff model is compared with real world measurements of diffuse scatter from a rough wall at 60GHz. This work is the first use the model to replicate an electromagnetic interaction with a large section of exterior wall at a frequency of interest for 5G communications networks where the geometry of the interaction is such that, theoretically, some of the assumptions on which the model are based are violated. This is important because this is a regime in which the model will commonly be called upon in a propagation modelling program, and thus determines the usefulness of the model. We present an experiment in which power is scatted in a 90 degree arc. There is a 3.21dB difference between the measured and modelled mean received power. The difference in the standard deviation is less than 1dB, suggesting that the multipath characteristic of the diffuse reflection has been captured by the model. In a sensitivity study, it is shown that uncertainty in permittivity, which was the most difficult parameter to measure, would not cause errors in peak scattered power of above 2dB.
Original languageEnglish
Number of pages5
Publication statusPublished - 9 Apr 2018
Event12th European Conference on Antennas and Propagation - London, United Kingdom
Duration: 9 Apr 201813 Apr 2018
Conference number: 12
http://www.eucap2018.org/

Conference

Conference12th European Conference on Antennas and Propagation
Abbreviated titleEuCAP
Country/TerritoryUnited Kingdom
CityLondon
Period9/04/1813/04/18
Internet address

Fingerprint

Dive into the research topics of 'A Kirchhoff Scattering Model for Millimetre Wavelength Wireless Links'. Together they form a unique fingerprint.

Cite this