IEEE Communications Society - Vancouver Chapter
 

 Open Source Development of
Terrain-Based Path Loss Models for RF Planning Tools

Dave Michelson
IEEE VT-S Propagation Committee

Chris Cobbold
Telus Mobility

Location:
BCIT, Bldg NW1 - Boardroom

Monday, April 10, 2000
7:30-9:00 pm

Abstract

The vast majority of RF planning tools used by designers of point-to-point, mobility, and broadcast radio systems rely on variants of the classic Okumura-Hata, Longley-Rice (ITM), and TIREM models to predict path loss.  Since such models only account for an abstraction of the terrain profile between the transmitter and receiver, they are inherently less accurate than schemes which use the true profile.  Although brute-force solution of the wave equation is computationally untenable in such cases, clever approximations can yield useful results.  Three such approaches have been considered over the years: integral equation methods, parabolic equation methods, and Fresnel-Kirchhoff methods.  Initially, the relatively high computational demands of such methods and a general lack of accurate terrain data were the principal deterrents to adoption of such schemes.  However, these issues have become far less important in recent years.  Instead, the lack of generally accepted implementations which can be folded into existing or future planning tools and concern regarding intellectual property issues have become the principal factors which have held back further development and adoption.

Open Source development may hold the key to overcoming these problems.  The IEEE VT-S Propagation Committee has proposed sponsoring development of an Open Source version of the Fresnel-Kirchhoff path loss model as a method for evaluating the effectiveness of this development approach and to encourage community participation.  After reviewing the Fresnel-Kirchhoff model in detail, the tools used to facilitate Open Source development, including FWEB, a system for literate programming which enables one to maintain both documentation and source code in a single document (the .web file), and CVS, a version control system which provides network-transparent source control for groups of developers, will be described.

About the Speakers:

David G. Michelson received the B.A.Sc., M.A.Sc., and Ph.D., all in Electrical Engineering, from the University of British Columbia.  From 1994-1996, he was a postdoctoral research fellow in the MDA/NSERC Radar Remote Sensing Group at UBC where, with support from the Canada Centre for Remote Sensing and the Imaging Radar group at NASA/JPL, he studied electromagnetic scattering from terrain and ground cover, methods for classifying and interpreting polarimetric SAR imagery, and RF coverage prediction at VHF/UHF wavelengths.  Since September 1996, he has been a full-time research consultant to AT&T and has worked with teams from AT&T Wireless Services (Redmond, WA) and AT&T Labs-Research (Red Bank, NJ) to develop both analytical and measurement-based mathematical models of radiowave propagation for use in system planning and simulation. He chairs the IEEE VT-S Propagation Committee, moderates NEC-LIST (a 600-member Internet mailing list devoted to computational electromagnetics) and chairs the Vancouver Linux Users Group.  He is a senior member of the IEEE.

Chris Cobbold received an M.A.Sc in Electrical Engineering from the University of British Columbia in 1995 and a B.Sc. in Electrical Engineering from Queen's University in 1991.  Chris has worked in the mobile communications industry since 1995, with particular emphasis on CDPD, CDMA and GSM technologies. He has helped plan networks for PCS/GSM operators in South Korea, Belgium and Spain.  Chris currently manages the RF Planning and Optimization group for B.C. at Telus Mobility in Burnaby.

No-Host Dinner

You can meet Dave Michelson and Chris Cobbold at a no-host dinner before the talk. Please join us at 6:00 PM at Earl's, 4361 Kingsway (one block west of Willingdon) in Burnaby. All are welcome, although advance notice by e-mail to Dave Michelson <dmichelson@ieee.org> would be appreciated so that we can ask for a table of appropriate size.