Benedetto Villa 
  
URBAN 3-D MODELS FOR PLANNIG OF MOBILE TELEPHONY SYSTEMS 
Francesco COPPOLA", Mauro LO BRUTTO ' 
Patrizia MIDULLA ', Benedetto VILLA" 
"University of Palermo, Italy 
Department of Representation 
bevilla 9 unipa.it 
"University of Palermo, Italy 
Town and Land Department 
pmidulla@unipa.it 
Working Group V/3 
KEY WORDS: CAD, Modelling, Photogrammetry, Rendering, Texture Mapping. 
ABSTRACT 
This paper deals with an application of numerical cartography, carried out by means of digital photogrammetry 
techniques, aimed at the planning of mobile telephony cells. Particularly, urban 3-D vector models (wireframe), got by 
digital restitution and integrated by raster images referred to rectified buildings facades, were carried out in a sample 
area selected inside the Orleans Park, seat of the Palermo University. Then such models were processed by computer 
graphics techniques owing to the operations of rendering and animation. These 3-D representations could come in very 
useful for study and planning of mobile telephony networks in urban areas. 
1. INTRODUCTION 
Problems referred to 3-D urban models are now the subject of numerous researches aimed, particularly, at carrying out 
automatic techniques for extraction of buildings from DTM (Brenner, 1996; Forstner, 1996; Gruen and Wang, 1996) 
and visualization of urban 3-D data (Gruber, 1996). 
In various applications - such as the urban planning, the positioning of telecommunication antennas, or the simulation of 
radio waves propagation - data obtained by a 3-D rendered model, which have both the metric information of the 
cartography and the qualitative one of the photographs (kind of building, features of materials), are of great interest. 
Integration between simple volumetric reconstruction of buildings and rendering allows virtual reality model; besides, 
animations carried out along fixed paths give an opportunity to move on the inside of the same model. 
For many years rendering techniques have been prerogative of fields, such as publicity graphics, whose metric side was 
far less important than the qualitative one. Therefore it’s obvious that rendering software favours the qualitative side of 
the representation. 
Sample area is located besides Orleans Park, the campus site of Palermo University (fig. 1), which has been chosen for 
a long time by researchers of topographic sector as the laboratory in which to experiment with new techniques of 
surveying and representation. 
2. TOPOGRAPHIC SURVEY 
Photogrammetric survey, both terrestrial and aerial, required a topographic basic network aimed at locating planimetric 
and/or altimetric control points. This network, surveyed by a Topcon GTS 702 total station, was adjusted according to a 
local reference system (fig. 2) and afterwards fitted in the national one (Gauss-Boaga) using a strict adjustment software 
(ReTop rel. 2.0), worked out with Windows 95 at the Department of Representation. To fit in with the national 
reference system two fixed points were used, belonging to a GPS network carried out for a mapping on a large scale of 
Orleans Park. 
Measurements of the necessary control points to survey the facades were carried out by a Leica DIOR 3002S electronic 
distantiometer, integrated to a Leica T1010 electronic theodolite. 
In the facades without well identifiable architectural particulars on the photos, presignalization was used, resorting to 
adhesive black-and-white signals having size of 10cm x 10cm. About 200 control points were surveyed in all. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000. 919