Cneter for PersonKommunikation 12/09/2018 (c) Patrick Eggers 1 APNET (c) Patrick Eggers 12/09/2018 1 Antennas & Propagation, 9sem WCS MM7 : Ray tracing and UTD Patrick Eggers 13/9-2018, A3-207, 8.15-12.00 APNET (c) Patrick Eggers Self study presentation ‘Int. equation’ • Part1 : Boundary conditions / materials – Dielectric interface -> show derivation of expression for reflection coefficient (Vertical polarization) – PEC, PMC: what is it/ what does it ‘do’ (EM wise) • Remember wrt exercises: we need some documentation back from you (written form, showing understanding of some of the core aspects) 12/09/2018 2 APNET (c) Patrick Eggers 12/09/2018 3 Contents • I: MANAGEMENT (What we do with all those rays) – Data base • Raster • Vector • Content (materials, detail) – Ray tracing – Ray launcing – 2D,2½D vs full 3D • II : ‘The ENGINE’ (How we do it pr single ray) – UTD & Diffraction APNET (c) Patrick Eggers 12/09/2018 4 Objects -> polygons • Facets - objects – # facets – # vertices – Coordinates of vertices – Material types (,,+ surface roughness) – Possible object composition (window vs brick ratio etc) • Scenery – data base – # objects – Placement/coordinate of objects – # facets for ground – Ground data like for objects
11
Embed
Self study presentation ‘Int. equation’kom.aau.dk/~pe/education/menu/9sem/AP_MM7_18... · ’spill rays’ (as is shooting ’in blind’) • Tree branches need ’death criteria’
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
– Dielectric interface -> show derivation of expression for reflection coefficient (Verticalpolarization)
– PEC, PMC: what is it/ what does it ‘do’ (EM wise)
• Remember wrt exercises: we need some documentation back from you (written form, showing understanding of some of the core aspects)
12/09/2018 2
APNET
(c) Patrick Eggers12/09/2018 3
Contents• I: MANAGEMENT (What we do with all those
rays)– Data base
• Raster• Vector• Content (materials, detail)
– Ray tracing– Ray launcing– 2D,2½D vs full 3D
• II : ‘The ENGINE’ (How we do it pr single ray)– UTD & Diffraction
APNET
(c) Patrick Eggers12/09/2018 4
Objects -> polygons• Facets - objects
– # facets– # vertices– Coordinates of vertices– Material types (,, + surface roughness)– Possible object composition (window vs brick ratio
etc)• Scenery – data base
– # objects– Placement/coordinate of objects– # facets for ground– Ground data like for objects
Cneter for PersonKommunikation 12/09/2018
(c) Patrick Eggers 2
APNET
(c) Patrick Eggers12/09/2018 5
Ray tracing (’tracking’ possible exact)• Connecting paths Tx-Rx• Image = efficient for for small problems
APNET
(c) Patrick Eggers12/09/2018 6
Ray shooting &bouncing (launching ’blindly’)
• Foreward (direct) tracing from Tx -> Rx• Ray tubes -> diverge -> large overhead of
’spill rays’ (as is shooting ’in blind’)• Tree branches need ’death criteria’ -> acc.
Techniques. + RECEPTION AREA
APNET
(c) Patrick Eggers12/09/2018 7
Ray disturbances - tracking• Example : limit possibilities to most dominating
contributions1. Direct ray2. Single reflection3. Double reflection4. Single diffraction5. Triple reflection6. Single reflection + single diffraction7. Double diffraction8. Double reflection + single diffraction9. (x flections or possible scattering = diffuse
reflection)
– Transmission (penetration ? If out to indoors )
APNET
(c) Patrick Eggers12/09/2018 8
Class 1:4 in urban scenario
www.awe-communications.com
Cneter for PersonKommunikation 12/09/2018
(c) Patrick Eggers 3
APNET
(c) Patrick Eggers12/09/2018 9
Class 5:8 in urban scenario
www.awe-communications.comMore components -> more detail and extension/ ‘contrast’
APNET
(c) Patrick Eggers12/09/2018 10
Search tree
Verifying path loss and delay spread predictions of a 3D ray tracing propagation model in urban environmentTerhi Rautiainen1, Gerd Wölfle2, Reiner Hoppe356th IEEE Vehicular Technology Conference (VTC) 2002 - Fall, Vancouver (British Columbia, Canada), Sept. 2002
• Determination of all visible objects
• Computation of the angles
• Recursively processing of angular conditions
• Tree structure– Fast and
efficient
APNET
(c) Patrick Eggers12/09/2018 11
Building data base• Structure• Resolution (raster, fx 10m)• Content (matrials, detail (windows metal
etc))• Complexity (2D, 2½D, 3D)
Fast 3-D Ray Tracing for the Planning of Microcells by Intelligent Preprocessing of the Data BaseR. Hoppe, G. W¨olfle, and F. M. Landstorfer3rd European Personal and Mobile Communications Conference (EPMCC) 1999, Paris, Mar. 1999
Raster Vector
APNET
(c) Patrick Eggers12/09/2018 12
Data base accuracy
www.awe-communications.com
Cneter for PersonKommunikation 12/09/2018
(c) Patrick Eggers 4
APNET
(c) Patrick Eggers12/09/2018 13
Material data base
www.awe-communications.com
APNET
(c) Patrick Eggers12/09/2018 14
Propagation models
www.awe-communications.com
”..most cases one propagation path contributes more than 90% of the total energy..”
Only power..’sell’ dispersion ’capability’ of full ray tracing to gain simplicity
APNET
(c) Patrick Eggers12/09/2018 15
Dominat path field strength – empirical input• Path length d• Path loss exponents before and after breakpoint p• individual interaction losses f(φ,i) for each
interaction i of all n interactions• Gain due to waveguiding wk at c pixels along the
path• Gain gt of base station antenna• Power pt of transmitter
Break point pathlosswaveguideinteractions
www.awe-communications.com
APNET
(c) Patrick Eggers12/09/2018 16
Computation time
• (1 GHZ PENTIUM® III, desktop relevant year 2002)
• Scenario Helsinki Downtown
• No. of buildings 1150
• Area 2 km2
• Resolution 8 m -> raster
• Database preprocessing (only once) 312 min
• Prediction time: Tx above rooftop 20 s -> why?/horizon
• Prediction time: Tx below rooftop 10 s -> why?/horizon
Verifying path loss and delay spread predictions of a 3D ray tracing propagation model in urban environmentTerhi Rautiainen1, Gerd Wölfle2, Reiner Hoppe356th IEEE Vehicular Technology Conference (VTC) 2002 - Fall, Vancouver (British Columbia, Canada), Sept. 2002
Blue ->over estim. loss Yellow -> over est. loss More green?
APNET
(c) Patrick Eggers12/09/2018 20
Helsinki : statistical evaluation
www.awe-communications.com
Cneter for PersonKommunikation 12/09/2018
(c) Patrick Eggers 6
APNET
(c) Patrick Eggers12/09/2018 21
Example urban pathloss
Verifying path loss and delay spread predictions of a 3D ray tracing propagation model in urban environmentTerhi Rautiainen1, Gerd Wölfle2, Reiner Hoppe356th IEEE Vehicular Technology Conference (VTC) 2002 - Fall, Vancouver (British Columbia, Canada), Sept. 2002
Notice street guidance!, i.e. not solely range dependant loss
GO : Beams / rays• The beam direction : steepest change of phase in
every point, also direction of Poynting vector (i.e. of the direction of the energy flow) – Preservation of power– (infinitely) facet dS1 is chosen of the wave surface and a
beam is led through every point of the edge of this facet : i.e. a beam tube
– energy propagates along the beams, it cannot leave the tube through the side walls. In the lossless medium, the power passing facets dS1 and dS2 is identical
• variable s is curvilinear coordinate along the beam
– not valid where the beams cut (-> infinitely high field intensity). occur in the focus and on the surface called causticshttp://www.urel.feec.vutbr.cz/~raida/multimedia_en/chapter-2/2_3A.html
APNET
(c) Patrick Eggers12/09/2018 31
Geometrical Theory of Diffraction (GTD)• Ray Theory• Solves some GO difficulties
The total field at an observation point P is decomposed into GO and diffracted components
The behavior of the diffracted field is based on the following postulates of GTD:1. Wavefronts are locally plane and waves are TEM.2. Diffracted rays emerge radially from an edge.3. Rays travel in straight lines in a homogeneous medium4. Polarization is constant along a ray in an isotropic medium5. The diffracted field strength is inversely proportional to the cross sectional area of the flux tube6. The diffracted field is linearly related to the incident field at the diffraction point by a diffraction coefficient
Cneter for PersonKommunikation 12/09/2018
(c) Patrick Eggers 9
APNET
(c) Patrick Eggers12/09/2018 33
GTD calculationProperties: Conceptionally simple Local phenomena Tracing of diffracted rays Pinpoints flash points Predicts non-zero field in shadow
regions A higher order approximation than GO
in terms of frequency Uniform versions yield smooth and
continuous fields at and around shadow boundaries (transition regions)
Disadvantages: Requires searching for diffraction
points on the edge Requires finding of attachment and
launching points and geodesics on the surface
Fails at caustics where many diffracted rays merge
•notation is borrowed from optics: s = soft or parallel polarization; h = hard or perpendicular polarization
APNET
(c) Patrick Eggers12/09/2018 34
GTD, UTD half plane I
Diffracted ray
Etotal = EGO+Ed(GTD or UTD)
|E(GO)|
jkses
1
|Ed(GTD)|
|Ed(UTD)| Geometrical Optics (’B or W’)Geometrical theory of diffraction(smoother B vs W region Transition, but singularity)Uniform theory of diffraction(fix of GTD at singularity)