Multipath Propagation Simulation in Mobile and Wireless - - PowerPoint PPT Presentation

▶
multipath propagation simulation in mobile and wireless
SMART_READER_LITE
LIVE PREVIEW

Multipath Propagation Simulation in Mobile and Wireless - - PowerPoint PPT Presentation

Feb 22, 2023 110 likes •1.08k views

Multipath Propagation Simulation in Mobile and Wireless Communications Application of Ray-Tracing for the Propagation Prediction in Microcellar Environments Jun-ichi TAKADA and Houtao ZHU Tokyo Institute of Technology Microwave Simulator

slide-1
SLIDE 1

Multipath Propagation Simulation in Mobile and Wireless Communications

Application of Ray-Tracing for the Propagation Prediction in Microcellar Environments

Jun-ichi TAKADA and Houtao ZHU Tokyo Institute of Technology

Microwave Simulator Workshop, Mar. 17, 2003 – p.1/28

slide-2
SLIDE 2

Table of Contents

Motivation Environment under consideration Ray tracing simulation Electromagnetic theory Tracing of ray Treatment of phase Validation by field test Conclusion and future works

Microwave Simulator Workshop, Mar. 17, 2003 – p.2/28

slide-3
SLIDE 3

Mobile / Wireless Commuications

Mobile / Wireless Comm.

Buildings BS Antenna Diffraction Reflection

Multiple reflected and diffracted paths are arrival.

Microwave Simulator Workshop, Mar. 17, 2003 – p.3/28

slide-4
SLIDE 4

Impacts of Multipath Propagation

Large propagation loss compared with line-of-sight (LOS) scenarios Fast level fluctuation called fading Time dispersion of channel resulting intersymbol interference (ISI)

Microwave Simulator Workshop, Mar. 17, 2003 – p.4/28

slide-5
SLIDE 5

Purpose of Propagation Simulation

Two different purposes Pathloss prediction for cell site design — Site-specific information is necessary for smaller cells Channel modeling for transmission evaluation — Typical and realistic model is eagerly needed

Microwave Simulator Workshop, Mar. 17, 2003 – p.5/28

slide-6
SLIDE 6

Advanced Transmission Technologies

Dependent on the channel properties SIMO (SISO) systems Equalizer : removal of ISI Interleaver : homogenization of fading Diversity antenna : removal of fading Adaptive array antenna : removal of ISI and CCI (co-channel interference) MIMO (MISO) systems Multiuser detectior : separation of CCI Space division multiplex receiver : parallel spatial channels Transmit diversity by space-time coding : removal of fading

Microwave Simulator Workshop, Mar. 17, 2003 – p.6/28

slide-7
SLIDE 7

Environment under Consideration

This presentation focuses on Outdoor microcellular environment;

Microwave Simulator Workshop, Mar. 17, 2003 – p.7/28

slide-8
SLIDE 8

Environment under Consideration

This presentation focuses on Outdoor microcellular environment; Base station antenna below rooftop;

Microwave Simulator Workshop, Mar. 17, 2003 – p.7/28

slide-9
SLIDE 9

Environment under Consideration

This presentation focuses on Outdoor microcellular environment; Base station antenna below rooftop; Diameter less than 500 m.

Microwave Simulator Workshop, Mar. 17, 2003 – p.7/28

slide-10
SLIDE 10

Environment under Consideration

This presentation focuses on Outdoor microcellular environment; Base station antenna below rooftop; Diameter less than 500 m. Examples : PHS, hot spot wireless access.

Microwave Simulator Workshop, Mar. 17, 2003 – p.7/28

slide-11
SLIDE 11

Outdoor Microcellular Environment

Main scatterers are buildings and ground. Building database required Otherwise : simulation cost >> measurement cost Full utilization of vector database Useless pixel database ⇒ extraction of surfaces ZENRIN Z-map Commercial vector database Polygon plan + numerical height

Microwave Simulator Workshop, Mar. 17, 2003 – p.8/28

slide-12
SLIDE 12

Propagation Mechanisms in Ray Tracing

Implemented

Specular reflection: Fresnel reflection coefficient Edge diffraction: UTD + reflection coefficient (empirical UTD)

Under study Surface roughness Edge roughness Items to be modeled non-specular component — increase of computational cost loss and its fluctuation — stochastic model de-polarization

Microwave Simulator Workshop, Mar. 17, 2003 – p.9/28

slide-13
SLIDE 13

Specular Reflection

θ θ

Fresnel reflection coefficient for infinite thickness is used for simplicity. Finite thickness model does not result in accuracy improvement due to inhomogenious materials.

Microwave Simulator Workshop, Mar. 17, 2003 – p.10/28

slide-14
SLIDE 14

Edge Diffraction

Wedge Keller cone

Keller cone is considered for direction of diffracted waves. UTD diffraction coefficient for conductor or its empirical modefication for dielectric.

Microwave Simulator Workshop, Mar. 17, 2003 – p.11/28

slide-15
SLIDE 15

Ray-Tracing Simulation

Ray-Launching Method Launching to each direction from Tx Capture circle

Tx Rx

Image Method Image source of Tx Huge memory to store various orders of image sources Shadow testing

Rx Rx Tx

I1 I2

Microwave Simulator Workshop, Mar. 17, 2003 – p.12/28

slide-16
SLIDE 16

2D-3D Hybrid Ray-Tracing

  • 1. 2D ray-launching and then 3D ray-path formulation
  • 2. Diffraction edges: treated as new point sources
  • 3. Intersection with points: capture circle
  • 4. Ground reflection : image method

Tx Rx Tx Rx Tx Rx X1 X2 X1 X2 X1 X2

incidence plane diffraction plane unfolded plane

θ θ θ

hT hT hR hR

Microwave Simulator Workshop, Mar. 17, 2003 – p.13/28

slide-17
SLIDE 17

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-18
SLIDE 18

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-19
SLIDE 19

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-20
SLIDE 20

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-21
SLIDE 21

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-22
SLIDE 22

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-23
SLIDE 23

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-24
SLIDE 24

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-25
SLIDE 25

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-26
SLIDE 26

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-27
SLIDE 27

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-28
SLIDE 28

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-29
SLIDE 29

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-30
SLIDE 30

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-31
SLIDE 31

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-32
SLIDE 32

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-33
SLIDE 33

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-34
SLIDE 34

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-35
SLIDE 35

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-36
SLIDE 36

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-37
SLIDE 37

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-38
SLIDE 38

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-39
SLIDE 39

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-40
SLIDE 40

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-41
SLIDE 41

Ray Tracing

Launched ray

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-42
SLIDE 42

Ray Tracing

Launched ray

Candidate surfaces

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-43
SLIDE 43

Ray Tracing

Launched ray

Survivor surface

  • 1. Semi-infinite ray is drawn as a ray.
  • 2. Each wall is checked for crossing.
  • 3. All the candidate walls are determined.

Microwave Simulator Workshop, Mar. 17, 2003 – p.14/28

slide-44
SLIDE 44

Ray Acceleration Techniques

  • 1. Back-face culling
  • 2. Volume bounding
  • 3. Partition vector

Microwave Simulator Workshop, Mar. 17, 2003 – p.15/28

slide-45
SLIDE 45

Back-Face Culling

Only visible faces are tested.

Launched ray

Microwave Simulator Workshop, Mar. 17, 2003 – p.16/28

slide-46
SLIDE 46

Back-Face Culling

Only visible faces are tested.

Normal vectors of faces

Microwave Simulator Workshop, Mar. 17, 2003 – p.16/28

slide-47
SLIDE 47

Back-Face Culling

Only visible faces are tested.

Negative innter product with ray = visible face

Microwave Simulator Workshop, Mar. 17, 2003 – p.16/28

slide-48
SLIDE 48

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-49
SLIDE 49

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-50
SLIDE 50

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-51
SLIDE 51

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-52
SLIDE 52

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-53
SLIDE 53

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-54
SLIDE 54

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-55
SLIDE 55

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-56
SLIDE 56

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-57
SLIDE 57

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes Candidate boundaries

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-58
SLIDE 58

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Bounded volumes Survivor volume

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-59
SLIDE 59

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-60
SLIDE 60

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-61
SLIDE 61

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-62
SLIDE 62

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-63
SLIDE 63

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-64
SLIDE 64

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-65
SLIDE 65

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-66
SLIDE 66

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-67
SLIDE 67

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-68
SLIDE 68

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-69
SLIDE 69

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-70
SLIDE 70

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray Check within candidate volume

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-71
SLIDE 71

Volume Bounding

Instead of each of the buildings, bounded volumes are tested for crossing.

Launched ray

Survivor surface

  • 1. Semi-infinite ray is drawn as a ray.

Microwave Simulator Workshop, Mar. 17, 2003 – p.17/28

slide-72
SLIDE 72

Patition Vectors

Instead of searching all the region, only within the region bounded by partition vectors are tested.

Launched ray Bounded volumes Partition vectors Search area

Microwave Simulator Workshop, Mar. 17, 2003 – p.18/28

slide-73
SLIDE 73

Properties of Spatio-Temporal Channel Mod

Ray parameters Magnitude and phase Directions of arrival and departure Delay time Doppler frequency

(spatial / temporal properties)

Microwave Simulator Workshop, Mar. 17, 2003 – p.19/28

slide-74
SLIDE 74

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt)

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-75
SLIDE 75

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, Ω

R , ΩT)

=

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt)

R: Receiver

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-76
SLIDE 76

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, Ω

T )

=

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt)

T: Transmitter

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-77
SLIDE 77

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(j ψl ) exp(jfdlt) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt) ψl: Path phase between Tx and Rx origins

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-78
SLIDE 78

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(j fdl t) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt) fdl: Doppler frequency due to motion of scatterer

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-79
SLIDE 79

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk( ΩRl ) · vRt) exp(+jk(ΩTl) · vTt) k(ΩR): Propagation vector toward ΩR

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-80
SLIDE 80

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk(ΩRl) · vR t) exp(+jk(ΩTl) · vTt) vR: Velocity vector of receiver antenna

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-81
SLIDE 81

Model of Single Ray

Dyadic (vector to vector) complex path gain between Tx and Rx origins

γl(t, τ, ΩR, ΩT) =

  • ˆ

θR(ΩRl) ˆ ϕR(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    ˆ θT(ΩTl) ˆ ϕT(ΩTl)   · δ(ΩR − ΩRl)δ(ΩT − ΩTl)δ(τ − τl) · exp(jψl) exp(jfdlt) exp(−jk(ΩRl) · vRt) exp(+jk(ΩTl) · vTt) Result of ray tracing

Microwave Simulator Workshop, Mar. 17, 2003 – p.20/28

slide-82
SLIDE 82

Model of Multipath

Γ(t, τ, ΩR, ΩT) =

L

  • l=1

γl(t, τ, ΩR, ΩT)

Microwave Simulator Workshop, Mar. 17, 2003 – p.21/28

slide-83
SLIDE 83

Ray-Based Channel Response Model

h(t, τ) = eR(ΩR) · Γ(t, τ, ΩR, ΩT) · e∗

T(ΩT)dΩRdΩT

=

L

  • l=1
  • eRθ(ΩRl)

eRϕ(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    e∗

Tθ(ΩTl)

e∗

Tϕ(ΩTl)

 

Microwave Simulator Workshop, Mar. 17, 2003 – p.22/28

slide-84
SLIDE 84

Ray-Based Channel Response Model

h(t, τ) = eR(ΩR) · Γ(t, τ, ΩR, ΩT) · e∗

T(ΩT)dΩRdΩT

=

L

  • l=1
  • eRθ(ΩRl)

eRϕ(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    e∗

Tθ(ΩTl)

e∗

Tϕ(ΩTl)

  Receiver antenna vector complex directivity

Microwave Simulator Workshop, Mar. 17, 2003 – p.22/28

slide-85
SLIDE 85

Ray-Based Channel Response Model

h(t, τ) = eR(ΩR) · Γ(t, τ, ΩR, ΩT) · e∗

T(ΩT) dΩRdΩT

=

L

  • l=1
  • eRθ(ΩRl)

eRϕ(ΩRl)

 γθθ

l

γθϕ

l

γϕθ

l

γϕϕ

l

    e∗

Tθ(ΩTl)

e∗

Tϕ(ΩTl)

  Receiver antenna vector complex directivity Transmitter antenna vector complex directivity

Microwave Simulator Workshop, Mar. 17, 2003 – p.22/28

slide-86
SLIDE 86

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite.

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-87
SLIDE 87

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite. Four different approaches

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-88
SLIDE 88

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite. Four different approaches

Deterministic phase : Raytracing phase is used.

Not meaningful; limited position accuracy and terminal motion

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-89
SLIDE 89

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite. Four different approaches

Deterministic phase : Raytracing phase is used.

Not meaningful; limited position accuracy and terminal motion

Power summing : Rays are incoherently summed.

Estimation of average; no fading fluctuation

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-90
SLIDE 90

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite. Four different approaches

Deterministic phase : Raytracing phase is used.

Not meaningful; limited position accuracy and terminal motion

Power summing : Rays are incoherently summed.

Estimation of average; no fading fluctuation

Random phase : Each ray has a random value of phase.

Model of fading instant; analytical PDF

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-91
SLIDE 91

Treatment of Phase

Effect of phase can be found when bandwidth and baamwidth are finite. Four different approaches

Deterministic phase : Raytracing phase is used.

Not meaningful; limited position accuracy and terminal motion

Power summing : Rays are incoherently summed.

Estimation of average; no fading fluctuation

Random phase : Each ray has a random value of phase.

Model of fading instant; analytical PDF

Dynamic phase : Motion of terminal is considered.

Phase rotation due to Doppler; similar to “Jakes model”

Microwave Simulator Workshop, Mar. 17, 2003 – p.23/28

slide-92
SLIDE 92

Effect of Finite Beam/Bandwidth

Impulse response of system shall be convolved to the ray- tracing channel model.

Microwave Simulator Workshop, Mar. 17, 2003 – p.24/28

slide-93
SLIDE 93

Field Test

PN correlation sounder ⇒ delay spectrum Rotating parabolic antenna ⇒ angular spectrum Smoothing over 30cm × 30cm area to remove fading frequency

8.45 GHz

bandwidth

100 MHz

delay resolution

20 ns

Tx antenna vertical halfwave dipole Rx antenna V-pol 50 cm parabola beamwidth

4◦

Microwave Simulator Workshop, Mar. 17, 2003 – p.25/28

slide-94
SLIDE 94

Test Site Yokosuka Highland area, Japan

Residential area with wooden houses and concrete walls

North 20 m

90 deg 0 deg

Average Building Height: 8 m

Tx Rx

Rx: height 4.4 m Tx: height 2.7 m

Microwave Simulator Workshop, Mar. 17, 2003 – p.26/28

slide-95
SLIDE 95

Azimuth Delay Spectrum

Experiment

  • 100
  • 90
  • 80
  • 70
  • 60
  • 50

40 80 120 160 200 240 280 320 500 1000 1500 2000 2500 360

Path gain [dB] Azimuth angle [deg] Delay [ns]

Simulation

40 80 120 160 200 240 280 320 500 1000 1500 2000 2500 360

  • 100
  • 90
  • 80
  • 70
  • 60
  • 50

Path gain [dB] Azimuth angle [deg] D e l a y [ n s ]

Both results are in agreement w.r.t. dominant signals. Non-specular components are observed in experiment.

Microwave Simulator Workshop, Mar. 17, 2003 – p.27/28

slide-96
SLIDE 96

Conclusion and Future Works

Conclusion Ray-based model is flexibly applied to SIMO and MIMO transmission. Ray tracing simulator for microcell environment is presented. The simulator is validated by field test. Commercial softwares are available; rather few validation data. Future works Modeling of non-specular scattering effect 3D propagation mechanism (e.g. path over the roof top) Birth and death of ray ⇒ shadowing

Microwave Simulator Workshop, Mar. 17, 2003 – p.28/28