EE359 – Lecture 2 Outline Makeup lecture options

l Announcements l Next Thursday

st
l 1 HW posted, due next Friday at 4pm. l Monday (1/13): 11:30-12:50pm w/lunch
l Discussion section and TA OHs starts next week l Tuesday (1/14): 3:00-4:20 or 4:30-5:50pm (+lecture 1:30-2:50)
l Will finalize makeup lecture schedule today l Wednesday (1/15): 9:00-10:20am or 11:30-12:50pm*
l Review of Last Lecture l Regular Time Thu 1:30-2:50pm with Tom giving lecture
l Friday (1/17): 9:00-10:20am* or 3:30-4:50pm
l TX and RX Signal Models l Following Monday is MLK 3-day weekend
l Path Loss Models l Thu 2/20 moved to Fri 2/21 from 3:30-4:50pm
l Free-space and 2-Ray Models l MT week, could do Fri 9am instead Regular
l General Ray Tracing l Tue 2/4 moved to Wed 2/5 11:30-12:50pm w/lunch time
l Single-Slope Path Loss Exponent Model with Tom
l Empirical Models l Thu 3/5 moved to Wed 3/4 11:30-12:50pm w/lunch
l mmWave Models l Could do Wed am or Friday 9am or 3:30 instead
*Pending availability

1 2

Lecture 1 Review Propagation Characteristics

l Course Information l Path Loss (includes average shadowing)
l Wireless Vision l Shadowing (due to obstructions)
l Technical Challenges l Multipath Fading
l Current/Next-Gen Wireless Systems
Slow
l Spectrum Regulation and Standards Pt Pr
Pr/Pt Fast
Very slow
v
l Emerging Wireless Systems
d=vt
Emerging systems can be covered in a bonus lecture d=vt

3 4

1
 Path Loss Modeling Free Space (LOS) Model
l Maxwell’s equations
l Complex and impractical
d=vt
l Free space and 2-path models
l Too simple
l Ray tracing models l Path loss for unobstructed LOS path
l Requires site-specific information l Power falls off :
l Single-slope path loss exponent model l Proportional to 1/d2
l Main characteristics: good for high-level analysis l Proportional to l2 (inversely proportional to f2)
l Measurement-based and Standards Models l This is due to the effective aperture of the antenna
l Not accurate; used to assess different designs

5 6

Two Ray Model General Ray Tracing

l Models signal components as particles
l Reflections
l Path loss for one LOS path and 1 ground (or l Scattering
reflected) bounce l Diffraction
l Ground bounce approximately cancels LOS
path above critical distance Reflections generally dominate
l Power falls off l Requires site geometry and dielectric properties
l Proportional to d2 (small d) l Easier than Maxwell (geometry vs. differential eqns)
l Proportional to d4 (d>dc)
l Independent of l (fc) l Computer packages often used
l Two-path cancellation equivalent to 2-element array, i.e. the
effective aperature of the receive antenna is changed.
10-ray reflection model explored in HW

7 8

2
 Simplified Path Loss Model mmWave: What’s the big deal?
l Used when path loss dominated by
reflections.
l Most important parameter is the path loss
exponent g, determined empirically.

g
éd ù
Pr = Pt K ê 0r ú , 2£g £8
ëd û
All existing commercial systems fit into a small fraction of the mmWave band

9 10

mmWave Propagation mmW Empirical Channel Models

Massive
(60-100GHz) MIMO (not covered in lecture, not on HW/exams)
l Channel models immature l Early cellular empirical models:
l Empirical path loss models for early cellular systems were based on
l Based on measurements, few accurate analytical models extensive measurements.
l Path loss proportion to l2 (huge) l Okumura model: empirically based (site/freq specific), uses graphs
l Hata model: Analytical approximation to Okumura
l Also have oxygen and rain absorbtion
l Cost 231 Model: extends Hata to higher freq. (2 GHz)
l l is on the order of a water molecule l Multi-slope model
l Walfish/Bertoni: extends Cost 231 to include diffraction
l Current cellular models (LTE and 5G):
l Detailed path loss models for UE (3GPP TS 36.101) and base stations (3GPP
TS 36.104) for different multipath delay spreads, user speeds and MIMO
antenna correlations.
l The 5G model includes higher frequencies (up to 100 GHz).
l WiFi channel models: TGn and TGac
l Indoor and outdoor path loss models with MIMO (4x4 & greater), 40
MHz channels (& greater), and different multipath delay spread.
mmWave systems will be short range or require “massive MIMO” Commonly used in cellular and WiFi system simulations
11 12

3
 Main Points
l Path loss models simplify Maxwell’s equations
l Models vary in complexity and accuracy
l Power falloff with distance is proportional to d2 in
free space, d4 in two path model
l Main characteristics of path loss captured in
single-slope exponent model Pr=PtK[dr/d]g
l mmWave propagation models still immature
l Path loss large due to frequency, rain, and oxygen

l Empirical models used in simulations

13