Upload
rozene
View
23
Download
0
Embed Size (px)
DESCRIPTION
EE359 – Lecture 3 Outline. Announcements HW posted, due Friday 10/4 at noon (extended deadline) . Regular OHs today – 10:45-11:45. Extra OHs for me Thursday 5-6pm. Mainak’s 2 nd OH: Wed 7-8 pm (via Hangout/ Webex ; see web) - PowerPoint PPT Presentation
Citation preview
EE359 – Lecture 3 Outline
Announcements HW posted, due Friday 4pm Discussion section starts tomorrow, 4-5pm, 364
Packard TA OHs start this week We will have to reschedule 1 more class, on election
day (Nov. 8), MT is also that week. Options are M am, W or F. Will send Piazza poll.
Log Normal ShadowingCombined Path Loss and
ShadowingOutage ProbabilityModel Parameters from
Measurements
Lecture 2 Review Signal Propagation Overview Ray Tracing Models Free Space Model
Power falloff with distance proportional to d-2
Two Ray ModelPower falloff with distance proportional
to d-4
Simplified Model: Pr=PtK[d0/d]g, 28.gCaptures main characteristics of path
loss
mmWave propagationEmpirical Models
Shadowing
Models attenuation from obstructions Random due to random # and type of
obstructions Typically follows a log-normal
distributiondB value of power is normally distributedm=0 (mean captured in path loss), 4<s<12
(empirical)CLT used to explain this modelDecorrelates over decorrelation distance
Xc
Xc
Combined Path Loss and Shadowing
Linear Model: y lognormal
dB Model
d
dK
P
P
t
r 0
,log10log10)(0
1010 dBt
r
d
dKdB
P
P
Pr/Pt
(dB)
log d
Very slow
Slow10logK
-10g
),0(~ 2 NdB
Outage ProbabilityPath loss only: circular “cells”;
Path loss+shadowing: amoeba-shaped cells
Outage probability: probability received power falls below given minimum:
For log-normal shadowing
model
rP
Model Parameters from
Empirical Measurements
Fit model to data
Path loss (K,g), d0 known:“Best fit” line through dB dataK obtained from measurements at d0.
Or can solve for (K,g) simultaneously (least squares fit)
Exponent is MMSE estimate based on data
Captures mean due to shadowingShadowing variance
Variance of data relative to path loss model (straight line) with MMSE estimate for g
Pr(dB)
log(d)10g
K (dB)
log(d0)
sy2
Main Points
Random attenuation due to shadowing modeled as log-normal (empirical parameters)
Shadowing decorrelates over decorrelation distance
Combined path loss and shadowing leads to outage and non-circular coverage areas for WiFi/cellular
Path loss and shadowing parameters obtained from empirical measurements through a least-squares fitMatches environment in which
measurements are taken.Can do a 1D fit with K fixed or a 2D fit over
K and .g