Wireless Communication

EC4011D

Dr. Nithin V. Sabu

Department of Electronics and Communication Engineering
National Institute of Technology Calicut

1 / 77
References

• Wireless Communications - Andrea Goldsmith

• Wireless Communications: Principles and Practice - Theodore Rappaport

2 / 77
Introduction to Propagation Modeling

• Predict how radio waves travel through different

environments.
• Essential for designing and optimizing wireless
communication systems.
• Applications:
• Mobile Network Planning
• Interference Analysis
• Coverage Prediction
• Spectrum Management

3 / 77
Types of Propagation Models

Small Scale Propagation Models

Large Scale Propagation Models
• Focus on rapid changes in signal
• Focus on predicting the average signal
strength over short distances(a few
strength over large distances.
wavelengths) or short time intervals.
• Received signal varies over distance due
• Influenced by multipath propagation,
to path loss and shadowing effects.
Doppler shift, and time dispersion.
• Received power variations due to path
• Statistical models are used when the
loss and shadowing occurs over a large
number of multipath components is
distance
large or the environment is unknown.

4 / 77
Path Loss and Shadowing in Large Scale Models

Shadowing
Path Loss • Caused by obstacles between the trans-
• Caused by dissipation of the power ra- mitter and receiver that attenuate sig-
diated by the transmitter as well as by nal power through absorption, reflec-
effects of the propagation channel. tion, scattering, and diffraction.
• Same at a given transmit–receive dis- • Variation occurs over distances that are
tance. proportional to the length of the ob-
• Power variation due to path loss occurs structing object.
over long distances (100–1000 m). • 10–100 m in outdoor environments and
less in indoor environments.

5 / 77
Comparison

Figure: Path loss, Shadowing and Multipath

6 / 77
Signal Modeling

• All the transmitted and received signals that we consider are real.
• We model communication channels using a complex frequency response for analytical
simplicity.
• We model the transmitted signal as

s(t) = Re{u(t)e j2πfc t }

• u(t) : complex envelope or equivalent lowpass signal of s(t).

7 / 77
Free Space Propagation Model

• To predict received signal strength when there is a clear direct path between the Tx
& Rx.
• Eg: Satellite links, Microwave transmission tower etc.
• The channel model associated with this transmission is called a line-of-sight (LOS)
channel, and the corresponding received signal is called the LOS signal or ray.
• Free-space path loss introduces a complex scale factor , resulting in the received signal
 √ −j2πd/λ 
λ Gl e j2πfc t
r (t) = Re u(t)e .
4πd

8 / 77
Free Space Propagation Model
√
• Gl : product of the transmit and receive antenna field radiation patterns in the
LOS direction.
• e −j2πd/λ : Phase shift due to the distance d that the wave travels.
•
√ 2
Pr Gl λ
= .
Pt 4πd
Gt , Gr : antenna gains of transmitter and receiver.
• Received signal power falls off in inverse proportion to the square of the distance.
• Also proportional to the square of the signal wavelength. (dependence of received
power on the λ is due to the effective area of the receive antenna)

9 / 77
Free Space Propagation Model

• The power density (in Watts per square meter) of the plane wave incident on the
receive antenna a distance d is
Pt
p= . (Pt : Transmitted power)
4πd 2
• If the transmit antenna has an antenna gain in the direction of the receive antenna
given by Gt , then,
Pt
p= Gt .
4πd 2
10 / 77
Free Space Propagation Model

• The power received by the antenna having an effective aperture Ae is

Pt
Pr = Gt Ae ,
4πd 2
2
λ c 2πc
where Ae = 4π G (G : Gain of the antenna) and λ = f = ωc .
• The resulting received power is,
Pt Gt Gr λ2
Pr = . (L : System loss factor.)
(4πd)2 L
Friis Free Space Equation
11 / 77
Free Space Propagation Model

• L is due to transmission line attenuation, filter losses, and antenna losses.

• Power decays with distance at a rate of 20 dB/decade.
• The path loss, which represents signal attenuation, defined as the difference(in dB)
between the Pt and Pr is

Gt Gr λ2
 
Pt
PL (dB) = 10 log = −10 log .
Pr (4π)2 d 2

(may or may not include the effect of antenna gains)

12 / 77
Validity of Friis Equation

• Valid only in the far-field region.

• Far-field or Fraunhofer region: The region beyond the far-field distance df .
• df is related to the largest linear dimension of the Tx antenna aperture D and the λ
as,

2D 2
df = .
λ

• Also df must satisfy df >> D & df >> λ.

13 / 77
Problem (Rap. Example 3.2)

If a transmitter produces 50 watts of power, express the transmit power in units of (a)
dBm, and (b) dBW. If 50 watts is applied to a unity gain antenna with a 900 MHz carrier
frequency, fnd the received power in dBm at a free space distance of 100 m from the
˙ km)? Assume unity gain for the receiver antenna.
antenna. What is Pr (l0

14 / 77
Solution:
Given :
Given that Pt = 50W. and, fc = 900 MHz.
(a) Transmitter power in dBm,

Pt (dBm) = 10 log [Pt (mW) / (1 mW)] = 10 log 50 × 103 = 47.0 dBm.

 

(b) Transmitter power in dBW,

Pt (dBW) = 10 log [Pt (W) / (1W)] =10 log [50] = 17.0 dBW.

The received power is:

Pt Gt Gr λ2 50 (1) (1) (1/3)2
Pr = = = 3.5 × 10−6 W = 3.5 × 10−3 mW
(4π)2 d 2 L (4π)2 (100)2 (1)
Pr (dBm) = 10 log Pr (mW) = 10 log 3.5 × 10−3 mW = −24.5dBm

Similarly you can find Pr for 10km.

15 / 77
Terrestrial Propagation Model

• The LOS path might have obstructions (Buildings, terrain, vegetation, etc.)
• Tx - Rx link is established mainly through:
• Reflection of EM waves from various objects.
• Diffraction around various objects
• Multiple paths arise between Tx and Rx - Results in constructive or destructive
interference
• The two-ray ground reflected model is found to be reasonably accurate. ( Over
distances of several kilometers and tall Tx tower (>50m))
• Used when a single ground reflection dominates the multipath effect.
• Curvature of earth is neglected: Flat-earth model is used.

16 / 77
Recap

• Large Scale Propagation Models

• Predicts the average signal strength over large distance.
• Path loss and shadowing effects
• Small Scale Propagation Models
• Focus on rapid changes in signal strength over short distances.
• Multipath propagation, Doppler shift, and time dispersion.
• Path Loss
• Caused by dissipation of the power radiated.
• Shadowing
• Caused by obstacles between the transmitter and receiver.

17 / 77
Recap

• Transmitted signal:

s(t) = Re{u(t)e j2πfc t }

• Free Space Propagation Model:
• Received signal
 √ −j2πd/λ 
λ Gl e j2πfc t
r (t) = Re u(t)e .
4πd
• Received power is,

Pt Gl λ2
Pr = 2.
(4πd)

18 / 77
Ground Reflection (Two-Ray) Model

• The received signal for the two-ray model is

 √ √ ′
Gl u(t)e −j2πl/λ R Gr u(t − τ )e −j2π(x+x )/λ j2πfc t
 
λ
r2−ray (t) = Re + e ,
4π l x + x′

• τ = (x + x ′ − l)/c : Time delay between ground reflected ray and the LOS ray.
√ √
• Gl = Ga Gb : Product of the transmit and receive antenna field radiation patterns
in the LOS direction.
19 / 77
Two-Ray Model
√ √
• Gr = Gc Gd : Product of the transmit and receive antenna field radiation
patterns corresponding to the rays of length x and x ′ .
• R : Ground reflection coefficient.
sin θ − Z
R= ,
sin θ + Z
where (p
εr − cos2 θ/εr for vertical polarization,
Z= p
εr − cos2 θ for horizontal polarization,
and εr is the dielectric constant of the ground.
• The delay spread of the two-ray model equals the delay between the LOS ray and the
reflected ray: (x + x ′ − l)/c.

20 / 77
Two-Ray Model

TX

RX

21 / 77
Two-Ray Model
• Path difference:
q q
x + x′ − l =(ht + hr )2 + d 2 − (ht − hr )2 + d 2 .
1/2 1/2
(ht + hr )2 (ht − hr )2
 
=d 1+ −d 1+
d2 d2

• When d is very large compared to ht + hr , (when x is small, (1 + x)1/2 ≈ 1 + x/2)

(ht + hr )2 (ht − hr )2
   
′ 2ht hr
x +x −l ≈d 1+ 2
−d 1+ 2
=
2d 2d d
• Phase difference:
2π(x + x ′ − l) 4πht hr
∆ϕ = ≈ .
λ λd

22 / 77
Two-Ray Model
• u(t) ≈ u(t − τ ). when transmitted signal is narrowband relative to τ .
• With this approximation,
 2 √ √
λ Gl R Gr e −j∆ϕ 2
Pr = Pt + ,
4π l x + x′
where ∆ϕ = 2π(x + x ′ − l)/λ is the phase difference.
• For asymptotically large d, x + x ′ ≈ l ≈ d, θ ≈ 0, Gl ≈ Gr , and R ≈ −1.
 √ 2  √
4πht hr 2 Gl ht hr 2
 
λ Gl
Pr ≈ Pt = Pt ,
4πd λd d2

• In dB,

Pr dBm = Pt dBm + 10 log10 (Gl ) + 20 log10 (ht hr ) − 40 log10 (d).

23 / 77
Two-Ray Model

• For small distances (d < ht ) the two

rays add constructively and the path
loss is slowing increasing.
• For ht < d < dc : sequence of maxima
and minima. (dc : Critical distance)
• At dc , final maximum is reached.
• After that, Pr ∝ 1
d4
• dc = 4ht hr
λ .

24 / 77
Difference Between Free Space Model and Two
Ray Model

√
• In two-ray model, at large distances (d ≫ ht hr ), Pr ∝ d14 , while in free space
model Pr ∝ d12 .
• As a consequence of the assumption that d ≫ hT , hR , The angle ∆ϕ is small and λ
cancels out of the equation, it becomes frequency independent.
• As ht and hr increases, Pr increases.

25 / 77
Problem( AG: 2.5)

Find the critical distance dc under the two-ray model for a large macrocell in a suburban
area with the base station mounted on a tower or building (ht = 20 m), the receivers at
height hr = 3 m, and fc = 2 GHz. Is this a good size for cell radius in a suburban
macrocell? Why or why not?

26 / 77
General Ray Tracing (GRT)
• GRT predicts field strength and delay spread for any building configuration and
antenna placement.
• Requires specific information:
• Building database (height, location, dielectric properties)
• Transmitter and receiver locations relative to buildings
• Site specific: explains the basic mechanism of urban propagation.
• Utilizes geometrical optics to trace:
• Line-of-Sight (LOS) paths
• Reflected signal components
• Diffracted and scattered signal components
• Dominant components:
• LOS and reflected paths
• Diffraction and scattering are significant near diffracting/scattering surfaces

27 / 77
GRT: Diffraction Modeling

Figure: Knife-edge diffraction

• Diffraction occurs when signals bend around obstacles.

• Diffraction results from many phenomena, including
• The curved surface of the earth,
• Hilly or irregular terrain,
• Building edges, or obstructions blocking the LOS path between the transmitter and
receiver

28 / 77
GRT: Diffraction Modeling

• Diffraction can accurately be modelled using the Geometrical Theory of Diffraction

(GTD).
• But is quite complex in modelling wireless channels.
• Diffraction is most commonly modelled by the Fresnel knife-edge diffraction model
because of its simplicity.
• Diffracting object is assumed to be very thin.
• Factors not considered in this model: polarization, conductivity, surface roughness.

29 / 77
GRT: Diffraction Modeling

Figure: Knife-edge diffraction

• As compared to the LOS signal, the diffracted signal travels an additional distance.
• The resulting path difference is given by,
hp p i
∆d = (d + d ′ ) − d 2 − h2 + d ′ 2 − h2

• For small h relative to d and d ′ , (when x is small, (1 − x)1/2 ≈ 1 − x/2)

h2 h2 h2 d + d ′
    
′ ′
∆d ≈ (d + d ) − d 1 − 2 + d 1 − ′ 2 = (1)
2d 2d 2 dd ′
30 / 77
GRT: Diffraction Modeling
• The corresponding phase difference relative to LOS path is given by,

2π∆d π
∆ϕ = ≈ ν 2,
λ 2
where
r
2(d + d ′ )
ν=h : Fresnel–Kirchhoff diffraction parameter.
λdd ′

• The path loss associated with knife-edge diffraction is a function of ν.

• Computation of this path loss is very difficult.
Requires the use of:
• Huygens’s principle, Fresnel zones, and the complex Fresnel integral.

31 / 77
GRT: Diffraction Modeling
• Diffraction loss cannot generally be found in closed form.
• Approximate values of knife-edge diffraction path loss (dB) relative to LOS path loss
are given by,


 20 log10 [.5 − .62ν] −.8 ≤ ν < 0






20 log [.5e −.95ν ]

0≤ν<1

 10
L(ν)dB =
 p
20 log10 [.4 − .1184 − (.38 − 0.1ν)2 ] 1 ≤ ν ≤ 2.4










20 log10 [.225/ν] ν > 2.4


32 / 77
GRT: Diffraction Modeling

• The knife-edge diffraction model yields the following formula for the received
diffracted signal:
n ′
o
r (t) = Re L(v ) Gd u(t − τ )e −j2π(d+d )/λ e j2πfc t ,
p

√
Gd : antenna gain,τ = ∆d/c: delay associated with the diffracted ray relative to
the LOS path.

33 / 77
Scattering in Cellular Communication

• Scattering occurs when the transmitted signal encounters rough surfaces or small
objects, causing the signal to be reflected in many directions.
• Causes of Scattering:
• Buildings with irregular surfaces
• Trees and foliage
• Vehicles and other small obstructions

34 / 77
Scattering in Cellular Communication

• Effects of Scattering:
• Multipath propagation: Multiple signal paths reaching the receiver
• Signal dispersion: Spreading of the signal in time and frequency
• Increased delay spread: Variability in signal arrival times
• Importance in Cellular Communication:
• Can improve signal coverage in urban environments
• May cause interference and reduce signal clarity
• Must be considered in network design and optimization

35 / 77
GRT: Scattering

• A scattered ray, shown by the segments s and s ′ , has a path loss proportional to the
product of s and s ′ .
• This multiplicative dependence is due to the extra spreading loss that the ray
experiences after scattering.

36 / 77
GRT: Scattering
• The received signal due to a scattered ray is given by the bistatic radar equation
√ ′
λ Gs σe −j2π(s+s )/λ j2πfc t
 
r (t) = Re u(t − τ ) e ,
(4π)3/2 ss ′

where τ = (s + s ′ − l)/c : delay associated with the scattered ray; σ : radar

cross-section of the scattering
√ object, which depends on the roughness, size, and
shape of the scatterer; and Gs : antenna gain.
• This model assumes that the signal propagates from the Tx to the scatterer based on
free-space propagation.
• It is then re-radiated by the scatterer with transmit power equal to σ times the
received power at the scatterer.

37 / 77
GRT: Scattering

• The received power in the event of scattering can be written as,

Pr dBm = Pt dBm + 10 log10 (Gs ) + 20 log10 (λ) + 10 log10 (σ)

− 30 log(4π) − 20 log10 (s) − 20 log10 (s ′ ).

• Empirical values of 10 log10 σ : Ranges from - 4. 5 dBm2 to 55. 7 dBm2 .

• The received signal is determined from the superposition of all the components due
to multiple rays.

38 / 77
GRT: Combined Effect
• If we have a LOS ray, Nr reflected rays, Nd diffracted rays, and Ns diffusely scattered
rays, the total received signal is,
  √
q ′′
j2πd ′ /λ Nr Rd ′′ Gd ′′ u(t − τi )e −j2πdi /λ
λ Gd u(t)e
′ X i i
rtotal (t) = Re +
4π l di′′
i=1
Nd
4π q ′
Lj (v ) Gdj u(t − τj )e −j2π(dj +dj )/λ
X
+
λ
j=1
Ns p ′
Gsk σk u(t − τk )e −j2π(sk +sk )/λ j2πfc t
X  
+ √ ′
e ,
k=1
4πs k sk
τi , τj ,and τk : time delays of the given reflected, diffracted, and scattered rays-
normalized to the delay of the LOS ray.
• The received power Pr of rtotal (t) and the corresponding path loss Pr /Pt are then
obtained from the above equation. 39 / 77
GRT: Combined Effect

√
• Gd ′ : Product of the transmit and receive antenna field radiation patterns in the
LOS direction.
• Rd ′′ : Ground reflection coeff. for the i th reflected path,
qi
• Gdi′′ : Product of Tx and Rx antenna field radiation patterns corresponding to the
length di′′
• Gdj is the Tx antenna gain for the j th diffracted ray,
• Lj (u) is the knife-edge diffraction loss parameter.

40 / 77
Introduction to Empirical Path-Loss Models

• Mobile communication systems operate in complex propagation environments.

• Free-space path loss and ray tracing are often insufficient for accurate modeling.
• Empirical path-loss models are developed based on measurements over specific
distances and frequency ranges.
• These models are used for various environments: urban macrocells, urban microcells,
and indoor settings.

41 / 77
Analytical vs. Empirical Models

• Analytical Models
• Characterize Pr /Pt as a function of distance.
• Path loss is well-defined.
• Empirical Models
• Include effects of path loss, shadowing, and multipath.
• Received power measurements and corresponding path loss are averaged over several
wavelengths to remove multipath effects.
• This average path loss is called Local Mean Attenuation (LMA) (at a distance d).

42 / 77
Empirical Path-Loss Models

• LMA depends on the specific location of the transmitter and receiver.

• Measurements are taken throughout the environment and averaged to characterize
LMA.
• Empirical path loss PL(d) is the average of LMA measurements at distance d.
• Example: Empirical path loss for a generic downtown area with a rectangular street
grid might be obtained by averaging LMA measurements in New York City,
downtown San Francisco, and downtown Chicago.
• Models: Okumura Model, Hata Model,COST 231 Extension, etc.

43 / 77
Okumura Model

• Most common model for signal prediction in large urban macrocells.

• Applicable over distances of 1-100km and freq. range of 150-1500MHz (and is
extrapolated upto 3GHz).
• Okumura did extensive measurements of BS-to-MS link over Tokyo.
• Developed a set of curves giving median attenuation relative to free space
propagation in irregular terrain.
• The BS heights were 30-100m.

44 / 77
Okumura Model

• The empirical path loss formula of Okumura at distance d at a carrier freq. fc is

given by,

PL (d) dB = L(fc , d) + Aµ (fc , d) − G (ht ) − G (hr ) − GAREA ,

where, L(fc , d): free-space path loss at distance d and carrier frequency fc ,
Aµ (fc , d): median attenuation in addition to free-space path loss across all
environments, G (ht ) : base station antenna height gain factor, G (hr ) : mobile
antenna height gain factor, and GAREA : gain due to the type of environment.

45 / 77
Okumura Model
• Aµ (fc , d) and GAREA are obtained from Okumura’s empirical plots.

46 / 77
Okumura Model

• Okumura derived empirical formulas for G (ht ) and G (hr ) as follows:

G (ht ) = 20 log10 (ht /200), 30m < ht < 1000m;


10 log10 (hr /3) hr ≤ 3m,
G (hr ) =
20 log10 (hr /3) 3m < hr < 10m.
• Some correction factors related to terrain are also developed to increase the accuracy
of the model.

47 / 77
Hata Model

• Empirical formulation of the graphical path-loss data provided by Okumura.

• Valid over roughly the same range of frequencies, 150–1500 MHz.
• Due to closed form, it simplifies calculation of path loss and is not based on
empirical curves for the different parameters.
• The standard formula for empirical path-loss as given by the Hata model is,

PL,urban (d)dB = 69.55 + 26.16 log10 (fc ) − 13.82 log10 (ht ) − a(hr )
+ (44.9 − 6.55 log10 (ht )) log10 (d).

• The parameters in this model are the same as under the Okumura model.

48 / 77
Hata Model
• a(hr ) :Correction factor for the mobile antenna height based on the size of the
coverage area.
• For small to medium-sized cities,
a(hr ) = (1.1 log10 (fc ) − .7)hr − (1.56 log10 (fc ) − .8)dB.
• For larger cities at frequencies fc > 300MHz,
a(hr ) = 3.2(log10 (11.75hr ))2 − 4.97dB.
• Corrections to the urban model are made for suburban and rural propagation, so that
these models are (respectively)
PL,suburban (d)dB = PL,urban (d)dB − 2[log10 (fc /28)]2 − 5.4

and
PL,rural (d) dB = PL,urban (d) dB − 4.78[log10 (fc )]2 + 18.33 log10 (fc ) − K ,
where K ranges from 35.94 (countryside) to 40.94 (desert). 49 / 77
Hata Model

• Hata model doesn’t require any other path specific correction factors (as used in
Okumura
• It approximates Okumura model for d > 1km
• A good model for first and even second generation cellular systems.
• But does not model propagation well in cellular systems with smaller sizes and higher
frequencies (like 3G and beyond).

50 / 77
COST 231 Extension to Hata Model

• The Hata model was extended by the European Cooperation in the field of Scientific
and Technical Research (EURO-COST) to 2 GHz as,

PL,urban (d)dB = 46.3 + 33.9 log10 (fc ) − 13.82 log10 (ht ) − a(hr )
+ (44.9 − 6.55 log10 (ht )) log10 (d) + CM

where a(hr ) is the same correction factor as before and where CM is 0dB for
medium sized cities and suburbs and is 3dB for metropolitan areas.
• Referred to as the COST 231 extension to the HATA model.
• Restricted to the following range of parameters:
1.5GHz < fc < 2GHz; 30m < ht < 200m; 1m < hr < 10m, and 1km < d < 20km.

51 / 77
Simplified Path Loss Model

• The complexity of signal propagation makes it difficult to obtain a single model that
accurately characterizes path loss across different environments.
• Accurate path-loss models can be obtained from:
• Complex analytical models
• Empirical measurements
• These models are essential when:
• Tight system specifications must be met.
• Determining the best locations for base stations or access-point layouts.
• For general trade-off analysis of system designs:
• Simple models capturing the essence of signal propagation are preferred.
• Complex path-loss models are approximations to the real channel anyway.

52 / 77
Simplified Path Loss Model
• A simplified model for path loss as a function of distance is given as,
 γ
d0
Pr = Pt K .
d

• The received power is thus

 
d
Pr dB = Pt dB + K dB − 10γ log10 .
d0
K : unitless constant that depends on the antenna characteristics and the average
channel attenuation, d0 : reference distance for the antenna far field, and γ: path-loss
exponent.
• K , d0 , and γ can be obtained to approximate either an analvtical or empirical model.
• Valid only at transmission distances d > d0 , where d0 is typically assumed to be
1 − 10 m indoors and 10 − 100 m outdoors.
53 / 77
Problem (AG: 2-13)

Consider a receiver with noise power -160 dBm within the signal bandwidth of interest.
Assume a simplified path-loss model with d0 = 1 m, K = 5.7 × 10−4 and fc = 1 GHz, and
γ = 4. For a transmit power of Pt = 10 mW, find the maximum distance between the
transmitter and receiver such that the received signal-to-noise power ratio is 20 dB.

54 / 77
Solution

• SNR= Signal Power(Pr )

Noise Power =20dB=100
Noise Power= -160dBm= 10−16 mW= 10−19 W.
Therefore, Pr = 10−17 W.
 γ
• Simplified path-loss model: Pr = Pt K dd0

 4
1
10−17 = 10 × 10−3 × 5.7 × 10−4 ×
d
 4
1
=⇒ = 1.754 × 10−12
d
=⇒ d = 868.89m

55 / 77
Log-Normal Shadowing

• Signal varies due to blockage by objects in the signal path

• Causes of random variations in received power:
• Blockage from objects
• Changes in reflecting surfaces
• Scattering objects
• Need for a model for random attenuation:
• Location, size, and dielectric properties of blocking objects
• Statistical models to characterize attenuation
• Log-normal shadowing model:
• Common model for additional attenuation
• Empirically confirmed for both outdoor and indoor environments.

56 / 77
Log-Normal Shadowing

• The ratio of transmit-to-receive power ψ = Pt /Pr is assumed to be random with a

log-normal distribution given by

(10 log10 ψ − µψdB )2

 
ξ
p(ψ) = √ exp − , ψ > 0,
2πσψdB ψ 2σψ2 dB

ξ = 10/ ln 10, µψdB : mean of ψdB in decibels, σψdB : SD of ψdB .

57 / 77
Log-Normal Shadowing

• The mean and standard deviation can be obtained through analytical modelling or
empirical measurements.
• If ψ is log-normal then the received power and received SNR will also be log-normal,
since these are just constant multiples of ψ.
• For outdoor channels the standard deviation ranges from 4dB to 13dB.

58 / 77
Combined Path Loss and Shadowing
• Path loss and shadowing models can be superimposed
• Combined model captures:
• Power falloff versus distance
• Random attenuation from shadowing
• Components of the combined model:
• Average dB path loss (µψdB ) characterized by the path-loss model
• Shadow fading with a mean of 0 dB causing variations about this path loss

59 / 77
Combined Path Loss and Shadowing

• Combination of the simplified path-loss model, and the log-normal shadowing.

• Ratio of received to transmitted power in dB is given by

Pr d
dB = 10 log10 K − 10γ log10 − ψdB
Pt do

ψdB : Gaussian-distributed random variable with mean zero and variance σψ2 dB .
• Path loss decreases linearly relative to log10 d with a slope of 10γ dB/decade, where
γ is the path-loss exponent.
• The variations due to shadowing change more rapidly.

60 / 77
Outage Probability under Path Loss and Shadowing

• Generally, there is a target min received power level.

• Below this the propagation becomes unacceptable.
• With shadowing, the received power will be log-normal distributed.
• There will be some probability that the received power at a given distance may fall
below, Pmin .
• This is denoted as outage probability,

Pout (Pmin , d) = P [Pr (d) < Pmin ] .

61 / 77
Derivation of Outage Probability

• From combined path loss and shadowing model:

d
Pr dB = Pr dB + 10 log10 K − 10γ log10 − ψdB ,
do
where ψdB is a Gaussian-distributed random variable with mean zero and variance
σψ2 dB .
• Mean of Pr dB : Pr dB + 10 log10 K − 10γ log10 d
do .
• Variance of Pr dB : σψ2 .
dB

62 / 77
Derivation of Outage Probability
• For the combined path loss and shadowing model, the outage probability is given by,

Pout (Pmin , d) = P [Pr (d) ≤ Pmin ]

= 1 − P [Pr (d) > Pmin ]
 
Pmin − (Pt + 10 log10 K − 10γ log10 (d/d0 ))
=1−Q
σψdB

Q-function: probability that a Gaussian random variable X with mean 0 and variance
1 is greater than z:
Z ∞
∆ 1 2
Q(z) = p(X > z) = √ e −y /2 dy .
z 2π
• An outage probability of 1% is a typical target in wireless system design.
63 / 77
Problem

Find the outage probability at 150 m for a channel based on the combined path loss and
shadowing with K dBm=-31.54 dBm, γ=3.71 and σψdBm = 3.65, assuming a transmit
power of 10 mW and minimum power requirement of Pmin = −110.5 dBm.

64 / 77
Solution

• Given that σψdBm = 3.65, γ = 3.71, K = −31.54 dBm, Pmin = −110.5 dBm. and
Pt = 10mW = 10 dBm.
•

Pout (−110.5dBm, 150 m) = P [Pr (150m) < −110.5dBm]

 
Pmin − (Pt dBm + K dBm − 10γ log10 (d/d0 ))
=1−Q
σψdB
 
−110.5 − (10 − 31.54 − 37.1 log10 (150))
=1−Q
3.65
= .0121.

65 / 77
Cell Coverage Area

• Defined as the expected percentage of area within a cell that has received power
above a given minimum.
• Consider a BS inside a circular cell of radius R.
• All mobiles within the cell require some minimum received SNR for acceptable
performance.
• SNR requirements translate to a min received power Pmin throughout the cell.
(Assuming reasonable noise and interference model)
• Tx power at the BS is designed for an average Rx power P̄R at the cell boundary.
( P̄R is averaged over shadowing variations.)
• Shadowing will cause some locations within the cell to have Rx power below P̄R .
• Whereas some other areas will have Rx power exceeding P̄R .

66 / 77
Cell Coverage Area

• Figure shows contours of constant received power based on a fixed transmit power at
the base station for
• Path loss and average shadowing and,
• Path loss and random shadowing.

67 / 77
Cell Coverage Area

• For path loss and average shadowing,

• Constant power contours form a circle around the base station.
(Since combined path loss and average shadowing is the same at a uniform distance
from the base station.)
• For path loss and random shadowing,
• The contours form an amoeba-like shape
(Due to the random shadowing variations about the average.)
• It is not possible for all users at the cell boundary to receive the same power level.
• Some users within the cell will not meet their minimum received power requirement.

68 / 77
Cell Coverage Area

• BS can transmit extra power to insure users affected by shadowing receive their
minimum required power Pmin .
• Causes excessive interference to neighbouring cells
• Since the Gaussian distribution has infinite tails,
• There is a Nonzero probability that any mobile within the cell will have a received
power that falls below the minimum target.
(A mobile may be in a tunnel or blocked by a large building, regardless of its proximity
to the base station.)

69 / 77
Cell Coverage Area
• To compute cell coverage area under path loss and shadowing:
• Consider an incremental area dA at radius r from the BS in the cell.
• Let Pr (r ) be the received power in dA from combined path loss and shadowing.
• Cell coverage area is obtained by integrating over all incremental areas where minimum
power requirement is exceeded.
• Fraction of area within the cell where the the minimum power requirement is
exceeded is
 Z 
1
C =E 1[Pr (r ) > Pmin in dA]dA
πR 2 cellarea
Z
1
= E[1[Pr (r ) > Pmin in dA]]dA,
πR 2 cellarea

where 1[] denotes the indicator function.

70 / 77
Cell Coverage Area
• Let PA = p(Pr (r ) > Pmin ) in dA. Then PA = E[1[Pr (r ) > Pmin in dA]].
• This gives
Z Z 2π Z R
1 1
C= PA dA = PA r dr dθ.
πR 2 cell area πR 2 0 0
• The outage probability of the cell is defined as the percentage of area within the cell
that does not meet its minimum power requirement Pmin that is, Pout cell = 1 − C .

• Given the log-normal distribution for the shadowing,

 
Pmin − (Pt + 10 log10 K − 10γ log10 (r /d0 ))
PA = p(Pr (r ) ≥ Pmin ) = Q
σψdB
= 1 − Pout (Pmin , r ),

where Pout is the outage probability defined in previous slides with d = r .

71 / 77
Cell Coverage Area
• Combining the equations give
Z R
2π
C= PA r dr
πR 2 0
Z R  
2 Pmin − (Pt + 10 log10 K − 10γ log10 (r /d0 ))
= 2 rQ dr
R 0 σψdB
Z R  
2 r
= 2 rQ a + b ln dr ,
R 0 R
where
Pmin − P̄r (R) 10γ log10 (e)
a= , b= ,
σψdB σψdB
P̄R = Pt + 10 log10 K − 10γ log10 (R/d0 ) is the received power at the cell boundary
(distance R from the base station) due to path loss alone.
72 / 77
Cell Coverage Area
• This integral yields a closed-form solution for C in terms of a and b:
   
2 − 2ab 2 − ab
C = Q(a) + exp Q .
b2 b

• If the target minimum received power equals the average power at the cell boundary,
Pmin = P̄r (R), then a = 0 and the coverage area simplifies to
   
1 2 2
C = + exp 2 Q .
2 b b

• Here, C depends only on the ratio γ/σψdB .

• Owing to the symmetry of the Gaussian distribution, under this assumption the
outage probability at the cell boundary Pout (P̄r (R), R) = .5.
73 / 77
Cell Coverage Area
• Show that Pout (P̄r (R), R) = .5.

74 / 77
Cell Coverage Area
• Show that Pout (P̄r (R), R) = .5.
•
 
P̄r (R) − (Pt + 10 log10 K − 10γ log10 (R/d0 ))
Pout (P̄r (R), R) = 1 − Q
σψdB
 
P̄r (R) − P̄r (R)
=1−Q
σψdB
= 1 − Q(0)
= 1 − 0.5
= 0.5.

75 / 77
Problem

Find the coverage area for a cell with the combined path loss and shadowing with K
dBm=-31.54 dBm, γ=3.71, σψdBm = 3.65, and cell radius of 600 m. The base station
transmit power of P¯t = 100 mW̄ = 20 dBm, and a minimum received power requirement
of Pmin = −110 dBm and also one of Pmin = −120 dBm.

76 / 77
Solution
• P̄r (R) = Pt + K − 10γ log10 (600) = 20 − 31.54 − 37.1 log10 (600) = −114.6dBm
• a = (−110 + 114.6)/3.65 = 1.26
• b = (37.1 × .434)/3.65 = 4.41.
• Therefore,
   
2 − 2(1.26 × 4.41) 2 − (1.26)(4.41)
C = Q(1.26) + exp Q = .59,
4.412 4.41

which would be a very low coverage value for an operational cellular system (lots of
unhappy customers).
• Now considering the less stringent received power requirement Pmin = −120 dBm
yields a = (−120 + 114.9)/3.65 = −1.479 and the same b = 4.41. Substituting these
values yields C = .988, a much more acceptable value for coverage area.

77 / 77