2004 16-Propagation.

doc

PROPAGATION EFFECTS
ELECTROMAGNETIC WAVES

Signal Attenuation and the Range Equation

• As an electromagnetic signal propagates through the atmosphere,
molecular interactions with the wave absorb energy, and the signal
amplitude decreases with range.
• For the attenuation α per unit metre, the relationship between the received
and transmitted power can be written as follows:
Pr = Pt e −2αR
• Using the change of base theorem
log b N
log a N = ,
log b a
it can be shown that log10N = 0.43429 logeN

• This allows the attenuation to be expressed as αdB = 4342.9α in dB/km

(one way), making the relationship:

10log10(Pr) = 10log10(Pt) - 2αdBRkm

where: Pr – Received power (w)
Pt – Transmitted power (w)
αdB.– Attenuation (dB/km)
Rkm – Range (km)

• This makes the range equation difficult to solve in a closed form and
numerical or graphical methods are normally resorted to.
• Writing the range equation for a monostatic radar system in dB

 λ2 
10 log10 ( Pr ) = 10 log10 ( Pt ) + 10 log10   + 20 log10 (G ) + 10 log10 (σ )
3 
 (4π ) 
− 10 log10 ( L) − 40 log10 ( R) − 2αRkm

Attenuation of Electromagnetic Waves

ƒ Clear Weather Attenuation
• Primary absorption by O2 and H2O
• Because absorption is proportional to the absolute humidity, the graph is
adjusted to be representative of a water content of 7.5g/m3 at 20°C
• Below 60GHz there are two main absorption bands, that from the 1.35cm
(22.235GHz) water vapour line and its skirts, and a series of oxygen lines
centred around 0.5cm (60GHz)

MECH4721 SENSORS & SIGNALS 1of26

2004 16-Propagation.doc

• These two peaks bracket the 35GHz window

• The atmosphere becomes progressively more opaque at higher
frequencies, with windows at 94, 140, 220 and 360GHz between the O2
and H2O lines.
• Totally opaque between 1THz an 10THz (0.3mm and 30µm)
• Windows appear between absorption lines of H2O and CO2 in the IR
region starting at 10µm
• Attenuation is very low in the visible region around 500nm

Figure 1: Clear Weather Attenuation as a function of Frequency

ƒ Effect of Atmospheric Pressure (air density)

• Because absorption is proportional to air density, this graph above is
adjusted to be representative of the pressure at sea level.
• This graph shows the difference in the attenuation (for micro and
millimetre waves) at sea level and 4km up

MECH4721 SENSORS & SIGNALS 2of26

2004 16-Propagation.doc

Figure 2: Effect of Pressure on Attenuation

ƒ Effect of Rain
• The effect of light rain that has a uniform spatial distribution can be
predicted with relative ease.
• Heavier rain normally comes in squalls with unpredictable intensity and
drop size distribution
• Mie (resonance) scattering theory is generally applied for frequencies up
to the sub-millimetre wave band.

αr = aRb

where: αr – Attenuation (dB/km)

R – Rain rate (mm/hr)
a,b – Coefficients dependent on frequency and temperature

Coefficient 35GHz 94GHz

a 0.27 1.6
b 0.985 0.64

• The Laws Parsons model is used to determine drop-size distributions if

they are required.
• The mean raindrop size is proportional to rainfall rate and the atmospheric
conditions.
Condition Drop Size (µm)
Haze 0.01-3
Fog 0.01-100
Clouds 1-50
Drizzle 3-800
Moderate rain (4mm/hr) 3-1500
Heavy rain (16mm/hr) 3-3000

MECH4721 SENSORS & SIGNALS 3of26

2004 16-Propagation.doc

• For typical conditions, the attenuation with frequency up to 300GHz is

plotted here at a number of different rainfall rates.

Figure 3: Rainfall Attenuation as a function of Rainfall Rate

• At IR and visible wavelengths, the drop size is much larger than the
wavelength, so the attenuation will be independent of drop diameter.
• At these frequencies attenuation is proportional to water content in g/m3,
and hence proportional to rain rate

α = cR

where: c – Scaling coefficient

R – Rain rate (mm/hr)

MECH4721 SENSORS & SIGNALS 4of26

2004 16-Propagation.doc

Figure 4: Specific Attenuation (dB/km)/(mm/hr) in Rain

ƒ Effect of Fog and Clouds

• Condensation of atmospheric water vapour into water droplets suspended
in the air.
• Fog is defined as a water droplet density that restricts the visibility to less
than 1km.
• Advection fog is caused by a horizontal movement of a warm air mass
over cold water.
• Radiation fog is caused by the cooling of air overnight (often occurs over
rivers and swamps)
• Clouds are similar in nature to fog with droplets less than 100µm in
diameter.

Radiation Fog Advection Fog

Average Drop Diameter 10µm 20µm
Typical Drop Size Range 5-35µm 7-65µm
Liquid water Content 0.11g/m3 0.17g/m3
Droplet Concentration 200/cm3 40/cm3
Visibility 100m 200m

• An empirical relationship that relates visibility to liquid water content for

advection fog is

M = 1660V −1.43

where M – Liquid water content (g/m3)

V – Visibility (feet) (1foot=0.308m)

• For a visibility of 125m (V=400ft) M = 0.32g/m3

MECH4721 SENSORS & SIGNALS 5of26

2004 16-Propagation.doc

• Because the drop diameter of both fog and clouds is less than 100µm, the
Rayleigh (low frequency) approximation can be used to evaluate the
attenuation at microwave and millimetre wave frequencies

αf,c = kρo dB/km

where: k – Attenuation coefficient (dB/km per g/m3) at a specific

frequency (See Figure 5)
ρo – Water content (g/m3)

Figure 5: Specific Attenuation through Clouds and Fog

• Using formula and taking k from Figure 5, for f = 100GHz, gives an

attenuation of 4dB/km per g/m3 at 20°C.
• The attenuation in a fog with a visibility of 125m (400ft) is therefore
α = 4x0.32 = 1.28dB/km
• For a moderate fog with liquid water content of 0.1g/m3, this equates to an
attenuation of 0.4dB/km. This ties in quite well with Figure 6.even though
the predicted visibility does not!!

Cloud Type Liquid Water

Content (g/m3)
Heavy Fog 0.15-0.4
Moderate Fog 0.01-0.1
Cumulus 0.5-1
Altostratus 0.35-0.45
Stratocumulus 0.3-0.6
Nimbostratus 0.6-0.7
Stratus 0.15-0.3

MECH4721 SENSORS & SIGNALS 6of26

2004 16-Propagation.doc

• As the frequency increases, the wavelength becomes an appreciable

percentage of the circumference of the drop and resonance occurs (near
IR region).
• Because there is such a large variation in drop size and LWC for mist and
fog, the actual attenuation that will occur in the IR region around this
resonance is difficult to determine accurately. As a rule of thumb, the
following formulae can be applied.
• Figure 6 shows the laser attenuation (dB/km per mg/m3) for various drop
diameters Rm in µm (NB note the units for LWC are different to those in Figure 5)

αf,c = kρo dB/km

where: k – Attenuation coefficient (dB/km per mg/m3) at a specific

frequency and a specific Rm (See Figure 6)
ρo – Water content (mg/m3)

Figure 6: Comparison of Radar and Laser Fog Attenuation

• For radiation fog with a drop diameter D=8µm (Rm = 4µm) and a LWC of
0.1g/m3 (100mg/m3) the attenuation is 0.2x100 = 20dB/km (one way).
• For a heavy advection fog (visibility 50m), the LWC was calculated to be
1.2g/m3 (1200mg/m3) for the same drop diameter (8µm ) the attenuation is
0.2x1200 = 240dB/km

MECH4721 SENSORS & SIGNALS 7of26

2004 16-Propagation.doc

Figure 7: Summary of Atmospheric Attenuation

• Effect of rain and fog as a function of frequency. Note the resonance

effects due to fog particle size at wavelengths between 30 and 3µm.
• Note the peak in the rain attenuation at wavelengths between 3mm and
1mm depending on the rainfall rate.

ƒ Other Solid Materials

• Absorption of EM radiation by solids and liquids is determined by their
relative dielectric constant εr and loss tangent tanδ of the material.
• For most materials this is a function of frequency.
• For normal incidence, the reflection coefficient is

Z L − Zo εr −1
Γ = =
Z L + Zo εr +1

where: Zo – Characteristic impedance of free space µo / ε o

ZL – Impedance of the material µ o / ε oε r

• The attenuation in dB for propagation through an unbounded dielectric

material is

MECH4721 SENSORS & SIGNALS 8of26

2004 16-Propagation.doc

d
α d = 27.3 ε r tan δ
λo
where; αd – One way attenuation (dB)
εr – Relative dielectric constant
tanδ - Loss tangent
d – Distance (m)
λo – Wavelength (m)

• Material characteristics are important for the design of radomes and lens
antennas or to determine the propagation characteristics for ground
penetrating radar (GPR).

Figure 8: Potential GPR Penetration Depth

ƒ Dust and Smoke

• Absorption of EM waves by dust and smoke is determined using Mie
scattering theory.
• For particles with a radius a, and relative dielectric constant εr , the
scattering cross section σ(a)
2
128π 5 a 6 ε r − 1
σ (a) =
3 λ4 ε r + 2

this reduces to πa2 at 2πa/λ = 0.8

according to the theory

• If δ(a) is the number of particles per unit volume, and αp the attenuation in
dB per unit length, then:

MECH4721 SENSORS & SIGNALS 9of26

2004 16-Propagation.doc

∞
α p = ∫ δ (a )σ (a)da
0

For δ(a) = µ/ρV µ - Mass concentration (µgram/cm3)

ρ - Density (gram/cm3)
V – Volume of the particle of diameter D (cm3)

2
µD 3 ε − 1
α p = 169.2 dB/km
ρλ4 ε + 2

Note: Uses cgs so all the lengths are in cm

• In the optical region where the scattering cross section has stabilised at
πa2, the attenuation equation reduces to the following:

2
0.65µ ε − 1
αp = dB/km
ρD ε + 2

Source Material Density Dielectric Diameter Mass conc Attn (dB/km)

3 3
(g/cm ) constant (cm) (g/m )
λ=4mm λ=1-10µm
Dust: Winds Quartz 2.6 2.25 1.E-03 10 2.20E-06 216.3
Stack: Stone Mill Quartz 2.6 2.25 1.E-03 80 1.76E-05 1730.1
Stack: Steel Mill Carbon 2 1000 1.E-03 10 3.28E-05 3230.6
Volcanic Dust Quartz 2.6 2.25 2.E-04 5 8.80E-09 540.7
Bacteria Water 1 80 2.E-04 5.E-05 2.45E-12 0.2
Pollen Biological 0.5 1.5 1.E-03 60 1.62E-05 1591.8
Light fog Water 1 80 1.E-03 1.E-03 6.13E-09 0.6
Heavy fog Water 1 80 2.E-03 1 4.91E-05 301.7

MECH4721 SENSORS & SIGNALS 10of26

2004 16-Propagation.doc

Refraction of Electromagnetic Waves

• Refraction effects are independent of wavelength at frequencies below
100GHz.
• The decrease in air density, and hence refractive index with height results
in a downward curvature of horizontally launched radar beams.
• Bending is about ¼ of the curvature of the earth

Figure 9: Refraction by the Earth’s Atmosphere

• To plot ray paths as straight lines, the effective earth radius is 4/3 of the
actual value.
• The radar horizon can be approximated by the following formula:

d = 2kah

where: d – radar horizon (m)

k – 4/3
a – actual radius (m)
h – radar height (m)

d (km) = 130 h(km)

MECH4721 SENSORS & SIGNALS 11of26

2004 16-Propagation.doc

ACOUSTICS AND VIBRATION

Alternate compression and expansion of the medium (solid, liquid or gas) with
certain frequencies are called sound waves.

Table: Common Units of Pressure

Units Atmospheric
Pressure
Atmospheres (atm) 1
Bar 1.01325
Pascal (Pa) 1.01325x105
2
Pounds/in (psi) 14.6960
Inches Hg 29.9213
Mm of Hg (torr) 760
Inches of water 406.8
Dynes/cm2 1.01325x106

ƒ Sound Propagation in Gases

The speed of sound in an ideal gas can be written in terms of the equations of
state (PV=nRT)

RTγ Pγ K
c= = = cm/s
M ρo ρo

where: c – Velocity of sound in the gas (cm/s)

R – Universal gas constant (8.314x107 ergs/mole/K)
T – Temperature (K)
M – Molecular Weight of the gas (g/mole)
γ - Ratio of specific heats (adiabatic exponent) Cp/Cv for the gas
P – Pressure (1atm = 1.013x106 dyne/cm2)
ρo - Density (g/cm3)
K – Bulk modulus (K=∆P/∆V)

The adiabatic exponent γ can be estimated as follows:

• 1.66 for monatomic gases (He, Ne, Ar)
• 1.40 for diatomic gases (H2, O2, N2)
• 1.33 for triatomic and more complex gases (NH3, CH4, C7H8)
• 1.286 for very long molecules

The table two pages on gives a more accurate measure of the adiabatic
exponent for various gases.

At atmospheric pressure, the volume of one mole of gas is 22414 cm3

Effect of Molecular Weight

The speed of sound in any gas mixture can be calculated by using the
molecular weight to determine the density as follows:
Air is made up of O2, N2, and CO2 in proportions (21:78:1)
The molecular weights are 16+16=32, 14+14=28, 12+16+16=44

MECH4721 SENSORS & SIGNALS 12of26

2004 16-Propagation.doc

21 × 32 78 × 28 1 × 44
M = + + = 29
100 100 100

29
ρo = = 0.0013 g/cm3
22414
1.01325 × 10 6 × 1.404 cm/s
c= = 3.3147 × 10 4
0.0013

Effect of Temperature and Pressure

The density of the gas at any temperature and pressure is related to that at
STP for the pressure in mm of mercury and the temperature in Kelvin.

 P  273.16 
ρ = ρo   
 760  T 

For rough calculations of the speed of sound (ignoring changes in Cv) the
changes in density are assumed to be the only variable affecting the velocity.

MECH4721 SENSORS & SIGNALS 13of26

2004 16-Propagation.doc

This table lists the constants that are required to calculate the speed of sound
of various gases at different temperatures and pressures.

MECH4721 SENSORS & SIGNALS 14of26

2004 16-Propagation.doc

ƒ Sound Propagation in Water

As with gases, the speed is equal to the bulk modulus of elasticity divided by
the density of the liquid.
B
c= m/s
ρ
Because the modulus of elasticity and the density are functions of the
pressure (depth), temperature and salinity, an empirical equation derived by
Wilson (1960) is usually used to calculate the speed.

c = 1449.2 + 4.62T − 0.055T 2 + 1.4( S − 35) + 0.017 z m/s

where: T – Temperature (°C)
S – Salinity (parts per thousand)
z – Depth (m)

Table: Speed of Sound in Fresh Water

Temp Speed Pressure Speed
(°C) p=1atm (psi) T=30°C
(m/s) (m/s)
0 1,403.0 14.7 1,509.7
10 1,447.8 2,000 1,532.9
20 1,482.9 4,000 1,556.2
30 1,509.7 6,000 1,579.4
40 1,529.3 8,000 1,602.4
50 1,542.9 10,000 1,625.3
60 1,551.3 12,000 1,648.1
70 1,555.1 14,000 1,670.9
80 1,554.7
90 1,550.5
100 1,542.5

14.7lbf/in2 = 101kN/m2 = 1.01bar

1000lbf/in2 = 6.89MN/m2 = 68.9bar
1 dyne = 10µN
1 kgf = 9.807N
1 Pa = 1 N/m2

10m depth = 1bar increase in pressure

Note that the speed of sound at the sea surface (35ppt salt) at 0°C is
1449.2m/s as contrasted with 1403m/s for fresh water

MECH4721 SENSORS & SIGNALS 15of26

2004 16-Propagation.doc

ƒ Sound Propagation in Solids

In an isotropic solid for which the wave-front is a large number of
wavelengths, longitudinal and shear waves can exist which have velocities

λ + 2µ
clong = m/s
ρo
µ
c shear = m/s
ρo
where: µ,λ - Lame Elastic Moduli for the material (N/m2)

For a rod, whose diameter is a small fraction of the wavelength, extensional

and torsion waves can propagate

µ
ctor = m/s
ρo
Yo
cext = m/s
ρo

3λ + 2µ
where: Yo = µ N/m2
λ+µ

MECH4721 SENSORS & SIGNALS 16of26

2004 16-Propagation.doc

MECH4721 SENSORS & SIGNALS 17of26

2004 16-Propagation.doc

ƒ Characteristic Impedance (Z)

As sound propagates through a medium, each small volume element of that
medium oscillates about its equilibrium position. For pure harmonic motion,
the displacement is
2π
y = y m cos ( x − vt )
λ
It can be shown that the pressure exerted by the sound wave is

p = (kρ o v 2 y m ) sin(kx − ϖt )
where k - 2π/λ - The wave number
ϖ - Angular frequency

The amplitude of the wave is the sound pressure pm = kρoν2ym so a sound

wave may be considered to be a pressure wave with a phase displacement of
90° to that of the displacement wave.

The pressure at any point changes sinusoidally with time around a mean
value. The RMS (effective) value of this fluctuating component is known as
the acoustic pressure P. The acoustic pressure can be measured using an
appropriately calibrated microphone or hydrophone.

The characteristic impedance is defined as the ratio of this acoustic pressure

and the RMS volume velocity (ξ) (this is NOT the wave velocity)
P
Z=
ξ
This is a complex quantity. For an idealised medium (with no loss) Z is real
and reduces to a proportionality factor that equates intensity to the square of
the sound pressure.

Z = ρc
where: ρ - Density (kg/m3)
c – Speed (m/s)

For air ρ=1.3kg/m2 and c=332m/s, Zair ≈ 400 acoustic ohms

ƒ Sound Intensity (I)

Intensity of a sound is defined as the power transferred per unit area, and can
be expressed through the acoustic impedance (similar to ohms law)
P2
I = Pξ =
Z

where: I – Sound Intensity (sound power density) (W/m2)

P – Acoustic Pressure (N/m2 or Pascal)
ξ - Volume velocity (m3/s)

It is common to specify the sound not as an intensity, but as a ratio β in dB

relative to Io = 10-12 W/m2

MECH4721 SENSORS & SIGNALS 18of26

2004 16-Propagation.doc

I
β = 10 log10
Io

Pressure levels can also be expressed in dB as follows

P
∏ = 20 log10
Po

where Po – 2x10-5 (N/m2) = 2.9x10-9 (psi) is the Minimum Audible

Threshold (MAC)

Io =
Po2
=
(
2 × 10 −5 )
2

= 10 −12 W / m 2
Z 400

Sound Level
(dB)
Safe Range
Stream flow, rustling leaves 15
Watch ticking, soft whisper 20-30
Quiet street noises 40
Normal conversation 45-60
Normal city or freeway traffic 70
Vacuum cleaner 75
Hair dryer 80
Motorcycle, electric shaver 85
Lawn mower, heavy equipment 90
Garbage truck 100
Screaming baby 115
Injury Range
Race car, loud thunder, rock band 120-130
Jet airplane's takeoff from 120 feet 120
Jet airplane's takeoff from 120 feet 120
Pain threshold 130
Rocket launch from 150 feet 180

Attenuation of Acoustic Waves in Air

• For a plane acoustic wave, a small attenuation will occur because of the
viscosity. The attenuation factor is e-αηx for the pressure (or particle
velocity).
2ηϖ 2
αη =
3 ρc 3
where: η - coefficient of viscosity for the gas (dyne-s/cm2)
ω - Angular frequency (rad/s)
ρ - Density (g/cm3)
c – Speed of sound (cm/s)

MECH4721 SENSORS & SIGNALS 19of26

2004 16-Propagation.doc

• A plane acoustic wave will be attenuated slightly due to thermal

conductivity effects.
k (γ − 1)ϖ 2
αt =
2γρC c 3
v
where: k – Thermal conductivity (cal/cm-s-deg)
γ - Ratio of specific heats (defined previously)
Cv – Specific heat at constant volume (cal/g-deg)

• In addition to the dispersion of sound due to wind, turbulence in the

atmosphere and temperature gradients, two properties combine to
attenuate the sound. The first is caused by molecular absorption due to
collisions, and the second due to viscosity and heat conduction described
previously.

• The attenuation due to heat conduction and viscosity together is given by

the following equation
A
α c = α η + α T = 0.0437 2 dB/m
λ
where λ - Wavelength (m)
A – taken from the following curves

Figure 10: Coefficient used to Determine Sound Attenuation in Air

• Total attenuation is very sensitive to humidity and temperature, and can be

calculated using the following empirical relationship derived by Knudsen
and Harris from measured data (accurate for relative humidity >30%).

3/ 2
 f  0.283
αA =   dB/m
 1000  20 + φ t

where: f – frequency (Hz)

φt - φ20(1+0.067∆t)
φ20 – Relative humidity % at 20°C
∆t – Denotes the temp difference from 20°C

MECH4721 SENSORS & SIGNALS 20of26

2004 16-Propagation.doc

The following graphs show the attenuation of sound in air as a function of

frequency with relative humidity as a parameter

Figure 11: Sound Attenuation as a function of Frequency in Air

Attenuation of Acoustic Waves in Water

• The absorption of sound as it propagates through pure water has a strong
frequency dependence

0.036 f 2
α dB = + 3.2 × 10 −7 f 2
dB/m
f 2 + 3600
where: αdB - Attenuation (dB/m)
f – Frequency (kHz)

Figure 12: Sound Attenuation in Water

MECH4721 SENSORS & SIGNALS 21of26

2004 16-Propagation.doc

• Because of the strong frequency dependence, tables are often constructed

to exclude this characteristic. For the tables shown below the attenuation
coefficient applies to the following formula:

A = Ao e −αx

Applying the change of base theorem

 A
( )
20 log 10   = 20 log 10 e −αx = −8.686αx
 Ao 

The attenuation coefficients are related as follows

α dB = 8.686α dB/m

Transmission Loss Anomaly (A): Other losses occur due to scattering off
suspended particles in the water. These and the losses due to refractive
effects are lumped together and called the transmission loss anomaly

Reflection and Refraction of Acoustic Waves

Reflection and refraction occur where changes in the characteristic
impedance of the medium.

Waves normal to the Interface

If a wave with intensity Io is propagating through medium-1 with a
characteristic impedance of Z1 through a boundary normal to the direction of
travel into medium-2 with characteristic impedance Z2 then the transmitted
and reflected sound intensities are as follows:

MECH4721 SENSORS & SIGNALS 22of26

2004 16-Propagation.doc

 Z1 − Z 2 
I ref = I o  

 Z1 + Z 2 

 2Z 2 
I trans = I o  

 Z1 + Z 2 

where: Iref – Reflected sound intensity

Itrans – Transmitted sound intensity
Io – Incident sound intensity

It can be seen that the reflected intensity increases as a function of the

difference in characteristic impedance between the two media.

Waves at an angle to the Interface

When the incident wave strikes the interface at an angle θI to the normal, a
reflected wave moves away from the interface at an angle θr = θI while the
transmitted wave is refracted at an angle θt to the normal in the second
medium. The angle of refraction is given by Snell's law

 c 2 sin θ i 
θ t = A sin  
 c1 

With incident and reflected intensities as follows:

 Z 1 cosθ t − Z 2 cosθ i 
I ref = I o  

 Z 1 cosθ t + Z 2 cosθ i 

 2Z 2 cosθ i 
I trans = I o  

 Z1 + Z 2 

The effects of reflection and refraction are generally visualised using a

technique called ray acoustics that is analogous to standard electromagnetic
ray-tracing techniques. Speed
θ1
If a wave is propagating through n horizontal Ray Path c1
layers in each of which, the speed is θ2
considered to be constant, then Snell’s law c2
can be rewritten
θ3
c3
c1 c c
= 2 = n θ4
cos φ1 cos φ 2 cos φ n c4

where: φn – Angle made with the horizontal at that point

MECH4721 SENSORS & SIGNALS 23of26

2004 16-Propagation.doc

Note than φn is the complement of the angle usually associated with Snell’s
law. It is called the grazing angle.

In practise, the temperature does not change abruptly, but rather the gradient
will increase or decrease at a measurable rate.

c( z ) = c0 + gz
where: c0 – speed at the surface (or transducer depth)
g – Speed gradient dc/dz between the surface and depth z

The net result is that ray traces appear to be curves rather than straight lines
with a radius of curvature at any point

c0 c
R=− =
g g cos φ

The actual path followed by sound waves in the sea under different conditions
is discussed below.

ƒ Refraction Examples
The SOFAR Channel
At depths of approximately 1,000 metres, pressure becomes the important
factor: it combines with temperature and salinity to produce a zone of
minimum sound speed. This zone has been named the SOFAR (sound fixing
and ranging) channel. If a sound is generated by a point source in the SOFAR
zone, it becomes trapped by refraction as shown in the figure. Dispersed
horizontally rather than in three directions, the sound is able to travel for great
distances

The following figure shows the different effects created by different velocity
profiles.

MECH4721 SENSORS & SIGNALS 24of26

2004 16-Propagation.doc

Figure 13: Sound Propagation Effects due to Refraction

MECH4721 SENSORS & SIGNALS 25of26

2004 16-Propagation.doc

ƒ Other Effects
Propagation is also affected by sound waves reflecting from the surface or the
sea-bottom and other effects as shown.

Figure 14: Sound Propagation Effects due to Refraction and Reflection

References
[1] E.Brookner, Radar Technology, Artech, 1980
[2] P.Bhartia & I.Bahl, Millimeter Wave Engineering and Applications, Wiley-Interscience, 1984.
[3] C.Currie (ed), Principles and Applications of Millimeter Wave Radar, Artech, 1987.
[4] Richard Cook (editor), Acoustics, National Bureau of Standards, ?
nd
[5] L.Kinsler & A. Frey, Fundamentals of Acoustics 2 ed., John Wiley & Sons, 1967

MECH4721 SENSORS & SIGNALS 26of26