Mobile Radio Propagation: Small-Scale Fading and Multipath
Mobile Radio Propagation: Small-Scale Fading and Multipath
1 ∆φ v
fd = ⋅ = cosθ
2π ∆t λ
4.2 Impulse Response Model of a
Multipath Channel
• A mobile radio channel may be modeled as a linear filter with a time
varying impulse response
– time variation is due to receiver motion in space
– filtering is due to multipath
real response
Tbb t
imaginary response
TREP
• The signal p(t) is a repetitive baseband pulse train with very narrow
pulse width Tbb and repetition period TREP , with TREP >> τ max .
• Now, let
p (t ) = 2 τ max / Tbb 0 ≤ t ≤ Tbb
• The channel output r(t) closely approximates the impulse response and
is given by
N −1
∑ a exp(− jθ ) ⋅ p(t −τ )
1
r (t ) = i i i
2 i =0
N −1
τ max
= ∑
i =0
ai exp(− jθ i ) ⋅
Tbb
T
2
rect t − bb − τ i
• Instantaneous multipath power delay profile
τ max
1
∫
2
r (t0 ) = r (t )r * (t )dt
τ max 0
N −1 N −1
∑∑
τ max
1 1
=
τ max ∫
0
Re
4 j =0
i =0
a j (t0 )ai (t0 ) p (t − τ j ) p (t − τ i ) exp(− j (θ j − θ i ))dt
• If all the multipath components are resolved by the probe p(t), then
τ j − τ i > Tbb ∀ j≠i
• Then we have
1
N −1
∑
τ max
1
∫ a (t0 ) p (t − τ k ) dt
2
r (t0 ) = 2 2
τ max 0 4
k
k =0
N −1 τ max
τ
∑
1 T
=
∫ rect t − − τ k dt
2 max bb
ak (t0 )
τ max k =0 0 Tbb 2
N −1
= ∑
k =0
ak2 (t0 )
• The total receiving power is related to the sum of the powers in the
individual multipath components.
• Assuming that the received power from the multipath components
forms a random process where each component has a random
amplitude and phase at any time t, the average small-scale received
power is
N −1 2
N −1
Ea ,θ [ PWB ] = Ea ,θ ∑
i =0
ai exp( jθ i ) =
∑
i =0
ai2
• Now, consider a CW signal which is transmitted into the exact same
channel, and let the complex envelope be given by c(t)=2. Then the
received signal can be expressed as
N −1
r (t ) = ∑ a exp( jθ (t,τ ))
i =0
i i
Ea ,θ [PCW ] = Ea ,θ
∑ ai exp( jθ i (t ,τ ))
i =0
N −1 N −1 N
≈ ∑
i =0
ai2 + 2 ∑∑ r cos(θ −θ )
i =0 i , j ≠i
ij i j
where
rij = Ea [ai a j ]
• The received power for CW wave has large fluctuations than that for
WB signal.
4.3 Small-Scale Multipath Measurement
• Multipath channel measurement techniques
– Direct pulse measurements
– Spread spectrum sliding correlator measurements
– Swept frequency measurements
4.3.1 Direct RF Pulse System
• Direct RF pulse system
– This system transmits a repetitive pulse of width τ bb , and uses a
receiver with a wideband filter with bandwidth BW = 2 / τ bb
– Envelope detector to detect the amplitude response.
• Minimum resolvable delay τ bb
Observed Time
Actual Propagation Time =
r
τ
expansion by a factor of r
t
display from oscilloscope
4.3.3 Frequency Domain Channel Sounding
• Dual relationship between time domain and frequency domain.
• It is possible to measure the channel impulse response in the frequency
domain.
• Measure the frequency domain response and then converted to the time
domain using inverse discrete Fourier transform (IDFT).
4.4 Prameters of Mobile Multipath
Channels
• Power delay profiles for different types of channels are different
Outdoor Indoor
4.4.1 Time Dispersion Parameters
• Time dispersion parameters
– mean excess delay
– RMS delay spread
– excess delay spread
• Mean excess delay
∑a τ 2
k k ∑ P(τ )τ k k
τ = k
= k
∑a k
2
k ∑ P(τ ) k
k
σ τ = τ 2 − (τ 2 )
where
∑a τ 2 2
k k ∑ P(τ )τ k
2
k
τ 2
= k
= k
∑a k
2
k ∑ P(τ )
k
k
• Depends only on the relative amplitude of the multipath components.
• Typical RMS delay spreads
– Outdoor: on the order of microseconds
– Indoor: on the order of nanoseconds
• Maximum excess delay (X dB) is defined to be the time delay during
which multipath energy falls to X dB below the maximum.
excess delay = τ X − τ 0
τ X : maximum delay at which a multipath component is within X dB
τ 0 : delay for the first arriving signal
• Example of an indoor power delay profile; rms delay spread, mean
excess delay, maximum excess delay (10dB), and the threshold level
are shown
4.4.2 Coherent Bandwidth
• Coherent bandwidth, Bc , is a statistic measure of the range of
frequencies over which the channel can be considered to be “flat”.
• Two sinusoids with frequency separation greater than Bc are affected
quite differently by the channel.
• If the coherent bandwidth is defined as the bandwidth over which the
frequency correlation function is above 0.9, then the coherent
bandwidth is approximately
1
Bc ≈
50σ τ
• If the frequency correlation function is above 0.5
1
Bc ≈
5σ τ
4.4.3 Doppler Spread and Coherent Time
• Doppler spread and coherent time are parameters which discribe the
time varying nature of the channel in a small-scale region.
• When a pure sinusoidal tone of f c is transmitted, the received signal
spectrum, called the Doppler spectrum, will have components in the
range f c − f d and f c + f d , where f d is the Doppler shift.
Channel
fc fc − fd fc fc + fd
channel response
f
BC
f
• Frequency selective fading is due to time dispersion of the transmitted
symbols within the channel.
– Induces intersymbol interference
• Frequency selective fading channels are much more difficult to model
than flat fading channels.
• Statistic impulse response model
– 2-ray Rayleigh fading model
– computer generated
– measured impulse response
• For frequency selective fading
BS > BC
and
TS > σ τ
• Frequency selective fading channel characteristic
4.5.2 Fading Effects Due to Doppler
Spread
• Fast Fading: The channel impulse response changes rapidly within the
symbol duration.
– The coherent time of the channel is smaller then the symbol period of the
transmitted signal.
– Cause frequency dispersion due to Doppler spreading.
• A signal undergoes fast fading if
TS > TC
and
BS < BD
• Slow Fading: The channel impulse response changes at a rate much
slower than the transmitted baseband signal s(t).
– The Doppler spread of the channel is much less then the bandwidth of the
baseband signal.
• A signal undergoes slow fading if
TS << TC
and
BS >> BD
4.6 Rayleigh and Ricean Distributions
• Rayleigh Fading Distribution
– The sum of two quadrature Gaussian noise signals
• Consider a carrier signal at frequency ω0 and with an amplitude a
p( r , θ ) = J p( x, y )
∂x / ∂r ∂x / ∂θ cosθ − r sin θ
J = = =r
∂y / ∂r ∂y / ∂θ sin θ r cosθ
• We have
r r2
p( r, θ ) = exp − 2
2πσ 2 2σ
• The Rayleigh distribution has a pdf given by
r r2
2π 2 exp − 2 r≥0
p( r ) = ∫ p( r, θ )dθ = σ 2σ
0
0 otherwise
• pdf of Rayleigh distribution
r r2
exp − 2 0 ≤ r ≤ ∞
p( r ) = σ 2 2σ
0 r<0
σ: rms value of the received signal before envelop detection
σ 2: time - average power of the received signal before envelop detection
• Cumulative distribution function (CDF)
R R2
P( R ) = Pr( r ≤ R ) = ∫ p( r )dr = 1 − exp − 2
0
2σ
• The mean value of the Rayleigh distribution is given by
∞ π
rmean = E[r ] = ∫ rp( r )dr = σ = 1.2533σ
0 2
• The variance of the Rayleigh distribution is given by
∞ σ 2π
σ = E[r ] − E [r ] = ∫ r p( r )dr −
2
r
2 2 2
0 2
π
= σ 2 2 − = 0.4292σ 2
2
• Ricean Fading Distribution: When there is a dominant stationary (non-
fading) signal component present, such as a line-of-sight propagation
path, the small-scale fading envelope distribution is Ricean.
Scattered waves Direct wave
where
Ar 1 2π Ar cosθ
I0 2 =
σ 2π
∫0
exp
σ
2 dθ
is the modified Bessel function of the first kind and zero-order.
• The Ricean distribution is often described in terms of a parameter K
which is defined as the ratio between the deterministic signal power
and the variance of the multipath. It is given by K = A2 /(2σ 2 ) or in
terms of dB 2
A
K (dB) = 10 log dB
2σ 2
• The parameter K is known as the Ricean factor and completely
specifies the Ricean distribution.
• As A → 0 , we have K → −∞ dB. The dominant path decrease in
amplitude, the Ricean distribution degenerates to a Rayleigh
distribution.
4.7 Statistical Models for Multipath
Fading Channels
4.7.1 Clarke’s Models for Flat
Fading
• Clark developed a model where the statistical characteristics of the
electromagnetic fields of the received signal are deduced from
scattering.
• The model assumes a fixed transmitter with a vertically polarized
antenna.
• The received antenna is assumed to comprise of N azimuthal plane
waves with arbitrary carrier phase., arbitrary angle of arrival, and each
wave having equal average amplitude.
• Equal amplitude assumption is based on the fact that in the absence of
a direct line-of-sight path, the scattered components arriving at a
receiver will experience similar attenuation over small-scale distance.
• Doppler shift due to the motion of the receiver.
• Assume no no excess delay due to multipath.
– Flat fading assumption.
• For the nth wave arriving at an angle α n to the x-axis, the Doppler
shift is given by
ν
fn = cos α n
λ
• The vertically polarized plane waves arriving at the mobile have E
field components given by (assume a single tone is transmitted)
N
E z (t ) = E0 ∑ Cn cos(2πf ct + θ n )
n =1
∑ n =1
C 2
i =1
• E z (t ) can be modeled as a Gaussian random process if N is sufficient
large.
• Since the Doppler shift is very small when compared to the carrier
frequency, the three field components may be modeled as narrow band
random process.
E z (t ) = Tc (t ) cos(2πf ct ) + Ts (t ) sin(2πf ct )
where N
Tc (t ) = E0 ∑ Cn cos(2πf nt + φn )
iN
=1
Ts (t ) = E0 ∑ Cn sin(2πf nt + φn )
i =1
E z (t ) = Tc2 (t ) + Ts2 (t ) = r (t )
• It can be shown that the random received signal envelope r has a
Rayleight distribution given by
r r2
exp − 2 0 ≤ r ≤ ∞
p ( r ) = σ 2 2σ
0 r<0
where σ 2 = E02 / 2
• Let p(α )dα denote the function of the total incoming power within dα
of the angle α , and let A denote the average received power with
respect to an isotropic antenna.
• As N → ∞ , p(α )dα approached a continuous distribution.
• If G (α ) is the azimuthal gain pattern of the mobile antenna as a
function of the angle of arrival, the total received power can be
expressed as
2π
Pr = ∫ AG(α ) p(α )dα
0
where S ( f ) = 0, f − fc > fm
• The spectrum is centered on the carrier frequency and is zero outside
the limits f c ± f m .
• Each of the arriving waves has its own carrier frequency (due to its
direction of arrival) which is slightly offset from the center frequency.
• Vertical λ / 4 antenna ( G (α ) = 1.5 ).
• Uniform distribution p (α ) = 1 /( 2π ) over 0 to 2π .
• The output spectrum
1 .5
S( f ) =
2
f − fc
πf m 1 −
fm
S( f )
4.7.2 Simulation of Clarke Fading
Model
• Produce a simulated signal with spectral and temporal characteristics
very close to measured data.
• Two independent Gaussian low pass noise are used to produce the in-
phase and quadrature fading branches.
• Use a spectral filter to sharp the random signal in the frequency
domain by using fast Fourier transform (FFT).
• Time domain waveforms of Doppler fading can be obtained by using
an inverse fast Fourier transform (IFFT).
• Smith simulator using N carriers to generate fading signal
1. Specify the number of frequency domain points N used to
represent S ( f ) and the maximum Doppler frequency
shift f m .
2. Compute the frequency spacing between adjacent spectral lines
as ∆f = 2 f m /( N − 1) . This defines the time duration of a fading
waveform, T = 1 / ∆f .
3. Generate complex Gaussian random variables for each of the
N/2 positive frequency components of the noise source.
4. Construct the negative frequency components of the noise
source by conjugating positive frequency and assigning these at
negative frequency values.
5. Multiply the in-phase and quadrature noise sources by the fading
spectrum S ( f ) .
6. Perform an IFFT on the resulting frequency domain signal from
the in-phase and quadrature arms, and compute the sum of the
squares of each signal.
7. Take the square root of the sum.
• Frequency selection fading model
4.7.3 Level Crossing and Fading
Statistics
• The level crossing rate (LCR) is defined as the expected rate at which
the Rayleigh fading envelope crosses a specified level in a positive-
going direction.
• Useful for designing error control codes and diversity.
• Relate the time rate of change of the received signal to the signal level
and velocity of the mobile.
• The number of level crossing per second to the level R is given by
∞
N R = ∫ r&p ( R, r&)dr& = 2π f m ρe −ρ 2
(A)
0