Introduction to Spread Spectrum
m a
r
• Problems such as capacity limits, propagation
a
effects, synchronization occur with wireless systems
h
S
• Spread spectrum modulation spreads out the
h
modulated signal bandwidth so it is much greater
s
than the message bandwidth
h i
• Independent code spreads signal at transmitter and
s
despreads signal at receiver
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spread Spectrum
m a
a r
S h
i sh
s h
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spread Spectrum
m a
r
• Input is fed into a channel encoder
a
• Produces analog signal with narrow bandwidth
S
• Signal is further modulated using sequence of digits
• Spreading code or spreading sequence
h
h
• Generated by pseudonoise, or pseudo-random number generator
i s
• Effect of modulation is to increase bandwidth of signal to be transmitted
h
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spread Spectrum
m a
r
• On receiving end, digit sequence is used to demodulate the spread spectrum signal
h
• Signal is fed into a channel decoder to recover data
a
h S
h i s
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spread Spectrum
m a
r
• What can be gained from apparent waste of spectrum?
h
• Immunity from various kinds of noise and multipath distortion
a
S
• Can be used for hiding and encrypting signals
• Several users can independently use the same higher bandwidth with very little
sh
interference
h i
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spread Spectrum Technology
• Side effects:
m a
a r
• coexistence of several signals without dynamic coordination
h
• tap-proof
S
• Alternatives: Direct Sequence (DS/SS), Frequency Hopping
(FH/SS)
sh
• Spread spectrum increases BW of message signal by a
i
factor N, Processing Gain
s h
A Processing Gain =
Bss Bss
N = 10 log10
B B
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Effects of spreading and interference
m a
a r
user signal
h
broadband interference
S
narrowband interference
P P
i)
i sh
f
ii)
f
h
sender
s
P P P
A
iii) iv) v)
f f f
receiver
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Spreading and frequency selective fading
channel
m a
r
quality
a
narrowband
h
1 2 5 6
3 channels
S
4
h
Narrowband signal frequency
guard space
channel
h i s
s
quality
2
A
2
2
2
2
1 spread spectrum
channels
spread frequency
spectrum
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Direct Sequence Spread Spectrum (DSSS)
m a
r
• Each bit in original signal is represented by multiple bits in the transmitted signal
h
• Spreading code spreads signal across a wider frequency band
a
S
• Spread is in direct proportion to number of bits used
h
• One technique combines digital information stream with the spreading code bit
s
stream using exclusive-OR
h i
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Direct Sequence Spread Spectrum (DSSS)
m a
a r
S h
i sh
s h
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
DSSS (Direct Sequence Spread Spectrum) I
• XOR the signal with pseudonoise (PN) sequence (chipping sequence)
m a
• Advantages
a r
h
• reduces frequency selective
fading Tb
S
• in cellular networks user data
h
• base stations can use the
0 1 XOR
s
same frequency range
i
• several base stations can Tc
h
detect and recover the signal chipping
s
• But, needs precise power control sequence
0 1 1 0 10 1 0 1 1 010 1
A
=
resulting
signal
0 1 1 0 10 1 1 0 0 101 0
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
DSSS (Direct Sequence Spread Spectrum) II
transmitter
m a
r
Spread spectrum
transmit
a
Signal y(t)=m(t)c(t)
signal
h
user data
X modulator
m(t)
chipping
h
sequence, c(t)
radio
S
s
carrier
receiver
h i correlator
s
sampled
received products
A
sums data
signal demodulator X integrator decision
radio
carrier
Chipping sequence,
c(t)
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
DS/SS Comments III
m a
r
• Pseudonoise(PN) sequence chosen so that its autocorrelation is very
a
narrow => PSD is very wide
• Concentrated around τ < Tc
S h
• Cross-correlation between two user’s codes is very small
i sh
s h
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
DS/SS Comments IV
m a
r
• Secure and Jamming Resistant
• Both receiver and transmitter must know c(t)
• Since PSD is low, hard to tell if signal present
h a
• Since wide response, tough to jam everything
h S
s
• Multiple access
h i
• If ci(t) is orthogonal to cj(t), then users do not interfere
• Near/Far problem
A s
• Users must be received with the same power
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
FH/SS (Frequency Hopping Spread Spectrum)
I
m a
r
• Discrete changes of carrier frequency
a
• sequence of frequency changes determined via PN sequence
h
• Two versions
S
• Fast Hopping: several frequencies per user bit (FFH)
h
• Slow Hopping: several user bits per frequency (SFH)
i s
• Advantages
h
• frequency selective fading and interference limited to short period
s
• uses only small portion of spectrum at any time
A
• Disadvantages
• not as robust as DS/SS
• simpler to detect
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Frequency Hoping Spread Spectrum (FHSS)
m a
r
• Signal is broadcast over seemingly random series of radio frequencies
a
• A number of channels allocated for the FH signal
h
• Width of each channel corresponds to bandwidth of input signal
h
• Transmitter operates in one channel at a time
S
• Signal hops from frequency to frequency at fixed intervals
i s
• Bits are transmitted using some encoding scheme
h
• At each successive interval, a new carrier frequency is selected
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
FHSS (Frequency Hopping Spread Spectrum) II
m a
r
Tb
a
user data
h
0 1 0 1 1 t
S
f
Td
h
f3 slow
s
f2 hopping
i
(3 bits/hop)
f1
s hTd t
f
A
f3 fast
f2 hopping
(3 hops/bit)
f1
t
Tb: bit period Td: dwell time
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
FHSS (Frequency Hopping Spread Spectrum) III
m a
r
transmitter narrowband Spread transmit
a
signal signal
user data
h
modulator modulator
S frequency hopping
h
synthesizer sequence
s
receiver
received
h i data
s
signal demodulator demodulator
A
hopping
sequence
frequency
synthesizer
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Applications of Spread Spectrum
m a
r
• Cell phones
• IS-95 (DS/SS)
• GSM
h a
• Global Positioning System (GPS)
h S
i s
• Wireless LANs
h
• 802.11b
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Performance of DS/SS Systems
m a
r
• Pseudonoise (PN) codes
• Spread signal at the transmitter
• Despread signal at the receiver
h a
• Ideal PN sequences should be
h S
i s
• Orthogonal (no interference)
h
• Random (security)
s
• Autocorrelation similar to white noise (high at τ=0 and low for τ not equal 0)
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
PN Sequence Generation
m a
r
• Codes are periodic and generated by a shift register and XOR
a
• Maximum-length (ML) shift register sequences, m-stage shift
h
register, length: n = 2m – 1 bits
h S
s
R(τ)
h i τ −>
A s
-nTc
-1/n Tc nTc
Output
+
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Generating PN Sequences
m a
r
m Stages connected
a
Output
h
+ to modulo-2 adder
S
• Take m=2 =>L=3 2 1,2
h
• cn=[1,1,0,1,1,0, . . .],
s
usually written as bipolar 3 1,3
h i
cn=[1,1,-1,1,1,-1, . . .] 4 1,4
Rc (m )
A s
1 L
= ∑ cn cn + m
L n =1
5
6
1,4
1,6
1 m=0 8 1,5,6,7
=
− 1 / L 1 ≤ m ≤ L − 1 ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Problems with m-sequences
m a
r
• Cross-correlations with other m-sequences generated by different
a
input sequences can be quite high
S h
• Easy to guess connection setup in 2m samples so not too secure
h
• In practice, Gold codes or Kasami sequences which combine the
i s
output of m-sequences are used.
s h
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Detecting DS/SS PSK Signals
transmitter
m a
r
Spread spectrum
transmit
a
Signal y(t)=m(t)c(t)
Bipolar, NRZ signal
h
m(t) X X
PN
h S
s
sequence, c(t) sqrt(2)cos (ωct + θ)
receiver
h i
s
received z(t) w(t)
A
data
signal
X X LPF integrator decision
x(t)
sqrt(2)cos (ωct + θ) c(t)
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Optimum Detection of DS/SS PSK
m a
r
• Recall, bipolar signaling (PSK) and white noise give the optimum error
a
probability
Pb = Q
S h
2 Eb
ℵ
h
i
• Not effected by spreading
h s
s
• Wideband noise not affected by spreading
A
• Narrowband noise reduced by spreading
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Signal Spectra
m a
Processing Gain =
Bss Bss Tb
N = 10 log10 =
a r
h
B B Tc
h S
• Effective noise power is channel noise power plus
jamming (NB) signal power divided by N
h i s
s
Tb
A Tc
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Multiple Access Performance
• Assume K users in the same frequency band,
m a
• Interested in user 1, other users interfere
a r
S h
i sh 4 6
s h
5
A 3 2
1
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Signal Model
m a
r
• Interested in signal 1, but we also get signals from other K-1 users:
(ha
xk (=
t)
• At receiver,
h S
2 mk ( t − τ k ) ck ( t − τ k ) cos ωc ( t − τ k ) + θ k )
s
2 mk ( t − τ k ) ck ( t − τ k ) cos (ωct + φk )
= φk =
θ k − ωcτ k
h i
A s x ( t ) x1 ( t ) +
=
K
∑ xk ( t )
k =2
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Interfering Signal
m a
• After mixing and despreading (assume τ1=0)
a r
h
zk ( t ) =2 mk ( t − τ k ) ck ( t − τ k ) c1 ( t ) cos (ωct + φk ) cos (ωct + θ1 )
• After LPF
h S
i s
wk ( t ) =mk ( t − τ k ) ck ( t − τ k ) c1 ( t ) cos (φk − θ1 )
h
• After the integrator-sampler
Ik
A s =cos (φk − θ1 )
Tb
0
mk ( t − τ k ) ck ( t − τ k ) c1 ( t ) dt
∫
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
At Receiver
• m(t) =+/-1 (PSK), bit duration Tb
m a
• Interfering signal may change amplitude at τk
a r
Ik =
τk
S h Tb
cos (φk − θ1 ) b−1 ∫ ck ( t − τ k ) c1 ( t ) dt +b0 ∫ ck ( t − τ k ) c1 ( t ) dt
τk
h
0
i s
• Tb
At User 1: I1 =∫ m1 ( t ) c1 ( t ) c1 ( t ) dt
h
0
• Ideally, spreading codes are Orthogonal:
∫
Tb
0 A s
c1 ( t ) c1 ( t ) dt = A ∫
Tb
0
ck ( t − τ k ) c1 ( t ) dt = 0
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Multiple Access Interference (MAI)
m a
a r
h
1
Pb = Q
S
( K − 1) 3N +ℵ 2 Eb
i sh
h
• If the users are assumed to be equal power interferers, can be
s
analyzed using the central limit theorem (sum of IID RV’s)
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Example of Performance Degradation
m a
a r
S h
i sh
s h
A N=8 N=32
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Near/Far Problem (I)
m a
r
• Performance estimates derived using assumption that all users have same power
a
level
• Adjust power levels constantly to keep equal
S h
• Reverse link (mobile to base) makes this unrealistic since mobiles are moving
i sh
s h
A k 1
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Near/Far Problem (II)
m a
a r
h
(1)
Pb = Q
1
S
∑ k =2 b
K ( k ) 3E (1) N +ℵ 2 E (1)
h
E
b b
h i s
s
• K interferers, one strong interfering signal dominates
A
performance
• Can result in capacity losses of 10-30%
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Multipath Propagation
m a
a r
S h
i sh
s h
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
RAKE Receiver
m a
a r
S h
i sh
• Received signal sampled at the rate 1/Ts> 2/Tc for detection and
h
synchronization
s
• Fed to all M RAKE fingers. Interpolation/decimation unit provides a data
A
stream on chiprate 1/Tc
• Correlation with the complex conjugate of the spreading sequence and
weighted (maximum-ratio criterion)summation over one symbol
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
RAKE Receiver
m a
r
• RAKE Receiver has to estimate:
• Multipath delays
Phase of multipath components
h a
S
•
• Amplitude of multipath components
•
sh
Number of multipath components
i
h
• Main challenge is receiver synchronization in fading channels
A s
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Orthogonal Codes
m a
r
• Orthogonal codes
a
• All pairwise cross correlations are zero
h
• Fixed- and variable-length codes used in CDMA systems
S
• For CDMA application, each mobile user uses one sequence in the set as a spreading code
h
• Provides zero cross correlation among all users
i s
• Types
h
• Walsh codes
s
• Variable-Length Orthogonal codes
A
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
Walsh Codes
• Set of Walsh codes of length n consists of the n rows of
m a
r
an n ´ n Walsh matrix:
• W1 = (0)
W2n
Wn
=
h
Wn
a
S
Wn Wn
sh
• n = dimension of the matrix
i
• Every row is orthogonal to every other row and to the logical
h
not of every other row
s
• Requires tight synchronization
A
• Cross correlation between different shifts of Walsh sequences is
not zero
ETEC 405: Wireless Communications
Faculty: ASHISH SHARMA,CSE,MAIT
