Frequency Hopping Direct Sequence 1 Shannons Theorem and Nyquists Theorem channel capacity bps channel bandwidth Hz SNR is the signaltoNoise ratio Unitless ID: 612019
Download Presentation The PPT/PDF document "Spread Spectrum Techniques" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Spread Spectrum Techniques
Frequency HoppingDirect Sequence
1Slide2
Shannon’s Theorem and
Nyquist’s Theorem
channel capacity,
[bps]. channel bandwidth, [Hz]. SNR is the signal-to-Noise ratio [Unitless].Shannon’s theorem has some interesting implications:For a given capacity, :There is no limit to how small the bandwidth can be, provided that SNR is sufficiently large.There is no limit to how small the SNR can be, provided that the bandwidth is sufficiently large.For a given Bandwidth, :There is no limit to the capacity provided that SNR is sufficiently large.There is no limit to how small SNR can be, provided that the Capacity is reduced accordingly.
2Slide3
Cont.
Rearranging Shannon’s equation gives:
For large
:
3Slide4
Graph C/B vs. SNR (dB)
4
SNR (dB)
C/BSlide5
Negative SNR
Shannon’s theorem acknowledges the possibility that the noise power can be at a higher level than the signal’s power, this situation does not make data transmission impossible.
Considering the equation, data transmission is possible as long as
is positive, that is when there is some signal power.
Shannon’s theorem therefore tells us that data transmission is possible under these conditions.
For negative values of however, will be less than 1. e.g., for an SNR of -30 dB, a bandwidth of 714 kHz would be required to carry data at 1 kbps. 5Slide6
Nyquist’s
Theorem Although Shannon’s theorem places a limit on the data that can be carried, it is not the only limit.Nyquist’s
theorem places an absolute limit on the symbol (baud) rate for a channel of a particular bandwidth regardless of noise levels.According to Nyquist’s theorem a channel of bandwidth
can carry a maximum of
symbols per second.
6Slide7
Example
It is required to transmit data at a rate of 9600 bps over a channel of bandwidth 2000 Hz. Use Shannon’s theorem to determine the minimum signal to noise ratio required and, referring to Nyquist’s theorem. Comment on a suitable modulation scheme.
Solution:
Therefore a SNR of at least 14.3 dB is required.The bandwidth 2000 Hz limits the baud rate to 4000 Baud
Thus there must be at least 3 bits per symbol.
7Slide8
Spread Spectrum
Spread spectrum is a communication technique that spreads a
narrowband communication signal over a wide range of frequencies for transmission then de-spreads it into the original data bandwidth at the receive. Makes jamming and interception harder
Two Techniques:
Frequency hoping
Signal broadcast over seemingly random series of frequenciesDirect SequenceEach bit is represented by multiple bits in transmitted signalChipping (spreading) code/sequence8Slide9
Spread Spectrum Concept
Input fed into channel encoder Produces narrow bandwidth signal around central frequencySignal modulated using sequence of digits
Spreading code/sequenceTypically generated by pseudonoise/pseudorandom number generator (PN)Increases bandwidth significantly
Spreads spectrum
Receiver uses same sequence to demodulate signal
Demodulated signal fed into channel decoder9Slide10
General Model of Spread Spectrum System
10Slide11
11
Spread spectrum
ConceptSlide12
12Slide13
13Slide14
14Slide15
Frequency Hopping Spread Spectrum (FHSS)
Signal broadcast over seemingly random series of frequenciesReceiver hops between frequencies in sync with transmitterEavesdroppers hear unintelligible blips
Jamming on one frequency affects only a few bits
Hedy
Lamarr15Slide16
16
Slide17
17
Slide18
Basic Operation
Typically 2k carriers frequencies forming 2k channelsChannel spacing corresponds with bandwidth of inputEach channel used for fixed interval
300 ms in IEEE 802.11Some number of bits transmitted using some encoding schemeMay be fractions of bit (see later)
Sequence dictated by spreading code
18Slide19
FHSS
19
Frequency
Time
f4
f3
f2
f1
Repeating period
Spreading Factor
Slide20
FHSS
20
Frequency
Time
f4
f3
f2
f1
Repeating periodSlide21
Multiplexing
21
Frequency
Time
f4
f3
f2
f1
Repeating period
User1
User2Slide22
Multiplexing
22
Frequency
Time
f4
f3
f2
f1
Repeating period
User1
User2Slide23
Frequency Hopping Spread Spectrum System
23Slide24
Frequency Hopping Spread Spectrum System
24Slide25
Slow and Fast FHSS
Frequency shifted every Tc secondsDuration of signal symbol is TM
secondsSlow FHSS has Tc T
M
Fast FHSS has
Tc < TMGenerally fast FHSS gives improved performance in noise (or jamming)25Slide26
Slow Frequency Hop Spread Spectrum Using MFSK (M=4, k=2)
26
Slide27
Fast Frequency Hop Spread Spectrum Using MFSK (M=4, k=2)
27
Slide28
Direct Sequence Spread Spectrum (DSSS)
Each bit represented by multiple bits using spreading (chipping
)code/sequence. This process
is called
Processing
Gain.The bits resulting from combining the information bits with the chipping code are called chips - the result- which is then transmitted.Spreading code spreads signal across wider frequency bandIn proportion to number of bits used10 bit spreading code spreads signal across 10 times bandwidth of 1 bit code Processing Gain Spreading Factor (SF) =10SF is the number of chips within each symbol duration , let be the chip duration
: chipping rate,
: Baud rate.
One method
:Combine input with spreading code using XORInput bit 1 inverts spreading code bitInput zero bit doesn’t alter spreading code bitData rate equal to original spreading codePerformance similar to FHSS 28Slide29
Direct Sequence Spread Spectrum Example
: XOR
29Slide30
Another method (polar)
30
Slide31
Approximate Spectrum of
DSSS Signal
The BW of
is
BW:
31Slide32
Direct Sequence Spread Spectrum Transmitter
32Slide33
Direct Sequence Spread Spectrum Receiver
Spread Signal
Original Signal
33Slide34
34
DSSS Overall Transmit/Receive
user data
chipping
sequence
modulator
radio
carrier
spread
spectrum
signal
transmit
signal
transmitter
demodulator
received
signal
radio
carrier
chipping
sequence
lowpass
filtered
signal
receiver
integrator
products
decision
data
sampled
sums
correlator
X
X
PN generator
PN generator
: we can look at it as a pulse shaping
Slide35
DSSS Using BPSK
Multiply BPSK signal,
by
[takes values +1, -1] to get
amplitude of signal
carrier frequency
digital baseband signal
pulse shaping signal
bit duration
At receiver, incoming signal multiplied by
Since,
, original signal is recovered
35Slide36
Direct Sequence Spread Spectrum Using BPSK Example
36Slide37
37
Narrowband
vs.
Spread Spectrum
Frequency
Power
Spread Spectrum
(Low Peak Power)
Narrowband
(High Peak Power)
The bandwidth increases with spreading but spectral power density necessary for transmission decreases. Spread spectrum needs only very small power densities, often below the level of natural background noise.Slide38
Gains
Immunity from various noise and multipath distortionIncluding jammingCan hide/encrypt signals
Only receiver who knows spreading code can retrieve signalAdvantagesreduces frequency selective fading
in cellular networks
base stations can use the
same frequency rangeseveral base stations can detect and recover the signalsoft handoverDisadvantagesprecise power control necessary38Slide39
39
Gains (cont.)
Problem of radio transmission: frequency dependent fading can wipe out narrow band signals for duration of the interference
Solution: spread the narrow band signal into a broad band signal using a special code - protection against narrow band interference
detection at
receiver
Interference
or noise
spread signal
signal
spread
Interference or noise
f
f
power
power
frequency
channel
quality
1
2
3
4
5
6
Narrowband signal
guard space
narrowband channels
2
2
2
2
2
frequency
channel
quality
1
spread
spectrum
spread spectrum channelsSlide40
40
Effects of spreading
on noise and interference
+
x
Rx
The noise is spreadSlide41
41
Effects of spreading and interference
PSD
f
i)
PSD
f
ii)
sender
PSD
f
iii)
PSD
f
iv)
receiver
f
v)
user signal
broadband interference
narrowband interference
PSDSlide42
Code Division Multiple Access (CDMA)
Multiplexing Technique used with spread spectrum
Start with data signal rate
Buad
rate (symbols per second)
Break each symbol into SF chips according to fixed pattern specific to each user (User’s spreading code)SF is the number of chips within each symbol duration , let be the chip duration New channel has chip data rate chips per second (cps). Spreading code repetition period (code length)
:
Short code:
an exact pattern of the PN will repeat each data symbol.
Long code:
Slide43
Short vs. Long Code
43
Short
Long
1 0 1 1 0 1 0 01 0 0 1 0 1 1 0 1 0 0 1 0 1 0 0 1 0 1 0 1 1 0 0 1 0 1 1 0 1 1 01 0 0 1 1 0 0 1 1 0 0 1 0 1 0 0 0 1 0 1 1 1 0 0 1 0 0 1 0 0 1 1Slide44
CDMA Example
e.g. SF=6, three users (A,B,C) communicating with base receiver RCode for A = <1,-1,-1,1,-1,1>
Code for B = <1,1,-1,-1,1,1>Code for C = <1,1,-1,1,1,-1>
Short codeSlide45
CDMA Explanation
Consider A communicating with baseBase knows A’s codeAssume communication already synchronized
A wants to send a 1Send chip pattern <c1, c2, c3, c4, c5, c6>, e.g.<1,-1,-1,1,-1,1>
A’s code
A wants to send 0
Send chip pattern <-c1, -c2, -c3, -c4, -c5, -c6>, e.g.<-1,1,1,-1,1,-1>Complement of A’s codeReceiver knows sender’s code and performs electronic decode function<d1, d2, d3, d4, d5, d6> = received chip pattern<c1, c2, c3, c4, c5, c6> = sender’s codeDecoder ignores other sources when using A’s code to decodeOrthogonal codes45Slide46
46
CDMA Example
User A code = <1,
–
1,
–1, 1, –1, 1>To send a 1 bit = <1, –1, –1, 1, –1, 1>To send a 0 bit = <–1, 1, 1, –1, 1, –1>User B code = <1, 1, –1, – 1, 1, 1>To send a 1 bit = <1, 1, –1,
–1, 1, 1>To send a 0 bit = <
–1,–1, 1, 1,
–
1,
–1>Receiver receiving with A’s code(A’s code) x (received chip pattern)User A ‘1’ bit: 6 1User A ‘0’ bit: -6 0User B ‘1’ bit: <1, –1, –1, 1, –1,
1>X<1, 1, –1, –1, 1, 1>=0
unwanted signal ignored
Slide47
CDMA for DSSS
n users each using different orthogonal PN sequenceModulate each users data streame.g. Using BPSKMultiply by spreading code of user
47Slide48
CDMA in a DSSS Environment
48Slide49
Seven Channel CDMA Encoding and Decoding
49Slide50
50
demodulator
received
signal
radio
carrier
chipping
sequence
lowpass
filtered
signal
receiver
integrator
products
decision
data
sampled
sums
correlator
X
PN generator
Slide51
Multiple Access Interference (MAI)
51
integrator
X
integrator
X
integrator
X
.
.
.
X
X
X
.
.
.
Channel
Compute the interference at the output of one receiver caused by the remaining M-1 users.
Focus on the time interval
and the
k
th
symbol of all users.
The
th
user transmit one symbol
over the time interval
.
Each user receives the
k
th
symbol of all users within
.
The
i
th
user’s transmitted power is
.
The
i
th
user’s channel has a gain
.
Slide52
MAI
Where
Is the time-varying cross-correlation between two spreading codes.
For short code
are periodic and the period is
, the spreading codes are identical over each interval
the cross correlation between two spreading codes is a constant (does not change with
k
)
52Slide53
MAI
Where
is the MAI
When
the spreading codes are selected to be orthogonal i.e.
So
53Slide54
Pseudorandom Numbers
Also called Pseudonoise(PN) because the autocorrelation of it resembles that of a white noise (an impulse).
Generated by algorithm using initial seedDeterministic algorithmNot actually randomIf algorithm good, results pass reasonable tests of randomness
Need to know algorithm and seed to predict sequence
Pseudonoise
(PN) sequence chosen so that its autocorrelation is very narrow PSD is very wide, in order to make the correlation between a code and a shifted version of it approximately zero.Concentrated around t < Tc toCross-correlation between two codes is very small, if it is zero Orthogonal. 54Slide55
PN Sequence Generation
Codes are periodic and generated by a shift register and XOR
Maximum-length (ML) shift register sequences (m-sequence), m-stage shift register, length:
bits
-
1/L
T
c
t
-
LT
c
LT
c
Output
1
-
T
c
D
Delay tap by
(shift register)
D
D
D
…
D
D
XOR (Modulo-2 adder)
55Slide56
Generating PN Sequences
Take m=2
L=3cn
=[1,1,0,1,1,0, . . .], usually written as bipolar
cn=[1,1,-1,1,1,-1, . . .] DD
+
Output
00
01
10
11
1
1
0
Finite State Machine (FSM)
Dead state
Slide
56Slide57
Properties of
m-sequencesThe cross correlation between an m-sequence and noise is low
This property is useful to the receiver in filtering out noiseThe cross correlation between two different m-sequences is low
This property is useful for CDMA applications
Enables a receiver to discriminate among spread spectrum signals generated by different m-sequences
Easy to guess connection setup in 2m samples so not too secureIn practice, Gold codes or Kasami sequences which combine the output of m-sequences are used.57Slide58
Gold Codes
Gold sequences constructed by the XOR of two m-sequences with the same clocking
Codes have well-defined cross correlation propertiesOnly simple circuitry needed to generate large number of unique codes
58Slide59
59
Orthogonal Variable Spreading Factor Codes (OVSF)
C
ch
,
i
,j
selected from this tree
Notes:
1) For fixed chip rate, desired information
rate determines length of spreading
sequence and therefore processing gain.
2) When a specific code is used, no other
code on the path from that code to the root
and
on
the
subtree
beneath that
code may be used.
3) All the codes at any depth into the tree
are the set of Walsh Sequences.
4) Code phase is synchronous with
information symbols.
5) FDD UL processing gain between 256 and 4
FDD DL processing gain between 512 and 4
TDD UL/DL processing gain between 16 and 1
6)
Multicode
used only for SF = 4
(
,
)
(
,
)
Walsh-
Hadamard
codes
,
,
integer,
Are perfect orthogonal such that:
,
,
Requires tight synchronization, Cross correlation between different shifts of Walsh sequences is not zero