chapter 7 spread-spectrum modulationshannon.cm.nctu.edu.tw/comtheory/chap7.pdf · 1 chapter 7...
TRANSCRIPT
1
Chapter 7 Spread-Spectrum Modulation
Spread Spectrum Technique simply consumes spectrum in excess of the minimum spectrumnecessary to send the data.
© Po-Ning [email protected] Chapter 7-2
7.1 Introduction
o Definition of spread-spectrum modulationn Weakly sense
o Occupy a bandwidth that is much larger than the minimum bandwidth (1/2T) necessary to transmit a data sequence.
n Strict senseo Spectrum is spreading by means of a pseudo-white
or pseudo-noise code.
2
© Po-Ning [email protected] Chapter 7-3
7.2 Pseudo-noise sequences
o A (digital) code sequence that mimics the (second-order) statistical behavior of a white noise.
o For example,n balance propertyn run propertyn correlation property
o From implementation standpoint, the most convenient way to generate a pseudo-noise sequence is to employ several shift-registers and a feedback through combinational logic.
© Po-Ning [email protected] Chapter 7-4
7.2 Pseudo-noise sequences
Exemplified block diagram of PN sequence generators
n Feedback shift register becomes “linear” if the feedback logic consists entirely of modulo-2 adders.
3
© Po-Ning [email protected] Chapter 7-5
7.2 Pseudo-noise sequences
o Example of linear feedback shift register
© Po-Ning [email protected] Chapter 7-6
7.2 Pseudo-noise sequences
o A PN sequence generated by a (possibly non-linear) feedback shift register must eventually become periodic with period at most 2m, where m is the number of shift registers.
o A PN sequence generated by a linear feedback shift register must eventually become periodic with period at most 2m - 1, where m is the number of shift registers.
o A PN sequence whose period reaches its maximum value is named the maximum-length sequence or simply m-sequence.
4
© Po-Ning [email protected] Chapter 7-7
7.2 Pseudo-noise sequences
o A maximum-length sequence generated from a linear shift register satisfies all three properties:n Balance property
o The number of 1s is one more than that of 0s.n Run property (total number of runs = 2m-1)
o ½ of the runs is of length 1o ¼ of the runs is of length 2o …
© Po-Ning [email protected] Chapter 7-8
n Correlation propertyo Autocorrelation of an
ideal discrete white process = a·d[t], where d[t] is the Kronecker delta function.
7.2 Pseudo-noise sequences
c(t)
5
© Po-Ning [email protected] Chapter 7-9
7.2 Pseudo-noise sequences
o Power spectrum view
Suppose c(t) is perfectly white with c2(t) = 1. Then,
m(t) = b(t)c(t) and b(t) = m(t)c(t).
Question: What will be the power spectrum of m(t)?
© Po-Ning [email protected] Chapter 7-10
7.3 A notion of spread spectrum
b(t)
c(t)
m(t) b(t)
c(t)
6
© Po-Ning [email protected] Chapter 7-11
7.2 Pseudo-noise sequences
o Please self-study Example 7.2 for m-sequences.n Its understanding will be part of the exam.
© Po-Ning [email protected] Chapter 7-12
7.3 A notion of spread spectrum
b(t)
c(t)
m(t) b(t)
c(t)
Make the transmitted signal to hide behind the background noise.
Spreading code
7
© Po-Ning [email protected] Chapter 7-13
o Example n Recall Slides 6-35 ~ 6-38
© Po-Ning [email protected] Chapter 7-14
n Noting that
we obtain
9
© Po-Ning [email protected] Chapter 7-17
7.3 A notion of spread spectrum
channel receivertransmitter
Lowpass filter that (only) allows signal b(t) to pass!
© Po-Ning [email protected] Chapter 7-18
7.4 Direct-sequence spread spectrum with coherent binary phase-shift keying
o DSSS system
transmitter
receiver
10
© Po-Ning [email protected] Chapter 7-19
© Po-Ning [email protected] Chapter 7-20
(As the conceptual system below.)
11
© Po-Ning [email protected] Chapter 7-21
7.5 Signal-space dimensionality and processing gain
o SNR before spreading
o SNR after spreading
o Orthonormal basis used at the receiver end
Assume “coherent detection”.In other words, perfect synchronization and no phase mismatch.
o SNR before spreading
© Po-Ning [email protected] Chapter 7-22
12
o SNR after spreading
© Po-Ning [email protected] Chapter 7-23
© Po-Ning [email protected] Chapter 7-24
13
© Po-Ning [email protected] Chapter 7-25
Mismatch with the text in SNRIo SNR before spreading
o SNR after spreading
o Orthonormal basis used at the receiver end
© Po-Ning [email protected] Chapter 7-26
Assume “coherent detection”.In other words, perfect synchronization but with phase mismatch.
14
o SNR before spreading
© Po-Ning [email protected] Chapter 7-27
© Po-Ning [email protected] Chapter 7-28
15
© Po-Ning [email protected] Chapter 7-29
o SNR after spreading
© Po-Ning [email protected] Chapter 7-30
Same as Slides 7-20 and 7-21, we derive
16
© Po-Ning [email protected] Chapter 7-31
7.6 Probability of error
The analysis for SNRI and SNRO assumes perfectly white noise j(t) for simplicity. Note that a white noise j(t) has infinite power such that j(t) and c(t)j(t) will have the same PSD !
© Po-Ning [email protected] Chapter 7-32
Slide 6-32 said that
17
© Po-Ning [email protected] Chapter 7-33
6.3 Coherent phase-shift keying – Error probability
o Error probability of Binary PSKn Based on the decision rule
© Po-Ning [email protected] Chapter 6-32
© Po-Ning [email protected] Chapter 7-34
7.6 Probability of error
o Comparing system performances with/without spreading, we obtain:
o With P = Eb/Tb, where P is the average signal power,
n J/P is termed the jamming margin (required for a specific error rate).
18
o Example 7.3n Without spreading, (Eb/N0) required for Pe = 10-5 is
around 10 dB.n PG = 4095n Then, Jamming margin for Pe = 10-5 is
n Information bits can be detected subject to the required error rate, even if the interference level is 409.5 times larger than the received signal power (in the price of the transmission speed is 4095 times slower).
© Po-Ning [email protected] Chapter 7-35
© Po-Ning [email protected] Chapter 7-36
7.7 Frequency-hop spread spectrum
o Basic characterization of frequency hoppingn Slow-frequency hopping
n Fast-frequency hopping
n Chip rate (The smallest unit = Chip)
19
© Po-Ning [email protected] Chapter 7-37
7.7 Frequency-hop spread spectrum
o A common modulation scheme for FH systems is the M-ary frequency-shift keying
© Po-Ning [email protected] Chapter 7-38
0
7
6
5
4
3
2
1
Slow-frequency hopping
The smallest unit= Chip
20
© Po-Ning [email protected] Chapter 7-39
7
6
5
4
3
2
1
Fast-frequency hopping
The smallest unit= Chip
© Po-Ning [email protected] Chapter 7-40
7.7 Frequency-hop spread spectrum
o Fast-frequency hopping is popular in military use because the transmitted signal hops to a new frequency before the jammer is able to sense and jam it.
o Two detection rules are generally used in fast-frequency hoppingn Make decision separately for each chip, and do majority
vote based on these chip-based decisions (Simple)n Make maximum-likelihood decision based on all chip
receptions (Optimal)
21
© Po-Ning [email protected] Chapter 7-41
7.8 Computer experiments: Maximum-length and gold codes
o Code-division multiplexing (CDM)n Each user is assigned a different spreading code.
© Po-Ning [email protected] Chapter 7-42
7.8 Computer experiments: Maximum-length and gold codes
o So, if
then signal one (i.e., s1) can be exactly reconstructed.o In practice, it may not be easy to have a big number of PN
sequences satisfying the above equality. Instead, we desire
22
© Po-Ning [email protected] Chapter 7-43
7.8 Computer experiments: Maximum-length and gold codes
o Gold sequences
© Po-Ning [email protected] Chapter 7-44
7.8 Computer experiments: Maximum-length and gold codes
o Gold sequencesn g1(x) and g2(x) are two maximum-length shift-register
sequences of period 2m - 1, whose “cross-correlation” lies in:
n Then, the structure in previous slide can give us 2m - 1 sequences (by setting different initial value in the shift registers).
n Together with the two original m-sequences, we have 2m
+ 1 sequences.
23
© Po-Ning [email protected] Chapter 7-45
7.8 Computer experiments: Maximum-length and gold codes
o Gold’s theoremn The cross-correlation between any pair in the 2m +1
sequences also lies in
o Experiment 1: Correlation properties of PN sequences
© Po-Ning [email protected] Chapter 7-46
24
n Autocorrelation
© Po-Ning [email protected] Chapter 7-47
27 - 1 = 127
n Cross-correlation of two maximum-length shift-register (PN) sequences, which are not Gold sequences.
© Po-Ning [email protected] Chapter 7-48
25
o Experiment 2: Correlation properties of Gold sequences
© Po-Ning [email protected] Chapter 7-49
n Cross-correlation
© Po-Ning [email protected] Chapter 7-50
15
-17
-1