m7 MQAM ErrorAnalysis
m7 MQAM ErrorAnalysis
!1
M-ary Modulation Types
• We have seen how to:
– Design an optimum M-ary Receiver
– Calculating the probability of bit and
symbol errors
– Considered some general transmission
signal types
• Now we wish to discuss specific but
popular modulation formats and
determine their properties and when to
use which one
!2
M-ary Phase-Shift Keying (MPSK)
• Assuming that
!3
Examples of MPSK constellation
!4
MPSK signal sets for M=2,4,8,16
!5
Error Performance of MPSK
• It can be shown that
!6
SER vs Eb/No for MPSK
!7
For large Es/No:
!9
Gray Coding
• Allows representation of symbols or bit-to-symbol
mapping
• In going from one symbol to an adjacent symbol,
only one bit out of the k (or n bits in text) bits
changes.
• An adjacent symbol error (i.e. the most likely
symbol error) will therefore be accompanied by one
and only one bit error.
• Thus, the bit error probability of Gray-coded
MPSK can be well approximated by
!10
M-QAM Modulation
Time Domain Description
sk (t) = ak cos( c t) + bk sin( c t) for t [0, Ts ], k = {1, 2, ..., M }
⌃
where ak , bk ⇤ {± , ±3 , ..., ±( M 1) } and M = {4, 16, 64, 256, ...}
!11
2
MQAM Performance
• Bit Rate: Rb = T1 log2 M
s
⇥
• Average Symbol Energy: Es = 2 + 2(3 ) + ...
2
M
2 2
⇥
– e.g. for 16QAM, Es = 2
4 2 + 2(3 ) = 10 2
2 2
• Remarks:
– Not all M points have the same energy.
– Information can be visualized as carried by
the amplitude and phase information of the
points
!12
MQAM Performance
• Symbol Error Probability (SER)
Analysis:
– M-QAM can be regarded as two
independent M-PAM, each having M = 2 k/2
points
2
!13
M-QAM Performance
• It can be shown that
⇥ ⇤ ⇥
P⇥M = 2 1 ⇥1
M
Q 3 Es
M 1 N0
• Hence, ⇥ ⇥
PM = 1 (1 P⇥M )2 ⇥ 2P⇥M = 4 1 ⇥1
M
Q 3 Es
M 1 N0
!15
Choices of M-ary Modulations
!16
Error Probability Performance Curves
• Allow us to design and set an operating point for a
system
• Consider MFSK
– Increasing M can provide an improvement in Pb,or
reduction in the Eb/N0 required, at the cost of increased
bandwidth
• Consider MPSK
– Increasing M can provide a reduction in bandwidth
requirement, at the cost of degraded Pb, or increase in
the Eb/N0 requirement
!17
Theoretical Limits on Performance
• Channel Capacity is the theoretical upper bound for the
maximum rate at which information could be transmitted
without error (Shannon 1948)
• For a bandlimited channel that is corrupted by AWGN
the maximum rate achievable is given by
!18
the ideal channel
Spectral Efficiency
Shannon Limit
• N=N0B, hence
• Then
!19
Bandwidth- efficiency plane
MPAM/MQAM
MPSK
Coherent
MFSK !20
Trade-Offs
• Power-Limited Systems: Power scarce but bandwidth
available
– Improved Pb by expanding bandwidth (for a
given Eb/N0 ) or required Eb/N0 can be reduced
by expanding bandwidth (for a given Pb)
• Bandwidth-Limited Systems: bandwidth scarce
– Maximize R over the bandlimited channel at the
expense of Eb/N0 (for a given Pb)
!21
Shannon Limit
• In the limit as R/B goes to 0, we get
!22
Summary of M-ary Modulation Schemes
M-FSK M-PSK M-QAM
Bit Rate 1 1 1
Rb = log2 (M ) Rb = log2 (M ) Rb = log2 (M )
Ts Ts Ts
BW (Bandpass) M 1 1
BW =
<latexit sha1_base64="63ex6CMatJbg99j4M8tf7vvI+Nc=">AAAB/XicdVDLSsNAFJ3UV62v+Ni5GWwFVyWplDYLodSNG6FCX9CGMJlO2qGTBzMToYbgr7hxoYhb/8Odf+OkraCiBy4czrmXe+9xI0aFNIwPLbeyura+kd8sbG3v7O7p+wddEcYckw4OWcj7LhKE0YB0JJWM9CNOkO8y0nOnl5nfuyVc0DBoy1lEbB+NA+pRjKSSHP2o1OzBCzj0OMLJdZq0HZGWHL1olK16tXpeh0bZMKyKVVPEsiyzZkJTKRmKYImWo78PRyGOfRJIzJAQA9OIpJ0gLilmJC0MY0EihKdoTAaKBsgnwk7m16fwVCkj6IVcVSDhXP0+kSBfiJnvqk4fyYn47WXiX94gll7dTmgQxZIEeLHIixmUIcyigCPKCZZspgjCnKpbIZ4gFYRUgRVUCF+fwv9Jt1I2Fb+pFBvNZRx5cAxOwBkwQQ00wBVogQ7A4A48gCfwrN1rj9qL9rpozWnLmUPwA9rbJ2LslI8=</latexit>
Ts
BW =
Ts
BW =
Ts
Average Transmit Es Es 4 2
log2 M/2
⇥ (2i 1)2
Power Ts Ts M Ts i=1
Average Symbol ⇤ ⇥ ⇤ ⇥
Es 2Es ⇥ ⇥
Error Probability Pe ⇥ (M 1)Q Pe 2Q sin( /M ) PM ⇥ 4 1 ⇥1 Q 3 Es
M 1 N0
N0 N0 M
(SER)
Remarks • Orthogonal Signaling • Equi-energy constellation • Points are NOT equi-energy
Schemes (Equi-energy points (information carried by phase
• Information is carried by
& Mutually orthogonal values only)
both amplitude and phase
signals)
• Dimension of the signal set • Enhance spectral efficiency
• Enhance Energy Efficiency is always 2 (I-Q modulator) at the expense of extra power
at the expense of extra BW
• Enhance spectral efficiency • Better than M-PSK for
at the expense of extra power
M>4.
!23