Module - 4

OFDM

Dr.Vetrivelan.P
VIT Chennai
FDM Vs OFDM
OFDM
Multicarrier Modulation
Multi-Carrier Modulation (MCM) is the
principle of transmitting data by dividing
the stream into several bit streams, each
of which has a much lower bit rate, and
by using these substreams to modulate
several carriers.
Single-carrier Basics
Consider a bandwidth B = 2W available for
communication, where W is the one-sided
bandwidth, or, the maximum frequency.
For a single carrier communication system, the
symbol time T is given as
This implies that symbols can be transmitted at
intervals of seconds each.
The symbol rate is given as
Single-carrier Basics
In a single-carrier system, a single carrier is
employed for the entire baseband bandwidth of
B.
Therefore, the symbols are transmitted as
symbol X(0) from 0 ≤ t < T, symbol X(1) from T
≤ t < 2T, and so on, i.e., roughly one symbol
transmitted every seconds.
 Multicarrier Basics
 Consider now dividing the total bandwidth B into N
sub-bands of bandwidth B/N each as shown in Figure.
 The subcarriers are placed at . . . , − B/N , 0, B/N , . . .,
as shown in the figure.
 For instance, consider the bandwidth B = 256 kHz with
N = 64 subcarriers.
 The bandwidth per sub-band is equal to 256/64 = 4 kHz,
which is also the frequency spacing between the
subcarriers.
Multicarrier Basics
 Consider the ith subcarrier at the frequency,
with .
 Let Xi denote the data transmitted on the ith subcarrier.
 Then, the signal si(t) corresponding to the ith subcarrier
is given as

where fi is the ith subcarrier centre frequency, and

is the ith subcarrier.
 The above equation shows the data modulation process
over the ith subcarrier.
 Multicarrier Transmission
Consider now the different modulated signals si(t)
corresponding to the N different subcarriers.
These signals are then superposed at the transmitter
to form the composite signal s(t) given as
 Multicarrier Transmission
This composite signal s(t) is then transmitted over
the wireless channels.
Thus, N different data streams are transmitted over
N subcarriers in parallel in this multicarrier system.
At the receivers, the individual data streams have
then to be isolated from the composite signal s(t).
 This is accomplished as follows.
Consider the signal y(t) received as
 Multicarrier Transmission
It can be seen that the expression on the right-hand
side is the Fourier series representation s(t),
corresponding to the fundamental frequency
f0=B/N and the various Xi representing the Fourier
coefficients.
To extract Xl, which is the Fourier coefficient
corresponding to the frequency one needs to
follow the procedure similar to compute the Fourier
series as,
Multicarrier Transmission
Multicarrier Transmission

 All the subcarriers other than the l th subcarrier are

orthogonal to the lth subcarrier.
 Multiplying with and integrating is coherent
demodulation, i.e., demodulation with the carrier
matched to the subcarrier frequency .
 Thus, Xl, the data modulated on the different subcarriers,
can be conveniently recovered by coherently
demodulating with each of the subcarriers corresponding
to
 The above scheme of transmission on multiple
orthogonal subcarriers and the associated data recovery
at the receiver is termed MultiCarrier Modulation
Multicarrier Transmission - OFDM
 The technique, where the MCM signal is generated
by employing the IFFT operation is termed
Orthogonal Frequency Division Multiplexing, or
OFDM.
 At the receiver, to recover the information symbols,
one can correspondingly employ an FFT operation.
OFDM - Basics
OFDM
OFDM
16- QAM Constellation
Contd…
Spectra of OFDM sun-channel and
OFDM signal
 Frequency Spectrum
Time domain ICI avoided
Orthogonality -Integer multiple of subcarriers

OFDM – Time and Frequency Domain

Multiple Channel vs Multicarrier
Communications
When one or more channels are
unreliable, signal fading due to multipath
Tropospheric and Ionospheric scatter
Multi-channel – Same information
transmitted over several channels
 Frequency band of the channel divided
into no. of sub-channels - Multicarrier
Effect of multipath
OFDM symbol with cyclic extension
OFDM – Two –ray multipath channel
 16-QAM constellation for 48 sub-
carrier OFDM link

(a) delay< guard time (b) delay(3%)>guard time

(c) delay(10%)>guard time
OFDM Transceiver
Coding
FEC- Channel Encoders
Redundancy to make the data more robust
for transmission through the channel
Trellis Coded Modulation – To Conserve
Bandwidth – gray encoded QAM
BER vs Eb/No AWGN Channel
BER vs Eb/No Rayleigh Fading Channel
 Interleaving
 Due to FSF(different sub-carrier different
amplitude), OFDM suffers bit errors occur as burst
 FEC –does not deal with error burst
 Interleaving is to randomize occurrence of bit errors
 In the transmitter, Coded bits are permutated
 Bit/Symbol Interleaving (Reed Solomon Codes)
Pilot Insertion
 Radio channels fade both in amplitude and
frequency
 Channel estimator – Estimate time-varying
amplitude and phase of all sub-carriers
 2-D channel estimator – estimate reference values
based on known pilot values
 By sampling theorem, Ts< Inverse of the double-
sided BW of the sampled signal
 Min. subcarrier spacing and min. symbol spacing
between pilots
 More pilots effective SNR small for data symbols
Pilots – Block of 9 OFDM symbols with
16 sub-carriers
Windowing
Sharp transitions-unfiltered QAM
subcarriers -Out-of band spectrum
decreases slowly acc. to sinc function
Windowing to individual OFDM symbols
Amplitude to go smoothly to zero at
boundaries
Contd..
Nc samples padded with zeros to get
N-samples –IFFT
Last Tprefix –samples inserted at start
First Tpostfix –samples inserted at end
OFDM symbol added to previous OFDM
symbol delay of Ts
 PSB with & without windowing
 PSD without Windowing  Spectra of RCW
 Windowing vs Filtering
OFDM symbol x window =
spectrum of window * impulse of subcarriers
OFDM symbol * impulse response of filter =
OFDM spectrum x filter frequency response
Filters introduce ripples -distort- less delay
spread tolerance
Filter – few multiplications per sample
Windowing - few multiplications per
symbol(roll-off region)
OFDM Transmitter
OFDM Receiver
Cyclic Prefix in OFDM
 Consider a frequency-selective channel modelled
with channel taps
 The received symbol y at a given time instant n can
be expressed as

 Considernow two OFDM symbols as follows.

 Let denote the IFFT
samples of the modulated symbols

denote the IFFT samples of the previous modulated

Cyclic Prefix in OFDM
 Thus, the samples corresponding to these two
blocks of OFDM symbols are transmitted
sequentially as

 Now, consider the received symbol y(0)

corresponding to the transmission of x(0).
 This can be expressed as
Cyclic Prefix in OFDM
 It can be seen from the above equation that the
received symbol y(0) experiences inter-symbol
interference from

 Thus, there is inter-OFDM symbol interference in

this new OFDM system.
 The initial samples of the current OFDM symbol
block are being subject to interference from the N
− 1 samples of the previous OFDM block.
Cyclic Prefix in OFDM
 Consider a modified transmission scheme as
follows.
 To each transmitted OFDM sample stream, we pad
the last Lc symbols to make the transmitted stream
as follows

 We are prefixing the transmitted sample block

of the current block with
the Lc samples
Cyclic Prefix in OFDM
 Thus, with the cyclic prefix of appropriate length,
, inter-OFDM symbol
interference can be avoided and inter-symbol
interference is restricted to samples from the same
OFDM symbol.
 The samples are given as
 It can now be clearly seen that the output y(n) is a
circular convolution between the channel filter h(n)
and the input x(n).
 This can be expressed as

 The output y can be written as

OFDM Transmitter with CP
OFDM Receiver with CP
Spectra of OFDM sun-channel and
OFDM signal
Design of ODFM
OFDM Example
 Thus, the total transmitted OFDM symbol duration
with cyclic prefix is 64 μs + 8 μs = 72 μs.
 The number of samples in the CP is

 Total number of samples is 256 + 32 = 288.

Problems - OFDM
 A 64 Kbps voice frame is to be modulated by OFDM scheme. The
duration of OFDM symbol is 1000 µs. Total of 32 subcarriers are to
be assigned to this frame. Compute the following:
◦ (a) Null-to-null sub-channel bandwidth assuming square signal
◦ (b) Total bandwidth occupied
◦ (c) Number of bits in OFDM frame
 OFDM - TX

https://www.scribd.com/presentation/375017376/29-
OfDM-19-Mar-2018-Reference-Material-I-OFDM-New-
18-3-2018
FFT & IFFT Processing
OFDM - Overall
OFDM System
 PAPR and its Reduction Methods
PAPR – Peak to Average Power Ratio
max x t 
2

PAPR  0t T
T
1/ T   x t  dt
2
0

 Envelope Fluctuation – Superposition of sub-carriers

 Causes Power Amplifiers to Saturate –leads to Non-
linear Distortion, Inter-modulation and Out-of Band
Radiation
 Increased Complexity of ADC and DAC
 BER high
 Spectral Spreading – Adjacent Channel Interference
(ACI) occurs
Distribution of PAPR
For a complex baseband signal

an – Modulating symbol
N – No. of sub-carriers
For Large N, real/imaginary value of x(t) –Gaussian
distributed
Amplitude – Rayleigh Distribution
Power Distribution – chi-square distribution
With zero mean and two degrees of freedom
z
F ( z ) 1  e
PAPR Reduction Techniques
Distortion Techniques
◦ Clipping
◦ Peak Windowing
◦ Peak Cancellation
Coding
Special FEC which exclude OFDM symbol with
large PAPR
Scrambling
Scramble OFDM to small PAPR
Clipping
Peak amplitude limited to desired level
Distorts the OFDM signal amplitude
(ie) Self-interference occurs degrades
BER
Non-linear distortion – leads to Out-of
band radiation
Peak Windowing
Clipping is OFDM signal x w(t)
(rectangular)
w(t) =  if OFDM amplitude< threshold
 <  if OFDM amplitude> threshold
(rectangular window only to analyze Out of Band)
Out-ofband – spectrum of rectangular
window
Contd…
Remedy to Out-of band
Clipping is OFDM signal x w(t)(non-
rectangular)
To minimize Out-of band interference,
Window NB
Not too long in time domain
Cosine, Hamming, Kaiser windows
Windowing of OFDM signal
Clipping vs Windowing
Out-of band characteristics
Contd…
Window length influencing Out-of band
Peak Cancellation(PC)
OFDM signal minus Time shifted and
scaled version reference function
Reference function – sinc function
Sinc – infinite time duration
Limited using raised cosine window
Sinc function windowed with raised
cosine window
Contd…
Accomplished by detector, comparator,
scaling of peak samples
PC on symbol-symbol basis
No stored reference
Impulse generated for each exceeded sample
Amplitude of impulse =peak amplitude – desired
maximum amplitude
Peak Cancellation effect

Peak amplitude reduced to maximum of 3dB above rms value

Scrambling
For each OFDM symbol, input sequence
scrambled by no.of scrambling sequences
O/p signal with smallest PAPR is transmitted
If a symbol has a probability ‘p’ > level(without
scrambling) then
Probability reduced to pk using ‘k’ scrambling
codes
Selected mapping – indp. Scrambling rotations to
all sub-carriers
Partial transmit sequence - Scrambling rotations
to group of sub-carriers
Contd…
Contd…
At -50dB BW= 2* -3dB BW
Scrambling gives 2dB gain for 10-scrambling codes
Benefits of Scrambling limited
Backoff –Power amplifier
To attain an acceptable level of out-of
band radiation
ratio of o/p power and max.o/p power
(sat.) with a sinusoidal i/p.
PAPR Reduction Codes
Codes produce OFDM symbols for which
PAPR<desirable level
Smaller desired level, smaller achievable
code rate
Golay Complementary sequences
 XnXn- = 0 delay shift,i0
Correlation properties of complementary
sequences translate into small PAPR of
3dB
 Golay Complementary Sequences
 Sequence x of length(N) is complementary to other sequence y

Taking Fourier Transform

 Guarantee that PAPR< 3dB for 16-channels

Contd..
Contd…
Generation of Complementary
sequences
Make kernel
◦ Derive complementary sequence from one
complementary pair
◦ A=  B=
◦ An Bn where An = An- Bn-
◦ Bn = An- -Bn-
◦ Codes of length 2n+  from codes of length 2n
Contd..
Determine no. of orthogonal subsets
◦ For length N codes, no. of orthogonal subsets log2N
– arbitrary phase offset
◦ N=16, log216 =4
◦ Subsets – elements, pairs, quads, octet
Contd…
Final Transformation
◦ Described as interleaving operation on underlying
short length codes that are used to make longer
length codes
◦ Code with length 2n interleaved n-times before
reproducing itself
◦ Interleaving st and 2nd half of original code

For a N-length code with M-possible phases,

amount of bits per codeword (1+log2N)log2M
Minimum Distance
 Min. Euclidean distance – det. performance in flat fading with
additive noise
 Min. distance between two code words is obtained if N/2
symbols have a minimum phase rotation 2 / M
Maximum Likelihood Decoding
 Length N = 2n code , n+ phases encoded to
2n output phases by adding the  st phase to
all code phases, 2nd to all odd code phases,
3rd to all odd pair code phases and so on.
 For large constellation sizes, maximum
likelihood decoding less trivial.
 Worst case it require M n+ Euclidean distance
calculations or correlations giving a total no.
of operations of NMn+(complex additions
and multiplications)
 Length 2/4 4-PSK fast Walsh transform

Length 2 4-PSK fast Walsh

transform
Standards
IEEE 802.11a – Wireless LAN
Define a high-speed (up to 54 Mbps) PHY
supplement in the 5-GHz band
Contd…
 U-NII– Unlicensed National Information Infrastructure
 Channel centre frequency = 5000+ 5* n ch (MHz)
 Where nch = 0,1,2, …., 200
Contd…
Contd…
Contd…
Contd…
Contd…
Contd..
Contd…
 PAPR Reduction Methods
Clipping – clip max. amplitude
- BER degrades, spectral efficiency
Filtering – reduce spectral splatter
Windowing – use windows of NB
Scrambling – Reduce the probability of large
peaks generated by bit patterns
Coding – Map the sequence into a large
sequence where high peak sequences
are not used; ECC
 OFDMA
Group of sub-carriers form a sub-channel
Sub-carriers with same color form a sub-channel
OFDMA
Grouped and Spread Interleaved Sub-
carrier Assignment
 Contd..
Grouped – less Inter-User Interference
More sensitive to fading
High PAPR
Difficulty in subcarrier synchronization
Need for coordinated subcarrier assignments
OFDMA Contd…
Nf bits in each frame parsed into Ñ
groups
ith group assigned ñ bits
Thank you…..
Contd…