BEC605: Wireless Communication

Module 4

Reference:
Wireless Communications
Andrea Goldsmith
Contents

4.1 Multicarrier Modulation

4.2 Data Transmission using Multiple Carriers
4.3 Multicarrier Modulation with Overlapping Subchannels
4.4 Mitigation of Subcarrier Fading- Adaptive loading, Precoding,
interleaving over time and frequency, Frequency Equalization
4.5 Challenges in Multicarrier systems
4.6 Discrete Implementation of OFDM
4.1 Multicarrier Modulation
• Multicarrier modulation means:
Instead of sending all data on one signal, we divide the data into many smaller
parts and send them at the same time on many small carrier frequencies called
subcarriers.
Example:
Suppose we want to send: ABCDEFGH
• Single-carrier system makes use of One fast stream: A B C D E F G H
• Multicarrier system uses:
Stream 1: A E
Stream 2: B F
Stream 3: C G
Stream 4: D H
• Each stream is slower.
Multicarrier Modulation
Why is this useful?
• Wireless channels create reflections called multipath.
• These reflections cause delayed copies of symbols.
• In a fast single-carrier system: Symbol 1 overlaps Symbol 2
• This overlap is called Intersymbol Interference (ISI)

• In multicarrier systems:

 Each stream is slower

 symbols are longer in time
 overlap becomes very small
 So, ISI reduces greatly.
Multicarrier Modulation

• OFDM is the most common multicarrier technique.

• In Orthogonal Frequency Division Multiplexing:
many subcarriers are used
subcarriers are orthogonal
they do not interfere with each other ideally
Multicarrier Modulation
• Each subcarrier bandwidth is kept smaller than the channel
coherence bandwidth.
• Then each subchannel experiences almost flat fading.
• So equalization becomes simple.
Condition:
𝐵subcarrier < 𝐵𝑐
Multicarrier Modulation
• OFDM also adds a cyclic prefix.
• A cyclic prefix is a copy of the end of the symbol added to the
beginning.
Example:
Original: ABCD
Add CP: CDABCD
• This removes ISI if the delay spread is smaller than the cyclic prefix
length.
• Condition:
𝑇𝐶𝑃 > 𝜏max
Multicarrier Modulation

• ABCD = actual OFDM data

• CD = cyclic prefix
Even though everything is delayed, the delayed copies still remain inside the CP region if
the delay is small enough.
Original signal:
CD|ABCD
Delayed echo:
CD|ABCD
Notice:
• the delayed copy shifts right
• overlap first happens in the CP region
Multicarrier Modulation
Receiver discards the first part:
• REMOVE → C D

Keep → A B C D

• Even though echoes existed, the useful part remains properly aligned.
• The CP absorbs the delay spread.
• Since the receiver throws away CP, the corrupted portion is discarded
4.2 Data Transmission using Multiple Carriers
Problem in Wideband Communication Main Idea of Multicarrier Modulation
• Suppose we transmit data very fast. High data rate means: Instead of one fast stream: ABCDEFGH
• very short symbol duration split into many slower streams:
• large bandwidth Carrier 1 → A E
Carrier 2 → B F
If channel has multipath delay spread: Carrier 3 → C G
• echoes of previous symbols overlap current symbol Carrier 4 → D H
Then:
ISI occurs Each stream now has:
because:
𝑇𝑠 ≈ 𝑇𝑚 • lower data rate
where: • longer symbol duration
• 𝑇𝑠 =symbol time • This reduces ISI.
• 𝑇𝑚 =channel delay spread
Data Transmission using Multiple Carriers
Bandwidth Division
• Original system bandwidth = 𝐵
• data rate = 𝑅
• Divide into 𝑁subchannels.
Then each subchannel has:
𝐵
𝐵𝑁 =
𝑁

and data rate:

𝑅
𝑅𝑁 ≈
𝑁
So each carrier becomes narrowband.

• 4. Why Narrowband Helps

Data Transmission using Multiple Carriers
Why Narrowband Helps? Smaller bandwidth means larger symbol time.
Relationship:
• Wireless channel has coherence 1
bandwidth 𝐵𝑐 . 𝑇𝑁 ≈
𝐵𝑁
Coherence bandwidth is inversely related to delay spread:
If: 1
𝐵𝑁 ≪ 𝐵𝑐 𝐵𝑐 ≈
𝑇𝑚

Since 𝐵𝑁 is small:
then each subchannel experiences: 𝑇𝑁 ≫ 𝑇𝑚
• almost flat fading
• very little ISI •symbol duration becomes much larger than delay spread
•echoes die out before next symbol interferes
• This is the MOST important idea. So ISI becomes very small.
Data Transmission using Multiple Carriers
All subcarrier signals are added:
𝑁−1

𝑠 𝑡 = ෍ 𝑠𝑖 𝑔 𝑡 cos 2𝜋𝑓𝑖 𝑡 + 𝜙𝑖
𝑖=0
where:
• 𝑠𝑖 =symbol on ith carrier
• 𝑓𝑖 =ith carrier frequency
• 𝑔 𝑡 =pulse shape function
Note: If we transmit abrupt rectangular signals:
• bandwidth becomes very large
• sidelobes increase
• neighboring channels get interference
• So we shape the symbols smoothly using 𝑔 𝑡 .
Data Transmission using Multiple Carriers
Receiver:
• separates subchannels using filters
• demodulates each carrier
• combines streams back together
Advantages
• Multicarrier modulation:
• reduces ISI
• simplifies equalization
• works well in multipath channels
Data Transmission using Multiple Carriers
Problems with Basic Multicarrier System
• The passage explains several issues.
(a) Large Bandwidth Requirement
• Real pulse shapes are not ideal.
• Extra bandwidth needed:
𝑁 1+𝛽+𝜖
𝐵=
𝑇𝑁
where:
• 𝛽= rolloff factor
• 𝜖= extra bandwidth due to practical pulse shaping
• So spectrum efficiency decreases.
Data Transmission using Multiple Carriers
(b) Expensive Filters

• Receiver needs very sharp filters to separate carriers.

• This increases complexity and cost

(c) Many Modulators/Demodulators

• Basic multicarrier needs one modulator per subcarrier and one
demodulator per subcarrier
• If 𝑁is large, hardware becomes expensive
• power consumption increases
Data Transmission using Multiple Carriers
Why OFDM Was Invented?
OFDM solves these problems by:
• allowing overlapping subcarriers
• maintaining orthogonality
• using FFT/IFFT instead of many modulators

This makes OFDM:

• efficient
• practical
• low complexity
Data Transmission using Multiple Carriers
Problem:
A multicarrier communication system has:
• Number of subchannels: 𝑁 = 64
• Symbol duration per subchannel: 𝑇𝑁 = 0.1 ms
• Raised cosine rolloff factor: 𝛽 = 0.5
• Additional bandwidth factor due to time-limited pulse shaping: 𝜖
= 0.2
Calculate the total bandwidth required for the system.
Data Transmission using Multiple Carriers
Using:
𝑁 1+𝛽+𝜖
𝐵=
𝑇𝑁

Substitute values:
64 1 + 0.5 + 0.2
𝐵=
0.0001

64 1.7
𝐵=
0.0001

108.8
𝐵=
0.0001

𝐵 = 1,088,000 Hz

𝐵 = 1.088 MHz
4.3 Multicarrier Modulation with Overlapping
Subchannels
• As the number of subcarriers increases, the wasted bandwidth becomes
very large.
• To improve spectral efficiency, OFDM allows the subchannels to overlap. At
first this seems impossible because overlapping frequencies should
interfere with each other. The key idea is that the subcarriers are chosen to
be orthogonal.
OFDM allows subcarriers to overlap:
• Even though spectra overlap:
• subcarriers remain orthogonal
• receiver can separate them perfectly
• This gives very high spectral efficiency
Multicarrier Modulation with Overlapping
Subchannels
The OFDM subcarriers are defined as

𝑖
cos 2𝜋 𝑓0 + 𝑡 + 𝜙𝑖
𝑇𝑁
where:
• 𝑓0 is the starting carrier frequency
• 𝑖is the subcarrier index
• 𝑇𝑁 is the symbol duration
• 𝜙𝑖 is the phase offset
Adjacent subcarriers are separated by
1
Δ𝑓 =
𝑇𝑁

• This spacing is extremely important because it guarantees orthogonality over the interval 0 𝑇𝑁 .
Multicarrier Modulation with Overlapping
Subchannels
Multicarrier Modulation with Overlapping
Subchannels
Subcarriers are chosen as:
𝑖
𝑓𝑖 = 𝑓0 +
𝑇𝑁
So spacing between adjacent carriers is:
1
Δ𝑓 =
𝑇𝑁
• This spacing guarantees orthogonality.
• Suppose one carrier completes exactly an integer number of cycles more than
another during symbol duration 𝑇𝑁 .
• Then positive and negative overlaps cancel out.
• Net overlap = 0
• So receiver can separate carriers perfectly.
Multicarrier Modulation with Overlapping
Subchannels
• Two signals are orthogonal if:
𝑇𝑁
න 𝑠𝑖 𝑡 𝑠𝑗 𝑡 𝑑𝑡 = 0
0
for 𝑖 ≠ 𝑗.
• average overlap energy is zero
• signals do not interfere after demodulation
• When two OFDM subcarriers are multiplied and integrated over one symbol
duration, the positive and negative parts cancel exactly. One carrier completes an
integer number of additional cycles compared to another carrier during 𝑇𝑁 .
Therefore their average overlap becomes zero even though they overlap
physically in frequency.
• This means overlapping subcarriers do not interfere after demodulation.
Multicarrier Modulation with Overlapping
Subchannels
• The actual transmitted basis functions are
𝑔 𝑡 cos 2𝜋𝑓𝑖 𝑡 + 𝜙𝑖

where 𝑔 𝑡 is the pulse-shaping function, usually a raised cosine pulse.

These basis functions form an orthonormal set over the interval
0 𝑇𝑁 .
The transmitted OFDM signal becomes
• 𝑠 𝑡 = σ𝑁−1𝑖=0 𝑠𝑖 𝑔 𝑡 cos 2𝜋𝑓𝑖 𝑡 + 𝜙𝑖
where 𝑠𝑖 is the symbol transmitted on the 𝑖-th subcarrier.
Multicarrier Modulation with Overlapping
Subchannels
• At the receiver, the received signal is multiplied by the desired basis
function and integrated over 𝑇𝑁 .The receiver computes
𝑇𝑁
𝑠Ƹ𝑖 = න 𝑠 𝑡 𝑔 𝑡 cos 2𝜋𝑓𝑖 𝑡 + 𝜙𝑖 𝑑𝑡
0
• Substituting the transmitted signal into this expression produces
many cross-product terms between subcarriers. Because the basis
functions are orthogonal, all cross terms become zero. Only the
desired term survives:
𝑠Ƹ𝑖 = 𝑠𝑖
• Thus the receiver can perfectly recover the desired symbol even
though all subcarriers overlap in frequency.
Multicarrier Modulation with Overlapping
Subchannels
• Since the subcarriers overlap, the total bandwidth becomes much smaller:
𝑁+𝛽+𝜖
𝐵=
𝑇𝑁

• For large 𝑁,

𝑁
𝐵≈
𝑇𝑁
• Thus overlapping OFDM subcarriers achieve much higher spectral efficiency than non-
overlapping systems.
• In a real wireless channel, each subcarrier experiences flat fading and noise. The received
symbol becomes
• 𝑠Ƹ𝑖 = 𝛼𝑖 𝑠𝑖 + 𝑛𝑖
Multicarrier Modulation with Overlapping
Subchannels
where:
• 𝛼𝑖 = 𝐻 𝑓𝑖 is the channel gain at subcarrier frequency
• 𝑛𝑖 is additive white Gaussian noise
• One important drawback is that orthogonality is very sensitive. Small
timing errors or frequency offsets destroy orthogonality and cause
intercarrier interference (ICI). Therefore OFDM systems require very
accurate synchronization
4.4 Mitigation of sub Carrier Fading
• Multicarrier modulation reduces the effect of delay spread because each subchannel is narrowband. Since the bandwidth of each
subcarrier is small, each subchannel experiences almost flat fading instead of frequency-selective fading. This greatly simplifies
equalization and reduces intersymbol interference.
• However, even though ISI is reduced, each individual subcarrier still suffers from flat fading. Flat fading can severely reduce the
received signal power on some subcarriers and increase the bit error rate.
• Suppose the transmit power on the 𝑖-th subcarrier is 𝑃𝑖 ,the channel fading coefficient is 𝛼𝑖 ,the noise spectral density is 𝑁0 ,and the
subchannel bandwidth is 𝐵𝑁 .Then the received signal-to-noise ratio on that subcarrier is
𝛼𝑖2 𝑃𝑖
𝛾𝑖 =
𝑁0 𝐵𝑁
• This equation shows that the received SNR depends directly on the fading coefficient 𝛼𝑖 .
• If 𝛼𝑖 becomes small because of deep fading, then
𝛾𝑖 → small
which means the received SNR becomes poor.
• Low SNR causes a high bit error rate on that subcarrier. Therefore some OFDM subchannels may become unreliable even
though the system successfully removed ISI.
• In wireless channels, the fading coefficient 𝛼𝑖 changes continuously with time according to the fading environment. Thus each
OFDM subcarrier behaves like an independent flat-fading channel. This is similar to flat fading in single-carrier systems, except that
OFDM has many parallel flat-fading subchannels
Mitigation of sub Carrier Fading
Methods to mitigate sub carrier fading:
1. Adaptive Loading
2. Coding with Interleaving over Time and Frequency
3. Frequency Equalization
4. Precoding
Adaptive Loading:
• Adaptive loading is a technique used in multicarrier systems and OFDM to
improve performance by dynamically changing the power and data rate of each
subcarrier according to channel conditions.
• In OFDM, different subcarriers experience different fading levels. Some
subcarriers may have very good channel conditions while others may be deeply
faded. If the same power and modulation are used for all subcarriers, then
badly faded subcarriers produce many errors and waste power.
Mitigation of sub Carrier Fading
Adaptive loading solves this problem by assigning:
• more power and higher data rate to good subcarriers
• less power or no transmission to poor subcarriers
Thus system resources are used efficiently.
• The transmitter must know the fading gain of every subcarrier:
{𝛼𝑖 ; i = 0, . . . ,N −1}.
• This information is called channel state information (CSI).
Suppose:
• there are 𝑁 independent subchannels , each subchannel has bandwidth 𝐵𝑁 , the gain of the 𝑖-th subchannel
is 𝛼𝑖 , power allocated to subcarrier 𝑖is 𝑃𝑖 , total available power is 𝑃
• Then the total capacity is
𝑁−1
𝛼𝑖2 𝑃𝑖
𝐶 = max ෍ 𝐵𝑁 log 2 1 +
𝑃𝑖 :෌ 𝑃𝑖 =𝑃 𝑁0 𝐵𝑁
𝑖=0
Mitigation of sub Carrier Fading
• The goal is to choose power allocation 𝑃𝑖 that maximizes total capacity while
satisfying the total power constraint.
• The optimal solution is called water-filling.
• The power allocation formula is
1 1
𝑃𝑖 − , 𝛾𝑖 ≥ 𝛾𝑐
= ൞𝛾𝑐 𝛾𝑖
𝑃
0, 𝛾𝑖 < 𝛾𝑐
where
𝛼𝑖2 𝑃
𝛾𝑖 =
𝑁0 𝐵𝑁
is the received SNR of the 𝑖-th subcarrier, and 𝛾𝑐 is the cutoff threshold.
Mitigation of sub Carrier Fading
• The cutoff value 𝛾𝑐 is chosen so that the total allocated power satisfies:
෍ 𝑃𝑖 = 𝑃
𝑖
• After water-filling, only subcarriers with
𝛾𝑖 ≥ 𝛾𝑐 are used.
• The resulting system capacity becomes
𝛾𝑖
𝐶 = ෍ 𝐵𝑁 log 2 𝛾𝑐
𝑖:𝛾𝑖 ≥𝛾𝑐
Thus adaptive loading significantly improves spectral efficiency and total
throughput by intelligently distributing power and modulation levels across
OFDM subcarriers according to channel quality.
Mitigation of sub Carrier Fading
(2) Coding with Interleaving over Time and Frequency
• To mitigate subcarrier fading, channel coding is applied to the data bits before transmission.
• The encoder converts the original data into coded bits containing redundancy, so that errors
introduced by fading can later be corrected at the receiver.
•The coded bits are then interleaved over both time and frequency. Interleaving spreads adjacent
coded bits across different subcarriers and different OFDM symbols instead of placing them
close together.
• Because of interleaving, all bits belonging to the same codeword are transmitted through different fading
conditions. Even if a few subcarriers are deeply faded, only some coded bits are affected, while most bits are
received correctly through other subcarriers with good channel quality.
• At the receiver, the decoder uses the redundancy introduced by channel coding to detect and correct errors
caused by faded subcarriers. Thus, the effect of severe fading on a small number of subcarriers can be
overcome
Mitigation of sub Carrier Fading
Frequency equalization: In multicarrier modulation systems such as Orthogonal Frequency-Division𝑡ℎ
Multiplexing, each subcarrier experiences flat fading independently. The fading coefficient of the 𝑖 subcarrier
is represented by 𝛼𝑖 .
• Frequency equalization is a technique used to compensate for the fading effect on each subcarrier at the
receiver.
• In this method, the receiver multiplies the received signal of the 𝑖𝑡ℎ subcarrier by the inverse of the channel
fading coefficient:
1
𝛼𝑖
• This operation removes the attenuation caused by the channel fading and restores the signal amplitude.
• If the received signal power on the 𝑖𝑡ℎ subcarrier is:
𝛼𝑖2 𝑃𝑖
• after equalization, the signal power becomes:
𝛼𝑖2 𝑃𝑖
= 𝑃𝑖
𝛼𝑖2
Mitigation of sub Carrier Fading
1
• However, the received noise is also multiplied by the same factor .
𝛼𝑖
• Therefore, the noise power after equalization becomes:
𝑁0 𝐵𝑁
𝛼𝑖2
• When the fading coefficient 𝛼𝑖 is very small during deep fading, the equalizer
greatly amplifies the noise along with the signal.
• As a result, although the fading effect on the signal is removed, the signal-to-
noise ratio (SNR) after equalization remains unchanged.
• Hence, frequency equalization does not improve the SNR degradation caused
by subcarrier fading; it only compensates for the signal attenuation.
• Therefore, frequency equalization alone cannot effectively mitigate
performance degradation in severely faded subcarriers.
Mitigation of sub Carrier Fading
Precoding is used to compensate for subcarrier fading by performing channel inversion at
the transmitter instead of the receiver.
• In this technique, the transmitter must know the fading gain 𝛼𝑖 of each subcarrier before
transmission. These channel gains are obtained through channel estimation and
feedback.
• If the desired received signal power in the 𝑖 𝑡ℎ subcarrier is 𝑃𝑖 ,and the channel
introduces a fading gain 𝜶𝒊 ,then the transmitter increases the transmitted power
according to:
𝑃𝑖
𝛼𝑖2
• The transmitted signal passes through the fading channel with gain 𝛼𝑖 .Therefore, the
received signal power becomes:
2 𝑃𝑖
𝛼𝑖 × 2 = 𝑃𝑖
𝛼𝑖
Mitigation of sub Carrier Fading
• Thus, the fading effect of the subcarrier is compensated before
transmission itself.
• Unlike frequency equalization, channel inversion occurs at the
transmitter, so the receiver noise is not amplified.
• The noise power at the receiver remains:
𝑁0 𝐵𝑁
• Hence, precoding avoids the noise enhancement problem associated
with receiver-side equalization.
Numerical:
A multicarrier communication system has a symbol duration 𝑇𝑁 = 0.2ms,
which is much greater than the channel delay spread, ensuring minimal ISI in
each subchannel. The system employs 𝑁 = 128 subchannels. Raised cosine
pulses with roll-off factor 𝛽 = 1 are used, and additional bandwidth factor
due to time limiting is 𝜖 = 0.1.
(i) Determine the total system bandwidth when subchannels are non-
overlapping.
(ii) Determine the total system bandwidth when subchannels are
overlapping.
(iii) Compare the two results and comment on the improvement in spectral
efficiency due to overlapping subchannels.
(i) Total bandwidth for non-overlapping subchannels
For non-overlapping subchannels, ii) Total bandwidth for overlapping subchannels
𝑁 1+𝛽+𝜖
𝐵= For overlapping subchannels,
𝑇𝑁 𝑁+𝛽+𝜖
• Substituting values, 𝐵 =
𝑇𝑁
128 1 + 1 + 0.1
𝐵= Substituting values,
0.0002 128 + 1 + 0.1
128 2.1 𝐵=
= 0.0002
0.0002 129.1
268.8
= =
0.00026 0.0002
= 1.344 × 10 Hz = 645.5 × 103 Hz
𝐵 = 1.344 MHz 𝑁
𝐵 = 645.5 kHz ≈ = 640 kHz
𝑇𝑁
 4.5 Challenges in Multicarrier Systems
(1) Very Large PAR
(2) Frequency and Timing offset

Very large PAR:

•In multicarrier systems such as Orthogonal Frequency-Division Multiplexing, the transmitted signal is formed by
adding many subcarrier signals together.
•At certain instants, the subcarrier components may add constructively and produce very large peaks in the
transmitted signal amplitude.
•At other instants, the subcarriers may partially cancel each other, resulting in low signal amplitudes.
•Because of this large variation between peak power and average power, OFDM signals exhibit a high Peak-to-
Average Power Ratio (PAR).
•A high PAR forces the transmitter power amplifier to operate with a large backoff in its linear region to avoid
nonlinear distortion of the signal.
Challenges in Multicarrier Systems
• This reduces the efficiency of the power amplifier and increases
power consumption.
• High PAR also increases the dynamic range requirement of the
receiver A/D converter, leading to greater hardware complexity and
higher power requirements.
• As the number of subcarriers increases, the probability of coherent
addition of subcarrier signals also increases, causing PAR to grow
approximately linearly with the number of subcarriers.
• Therefore, although multicarrier systems provide high spectral
efficiency and robustness against ISI, high PAR is a major practical
disadvantage.
Challenges in Multicarrier Systems
• In multicarrier systems such as Orthogonal Frequency-Division Multiplexing, the transmitted OFDM signal is
obtained by adding many subcarrier signals together using the IFFT operation:
𝑁−1
1
𝑥𝑛 = ෍ 𝑋 𝑖 𝑒 𝑗2𝜋𝑖𝑛/𝑁
𝑁 𝑖=0
where
• 𝑁= number of subcarriers
• 𝑋 𝑖 =data symbol transmitted on the 𝑖𝑡ℎ subcarrier
• The Peak-to-Average Power Ratio (PAR) for a discrete-time signal is defined as
max ∣ 𝑥 𝑛 ∣2
𝑃𝐴𝑅 = 𝑛
𝐸𝑛 ∣ 𝑥 𝑛 ∣2
where
• max 𝑛 ∣ 𝑥 𝑛 ∣2 =peak signal power
• 𝐸𝑛 ∣ 𝑥 𝑛 ∣2 =average signal power
Thus, the maximum PAR increases linearly with the number of subcarriers.
Challenges in Multicarrier Systems
•
The average power of the OFDM signal is approximately
𝐸 ∣ 𝑥 𝑛 ∣2 = 1
because of the normalization factor 1/ 𝑁.
• The maximum signal power occurs when all subcarrier components add coherently in phase.
• In that case,
1
𝑥𝑛 = 𝑋0 + 𝑋1 + ⋯ + 𝑋𝑁−1
𝑁
• If all subcarriers have equal amplitude and phase,
𝑁
𝑥𝑛 = = 𝑁
𝑁
• Hence,
∣ 𝑥 𝑛 ∣2 = 𝑁
• Therefore, the maximum PAR becomes
𝑁
𝑃𝐴𝑅𝑚𝑎𝑥 = =𝑁
1

Thus, the maximum PAR increases linearly with the number of

subcarriers.
Challenges in Multicarrier Systems
(2) Frequency and Timing offset:
• Another major challenge in multicarrier systems such as Orthogonal Frequency-Division Multiplexing is Intercarrier Interference
(ICI).
• In OFDM, data symbols are transmitted on orthogonal subcarriers.
• Orthogonality is maintained when the subcarrier spacing is
1
Δ𝑓 =
𝑇𝑁
• where:
• 𝑇𝑁 =OFDM symbol duration
• Due to this orthogonality, overlapping subcarriers do not interfere with each other under ideal conditions.

In practice, perfect orthogonality is difficult to maintain because of:

• carrier frequency mismatch
• Doppler frequency shift
• timing synchronization errors
• These imperfections introduce a frequency offset between transmitter and receiver.
• Because of the frequency offset, the FFT output of one subcarrier leaks into adjacent subcarriers, producing Intercarrier
Interference (ICI).
4.6 Discrete implementation of OFDM
Discrete implementation of OFDM
• Orthogonal Frequency-Division Multiplexing is a multicarrier modulation technique in which the total bandwidth is divided into
many orthogonal narrowband subcarriers.
• The input binary data stream is first mapped into complex modulation symbols using QAM modulation.
• These modulated symbols are represented as
𝑋 0 ,𝑋 1 ,…,𝑋 𝑁 − 1
where:
• 𝑁= number of subcarriers
• A serial-to-parallel converter divides the high-rate data stream into 𝑁parallel low-rate symbol streams.
• Each symbol is transmitted on a different orthogonal subcarrier.
• The OFDM time-domain signal is generated using the Inverse Fast Fourier Transform (IFFT):
𝑁−1
1
𝑥𝑛 = ෍ 𝑋 𝑖 𝑒𝑗2𝜋𝑛𝑖/𝑁 , 0 ≤ 𝑛 ≤ 𝑁 − 1
𝑁 𝑖=0
where:
• 𝑋 𝑖 =QAM symbol on the 𝑖 𝑡ℎ subcarrier
• 𝑥 𝑛 =OFDM time-domain sample
Discrete implementation of OFDM
• The exponential term
𝑒 𝑗2𝜋𝑛𝑖/𝑁
represents orthogonal subcarrier signals.
• Thus, the OFDM signal is obtained by adding many orthogonal subcarrier
components together.
• To eliminate Inter Symbol Interference (ISI), a cyclic prefix (CP) of length 𝜇 is
added before transmission.
• The cyclic prefix consists of the last 𝜇samples of the OFDM symbol appended to
the beginning:
𝑥෤ 𝑛 = 𝒙 𝑵 − 𝝁 , … , 𝒙 𝑵 − 𝟏 , 𝑥 0 , … , 𝑥 𝑁 − 1
• The cyclic prefix converts linear convolution of the channel into circular
convolution.
• This preserves subcarrier orthogonality even in multipath channels.
Discrete implementation of OFDM
• The transmitted signal passes through the wireless channel and is
corrupted by noise.
• The received signal is
𝑦 𝑛 = 𝑥෤ 𝑛 ∗ ℎ 𝑛 + 𝑣 𝑛
where:
• ℎ 𝑛 =channel impulse response
• 𝑣 𝑛 =additive noise
• At the receiver:
• cyclic prefix is removed
• FFT is applied to recover frequency-domain symbols
Discrete implementation of OFDM
• After FFT, the received symbols become
𝑌 𝑖 =𝐻 𝑖 𝑋 𝑖
where:
• 𝐻 𝑖 =channel gain of the 𝑖 𝑡ℎ subcarrier
• Thus, OFDM converts a frequency-selective wideband channel into many narrowband
flat-fading subchannels.
• Each subcarrier experiences only flat fading, simplifying equalization.
•
The original transmitted symbols can be recovered using frequency equalization:
𝑌𝑖
෠
𝑋𝑖 =
𝐻𝑖
• However, frequency equalization also amplifies noise when 𝐻 𝑖 becomes small during
deep fades.