This action might not be possible to undo. Are you sure you want to continue?
10/30/98/TUDTVS
OFDM as a possible modulation technique for multimedia
applications in the range of mm waves
Dušan Matiæ
Abstract  In this paper is given an overview of a multiple carrier modulation technique known as
OFDM (Orthogonal Frequency Division Multiplex). It focuses on problems that are specific for its
use in the future mobile multimedia communications (MMC) in the range of 60 GHz.
I Introduction
Multimedia is effectively an infrastructure technology with widely different origins in computing,
telecommunications, entertainment and publishing. New applications are emerging, not just in the wired
environment, but also in the mobile one. At present, only low bitrate data services are available to the
mobile users. However, demands of the wireless multimedia broadband system are anticipated within
both public and private sector.
This report discusses possible ways to enable multimedia communications in the mobile
environment. Multimedia communication has a rather large demands upon bandwidth and quality of
service (QoS) compared to what is available today to the mobile user. Bitrates for multimedia span from
a few Kb/s, for voice, to about 20 Mb/s for HDTV, or even more in the peaks.
When solving this problem, first question is how to put this large bit stream on air with
sufficient QoS guaranties, i.e. which modulation can compromise all contradicting requirements in the
best manner. The radio environment is harsh, due to the many reflected waves and other effects. Using
adaptive equalization techniques at the receiver could be the solution, but there are practical difficulties
in operating this equalization in realtime at several Mb/s with compact, lowcost hardware. A
promising candidate that eliminates a need for the complex equalizers is the Orthogonal Frequency
Division Multiplexing (OFDM), a multiple carrier modulation technique. This modulation system is
described, its applications and drawbacks are outlined, along with some important characteristics of
OFDM and singlecarrier techniques.
The second question is how to deal with a number of users wanting to exploit one
communication medium. Some way of sharing the common medium is needed. Multiple access
techniques are quite developed for the single carrier modulations (e.g. TDMA, FDMA), but there is
nothing in the literature concerning OFDM. Downlink is easy, but uplink poses serious troubles on the
system designer.
II The mobile environment
The main problem with reception of radio signals is fading caused by multipath propagation. Also, there
are intersymbol interference (ISI), shadowing, interference. This makes link quality vary. Further
constraints are limited bandwidth, low power consumption, network management and multicellular
operation.
Introduction to OFDM, II edition 2
10/30/98/TUDTVS
As a result of the multipath propagation, there are many reflected signals, which arrive at the
receiver at different times. Delayed signals are the result of reflections from terrain features such as
trees, hills or mountains, or objects such as people, vehicles or buildings. These echoes cause ISI.
Combined, these signals can produce fading. Some of these reflections can be avoided by using a
directional antenna (current trend is simple antennas), but it is impossible to use them for a mobile user.
A solution could be usage of antenna arrays, but this technology is still being developed.
A characteristic of frequency selective fading is that some frequencies are enhanced, whereas
others are attenuated. If there is a mobile reception, then the relative lengths and attenuations of the
various reception paths will change with time. A narrowband signal will vary in quality as the peaks and
the troughs of the frequency response move around in frequency domain. There will also be a noticeable
variation in phase response, which will affect all systems using phase as a means of signalling .
Fig 1. Typical frequency response of a channel suffering from multipath propagation. The frequency response will
vary with both time and position. Frequency selective attenuation is clearly present.
Now consider a signal which is of greater bandwidth. Some parts of the signal may suffer
from constructive interference and be enhanced in level, whereas others may suffer from destructive
interference and be attenuated, sometimes to the point of extinction. In general, frequency components
that are close together will suffer variations in signal strength that are strongly correlated. The
correlation (or coherence) is used as a measure of this phenomenon. There is no standard definition of
the correlation bandwidth. For a narrowband signal, distortion is usually minimised if the bandwidth is
less than the correlation bandwidth of the channel, because all frequencies in the band are usually
distorted in the same way. There is, however, a significant chance that the signal will be subject to
severe attenuation on some occasions. A signal which occupies a wider bandwidth, greater than
correlation bandwidth, will be subject to more distortion, but will suffer less variation in total received
power, even if it is subject to significant levels of multipath propagation. This comes from the fact that
variation averages out if the bandwidth is much larger than the correlation bandwidth, because different
parts of the band suffer different levels of distortion. One can often find following formula for
correlation bandwidth B
c
, where D is the RMS value of delay spread (not average).
c
B
D
≈
1
If we look at the temporal response of the channel, we see a number of echoes present. There
are many different types of echo environment, which are typical of different outdoor/indoor areas. This
range of delay can be measured and then processed to get statistical parameters. Different studies use the
total range of delay, or the average delay. Whichever is chosen, the inverse of this leads to a good
approximation for the correlation bandwidth.
amplitude
time or position
frequency
Introduction to OFDM, II edition 3
10/30/98/TUDTVS
Spread spectrum techniques are robust against fading and interference, but they set forth
impossible demands on the existing technology – for instance, if a user needs a speed of 20Mb/s on air
and the spreading factor is 128 (today’s typical), this results in 2.56 GB/s which have to processed in
realtime and have impracticably large bandwidth. Besides that, they have difficulty with the nearfar
effect and have a large powerconsumption.
Singlecarrier techniques are vulnerable to fading and multipath propagation, especially in the
case of very high bitrates. Improvements can be made with frequency equalization and directional
antennas, which can also be used to improve multicarrier techniques.
Recently, research and development of the OFDM have received considerable attention and
have made a great deal of progress in Europe. OFDM is a wideband modulation scheme that is
specifically able to cope with the problems of the multipath reception. This is achieved by transmitting
many narrowband overlapping digital signals in parallel, inside one wide band. Increasing the number
of parallel transmission channels reduces the data rate that each individual carrier must convey, and that
lengthens the symbol period. As a result, the delay time of reflected waves is suppressed to within 1
symbol time.
The 60 GHz band
The scarcity of spectrum and the new technical possibilities in recent years have drawn attention all over
the world to the millimetre band. It has become a hot topic as a research area for broadband
communications.
The low mmwave band from 2060 GHz, which is nearly unused and allows for large
bandwidth applications, combines the advantages of infrared (enough free bandwidth and UHF (good
coverage). Use of the area around 60 GHz is encouraged for the following reasons.
• The coherence bandwidth of a 60 GHz link is several MHz. [1]
• There is enough unused space for the multimedia needs (5 GHz) [2]. This frequency region is not in
use by any other medium, so to every user can be allocated a large bandwidth, i.e. bitrates of an order
of 155Mb/s are possible, thus a potential exists to support broadband service access, which is
especially relevant because of the advent of the Broadband Digital Network (BISDN).
• Systems operating particularly in the 60 GHz frequency band can have a small reuse distance,
because of the oxygen absorption at the rate of 14 dB/km [1]. Usually, this is a disadvantage for
many applications; however, this high attenuation over the propagation path creates a natural barrier
for cochannel interference in the mobile cellular system. Thus, frequency reuse is easy.
• Wavelength is as little as 5 mm. This enables crating of small antenna's and other parts of the radio
part of the system. This size of communication equipment makes it easy for wearing.
A major drawback of this frequency region is the fact that the technology for the transceivers
will be expensive in the early stages. If used indoor, the mmwave radio channel shows adverse
frequency selective multipath characteristics due to the highly reflective indoor environment, which
results in severe signal dispersion and limits the maximum usable symbol rate. It is worth mentioning
that no definitive evidence of any hazards has been shown to date to the general public arising from the
prolonged exposure in fields of less than 10 mW/cm
2
in the mmwaves.
There is a lot of fundamental investigation needed in this area, e.g. propagation modeling for
60 GHz, effects of antenna diversity, technology development, etc.
For the final choice between OFDM and singlecarrier modulation, one needs knowing the channel
properties at 60 GHz. Measurements and models can hardly be found in the literature. These properties
have to be built in the simulation model, which could then be used for evaluating modulation technique
candidates for the mobile multimedia communication.
Introduction to OFDM, II edition 4
10/30/98/TUDTVS
III OFDM
OFDM history
The concept of using parallel data transmission by means of frequency division multiplexing (FDM) was
published in mid 60s [3,4]. Some early development can be traced back in the 50s. A U.S. patent was
filled and issued in January, 1970. The idea was to use parallel data streams and FDM with overlapping
subchannels to avoid the use of high speed equalization and to combat impulsive noise, and multipath
distortion as well as to fully use the available bandwidth. The initial applications were in the military
communications. In the telecommunications field, the terms of discrete multitone (DMT), multichannel
modulation and multicarrier modulation (MCM) are widely used and sometimes they are
interchangeable with OFDM. In OFDM, each carrier is orthogonal to all other carriers. However, this
condition is not always maintained in MCM. OFDM is an optimal version of multicarrier transmission
schemes.
Fig. 2 Comparison of the bandwidth utilization for FDM and OFDM
For a large number of subchannels, the arrays of sinusoidal generators and coherent demodulators
required in a parallel system become unreasonably expensive and complex. The receiver needs precise
phasing of the demodulating carriers and sampling times in order to keep crosstalk between subchannels
acceptable. Weinstein and Ebert [5] applied the discrete Fourier transform (DFT) to parallel data
transmission system as part of the modulation and demodulation process. In addition to eliminating the
banks of subcarrier oscillators and coherent demodulators required by FDM, a completely digital
implementation could be built around specialpurpose hardware performing the fast Fourier transform
(FFT). Recent advances in VLSI technology enable making of highspeed chips that can perform large
size FFT at affordable price.
In the 1980s, OFDM has been studied for highspeed modems, digital mobile communications
and highdensity recording. One of the systems used a pilot tone for stabilizing carrier and clock
frequency control and trellis coding was implemented. Various fast modems were developed for
telephone networks.
In 1990s, OFDM has been exploited for wideband data communications over mobile radio FM
channels, highbitrate digital subscriber lines (HDSL, 1.6 Mb/s), asymmetric digital subscriber lines
(ADSL, 1,536 Mb/s), very highspeed digital subscriber lines (VHDSL, 100 Mb/s), digital audio
broadcasting (DAB) and HDTV terrestrial broadcasting.
Conventional Frequency Division Multiplex (FDM) multicarrier modulation technique
Orthogonal Frequency Division Multiplex (OFDM) multicarrier modulation technique
Saving of the bandwidth
frequency
frequency
Introduction to OFDM, II edition 5
10/30/98/TUDTVS
Qualitative description of OFDM
In multimedia communication, a demand emerges for highspeed, highquality digital mobile portable
reception and transmission. A receiver has to cope with a signal that is often weaker than desirable and
that contains many echoes. Simple digital systems do not work well in the multipath environment.
Fig. 3 The effect of adopting a multicarrier system. For a given overall data rate, increasing the number of carriers
reduces the data rate that each individual carrier must convey, and hence (for a given modulation system) lengthens
the symbol period. This means that the intersymbol interference affects a smaller percentage of each symbol as the
number of carriers and hence the symbol period increases (after [10[DM1]]). For example, on the picture is shown a
1 carrier
2 carrier
parallel
transmission
time
channel
responses
δ(t)
Step(t)
Step(t)
time
time
8 carrier
parallel
transmission
frequency
δ(t)
f
f
f
t
t
t
t
t
t
t
1st carrier
2nd carrier
7th carrier
8th carrier
1st carrier
2nd carrier
Introduction to OFDM, II edition 6
10/30/98/TUDTVS
8 bit long part of a data sequence. For a single carrier system, the responses of individual bits are overlapping, thus
creating ISI. Multicarrier system is robust against these physical effects.
In a conventional serial data system, the symbols are transmitted sequentially, with the
frequency spectrum of each data symbol allowed to occupy the entire available bandwidth. In a parallel
data transmission system several symbols are transmitted at the same time, what offers possibilities for
alleviating many of the problems encountered with serial systems.
In OFDM, the data is divided among large number of closely spaced carriers. This accounts
for the “frequency division multiplex” part of the name. This is not a multiple access technique, since
there is no common medium to be shared. The entire bandwidth is filled from a single source of data.
Instead of transmitting in serial way, data is transferred in a parallel way. Only a small amount of the
data is carried on each carrier, and by this lowering of the bitrate per carrier (not the total bitrate), the
influence of intersymbol interference is significantly reduced. In principle, many modulation schemes
could be used to modulate the data at a low bit rate onto each carrier.
It is an important part of the OFDM system design that the bandwidth occupied is greater
than the correlation bandwidth of the fading channel. A good understanding of the propagation statistics
is needed to ensure that this condition is met. Then, although some of the carriers are degraded by
multipath fading, the majority of the carriers should still be adequately received. OFDM can effectively
randomize burst errors caused by Rayleigh fading, which comes from interleaving due to paralellisation.
So, instead of several adjacent symbols being completely destroyed, many symbols are only slightly
distorted. Because of dividing an entire channel bandwidth into many narrow subbands, the frequency
response over each individual subband is relatively flat. Since each subchannel covers only a small
fraction of the original bandwidth, equalization is potentially simpler than in a serial data system. A
simple equalization algorithm can minimize meansquare distortion on each subchannel, and the
implementation of differential encoding may make it possible to avoid equalization altogether [5]. This
allows the precise reconstruction of majority of them, even without forward error correction (FEC).
In addition, by using a guard interval the sensitivity of the system to delay spread can be
reduced [8].
In a classical parallel data system, the total signal frequency band is divided into N non
overlapping frequency subchannels. Each subchannel is modulated with a separate symbol and, then, the
N subchannels are frequency multiplexed. There are three schemes that can be used to separate the
subbands:
1. Use filters to completely separate the subbands. This method was borrowed from the conventional
FDM technology. The limitation of filter implementation forces the bandwidth of each subband to be
equal to (1+α)f
m
, where α is the rolloff factor and f
m
is the Nyquist bandwidth. Another
disadvantage is that it is difficult to assemble a set of matched filter when the number of carriers is
large.
2. Use staggered QAM to increase the efficiency of band usage. In this way the individual spectra of the
modulated carriers still use an excess bandwidth, but the are overlapped at the 3 dB frequency. The
advantage is that the composite spectrum is flat. The separability or orthogonality is achieved by
staggering the data (offset the data by half a symbol). The requirement for filter design is less critical
than that for the first scheme.
3. Use discrete Fourier transform (DFT) to modulate and demodulate parallel data. The individual
spectra are now sinc functions and are not band limited. The FDM is achieved, not by bandpass
filtering, but by baseband processing. Using this method, both transmitter and receiver can be
implemented using efficient FFT techniques that reduce the number of operations from N
2
in DFT,
down to NlogN.
OFDM can be simply defined as a form of multicarrier modulation where its carrier spacing is carefully
selected so that each subcarrier is orthogonal to the other subcarriers. As is well known, orthogonal
signals can be separated at the receiver by correlation techniques; hence, intersymbol interference among
channels can be eliminated. Orthogonality can be achieved by carefully selecting carrier spacing, such as
letting the carrier spacing be equal to the reciprocal of the useful symbol period. Mathematical deduction
of the orthogonal carrier frequencies is given in [3].
In order to occupy sufficient bandwidth to gain advantages of the OFDM system, it would be
good to group a number of users together to form a wideband system, in order to interleave data in time
and frequency (depends how broad is one user signal).
Introduction to OFDM, II edition 7
10/30/98/TUDTVS
The importance of coding
The distribution of the data over many carriers means that selective fading will cause some bits to be
received in error while others are received correctly. By using an errorcorrecting code, which adds extra
bits at the transmitter, it is possible to correct many or all of the bits that were incorrectly received. The
information carried by one of the degraded carriers is corrected, because other information, which is
related to it by the errorcorrecting code, is transmitted in a different part of the multiplex (and, it is
hoped, will not suffer from the same deep fade). This accounts for the “coded” part of the name
COFDM.
There are many types of error correcting codes, which could be used.
The importance of orthogonality
The “orthogonal” part of the OFDM name indicates that there is a precise mathematical relationship
between the frequencies of the carriers in the system. In a normal FDM system, the many carriers are
spaced apart in such way that the signals can be received using conventional filters and demodulators. In
such receivers, guard bands have to be introduced between the different carriers (Fig. 2.), and the
introduction of these guard bands in the frequency domain results in a lowering of the spectrum
efficiency.
It is possible, however, to arrange the carriers in an OFDM signal so that the sidebands of the
individual carriers overlap and the signals can still be received without adjacent carrier interference. In
order to do this the carriers must be mathematically orthogonal. The receiver acts as a bank of
demodulators, translating each carrier down to DC, the resulting signal then being integrated over a
symbol period to recover the raw data. If the other carriers all beat down to frequencies which, in the
time domain, have a whole number of cycles in the symbol period (τ), then the integration process
results in zero contribution from all these carriers. Thus the carriers are linearly independent (i.e.
orthogonal) if the carrier spacing is a multiple of 1/τ.
Mathematically, suppose we have a set of signals ψ, where ψ
p
is the pth element in the set.
The signals are orthogonal if
( ) ( ) Ψ Ψ
p q
a
b
t t dt
K for p q
for p q
*
·
·
≠
¹
'
¹
∫
0
,
where the * indicates the complex conjugate and interval [a,b] is a symbol period. A fairly simple
mathematical proof exists, that the series sin(mx) for m=1,2,… is orthogonal over the interval π to π.
Much of transform theory makes the use of orthogonal series, although they are by no means the only
example.
Introduction to OFDM, II edition 8
10/30/98/TUDTVS
Mathematical description of OFDM
After the qualitative description of the system, it is valuable to discuss the mathematical definition of the
modulation system. This allows us to see how the signal is generated and how receiver must operate, and
it gives us a tool to understand the effects of imperfections in the transmission channel. As noted above,
OFDM transmits a large number of narrowband carriers, closely spaced in the frequency domain. In
order to avoid a large number of modulators and filters at the transmitter and complementary filters and
demodulators at the receiver, it is desirable to be able to use modern digital signal processing techniques,
such as fast Fourier transform (FFT).
Fig. 4 Examples of OFDM spectrum (a) a single subchannel, (b) 5 carriers
At the central frequency of each subchannel, there is no crosstalk from other subchannels
Mathematically, each carrier can be described as a complex wave:
( ) ( )
( )
c c
j t t
s A e
t t
c
c
·
+
]
]
]
ω φ
(1)
The real signal is the real part of s
c
(t). Both A
c
(t) and φ
c
(t), the amplitude and phase of the carrier, can
vary on a symbol by symbol basis. The values of the parameters are constant over the symbol duration
period τ.
OFDM consists of many carriers. Thus the complex signals s
s
(t) (Fig. 4) is represented by:
( ) ( )
( )
s N
j t t
n
N
s A e
t
N
t
n
n
·
+
]
]
]
·
−
∑
1
0
1
ω φ
(2)
where
ω ω ω
n
n · +
0
∆
This is of course a continuous signal. If we consider the waveforms of each component of the signal over
one symbol period, then the variables A
c
(t) and φ
c
(t) take on fixed values, which depend on the frequency
of that particular carrier, and so can be rewritten:
φ
n
(t) ⇒ φ
n
A
n
(t) ⇒ A
n
If the signal is sampled using a sampling frequency of 1/T, then the resulting signal is represented by:
OFDM spectrum
A spectrum of an OFDM subchannel (during a single bit)
Introduction to OFDM, II edition 9
10/30/98/TUDTVS
( )
( ) [ ]
s n
j n kT
n
N
s A e
kT
N
n
·
+ +
·
−
∑
1
0
0
1
ω ω φ ∆
(3)
At this point, we have restricted the time over which we analyse the signal to N samples. It is convenient
to sample over the period of one data symbol. Thus we have a relationship:
τ=NT
If we now simplify eqn. 3, without a loss of generality by letting ω
0
=0, then the signal becomes:
( )
( )
s n
j j n kT
n
N
s A e e
kT
N
n
·
·
−
∑
1
0
1
φ ω ∆
(4)
Now Eq. 4 can be compared with the general form of the inverse Fourier transform:
( ) g kT
N
G
n
NT
j nk N
n
N
e
·

.
`
,
·
−
∑
1
2
0
1
π /
(5)
In eq. 4, the function A
n
j
n
e
φ
is no more than a definition of the signal in the sampled frequency
domain, and s(kT) is the time domain representation. Eqns. 4 and 5 are equivalent if:
∆
∆
f
NT
· · ·
ω
π τ 2
1 1
(6)
This is the same condition that was required for orthogonality (see Importance of orthogonality). Thus,
one consequence of maintaining orthogonality is that the OFDM signal can be defined by using Fourier
transform procedures.
The Fourier transform
The Fourier transform allows us to relate events in time domain to events in frequency domain. There
are several version of the Fourier transform, and the choice of which one to use depends on the
particular circumstances of the work.
The conventional transform relates to continuous signals which are not limited to in either
time or frequency domains. However, signal processing is made easier if the signals are sampled.
Sampling of signals with an infinite spectrum leads to aliasing, and the processing of signals which are
not time limited can lead to problems with storage space.
To avoid this, the majority of signal processing uses a version of the discrete Fourier
transform (DFT) [6,7]. The DFT is a variant on the normal transform in which the signals are sampled
in both time and the frequency domains. By definition, the time waveform must repeat continually, and
this leads to a frequency spectrum that repeats continually in the frequency domain. [5]
The fast Fourier transform (FFT) is merely a rapid mathematical method for computer
applications of DFT. It is the availability of this technique, and the technology that allows it to be
implemented on integrated circuits at a reasonable price, that has permitted OFDM to be developed as
far as it has. The process of transforming from the time domain representation to the frequency domain
representation uses the Fourier transform itself, whereas the reverse process uses the inverse Fourier
transform.
The use of the FFT in OFDM
The main reason that the OFDM technique has taken a long time to become a prominence has been
practical. It has been difficult to generate such a signal, and even harder to receive and demodulate the
Introduction to OFDM, II edition 10
10/30/98/TUDTVS
signal. The hardware solution, which makes use of multiple modulators and demodulators, was
somewhat impractical for use in the civil systems.
The ability to define the signal in the frequency domain, in software on VLSI processors, and
to generate the signal using the inverse Fourier transform is the key to its current popularity. The use of
the reverse process in the receiver is essential if cheap and reliable receivers are to be readily available.
Although the original proposals were made a long time ago [5], it has taken some time for technology to
catch up.
At the transmitter, the signal is defined in the frequency domain. It is a sampled digital
signal, and it is defined such that the discrete Fourier spectrum exists only at discrete frequencies. Each
OFDM carrier corresponds to one element of this discrete Fourier spectrum. The amplitudes and phases
of the carriers depend on the data to be transmitted. The data transitions are synchronised at the carriers,
and can be processed together, symbol by symbol (Fig. 5).
Fig. 5 Block diagram of an OFDM system using FFT, pilot PN sequence and a guard bit insertion [9]
The definition of the (Npoint) discrete Fourier transform (DFT) is:
[ ] [ ]
( )
X k x n
p p
j N kn
n
N
e
·
−
·
−
∑
2
0
1
π
(DFT) (7)
and the (Npoint) inverse discrete Fourier transform (IDFT):
[ ] [ ]
( )
p p
j N kn
k
N
x e
n
N
X k ·
·
−
∑
1
2
0
1
π /
(IDFT) (8)
A natural consequence of this method is that it allows us to generate carriers that are orthogonal. The
members of an orthogonal set are linearly independent.
Consider a data sequence (d
0
, d
1
, d
2
, …, d
N1
), where each d
n
is a complex number d
n
=a
n
+jb
n
.
(a
n
, b
n
=t1 for QPSK, a
n
, b
n
=t1, t3 for 16QAM, … )
D d e d e
m n
j nm N
n
N
n
j f t
n
N
n m
· ·
−
·
−
−
··
−
∑ ∑
( / ) 2
0
1
2
0
1
π π
k=0,1,2, …, N1 (9)
where f
n
=n/(N∆T), t
k
=k∆t and ∆t is an arbitrarily chosen symbol duration of the serial data sequence d
n
.
The real part of the vector D has components
SerialtoParallel
converter
Signal
Mapper
Pilot
PN sequence
insertion
Guard bit
insertion
circuit
IFFT
D/A
LPF
Up
converter
CHANNEL
Parallelto Serial
converter
Signal
Mapper
Equalisation
& Pilot
extraction
Removal
guard
interval
FFT
LPF
A/D
Down
converter
Serial data
Input
Output
Serial data
x bits
d
0
d
1
d
n  1
Introduction to OFDM, II edition 11
10/30/98/TUDTVS
{ ¦ ( ) ( )
[ ]
Y a f t b f t
m m n n m n n m
n
N
D
· · +
·
−
∑
Re ( cos sin 2 2
0
1
π π , k=0,1,…,N1
(10)
If these components are applied to a lowpass filter at time intervals ∆t, a signal is obtained that closely
approximates the frequency division multiplexed signal
( ) ( )
[ ]
y t a f t b f t
n n m n n m
n
N
( ) ( cos sin · +
·
−
∑
2 2
0
1
π π , 0 ≤ t ≤ N∆t (11)
Fig. 5 illustrates the process of a typical FFTbased OFDM system. The incoming serial data is first
converted form serial to parallel and grouped into x bits each to form a complex number. The number x
determines the signal constellation of the corresponding subcarrier, such as 16 QAM or 32QAM. The
complex numbers are modulated in the baseband by the inverse FFT (IFFT) and converted back to serial
data for transmission. A guard interval is inserted between symbols to avoid intersymbol interference
(ISI) caused by multipath distortion. The discrete symbols are converted to analog and lowpass filtered
for RF upconversion. The receiver performs the inverse process of the transmitter. Onetap equalizer is
used to correct channel distortion. The tapcoefficients of the filter are calculated based on the channel
information.
Fig. 6 Example of the power spectral density of the OFDM signal with a guard interval ∆ = TS/4 (number of carriers
N=32) [12]
Fig 4a shows the spectrum of an OFDM subchannel and Fig. 4b and Fig. 6 present composite
OFDM spectrum. By carefully selecting the carrier spacing, the OFDM signal spectrum can be made flat
and the orthogonality among the subchannels can be guaranteed.
Guard interval and its implementation
The orthogonality of subchannels in OFDM can be maintained and individual subchannels can be
completely separated by the FFT at the receiver when there are no intersymbol interference (ISI) and
intercarrier interference (ICI) introduced by transmission channel distortion. In practice these conditions
can not be obtained. Since the spectra of an OFDM signal is not strictly band limited (sinc(f) function),
linear distortion such as multipath cause each subchannel to spread energy into the adjacent channels
1/T
s
dB
0
20
10
Introduction to OFDM, II edition 12
10/30/98/TUDTVS
and consequently cause ISI. A simple solution is to increase symbol duration or the number of carriers so
that distortion becomes insignificant. However, this method may be difficult to implement in terms of
carrier stability, Doppler shift, FFT size and latency.
Fig. 7 The effect on the timing tolerance of adding a guard interval. With a guard interval included in the signal, the
tolerance on timing the samples is considerably more relaxed.
Fig. 8 Example of the guard interval. Each symbol is made up of two parts. The whole signal is contained in the
active symbol (shown highlighted for the symbol M) The last part of which (shown in bold) is also repeated at the
start of the symbol and is called the guard interval
One way to prevent ISI is to create a cyclically extended guard interval (Fig. 7, 8), where each
OFDM symbol is preceded by a periodic extension of the signal itself. The total symbol duration is
T
total
=T
g
+T, where T
g
is the guard interval and T is the useful symbol duration. When the guard interval
is longer than the channel impulse response (Fig. 3), or the multipath delay, the ISI can be eliminated.
However, the ICI, or inband fading, still exists. The ratio of the guard interval to useful symbol duration
is applicationdependent. Since the insertion of guard interval will reduce data throughput, T
g
is usually
less than T/4.
The reasons to use a cyclic prefix for the guard interval are:
• to maintain the receiver carrier synchronization ; some signals instead of a long silence must always
be transmitted;
• cyclic convolution can still be applied between the OFDM signal and the channel response to model
the transmission system.
symbol M1
symbol M
symbol M1
a) correctly lined samples
b) Incorrectly timed, but decodable samples
c) Incorrectly timed samples giving intersymbol interference
time
time
symbol M1 symbol M symbol M+1
guard
interval active period
a
m
p
l
i
t
u
d
e
Introduction to OFDM, II edition 13
10/30/98/TUDTVS
Application of OFDM: Coded OFDM for Digital Audio Broadcast (DAB) [10]
In DAB, staggered quadrature phase shift keying (Π/4 QPSK) is used, with differential coding.
Main channel code is convolutional error correcting code, with a Viterbi decoder., some of the higher
priority data is precoded with a block code for additional security. These coding options have been
tailored to audio and signaling data, which is being broadcast.
The data is interleaved, both in frequency and time. The error correction process works best if
the errors in the incoming data are random.
The addition of the guard interval allows the system to cope with echoes of moderate duration,
and with small inaccuracies in the receiver.
IV Choice of the key elements
Useful symbol duration
The useful symbol duration T affects the carrier spacing and coding latency. To maintain the data
throughput, a longer useful symbol duration results in increase of the number of carriers and the size of
FFT (assuming the constellation is fixed). In practice, carrier offset and phase stability may affect how
close two carriers can be placed. If the application is for the mobile reception, the carrier spacing must
be large enough to make the Doppler shift negligible. Generally, the useful symbol duration should be
chosen so that the channel is stable for the duration of a symbol.
Number of carriers
The number of subcarriers can be determined based on the channel bandwidth, data throughput and
useful symbol duration.
N
T
·
1
The carriers are spaced by the reciprocal of the useful symbol duration. The number of carriers
corresponds to the number of complex points being processed in FFT. For HDTV applications, the
number of subcarriers are in the range of several thousands, so as to accommodate the data rate and
guard interval requirement.
Modulation scheme
The modulation scheme in an OFDM system can be selected based on the requirement of power or
spectrum efficiency. The type of modulation can be specified by the complex number d
n
=a
n
+jb
n
, defined
in section The use of FFT in OFDM. The symbols a
n
and b
n
can be selected to (t1, t3) for 16QAM and
t1 for QPSK. In general, the selection of the modulation scheme applying to each subchannel depends
solely on the compromise between the data rate requirement and transmission robustness. Another
advantage of OFDM is that different modulation schemes can be used on different subchannels for
layered services.
Introduction to OFDM, II edition 14
10/30/98/TUDTVS
Coded OFDM
By using frequency and time diversity OFDM provides a means to transmit data in a frequency selective
channel. However, it does not suppress fading itself. Depending on their position in the frequency
domain, individual subchannels could be affected by fading. This requires the use of channel coding to
further protect transmitted data. Among those channel techniques, trellis coded modulation (TCM),
combined with frequency and time interleaving is considered the most effective means for a selective
fading channel.
TCM combines coding and modulation to achieve a high coding gain without affecting the
bandwidth of the signal. In a TCM encoder, each symbol of n bits is mapped into constellation of n+1
bits, using a setpartitioning rule. This process increases the constellation size and effectively adds
additional redundancy to the signal. A TCM code can be decoded with a soft decision Viterbi decoding
algorithm, which exploits the soft decision nature of the received signal. The coding gain for a two
dimensional TCM code over a Gaussian channel is about 3 dB for a bit error rate (BER) of 10
5
.
It should be mentioned that one of the advantages of OFDM is that it can convert a wideband
frequency selective fading channel into a series of narrowband and frequency nonselective fading
subchannels by using parallel and multicarrier transmission. Coding OFDM subcarriers sequentially by
using specially designed TCM codes for frequency nonselective fading channel is the major reason for
using the COFDM for terrestrial broadcasting. However, the search of the best TCM code is still
ongoing.
Although trellis codes produce improvements in the signaltonoise ratio (S/N), they do not
perform well with impulsive or burst noise. In general, transmission errors have a strong time/frequency
correlation. Interleaving plays an essential role in channel coding by providing diversity in the time
domain. Interleaving breaks the correlation and enables the decoder to eliminate or reduce local fading
throughout the band and over the whole depth of the time interleaving. Interleaving depth should be
enough to break long straight errors.
Flexibility and scalability
Based on the information theory, the channel capacity is a function of the signaltonoise ratio and
channel bandwidth. The concept of graceful degradation has been implemented in the analog TV
systems. It is believed that the joint source/channel coding is the best way to achieve flexibility and
scalability. COFDM has been considered very flexible for the layered and scaleable transmission.
Different groups of COFDM subchannels can be assigned to different orders of modulation, power
levels, and channel coding schemes.
V COFDM performance expectations
The following expectations are based on the research for the digital terrestrial television broadcasting
[9].
It should be noted that for the additive white Gaussian channel, COFDM and single carrier modulation
have comparable performance. However, the broadcasting channel for HDTV consists of various other
impairments: random noise, impulse noise, multipath distortion, fading and interference. Also, at the
high end of the UHF band the wavelengths are short (around 0.5 m). Thus, characteristics of these holes
and peaks in this band are better modeled by a statistical distribution known as a Rayleigh distribution.
Multipath/fading
It is believed that with properly designed guard interval, interleaving and channel coding,
COFDM is capable of handling very strong echoes. The BER improvement, which resulted from the
multiple echoes, was indicated by the computer simulations and laboratory demonstrations. With the
assumption of withstanding strong multipath propagation, COFDM might allow the use of omni
directional antenna in urban areas and mobile reception where C/N is sufficiently high.
Introduction to OFDM, II edition 15
10/30/98/TUDTVS
In addition to channel fading, timevariant signals caused by transmitter tower swaying,
airplane fluttering and even tree swaying generate dynamic ghosts and consequently produce errors in
digital transmission. With its parallel transmission structure as well as the use of trellis coding, COFDM
systems might present advantages in fading and timeinvariant environments.
Phase noise and jitter
A COFDM system is much more affected by carrier frequency errors. A small frequency offset at the
receiver compromises the orthogonality between the subchannels, giving a degradation in a system
performance that increases rapidly with frequency offset and with the number of subcarriers. Phase noise
and jitter can be influenced by the transmitter upconverter and tuner. A possible solution is the use of
pilots which can be used to track phase noise in the demodulation. However, this is done under the
penalty of reducing the payload data throughput.
Carrier recovery / Equalization
In the severe channel conditions, such as low C/N, strong interference and fading, COFDM signal must
be designed to provide robust carrier recovery. Carrier frequency detection could be one of the biggest
limitations in COFDM design. The use of pilots and reference symbols are efficient methods for carrier
recovery and subchannel equalization. A pilot can be a sine wave or a known binary sequence. A
reference symbol can be a chirp or a pseudorandom sequence.
The twodimensional (time/frequency) signal feature in COFDM makes pilot and reference
symbol insertion very flexible. Pilots can be inserted in frequencydomain (fixed carriers) and reference
symbols in the time domain (fixed data packets). Because they are transmitted at the predetermined
positions in the signal frame structure, it can be captured in the receiver whenever the frame
synchronisation is recovered. In a frequencyselective channel, high correlation between the complex
fading envelopes of the pilots and data must be ensured. The appropriate complex correction can be
obtained by interpolating among the pilots. Cimini [8] reported that interpolation in real and imaginary
parts of the complex fading envelopes outperformed the interpolation in amplitude and phase.
For a single carrier system, equalization is done in the time domain. For a QAM system with
a Ntap equalizer, there are about N complex multiplication, or 4N real multiplicationaccumulation per
input symbol. For a VSB system, its symbol rate needs to be twice that of a QAM system for the same
data throughput. Assuming the same echo range as for the QAM system, a 2Ntap equalizer is required,
which is a computational complexity of about 2N multiplicationaccumulations per input symbol.
For a COFDM system, assuming multipath delay is less than the guard interval, a frequency
domain onetap equalizer could be used for each subchannel to correct the amplitude and phase
distortions. This corresponds to 4 real multiplicationaccumulations per data symbol. Additionally, the
FFT operations requires a computational complexity that is proportional to C*
2
log (M), where M is the
size of the FFT and C is the constant between 1.5 to 4 depending on the FFT implementation.
The number of pilots and reference symbols used in a COFDM system determines the trade
off between payload capacity and transmission robustness.
Simulation results indicated that an OFDM system with equalization performed better than
that of a single carrier system with a linear equalizer.
Impulse interference
COFDM is more immune to impulse noise than single carrier system, because a COFDM signal is
integrated over a long symbol period and the impact of impulse noise is much less than that for single
carrier systems. As a matter of fact, the immunity of impulse noise was one of the original motivations
for MCM. In a report submitted to the CCITT [11], which presented comparative performance results
for asymmetrical duplex V.32 (extended) and multicarrier modems, was shown that the threshold level
for the impulse noise, at which errors occur, can be as much as 11 dB higher for MCM than for a single
carrier system. Meanwhile, studies indicated that the best approach of impulse noise reduction for
OFDM involves a combination of soft and hard error protection.
Introduction to OFDM, II edition 16
10/30/98/TUDTVS
Peaktoaverage ratio
The peaktoaverage ratio for a single carrier system depends on the signal constellation and the rolloff
factor α of the pulse shaping filter (Gibbs’ phenomenon). For the Grand Alliance 8VSB system (single
carrier rival for the HDTV broadcast), α=11.5 %. The corresponding peaktoaverage power ratio is
about 7 dB for 99.99 % of the time.
Theoretically, the difference of the peaktoaverage power ratio between a multicarrier system
and a single carrier system is a function of the number of carriers as:
( ) ∆ dB N · 10
10
log
where N is the number of carriers. When N=1000, the difference could be 30 dB. However, this
theoretical value can rarely occur. Since the input data is well scrambled, the chances of reaching its
maximum value are very low, especially when the signal constellation size is large.
Since COFDM signal can be treated as a series of independent and identically distributed
carriers, the central limiting theorem implies that the COFDM signal distribution should tend to be
Gaussian when the number of carriers, N, is large. Generally, when N>20, which is the case for most of
the OFDM systems, the distribution is very close to Gaussian. Its probability of above three times (9.6
dB peaktoaverage ratio) of its variance, or average power, is about 0.1 %. For four times of variance, or
12 dB peaktoaverage power ratio, it is less than 0.01 %.
It should be pointed out that, for each COFDM subchannel, there is usually no pulse shaping
implemented. The peaktoaverage power ratio for each subchannel depends only on the signal
constellation.
In common practice, signals could be clipped because of limited quantization levels, rounding
and truncation during the FFT computation as well as other distribution parameters after D/A
conversion. It is safe to say that the Gaussian model can be used as the upper bound for the COFDM
signals.
Nonlinear distortion
Since a broadcast transmitter is a nonlinear device, clipping will always happen for COFDM signal.
However, clipping of a COFDM signal is similar to the impulse interference on which COFDM systems
have strong immunity. Tests show that when clipping occurs at 0.1% of the time, the BER degradation
is only 0.10.2 dB. Even at 1% of clipping, the degradation is 0.50.6 dB. However, the BER
performance of COFDM system under nonlinear distortion might not be the decisive factor. When
clipping occurs, energy would spill into the adjacent channels. More studies are required in this area. It
has been reported that, for an OFDM system, a 9 dB output backoff causes negligible BER degradation
and adjacent channel interference. Another study indicated that, for modern solidstate transmitters, a
prudent backoff level would be around 6 dB.
VI Conclusions
OFDM/COFDM has long been studied and implemented to combat transmission channel impairments.
Its applications have been extended from high frequency radio communications to telephone networks,
digital audio broadcasting and terrestrial broadcasting of digital television. The advantages of COFDM,
especially in the multipath propagation, interference and fading environment, make the technology a
promising alternative in digital communications including mobile multimedia.
The advantages of OFDM are:
¤ Efficient use of the available bandwidth since the subchannels are overlapping
¤ Spreading out the frequency fading over many symbols. This effectively randomizes the burst
errors caused by the Rayleigh fading, so that instead of several adjacent symbols (in time on a
Introduction to OFDM, II edition 17
10/30/98/TUDTVS
singlecarrier) being completely destroyed, (many) symbols in parallel are only slightly
distorted.
¤ The symbol period is increased and thus the sensitivity of the system to delay spread is reduced.
The disadvantages of the OFDM modulation are:
Z OFDM signal is contaminated by nonlinear distortion of transmitter power amplifier, because
it is a combined amplitudefrequency modulation (it is necessary to maintain linearity)
Z OFDM is very sensitive to carrier frequency offset caused by the jitter of carrier wave and
Doppler effect caused by moving of the mobile terminal.
Z At the receiver, it is very difficult to decide the starting time of the FFT symbol
Communications research and current development of COFDM around the world will certainly provide
us with valuable findings in theory and implementation. Further studies should be conducted on the
synchronization of OFDM signal, power demand, countermeasures against frequency offset, fading and
multiple access.
VII References
[1] R. Prasad, “An overview of millimetre waves for future personal wireless communication systems”,
Proc. IEEE First symposium. on communications and vehicular technology in the Benelux, K3, Delft,
Netherlands, Oct. 2728. 1993.
[2] Ministerie van Verkeer and Waterstaat, Hoofddirectie Telecommunicatie en Post, Frequency
allocations in the Netherlands, 2
nd
edition, Groningen, 1993.
[3] R.W. Chang, ”Synthesis of BandLimited Orthogonal Signals for Multichannel Data Transmission”,
Bell Syst. Tech. J., vol.45, pp. 17751796, Dec. 1966.
[4]B.R. Salzberg, “Performance of an efficient parallel data transmission system”, IEEE Trans.
Commun. Technol., vol. COM15, pp. 805813, Dec. 1967.
[5]S.B. Weinstein and P.M. Ebert, “Data transmission by frequencydivision multiplexing using the
discrete Fourier transform”, IEEE Trans. Commun. Technol., vol. COM19, pp. 628634, Oct. 1971.
[6]A.W.M. van den Enden and N.A.M. Verhoeckx, Discretetime signal processing: an introduction.
London: Prentice Hall Int., 1989., ISBN 0132167638
[7]A.V. Oppenheim and R.W. Schaffer, Discrete time signal processing, PrenticeHall International,
1989., ISBN 0132167719
[8]L.J. Cimini, Jr., “Analysis and simulation of a digital mobile channel using orthogonal frequency
division multiplexing”, IEEE Trans. Commun.,vol. COM33, pp. 665675, July 1985.
[9] W.Y. Zou and Y. Wu, “COFDM: An overview”, IEEE Trans. Broadc., vol. 41, no. 1, pp. 18,
March 1995
[10] P. Shelswell, “The COFDM modulation system: the heart of digital audio broadcasting”,
Electronics & communication engineering journal, pp. 127136, June 1995
[11] Telebit corporation, “Comparative performance results for asymmetrical duplex, V.32 (extended),
and multicarrier modems”, CCITT SG XVII, contribution D56, Sept. 1989.
Introduction to OFDM, II edition 18
10/30/98/TUDTVS
[12] M.Alard and R. Lassalle, “Principles of modulation and channel coding for digital broadcasting for
mobile receivers”, EBU Review, no. 224, pp. 325, Aug. 1987.
Some parts of the signal may suffer from constructive interference and be enhanced in level. sometimes to the point of extinction. B c ≈ 1 D If we look at the temporal response of the channel. which will affect all systems using phase as a means of signalling . Combined. This range of delay can be measured and then processed to get statistical parameters. these signals can produce fading. but this technology is still being developed. One can often find following formula for correlation bandwidth Bc. A characteristic of frequency selective fading is that some frequencies are enhanced. These echoes cause ISI. Frequency selective attenuation is clearly present. even if it is subject to significant levels of multipath propagation. greater than correlation bandwidth. which are typical of different outdoor/indoor areas. Whichever is chosen. because all frequencies in the band are usually distorted in the same way. a significant chance that the signal will be subject to severe attenuation on some occasions. which arrive at the receiver at different times. Some of these reflections can be avoided by using a directional antenna (current trend is simple antennas). A narrowband signal will vary in quality as the peaks and the troughs of the frequency response move around in frequency domain. For a narrowband signal. there are many reflected signals. . we see a number of echoes present. Typical frequency response of a channel suffering from multipath propagation. There is no standard definition of the correlation bandwidth. frequency components that are close together will suffer variations in signal strength that are strongly correlated. or objects such as people. the inverse of this leads to a good approximation for the correlation bandwidth. vehicles or buildings. The correlation (or coherence) is used as a measure of this phenomenon. Delayed signals are the result of reflections from terrain features such as trees. Now consider a signal which is of greater bandwidth. A solution could be usage of antenna arrays. because different parts of the band suffer different levels of distortion. A signal which occupies a wider bandwidth. There will also be a noticeable variation in phase response. There is. Different studies use the total range of delay. however. In general. but will suffer less variation in total received power. whereas others are attenuated. or the average delay. but it is impossible to use them for a mobile user. will be subject to more distortion. where D is the RMS value of delay spread (not average). There are many different types of echo environment. distortion is usually minimised if the bandwidth is less than the correlation bandwidth of the channel. amplitude time or position frequency Fig 1. If there is a mobile reception. then the relative lengths and attenuations of the various reception paths will change with time. The frequency response will vary with both time and position.Introduction to OFDM. II edition 10/30/98/TUDTVS 2 As a result of the multipath propagation. whereas others may suffer from destructive interference and be attenuated. This comes from the fact that variation averages out if the bandwidth is much larger than the correlation bandwidth. hills or mountains.
research and development of the OFDM have received considerable attention and have made a great deal of progress in Europe. if a user needs a speed of 20Mb/s on air and the spreading factor is 128 (today’ typical). There is a lot of fundamental investigation needed in this area. Improvements can be made with frequency equalization and directional antennas. Usually. The low mmwave band from 2060 GHz. because of the oxygen absorption at the rate of 14 dB/km [1]. This is achieved by transmitting many narrowband overlapping digital signals in parallel. Increasing the number of parallel transmission channels reduces the data rate that each individual carrier must convey. . effects of antenna diversity. • Wavelength is as little as 5 mm. this is a disadvantage for many applications.e. i. This enables crating of small antenna's and other parts of the radiopart of the system. this results in 2. so to every user can be allocated a large bandwidth. technology development. the delay time of reflected waves is suppressed to within 1 symbol time. If used indoor. inside one wide band. It has become a hot topic as a research area for broadband communications. however. Thus. Use of the area around 60 GHz is encouraged for the following reasons. • The coherence bandwidth of a 60 GHz link is several MHz. but they set forth impossible demands on the existing technology – for instance. which is nearly unused and allows for large bandwidth applications. e. Singlecarrier techniques are vulnerable to fading and multipath propagation. which could then be used for evaluating modulation technique candidates for the mobile multimedia communication. which results in severe signal dispersion and limits the maximum usable symbol rate. combines the advantages of infrared (enough free bandwidth and UHF (good coverage). Measurements and models can hardly be found in the literature. These properties have to be built in the simulation model. this high attenuation over the propagation path creates a natural barrier for cochannel interference in the mobile cellular system. It is worth mentioning that no definitive evidence of any hazards has been shown to date to the general public arising from the prolonged exposure in fields of less than 10 mW/cm2 in the mmwaves. the mmwave radio channel shows adverse frequency selective multipath characteristics due to the highly reflective indoor environment. which is especially relevant because of the advent of the Broadband Digital Network (BISDN). and that lengthens the symbol period.Introduction to OFDM. This size of communication equipment makes it easy for wearing. The 60 GHz band The scarcity of spectrum and the new technical possibilities in recent years have drawn attention all over the world to the millimetre band. especially in the case of very high bitrates. II edition 10/30/98/TUDTVS 3 Spread spectrum techniques are robust against fading and interference. etc. [1] • There is enough unused space for the multimedia needs (5 GHz) [2]. Besides that. bitrates of an order of 155Mb/s are possible. thus a potential exists to support broadband service access. which can also be used to improve multicarrier techniques. This frequency region is not in use by any other medium. Recently. one needs knowing the channel properties at 60 GHz. OFDM is a wideband modulation scheme that is specifically able to cope with the problems of the multipath reception. A major drawback of this frequency region is the fact that the technology for the transceivers will be expensive in the early stages. they have difficulty with the nearfar effect and have a large powerconsumption.g. propagation modeling for 60 GHz. For the final choice between OFDM and singlecarrier modulation. As a result. • Systems operating particularly in the 60 GHz frequency band can have a small reuse distance. frequency reuse is easy.56 GB/s which have to processed in s realtime and have impracticably large bandwidth.
Recent advances in VLSI technology enable making of highspeed chips that can perform large size FFT at affordable price. In OFDM. II edition 10/30/98/TUDTVS 4 III OFDM OFDM history The concept of using parallel data transmission by means of frequency division multiplexing (FDM) was published in mid 60s [3. Various fast modems were developed for telephone networks. multichannel modulation and multicarrier modulation (MCM) are widely used and sometimes they are interchangeable with OFDM. 2 Comparison of the bandwidth utilization for FDM and OFDM For a large number of subchannels. digital mobile communications and highdensity recording. the arrays of sinusoidal generators and coherent demodulators required in a parallel system become unreasonably expensive and complex. 1. each carrier is orthogonal to all other carriers. highbitrate digital subscriber lines (HDSL. frequency Conventional Frequency Division Multiplex (FDM) multicarrier modulation technique Saving of the bandwidth frequency Orthogonal Frequency Division Multiplex (OFDM) multicarrier modulation technique Fig.4]. OFDM has been exploited for wideband data communications over mobile radio FM channels. 1. a completely digital implementation could be built around specialpurpose hardware performing the fast Fourier transform (FFT). One of the systems used a pilot tone for stabilizing carrier and clock frequency control and trellis coding was implemented. OFDM is an optimal version of multicarrier transmission schemes. OFDM has been studied for highspeed modems. asymmetric digital subscriber lines (ADSL.Introduction to OFDM. In addition to eliminating the banks of subcarrier oscillators and coherent demodulators required by FDM. 100 Mb/s). digital audio broadcasting (DAB) and HDTV terrestrial broadcasting.536 Mb/s).S. . The receiver needs precise phasing of the demodulating carriers and sampling times in order to keep crosstalk between subchannels acceptable. Weinstein and Ebert [5] applied the discrete Fourier transform (DFT) to parallel data transmission system as part of the modulation and demodulation process. In the telecommunications field. A U. However. In the 1980s. Some early development can be traced back in the 50s. 1970. and multipath distortion as well as to fully use the available bandwidth. very highspeed digital subscriber lines (VHDSL. The initial applications were in the military communications.6 Mb/s). patent was filled and issued in January. In 1990s. this condition is not always maintained in MCM. The idea was to use parallel data streams and FDM with overlapping subchannels to avoid the use of high speed equalization and to combat impulsive noise. the terms of discrete multitone (DMT).
highquality digital mobile portable reception and transmission. A receiver has to cope with a signal that is often weaker than desirable and that contains many echoes. This means that the intersymbol interference affects a smaller percentage of each symbol as the number of carriers and hence the symbol period increases (after [10[DM1]]). For example. For a given overall data rate.Introduction to OFDM. increasing the number of carriers reduces the data rate that each individual carrier must convey. Simple digital systems do not work well in the multipath environment. a demand emerges for highspeed. 3 The effect of adopting a multicarrier system. δ (t) time Step(t) time Step(t) channel responses δ (t) frequency time 1 carrier t f 1st carrier t 2nd carrier t 2 carrier parallel transmission f 1st carrier t 2nd carrier t 8 carrier parallel transmission t 8th carrier t 7th carrier f Fig. on the picture is shown a . II edition 10/30/98/TUDTVS 5 Qualitative description of OFDM In multimedia communication. and hence (for a given modulation system) lengthens the symbol period.
The FDM is achieved. the data is divided among large number of closely spaced carriers. A good understanding of the propagation statistics is needed to ensure that this condition is met. This method was borrowed from the conventional FDM technology. down to NlogN. Since each subchannel covers only a small fraction of the original bandwidth. many symbols are only slightly distorted. OFDM can effectively randomize burst errors caused by Rayleigh fading. In addition. and by this lowering of the bitrate per carrier (not the total bitrate). equalization is potentially simpler than in a serial data system. orthogonal signals can be separated at the receiver by correlation techniques. Instead of transmitting in serial way. The limitation of filter implementation forces the bandwidth of each subband to be equal to (1+α)fm. Use discrete Fourier transform (DFT) to modulate and demodulate parallel data. thus creating ISI. which comes from interleaving due to paralellisation. Use staggered QAM to increase the efficiency of band usage. where α is the rolloff factor and fm is the Nyquist bandwidth. many modulation schemes could be used to modulate the data at a low bit rate onto each carrier. since there is no common medium to be shared. the N subchannels are frequency multiplexed. and the implementation of differential encoding may make it possible to avoid equalization altogether [5]. Using this method. A simple equalization algorithm can minimize meansquare distortion on each subchannel. the frequency response over each individual subband is relatively flat. Each subchannel is modulated with a separate symbol and. OFDM can be simply defined as a form of multicarrier modulation where its carrier spacing is carefully selected so that each subcarrier is orthogonal to the other subcarriers. Multicarrier system is robust against these physical effects. Another disadvantage is that it is difficult to assemble a set of matched filter when the number of carriers is large. Only a small amount of the data is carried on each carrier. In a conventional serial data system.Introduction to OFDM. the responses of individual bits are overlapping. but by baseband processing. The separability or orthogonality is achieved by staggering the data (offset the data by half a symbol). then. The advantage is that the composite spectrum is flat. In order to occupy sufficient bandwidth to gain advantages of the OFDM system. In this way the individual spectra of the modulated carriers still use an excess bandwidth. hence. As is well known. with the frequency spectrum of each data symbol allowed to occupy the entire available bandwidth. even without forward error correction (FEC). the total signal frequency band is divided into N nonoverlapping frequency subchannels. in order to interleave data in time and frequency (depends how broad is one user signal). In OFDM. by using a guard interval the sensitivity of the system to delay spread can be reduced [8]. the symbols are transmitted sequentially. It is an important part of the OFDM system design that the bandwidth occupied is greater than the correlation bandwidth of the fading channel. Orthogonality can be achieved by carefully selecting carrier spacing. So. In principle. Then. the majority of the carriers should still be adequately received. II edition 10/30/98/TUDTVS 6 8 bit long part of a data sequence. the influence of intersymbol interference is significantly reduced. The entire bandwidth is filled from a single source of data. Use filters to completely separate the subbands. data is transferred in a parallel way. not by bandpass filtering. Because of dividing an entire channel bandwidth into many narrow subbands. Mathematical deduction of the orthogonal carrier frequencies is given in [3]. intersymbol interference among channels can be eliminated. although some of the carriers are degraded by multipath fading. This is not a multiple access technique. both transmitter and receiver can be implemented using efficient FFT techniques that reduce the number of operations from N2 in DFT. The requirement for filter design is less critical than that for the first scheme. In a classical parallel data system. 2. In a parallel data transmission system several symbols are transmitted at the same time. but the are overlapped at the 3 dB frequency. 3. There are three schemes that can be used to separate the subbands: 1. it would be good to group a number of users together to form a wideband system. instead of several adjacent symbols being completely destroyed. . This accounts for the “frequency division multiplex” part of the name. For a single carrier system. This allows the precise reconstruction of majority of them. what offers possibilities for alleviating many of the problems encountered with serial systems. The individual spectra are now sinc functions and are not band limited. such as letting the carrier spacing be equal to the reciprocal of the useful symbol period.
in the time domain. suppose we have a set of signals ψ . In a normal FDM system. Mathematically. The importance of orthogonality The “orthogonal” part of the OFDM name indicates that there is a precise mathematical relationship between the frequencies of the carriers in the system. It is possible. There are many types of error correcting codes. have a whole number of cycles in the symbol period (τ). then the integration process results in zero contribution from all these carriers. which could be used. is transmitted in a different part of the multiplex (and. it is possible to correct many or all of the bits that were incorrectly received.2. the many carriers are spaced apart in such way that the signals can be received using conventional filters and demodulators. In such receivers.). that the series sin(mx) for m=1. The signals are orthogonal if ∫Ψ a b p ( t )Ψ q* ( t )dt = K 0 for p = q . which is related to it by the errorcorrecting code. Thus the carriers are linearly independent (i. orthogonal) if the carrier spacing is a multiple of 1/τ. guard bands have to be introduced between the different carriers (Fig. the resulting signal then being integrated over a symbol period to recover the raw data. however. will not suffer from the same deep fade). If the other carriers all beat down to frequencies which. because other information. 2. translating each carrier down to DC. Much of transform theory makes the use of orthogonal series. By using an errorcorrecting code. This accounts for the “coded” part of the name COFDM. for p ≠ q where the * indicates the complex conjugate and interval [a.Introduction to OFDM. The receiver acts as a bank of demodulators.e.b] is a symbol period. A fairly simple mathematical proof exists. which adds extra bits at the transmitter. The information carried by one of the degraded carriers is corrected. In order to do this the carriers must be mathematically orthogonal. although they are by no means the only example. to arrange the carriers in an OFDM signal so that the sidebands of the individual carriers overlap and the signals can still be received without adjacent carrier interference. and the introduction of these guard bands in the frequency domain results in a lowering of the spectrum efficiency. II edition 10/30/98/TUDTVS 7 The importance of coding The distribution of the data over many carriers means that selective fading will cause some bits to be received in error while others are received correctly. it is hoped. where ψ p is the pth element in the set. .… is orthogonal over the interval π to π.
closely spaced in the frequency domain. Thus the complex signals ss(t) (Fig. This allows us to see how the signal is generated and how receiver must operate. The values of the parameters are constant over the symbol duration period τ. which depend on the frequency c of that particular carrier. 4 Examples of OFDM spectrum (a) a single subchannel.Introduction to OFDM. the amplitude and phase of the carrier. OFDM consists of many carriers. Both Ac(t) and φ(t). As noted above. it is desirable to be able to use modern digital signal processing techniques. it is valuable to discuss the mathematical definition of the modulation system. can c vary on a symbol by symbol basis. there is no crosstalk from other subchannels Mathematically. then the variables Ac(t) and φ(t) take on fixed values. each carrier can be described as a complex wave: sc (t ) = j ω t + φ( t ) Ac (t )e c c (1) The real signal is the real part of sc(t). and it gives us a tool to understand the effects of imperfections in the transmission channel. (b) 5 carriers At the central frequency of each subchannel. and so can be rewritten: φ(t) ⇒ φ n n An(t) ⇒ An If the signal is sampled using a sampling frequency of 1/T. In order to avoid a large number of modulators and filters at the transmitter and complementary filters and demodulators at the receiver. A spectrum of an OFDM subchannel (during a single bit) OFDM spectrum Fig. such as fast Fourier transform (FFT). then the resulting signal is represented by: . If we consider the waveforms of each component of the signal over one symbol period. 4) is represented by: ss ( t ) = where 1 N ∑A n= 0 N− 1 N ( t )e j ω t + φ t n n ( ) (2) ω n = ω 0 + n∆ω This is of course a continuous signal. OFDM transmits a large number of narrowband carriers. II edition 10/30/98/TUDTVS 8 Mathematical description of OFDM After the qualitative description of the system.
signal processing is made easier if the signals are sampled. The process of transforming from the time domain representation to the frequency domain representation uses the Fourier transform itself.7]. the function An is no more than a definition of the signal in the sampled frequency domain. There are several version of the Fourier transform. [5] The fast Fourier transform (FFT) is merely a rapid mathematical method for computer applications of DFT. one consequence of maintaining orthogonality is that the OFDM signal can be defined by using Fourier transform procedures. The DFT is a variant on the normal transform in which the signals are sampled in both time and the frequency domains. and the technology that allows it to be implemented on integrated circuits at a reasonable price. The use of the FFT in OFDM The main reason that the OFDM technique has taken a long time to become a prominence has been practical. and the choice of which one to use depends on the particular circumstances of the work. then the signal becomes: N− 1 n= 0 ss (kT ) = 1 N ∑ Ae e jφ n n j ( n∆ω ) kT (4) Now Eq. Thus we have a relationship: τ=NT If we now simplify eqn. and this leads to a frequency spectrum that repeats continually in the frequency domain. It has been difficult to generate such a signal. To avoid this. and s(kT) is the time domain representation. By definition.Introduction to OFDM. the time waveform must repeat continually. 4 and 5 are equivalent if: e j φ ∆f = ∆ω 1 1 = = 2π NT τ (6) This is the same condition that was required for orthogonality (see Importance of orthogonality). It is the availability of this technique. we have restricted the time over which we analyse the signal to N samples. and the processing of signals which are not time limited can lead to problems with storage space. The Fourier transform The Fourier transform allows us to relate events in time domain to events in frequency domain. whereas the reverse process uses the inverse Fourier transform. It is convenient to sample over the period of one data symbol. II edition 10/30/98/TUDTVS 9 ss (kT ) = 1 N ∑ A e [( n=0 n N− 1 j ω 0 + n∆ω )kT + φ n ] (3) At this point. Eqns. However. without a loss of generality by letting ω 0=0. Thus. 3. the majority of signal processing uses a version of the discrete Fourier transform (DFT) [6. that has permitted OFDM to be developed as far as it has. Sampling of signals with an infinite spectrum leads to aliasing. 4 can be compared with the general form of the inverse Fourier transform: 1 g( kT ) = N ∑ G NT e n= 0 N− 1 n j 2 πnk / N (5) n In eq. 4. The conventional transform relates to continuous signals which are not limited to in either time or frequency domains. and even harder to receive and demodulate the .
… . ±3 for 16QAM. The hardware solution. The amplitudes and phases of the carriers depend on the data to be transmitted. II edition 10/30/98/TUDTVS 10 signal. (an. it has taken some time for technology to catch up. Each OFDM carrier corresponds to one element of this discrete Fourier spectrum. dN1).Introduction to OFDM.1. d2. N1 (9) where fn=n/(N∆T). bn=±1. The ability to define the signal in the frequency domain. It is a sampled digital signal.2.1 CHANNEL Serial data Parallelto Serial converter Output Signal Mapper Equalisation & Pilot extraction Removal guard interval FFT LPF A/D Down converter Fig. 5). in software on VLSI processors. an. The real part of the vector D has components . Although the original proposals were made a long time ago [5]. and to generate the signal using the inverse Fourier transform is the key to its current popularity. x bits Serial data Input SerialtoParallel converter Signal Mapper d d 0 1 Pilot PN sequence insertion Guard bit insertion circuit IFFT D/A LPF Up converter d n . Consider a data sequence (d0. where each dn is a complex number dn=an+jbn. The use of the reverse process in the receiver is essential if cheap and reliable receivers are to be readily available. … . and can be processed together. 5 Block diagram of an OFDM system using FFT. pilot PN sequence and a guard bit insertion [9] The definition of the (Npoint) discrete Fourier transform (DFT) is: X p []= ∑ x p [] k ne n= 0 N− 1 − j ( 2π N )kn (DFT) (7) and the (Npoint) inverse discrete Fourier transform (IDFT): n x p []= 1 N ∑ N− 1 k =0 X p [] k e j ( 2π / N ) kn (IDFT) (8) A natural consequence of this method is that it allows us to generate carriers that are orthogonal. which makes use of multiple modulators and demodulators. was somewhat impractical for use in the civil systems. d1. At the transmitter. and it is defined such that the discrete Fourier spectrum exists only at discrete frequencies. bn=±1 for QPSK. The members of an orthogonal set are linearly independent. … ) Dm = ∑ dn e n=0 N− 1 − j ( 2 πnm/ N ) = n== 0 ∑de n N− 1 − j 2πf n tm k=0. The data transitions are synchronised at the carriers. the signal is defined in the frequency domain. tk=k∆t and ∆t is an arbitrarily chosen symbol duration of the serial data sequence dn. symbol by symbol (Fig.
6 Example of the power spectral density of the OFDM signal with a guard interval ∆ = TS/4 (number of carriers N=32) [12] Fig 4a shows the spectrum of an OFDM subchannel and Fig. n=0 N− 1 [ ] 0 ≤t ≤N∆t (11) Fig. Guard interval and its implementation The orthogonality of subchannels in OFDM can be maintained and individual subchannels can be completely separated by the FFT at the receiver when there are no intersymbol interference (ISI) and intercarrier interference (ICI) introduced by transmission channel distortion. 1/T s dB 0 10 20 Fig. The discrete symbols are converted to analog and lowpass filtered for RF upconversion. 5 illustrates the process of a typical FFTbased OFDM system. By carefully selecting the carrier spacing. linear distortion such as multipath cause each subchannel to spread energy into the adjacent channels . Onetap equalizer is used to correct channel distortion. In practice these conditions can not be obtained.1. A guard interval is inserted between symbols to avoid intersymbol interference (ISI) caused by multipath distortion. The number x determines the signal constellation of the corresponding subcarrier. ] k=0.N1 (10) If these components are applied to a lowpass filter at time intervals ∆t. The complex numbers are modulated in the baseband by the inverse FFT (IFFT) and converted back to serial data for transmission. a signal is obtained that closely approximates the frequency division multiplexed signal y(t ) = ∑ (an cos(2πf n tm ) + bn sin (2πf n t m ) .Introduction to OFDM.… . The incoming serial data is first converted form serial to parallel and grouped into x bits each to form a complex number. The tapcoefficients of the filter are calculated based on the channel information. II edition 10/30/98/TUDTVS 11 Ym = Re ( {D }= ∑ [a m n= 0 N− 1 n cos(2πf n t m ) + bn sin(2πf n t m ) . The receiver performs the inverse process of the transmitter. such as 16 QAM or 32QAM. Since the spectra of an OFDM signal is not strictly band limited (sinc(f) function). 4b and Fig. 6 present composite OFDM spectrum. the OFDM signal spectrum can be made flat and the orthogonality among the subchannels can be guaranteed.
The reasons to use a cyclic prefix for the guard interval are: • to maintain the receiver carrier synchronization . Tg is usually less than T/4. The ratio of the guard interval to useful symbol duration is applicationdependent. time symbol M symbol M1 symbol M1 a) correctly lined samples b) Incorrectly timed. . Doppler shift. Each symbol is made up of two parts. where Tg is the guard interval and T is the useful symbol duration. symbol M1 amplitude symbol M symbol M+1 guard interval active period time Fig. 8 Example of the guard interval. When the guard interval is longer than the channel impulse response (Fig. The total symbol duration is Ttotal=Tg+T. With a guard interval included in the signal. still exists. • cyclic convolution can still be applied between the OFDM signal and the channel response to model the transmission system. the tolerance on timing the samples is considerably more relaxed. However. FFT size and latency. or the multipath delay. However. this method may be difficult to implement in terms of carrier stability. Since the insertion of guard interval will reduce data throughput. 7. or inband fading.Introduction to OFDM. the ICI. but decodable samples c) Incorrectly timed samples giving intersymbol interference Fig. A simple solution is to increase symbol duration or the number of carriers so that distortion becomes insignificant. II edition 10/30/98/TUDTVS 12 and consequently cause ISI. 3). some signals instead of a long silence must always be transmitted. the ISI can be eliminated. 8). where each OFDM symbol is preceded by a periodic extension of the signal itself. 7 The effect on the timing tolerance of adding a guard interval. The whole signal is contained in the active symbol (shown highlighted for the symbol M) The last part of which (shown in bold) is also repeated at the start of the symbol and is called the guard interval One way to prevent ISI is to create a cyclically extended guard interval (Fig.
The addition of the guard interval allows the system to cope with echoes of moderate duration. the selection of the modulation scheme applying to each subchannel depends solely on the compromise between the data rate requirement and transmission robustness. Another advantage of OFDM is that different modulation schemes can be used on different subchannels for layered services. Number of carriers The number of subcarriers can be determined based on the channel bandwidth. In practice. In general. some of the higher priority data is precoded with a block code for additional security.Introduction to OFDM. Main channel code is convolutional error correcting code. If the application is for the mobile reception. These coding options have been tailored to audio and signaling data. the number of subcarriers are in the range of several thousands. The error correction process works best if the errors in the incoming data are random. carrier offset and phase stability may affect how close two carriers can be placed. a longer useful symbol duration results in increase of the number of carriers and the size of FFT (assuming the constellation is fixed). which is being broadcast. and with small inaccuracies in the receiver. The number of carriers corresponds to the number of complex points being processed in FFT. Generally. N= 1 T The carriers are spaced by the reciprocal of the useful symbol duration. staggered quadrature phase shift keying (Π/4 QPSK) is used. the carrier spacing must be large enough to make the Doppler shift negligible. defined in section The use of FFT in OFDM. with a Viterbi decoder.. . the useful symbol duration should be chosen so that the channel is stable for the duration of a symbol. The data is interleaved. II edition 10/30/98/TUDTVS 13 Application of OFDM: Coded OFDM for Digital Audio Broadcast (DAB) [10] In DAB. IV Choice of the key elements Useful symbol duration The useful symbol duration T affects the carrier spacing and coding latency. For HDTV applications. The type of modulation can be specified by the complex number dn=an+jbn . data throughput and useful symbol duration. To maintain the data throughput. with differential coding. Modulation scheme The modulation scheme in an OFDM system can be selected based on the requirement of power or spectrum efficiency. ±3) for 16QAM and ±1 for QPSK. both in frequency and time. The symbols an and bn can be selected to (±1. so as to accommodate the data rate and guard interval requirement.
power levels. was indicated by the computer simulations and laboratory demonstrations. It should be mentioned that one of the advantages of OFDM is that it can convert a wideband frequency selective fading channel into a series of narrowband and frequency nonselective fading subchannels by using parallel and multicarrier transmission. characteristics of these holes and peaks in this band are better modeled by a statistical distribution known as a Rayleigh distribution. Depending on their position in the frequency domain. the broadcasting channel for HDTV consists of various other impairments: random noise.5 m). In general. COFDM has been considered very flexible for the layered and scaleable transmission. A TCM code can be decoded with a soft decision Viterbi decoding algorithm. interleaving and channel coding. individual subchannels could be affected by fading. II edition 10/30/98/TUDTVS 14 Coded OFDM By using frequency and time diversity OFDM provides a means to transmit data in a frequency selective channel. It is believed that the joint source/channel coding is the best way to achieve flexibility and scalability. Flexibility and scalability Based on the information theory. multipath distortion. COFDM and single carrier modulation have comparable performance. Interleaving plays an essential role in channel coding by providing diversity in the time domain. However. they do not perform well with impulsive or burst noise. trellis coded modulation (TCM). Interleaving depth should be enough to break long straight errors. fading and interference. which resulted from the multiple echoes. . COFDM might allow the use of omnidirectional antenna in urban areas and mobile reception where C/N is sufficiently high. the search of the best TCM code is still ongoing. COFDM is capable of handling very strong echoes. Although trellis codes produce improvements in the signaltonoise ratio (S/N). Different groups of COFDM subchannels can be assigned to different orders of modulation.Introduction to OFDM. at the high end of the UHF band the wavelengths are short (around 0. With the assumption of withstanding strong multipath propagation. However. Interleaving breaks the correlation and enables the decoder to eliminate or reduce local fading throughout the band and over the whole depth of the time interleaving. This process increases the constellation size and effectively adds additional redundancy to the signal. The coding gain for a twodimensional TCM code over a Gaussian channel is about 3 dB for a bit error rate (BER) of 105. V COFDM performance expectations The following expectations are based on the research for the digital terrestrial television broadcasting [9]. using a setpartitioning rule. and channel coding schemes. the channel capacity is a function of the signaltonoise ratio and channel bandwidth. Among those channel techniques. impulse noise. The concept of graceful degradation has been implemented in the analog TV systems. However. it does not suppress fading itself. transmission errors have a strong time/frequency correlation. Multipath/fading It is believed that with properly designed guard interval. TCM combines coding and modulation to achieve a high coding gain without affecting the bandwidth of the signal. which exploits the soft decision nature of the received signal. Thus. The BER improvement. Coding OFDM subcarriers sequentially by using specially designed TCM codes for frequency nonselective fading channel is the major reason for using the COFDM for terrestrial broadcasting. This requires the use of channel coding to further protect transmitted data. In a TCM encoder. combined with frequency and time interleaving is considered the most effective means for a selective fading channel. Also. It should be noted that for the additive white Gaussian channel. each symbol of n bits is mapped into constellation of n+1 bits.
which presented comparative performance results for asymmetrical duplex V. Carrier frequency detection could be one of the biggest limitations in COFDM design. A possible solution is the use of pilots which can be used to track phase noise in the demodulation. strong interference and fading. However.Introduction to OFDM. Cimini [8] reported that interpolation in real and imaginary parts of the complex fading envelopes outperformed the interpolation in amplitude and phase. For a QAM system with a Ntap equalizer. For a COFDM system. A small frequency offset at the receiver compromises the orthogonality between the subchannels. Meanwhile. COFDM signal must be designed to provide robust carrier recovery. there are about N complex multiplication. In a report submitted to the CCITT [11]. The use of pilots and reference symbols are efficient methods for carrier recovery and subchannel equalization. it can be captured in the receiver whenever the frame synchronisation is recovered. For a VSB system. assuming multipath delay is less than the guard interval. at which errors occur. In a frequencyselective channel. Carrier recovery / Equalization In the severe channel conditions. because a COFDM signal is integrated over a long symbol period and the impact of impulse noise is much less than that for single carrier systems. Assuming the same echo range as for the QAM system. A pilot can be a sine wave or a known binary sequence. As a matter of fact. airplane fluttering and even tree swaying generate dynamic ghosts and consequently produce errors in digital transmission. Phase noise and jitter can be influenced by the transmitter upconverter and tuner. A reference symbol can be a chirp or a pseudorandom sequence. Impulse interference COFDM is more immune to impulse noise than single carrier system. such as low C/N. its symbol rate needs to be twice that of a QAM system for the same data throughput. where M is the size of the FFT and C is the constant between 1.5 to 4 depending on the FFT implementation.32 (extended) and multicarrier modems. this is done under the penalty of reducing the payload data throughput. can be as much as 11 dB higher for MCM than for a single carrier system. II edition 10/30/98/TUDTVS 15 In addition to channel fading. equalization is done in the time domain. the immunity of impulse noise was one of the original motivations for MCM. For a single carrier system. which is a computational complexity of about 2N multiplicationaccumulations per input symbol. studies indicated that the best approach of impulse noise reduction for OFDM involves a combination of soft and hard error protection. The twodimensional (time/frequency) signal feature in COFDM makes pilot and reference symbol insertion very flexible. or 4N real multiplicationaccumulation per input symbol. COFDM systems might present advantages in fading and timeinvariant environments. Phase noise and jitter A COFDM system is much more affected by carrier frequency errors. This corresponds to 4 real multiplicationaccumulations per data symbol. giving a degradation in a system performance that increases rapidly with frequency offset and with the number of subcarriers. a 2Ntap equalizer is required. was shown that the threshold level for the impulse noise. . Pilots can be inserted in frequencydomain (fixed carriers) and reference symbols in the time domain (fixed data packets). a frequency domain onetap equalizer could be used for each subchannel to correct the amplitude and phase distortions. With its parallel transmission structure as well as the use of trellis coding. The appropriate complex correction can be obtained by interpolating among the pilots. high correlation between the complex fading envelopes of the pilots and data must be ensured. the FFT operations requires a computational complexity that is proportional to C* log 2 (M). Additionally. The number of pilots and reference symbols used in a COFDM system determines the tradeoff between payload capacity and transmission robustness. Simulation results indicated that an OFDM system with equalization performed better than that of a single carrier system with a linear equalizer. Because they are transmitted at the predetermined positions in the signal frame structure. timevariant signals caused by transmitter tower swaying.
make the technology a promising alternative in digital communications including mobile multimedia. It is safe to say that the Gaussian model can be used as the upper bound for the COFDM signals. When N=1000. However. it is less than 0. clipping of a COFDM signal is similar to the impulse interference on which COFDM systems have strong immunity. is about 0. there is usually no pulse shaping implemented. This effectively randomizes the burst errors caused by the Rayleigh fading. so that instead of several adjacent symbols (in time on a . More studies are required in this area. N. Generally. the central limiting theorem implies that the COFDM signal distribution should tend to be Gaussian when the number of carriers. It should be pointed out that. this theoretical value can rarely occur.1% of the time. the degradation is 0. especially in the multipath propagation. when N>20. rounding and truncation during the FFT computation as well as other distribution parameters after D/A conversion.50. or 12 dB peaktoaverage power ratio. For the Grand Alliance 8VSB system (singlecarrier rival for the HDTV broadcast).2 dB.1 %. for an OFDM system. the difference could be 30 dB. the BER performance of COFDM system under nonlinear distortion might not be the decisive factor. the difference of the peaktoaverage power ratio between a multicarrier system and a single carrier system is a function of the number of carriers as: ∆( dB) = 10 log10 N where N is the number of carriers. Since COFDM signal can be treated as a series of independent and identically distributed carriers.99 % of the time. the chances of reaching its maximum value are very low. Since the input data is well scrambled. Tests show that when clipping occurs at 0. which is the case for most of the OFDM systems.6 dB. or average power. Its probability of above three times (9. Nonlinear distortion Since a broadcast transmitter is a nonlinear device. is large. energy would spill into the adjacent channels. In common practice.5 %. Even at 1% of clipping. It has been reported that. II edition 10/30/98/TUDTVS 16 Peaktoaverage ratio The peaktoaverage ratio for a single carrier system depends on the signal constellation and the rolloff factor α of the pulse shaping filter (Gibbs’ phenomenon).Introduction to OFDM. Theoretically. VI Conclusions OFDM/COFDM has long been studied and implemented to combat transmission channel impairments. Another study indicated that. α=11. However. a prudent backoff level would be around 6 dB.10. The corresponding peaktoaverage power ratio is about 7 dB for 99. However. The peaktoaverage power ratio for each subchannel depends only on the signal constellation. interference and fading environment. clipping will always happen for COFDM signal. digital audio broadcasting and terrestrial broadcasting of digital television. The advantages of OFDM are: ¤ Efficient use of the available bandwidth since the subchannels are overlapping ¤ Spreading out the frequency fading over many symbols. The advantages of COFDM. signals could be clipped because of limited quantization levels. the BER degradation is only 0.6 dB peaktoaverage ratio) of its variance. for modern solidstate transmitters. When clipping occurs. for each COFDM subchannel. the distribution is very close to Gaussian.01 %. especially when the signal constellation size is large. Its applications have been extended from high frequency radio communications to telephone networks. a 9 dB output backoff causes negligible BER degradation and adjacent channel interference. For four times of variance.
Oppenheim and R. Chang. IEEE Trans. because it is a combined amplitudefrequency modulation (it is necessary to maintain linearity) Z OFDM is very sensitive to carrier frequency offset caused by the jitter of carrier wave and Doppler effect caused by moving of the mobile terminal. 628634. March 1995 [10] P. II edition 10/30/98/TUDTVS 17 singlecarrier) being completely destroyed. Delft.. Bell Syst. IEEE First symposium. Verhoeckx. pp.M. The disadvantages of the OFDM modulation are: Z OFDM signal is contaminated by nonlinear distortion of transmitter power amplifier. 2728. COM15.32 (extended). pp.45. Commun. pp. 2nd edition..A. 1993.B. [6]A. Cimini. Sept. “Performance of an efficient parallel data transmission system”. Oct. no. Tech. 18. ISBN 0132167638 [7]A. 1966. PrenticeHall International. 41. K3. J. Discretetime signal processing: an introduction. Discrete time signal processing. IEEE Trans. Z At the receiver. ¤ The symbol period is increased and thus the sensitivity of the system to delay spread is reduced.. and multicarrier modems”. pp. “COFDM: An overview”. Technol.M. van den Enden and N. Jr. 1993...vol. COM33. Salzberg. fading and multiple access.J. “An overview of millimetre waves for future personal wireless communication systems”. [2] Ministerie van Verkeer and Waterstaat. “Data transmission by frequencydivision multiplexing using the discrete Fourier transform”. Oct. July 1985. countermeasures against frequency offset. Netherlands.W. 127136. Proc. vol. [5]S. 805813. Further studies should be conducted on the synchronization of OFDM signal. “Analysis and simulation of a digital mobile channel using orthogonal frequency division multiplexing”.. Shelswell. 1989. 1989. 1967.. power demand. Dec.. June 1995 [11] Telebit corporation.Y. . ISBN 0132167719 [8]L. Ebert. Wu. Hoofddirectie Telecommunicatie en Post. [4]B. COM19. pp. 1989. London: Prentice Hall Int.V. pp. vol. [9] W. Electronics & communication engineering journal. CCITT SG XVII. Technol. Dec. Prasad.R. it is very difficult to decide the starting time of the FFT symbol Communications research and current development of COFDM around the world will certainly provide us with valuable findings in theory and implementation. 665675. “The COFDM modulation system: the heart of digital audio broadcasting”.W. on communications and vehicular technology in the Benelux. Schaffer. vol. Weinstein and P. Broadc. IEEE Trans. 1.. Frequency allocations in the Netherlands. ”Synthesis of BandLimited Orthogonal Signals for Multichannel Data Transmission”. 1971. “Comparative performance results for asymmetrical duplex. Commun.M. V. contribution D56.Introduction to OFDM. IEEE Trans. (many) symbols in parallel are only slightly distorted.W. Groningen. Zou and Y. 17751796. Commun. VII References [1] R. vol. [3] R.
II edition 10/30/98/TUDTVS 18 [12] M. 325. no. EBU Review. “Principles of modulation and channel coding for digital broadcasting for mobile receivers”.Introduction to OFDM. 224. pp. . Lassalle.Alard and R. Aug. 1987.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.