Fading Types

OFDM History

“OFDM was proposed as a mathematical possibility as far back as 1957 with “Kineplex,” a
multi-carrier high frequency (HF) modem designed by Mosier and Clabaugh, although the first
patented application was not until 1966 when Chang of Bell Labs filed US patent 3488445. The
first practical implementation of an OFDM system came in 1985 when Telebit introduced the
“Trailblazer” range of modems that reached speeds of 9600 bps. This highlighted one of the key
advantages of OFDM: its ability to perform well through a low quality channel—in this case
telephone lines—thereby outperforming existing solutions. From this early beginning, OFDM
has become the technology that now delivers up to 10 Mbps over digital subscriber lines (DSL).
It is also used in systems that communicate over domestic power lines.

The 1980s and early 1990s saw a number of experimental broadcast systems, with companies
including Thomson-CSF and TDF in France and BBC Research in the UK. The first international
standard to specify OFDM was digital audio broadcast (DAB) in 1995, the outcome of the
European Eureka147 project, and this was followed two years later by the digital video
broadcastterrestrial (DVB-T) standard. Both DAB and DVB-T are now in widespread use. In
addition to the use of OFDM in unidirectional broadcast technologies, parallel work throughout
the 1990s led in 1999 to the first OFDM-based wireless LAN (WLAN) standard, IEEE 802.11a.
This was followed in succession by 802.11g, 802.11n (adding MIMO), and 802.16d (fixed
WiMAX™), although the most widely deployed WLAN standard is still 802.11b, which uses
direct sequence spread spectrum.

The use of OFDM for cellular systems was first briefly considered back in the late 1980s as a
candidate technology for GSM but was quickly dropped due to lack of cost-effective computing
power. A decade later, OFDM was seriously considered as one of the candidates for 3GPP’s
UMTS but was ruled out in favor of wideband code division multiple access (W-CDMA). Again
the decision was influenced by the cost of computing power and the associated power
consumption in the terminals.

However, with today’s availability of small, low-cost, low-power chipsets, OFDM has become the
technology of choice for the next generation of cellular wireless. The first cellular system to
adopt OFDM was 802.16e (Mobile WiMAX™). It was followed soon after by 802.20, the basis
for 3GPP2’s Ultra-Mobile Broadband (UMB), and most recently by 3GPP for the long-term
evolution of UMTS. It now seems apparent that the evolution of these newest so-called 3.9G
systems toward 4G will not result in any change to the underlying air interface, so OFDM will
likely be the technology of choice for cellular wireless systems well into the future. The new
OFDM cellular systems all focus on delivering high-speed data services and have similar goals
in terms of improving spectral efficiency, with the widest bandwidth systems providing the
highest single-user data rates. [1]
OFDM Motivations

For a transmission channel with a delay spread τm, and a symbol duration TS ( = 1/Bchannel),
a reception free of intersymbol interference (ISI) is only possible if the condition “​τm << TS​​”
(flat-fading) is fulfilled. As a consequence, the possible bit rate Rb = log2(M)TS^−1 for a given
single carrier modulation scheme is limited by the delay spread of the channel.

We can overcome this limitation is to split the data stream into K substreams of lower data rate
and to transmit these data substreams on adjacent subcarriers. Each subcarrier has a
bandwidth B/K, while the symbol duration TS is increased by a factor of K, which allows for a K
times higher data rate for a given delay spread.

The factor K, however, cannot be increased arbitrarily, because too long symbol durations make
the transmission too sensitive against the time incoherence of the channel that is related to the
maximum Doppler frequency ​νmax.​ There, we state that the condition ​νmax*TS << 1​​ must be
fulfilled. One should expect that there exists a symbol duration TS that satisfies both
requirements together to give the best possible transmission conditions for that channel. [2]

Frequency diversity is attained by coding over independently faded sub-carriers. [5]

OFDM Basics
A basic OFDM signal x(t) during the time interval mTu to (m+1)Tu can thus be expressed as

The term Orthogonal Frequency-Division Multiplex is due to the fact that two modulated OFDM
subcarriers xk1 and xk2 are mutually orthogonal over the time interval mTu t (m 1)Tu, that is:
 [3]

OFDM Limitations

1. Delay spread which leads to Inter-symbol Interference (while we already assume

By selecting the bandwidth f of each subchannel to be very small, the symbol duration T
= 1/f is large compared with the channel time dispersion. One way to avoid the effect of ISI is to
insert a time guard band of duration νT/N between transmissions of successive blocks. An
alternative method that avoids ISI is to append a cyclic prefix to each block of N signal samples.
[4]
2. Doppler Spread which leads to Inter-channel Interference (solved by filters with
sharp transition)

The large spectral overlap of the OFDM signals has various ramifications when the
communication channel is a radio channel and the receiving terminal is mobile, as in the case of
cellular radio communications. In such mobile radio communications, the transmitted signal is
imparted with Doppler frequency shifts or Doppler spreading, which destroys the orthogonality
among the subcarriers and, as a consequence, results in interchannel interference (ICI). The ICI
produces a significant degradation in the performance (error probability) of the OFDM system.
The degree of performance degradation is proportional to the speed at which the receiving
terminal is moving. In general, the degradation is small when the terminal is moving at
pedestrian speed.
The detrimental effects of ICI in a multicarrier system, such as OFDM, can be
significantly reduced by employing a bank of parallel filters in the implementation of the system,
as illustrated in Figure 11.2–7. In such an implementation, the prototype filter H0( f ) and, hence,
its frequency-shifted versions Hk ( f ) = H0( f − k/T ) are designed to have sharp cutoff
frequency-response characteristics. Consequently, a Doppler frequency spread that is small
compared to 1/2T , or equivalently, compared to the bandwidth of the prototype filter H0( f ), will
result in negligible ICI.[4]

3. Peak to Average Power Ratio (PAPR) (solved by changing the phase of all/some of
the subcarriers)

A major problem with multicarrier modulation is the relatively high peak-to-average ratio
(PAR) that is inherent in the transmitted signal. In general, large signal peaks occur in the
transmitted signal when the signals in many of the various subchannels add constructively in
phase. Such large signal peaks may result in clipping of the signal voltage in a D/A converter
when the multicarrier signal is synthesized digitally, and/or it may saturate the power amplifier
and thus cause intermodulation distortion in the transmitted signal. When the number N of
subcarriers is large, the central limit theorem may be used to model the combined signal on the
N subchannels as a zero-mean Gaussian random process. In such a model, the voltage PAR is
proportional to √N.
Various methods have been devised to reduce the PAPR in multicarrier systems. One of
the simplest methods is to insert different phase shifts in each of the subcarriers. These phase
shifts can be selected pseudorandomly, or by means of some algorithm, to reduce the PAPR.
Another method that can be used to reduce the PAR is to modulate a small subset of the
subcarriers with dummy symbols which are selected to reduce the PAR. Since the dummy
symbols do not have to be constrained to take amplitude and phase values from a specified
signal constellation, the design of the dummy symbols is very flexible. The subcarriers carrying
dummy symbols may be distributed across the frequency band. Since modulating subcarriers
with dummy symbols results in a lower throughput in data rate, it is desirable to employ only a
small percentage of the total subcarriers for this purpose. [4]
References

[1] Romney, Muray, “Chapter2: Air Interface Concepts: Orthogonal Frequency Division
Multiplexing”, LTE and the Evolution to 4G Wireless, John Wiley & Sons Limited, 2013, pp
53-54 (Book)
[2] Schulze, Henry, Luders, Christian, “Chapter 4: OFDM”, Theory and Applications of OFDM
and CDMA Wireless Wideband Communications, John Wiley & Sons Limited, 2005, pp 145
(Book)
[3] Erik Dahlman, Stefan Parkvall, and Johan Sköld, “Chapter 3: OFDM Transmission”, 4G
LTE/LTE-Advanced for Mobile Broadband, Elsevier Ltd., 2011, pp 2 (Book)
[4] Proakis, John, Salehi, Masoud, “Chapter 11: Multichannel and Multicarrier Systems”, Digital
Communications, Fifth Edition, McGraw-Hill Higher Education, 2008, pp 754