OFDM Modulation Study for a Radio-over-Fiber System

for Wireless LAN (IEEE 802.11a)

Gurprakash Singh and Arokiaswami Alphones.
School of Electrical and Electronic Engineering
Nanyang Technological University, Singapore 639798.
Email: pa5385811@ntu.edu.sg, ealphones@ntu.edu.sg


AbstractRadio-over-Fiber(RoF) technology has
several benefits such as larger bandwidth, reduced power
consumption etc. that has made it an attractive
implementation option for various communication systems
including Wireless LANs. This paper investigates the
feasibility of Orthogonal Frequency Division Multiplexing
(OFDM) as a modulation technique for a ROF based
WLAN system in consistency with IEEE 802.11a [1].
Results from a Matlab system model, which is consistent
with the specifications of IEEE 802.11a, have been included
to show the BER performance of the OFDM modulation
technique. Next, the paper presents the results obtained from
measuring several of the main factors affecting the
performance of an OFDM system, including channel noise,
multipath delay spread, power clipping, and synchronization
requirements.

I. Introduction

RoF is a technology by which microwave (electrical) signals
are distributed by means of optical components and
techniques. A RoF system consists of a Central Site (CS)
and a Remote Site (RS) connected by an optical fiber link or
network. One of the major motivation and system
requirement for RoF technology is the use simple and cost-
effective RS [2]. The electrical signal distributed may be
baseband data, modulated IF, or the actual modulated RF
signal. The electrical signal is used to modulate the optical
source. The resulting optical signal is then carried over the
optical fiber link to the remote station. By delivering the
radio signals directly, the optical fiber link avoids the
necessity to generate high frequency radio carriers at the
antenna site. Since antenna sites are usually remote from
easy access, there is a lot to gain from such an arrangement.
However, the main advantage of RoF systems is the ability
to concentrate most of the expensive, high frequency
equipment at a centralized location, thereby making it
possible to use simpler remote sites [3].

The IEEE 802.11a standard is proposed for a range of data
rates from 6 up to 54 Mbps using the OFDM modulation
technique in the 5 GHz band. The 5 GHz band is
specifically designed for Wireless Broad-band Mobile
Communication Systems (WBMCS)". The IEEE 802.11a
WLAN standard is superior compared with current

technologies because of its greater scalability, better
interference immunity and significantly higher speed, and at
the same time allowing for higher bandwidth applications
and more users [4]. The 802.11a standard utilizes 300 MHz
of bandwidth in the 5 GHz Unlicensed National Information
Infrastructure (U-NII) band.

Orthogonal Frequency Division Multiplexing (OFDM) is
seen as the modulation technique for future broadband
wireless communications because it provides increased
robustness against frequency selective fading and
narrowband interference, and is efficient in dealing with
multi-path delay spread [5]. To achieve this, OFDM splits
high-rate data streams into lower rate streams, which are
then transmitted simultaneously over several sub-carriers.
By doing so, the symbol duration is increased. The
advantage of this is that the relative amount of dispersion in
time caused by multi-path delay spread is decreased
significantly.

As stated above, OFDM uses multiple sub-carriers to
transmit low rate data streams in parallel. The sub-carriers
are modulated by using Phase Shift Keying (PSK) or
Quadrature Amplitude Modulation (QAM) and are then
carried on a high frequency microwave carrier (e.g. 5 GHz).
This is similar to conventional Frequency Division
Multiplexing (FDM) or Sub-Carrier Multiplexing, except
for the stringent requirement of orthogonality between the
sub-carriers.

II. Overview

A. IEEE 802.11a Physical link Simulation Model

Figure 1 below depicts the Physical link model used for
BER study of OFDM in 802.11a.


ICICS-PCM 2003
15-18 December 2003
Singapore
3A6.4
0-7803-8185-8/03/$17.00 2003 IEEE
Scrambler
FEC
Convolu
tional
Coder
(or)
Puncturi
ng
Block
Interleaver
Mapping
OFDM
IFFT
Cyclic
Prefix
and
Filter
IQ
Modulation
DAC
RF
TX
Input
Bits
RF
RX
ADC
Timing and
Frequency
Synch.
Remove
Cyclic
Extension
OFDM
FFT
Channel
Correction
Demapping
Block
Deinterleaver
FEC
Decoder (or)
Depuncturing
Descrambler
Output
Bits
Symbol Timing
Frequency
corrected
signal



Figure 1: 802.11a Transmitter and Receiver

The Data Rate determines the choice of Modulation
technique and Coding Rate in OFDM transmission for
802.11a. A list of timing related parameters for OFDM in
the IEEE 802.11a standard is given in Table 1 [1].

Data Rate
(Mbits/sec)
Modulation Coding
Rate
Coded
Bits per
sub-
carrier
Coded
Bits per
OFDM
Symbol
6 BPSK 1/2 1 48
9 BPSK 3/4 1 48
12 QPSK 1/2 2 96
18 QPSK 3/4 2 96
24 16-QAM 1/2 4 192
36 16-QAM 3/4 4 192
48 64-QAM 2/3 6 288
54 64-QAM 3/4 6 288

Table 1:Rate Dependent Parameters


A MATALB model of the 802.11a physical link was used to
simulate the transmission and reception of data at the rates
of 6,9,12 and 18 Mbps. The standard specifies 52 sub-
carriers that are modulated using either binary or quadrature
phase shift keying (BPSK/QPSK); forward error correction
coding (convolution coding) is used with coding rates of
and .


The parameters of the 802.11a physical layer implemented
for this simulation are listed below
End-to-end 802.11a physical layer
BPSK and QPSK Modulations
All mandatory and optional rates for PSK: 6, 9,
12, 18 Mbps
OFDM transmission: 52 sub-carriers, 4 pilots,
64-pt FFTs, circular prefix.
Forward error correction coding (convolution;
code rates and )
Data interleaving
Viterbi decoding
AWGN Channel Model

B. OFDM Simulation Model

Figure 2 below shows the configuration for a basic OFDM
transmitter and receiver. The signal generated is at base-
band and so to generate an RF signal the signal must be
filtered and mixed to the desired transmission frequency [5].



Figure 2: Basic FFT, OFDM Transmitter and
Receiver


Figure 3 shows the OFDM system that was modeled using
MATLAB. The simulations were designed to measure the
performance of OFDM under different channel conditions,
and to allow for different OFDM configurations to be tested.
Three main criteria were used to assess the performance of
the OFDM system, which were its tolerance to multipath
delay spread, peak power clipping, and channel noise.


Figure 3: Simulation Model for OFDM



Some of the parameters and features of the model are listed
below
Since these simulations employ PSK modulation
schemes, the supported data rates in the simulation are up
to 18 Mbps (IEEE 802.11a Standard. Refer Table 1) [1]
A suitable guard period structure was used to allow for
symbol timing recovery using envelope detection [6]
The channel model allows for the signal to noise ratio,
multipath, and peak power clipping to be controlled

Table 2 shows the configuration used for the simulations
performed on the OFDM signal. An 800-carrier system was
used, as it would allow for up to 100 users if each were
allocated 8 carriers. The aim was that each user has multiple
carriers so that if several carriers are lost due to frequency
selective fading then the remaining carriers will allow the
lost data to be recovered using forward error correction.

In addition to DPSK and QPSK (802.11a standard), 16-PSK
modulation was also simulated to compare efficiency.


Parameter Value
Carrier Modulation
Used
DBPSK, DQPSK, D16PSK
FFT Size 2048
Guard Time 512 Samples (25%)
Guard Period Type
Half zero signal, half cyclic
extension of symbol
Number of Carriers
used
800

Table 2: Simulation System Parameters

III. Implementation and Results

Theoretical results generated for the IEEE 802.11a Phy link
(described in II A) are presented first. The simulation
calculates the BER for various EsNo values (where Es is
energy per symbol) for the selected PSK modulation as well
as data rate.

The results obtained using DBPSK modulation for a Data
Rate of 6 Mbps and DQPSK modulation for 12 Mbps are
shown below.


Figure 4: BER in 802.11a for 6Mbps



Figure 5: BER in 802.11a for 12Mbps


Shown next are the results for the OFDM transmission
simulation (described in II B). Differential encoding has
been considered to be an integral part of an OFDM
Transmission system and hence these simulations make use
of Differential PSK.



It was found that the SNR performance of OFDM is similar
to a standard single carrier digital transmission. This is to be
expected, as the transmitted signal is similar to a standard
Frequency Division Multiplexing (FDM) system.
BER vs Channel SNR
0.0001
0.001
0.01
0.1
1
0 5 10 15 20 25
Channel SNR (dB)
B
E
R
DBPSK
DQPSK
D16PSK

Figure 6:BER vs SNR for OFDM

The OFDM signal was tested with a multipath signal
containing a single reflected echo. The reflected signal was
made 3 dB weaker than the direct signal as reflections
weaker than this did not cause measurable errors.
BER vs Multipath Delay
0.001
0.01
0.1
1
0 100 200 300 400 500 600 700 800 900
Delay Spread (samples)
B
E
RDBPSK
DQPSK
D16PSK

BER vs Peak Power Clipping
0.00001
0.0001
0.001
0.01
0.1
1
0 3 6 9 12 15 18 21 24
Peak Power Clipping (dB)
B
E
R
DBPSK
DQPSK
D16PSK

Figure 7:Delay Spread tolerance for OFDM
Figure 8: Peak Power Clipping for OFDM

IV. Conclusions

The results shown in Figures 4 and 5 depict the theoretical
performance of PSK modulation using OFDM in the
802.11a physical model. The results obtained show that the
performance of PSK signals suffers during transmission
through the 802.11a PHY link. This can be compared to the
results generated in Figure 6 for OFDM Transmission using
PSK modulation through an AWGN channel. This drop in
performance is due to the higher data rates being used in this
simulation. Also some factor of the difference could be
attributed to the use of different OFDM data generators in
the two MATLAB simulations. IEEE 802.11a employs
Quadrature Amplitude Modulation (QAM), a coherent
modulation scheme, rather than DPSK for higher data rates
of 24, 54 Mbps. This is because QAM provides lower
detection error probability and better spectral efficiency.

Figure 6 shows that using QPSK, the transmission can
tolerate a SNR of >10-12 dB. The bit error rate BER gets
rapidly worse as the SNR drops below 6 dB. However,
using BPSK allows the BER to be improved in a noisy
channel, at the expense of transmission data capacity.

Figure 7 shows that the BER is very low for a delay spread
of less than approximately 256 samples. The results show
that the tolerable delay spread matches the time of the cyclic
extension of the guard period. For a delay spread that is
longer than the effective guard period, the BER rises rapidly
due to the inter-symbol interference.

From Figure 8 we can conclude that the transmitted OFDM
signal could be heavily clipped with little effect on the
received BER. In fact, the signal could the clipped by up to
9 dB without a significant increase in the BER. This means
that the signal is highly resistant to clipping distortions
caused by the power amplifier used in transmitting the
signal.

References

[1] IEEE Std 802.11a-1999 Supplement to IEEE standard for
information technology telecommunications and information
exchange between systems - local and metropolitan area networks
- specific requirements. Part 11: wireless LAN Medium Access
Control (MAC) and Physical Layer (PHY) specifications: high-
speed physical layer in the 5 GHz band.

[2] Jiunn-Shyen Wu, Jingshown Wu, Hen-Wai Tsao A Radio-
over-Fiber Network for Microcellular System Application IEEE
TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 47,
NO. 1, FEBRUARY 1998.

[3] OReilly J. J., Lane P. M., and Capstick M. H: Optical
Generation and Delivery of Modulated mm-waves For Mobile
Communications, in Analogue Optical Fiber Communications,
Editors: Wilson B., Ghassemlooy Z. and Darwazeh I.; The
Institution of Electrical Engineers, London, 1995, pp229-256.

[4] Hills A., Large-Scale Wireless LAN Design; IEEE
Communications Magazine; pp 98 - 104, Nov., 2001.

[5] Ramasami, V.C., (KUID 698659), \Orthogonal Frequency
Division Multiplexing",
http://www.ittc.ukans.edu/rvc/acads/ofdmreport.pdf

[6] Larsson, E.G., Liu, G., Li, J., and Giannakis, G.B., Joi t Symbol
Timing And Channel Estimation For OFDM Based WLANs", IEEE
Communications Letters, vol. 5, pp. 325327, Aug. 2001.