1
IEEE 802.11 OFDM Software Defined Radio
Beacon Frame Transmission
Joseph Wamicha, Member, IEEE, Simon Winberg, Member, IEEE
Faculty of Engineering and the Built Environment, University of Cape Town.
Abstract—A transmit chain test bed is used in the project
presented in this paper to test an 802.11g OFDM (Orthogonal
Frequency Division Multiplexing) Software Defined Radio (SDR)
base station beacon frame transmitter. Several prior Gnuradio
test bed setups show how a Gnuradio SDR Transmitter is able to
communicate with another Gnuradio SDR. This paper additionally
shows how a Gnuradio SDR Transmitter is able to visibly com-
municate with off-the-shelf 802.11 WLAN (Wireless Local Area
Network) off-the-shelf chips. Results of the implementation show
that a SDR base station is successfully able to communicate with
traditional WLAN off-the-shelf communication chips found inside
devices such as laptops, netbooks and mobile phones, which were
used as the OFDM beacon frame receptors.
Index Terms—Software defined radio, orthogonal frequency
division multiplexing, wireless local area network, 802.11, beacon
frame, medium access control, gnuradio, universal software radio
peripheral 2, flex 2400 radio frequency board, symbols, frames,
wifi, off the shelf.
I. I NTRODUCTION
R
ADIO communication systems in production today
are often implemented using application specific
integrated circuits (ASICs) which support a single modulation
method. Due to the rapidly evolving nature of wireless
communications and the aggressive competition from market
rivals, network operators are forced to upgrade their ASIC radio
communications equipment, without getting the opportunity
to fully realize their return on investment on the ASIC
hardware installation. Despite this disadvantage, ASIC radio
communications hardware devices are relatively cheap to mass
produce [1]
To address high radio communications hardware turnover
rate problem, in his classic paper Joe Mitola proposes
the partitioning of channel coding into the following
functional components: antenna, radio frequency, analog
to digital conversion (reception), digital to analog conversion
(transmission) and digital signal processing [2]. Signal
processing functions such as modulators, demodulators,
mixers, filters and amplifiers, which vary between different
communication protocols, could now be implemented using
software. This would now allow a network operator’s base
station to communicate using different communication protocols
by reconfiguring the abstract software application layer, thereby
overcoming the more prohibitively expensive option of having
to replace the underlying communications hardware. This
makes it possible for network operators to affordably upgrade
mobile networks in underserved African rural areas to the latest
wireless protocols, using the intial SDR hardware investment.
In order to accurately model real life network conditions
and to prove that software defined radios could replace the
ASIC communications chip in future, the OFDM software
defined radio implementation presented in this paper had to
be compatible with off-the-shelf ASIC chips. So as to ensure
this compatibility goal was met, an open standard had to be
implemented on the Gnuradio-USRP2 (Universal Software
Defined Radio Peripheral 2) software defined radio platform
from Ettus Research. The potential open standards which could
have been implemented were the IEEE 802.11 WLAN standard
[3], the IEEE 802.16 WiMAX (Worldwide Interoperability
for Microwave Access) standard [11] and the ETSI DVB-T2
(Digital Video Broadcasting - Second Generation Terrestrial)
standard [12]. In the end the IEEE 802.11 WLAN standard was
selected for the test implementation since off-the-shelf 802.11
devices are more readily available.
The paper begins by detailing the hardware used for the
test bed setup, followed by an explanation of the OFDM
WLAN protocol and implementation used and ending with
the presentation of the test bed results. The paper will often
refer to the gr wlan code which was used to generate the
802.11g WLAN OFDM beacon frames. The gr wlan code
wraps and refactors the Gnuradio OFDM python flow graph and
C++ signal processing blocks from FTW and originally written
by Andrea Constantini [17]. The python flow graph code in
turn makes use of underlying C++ Gnuradio signal processing
blocks by using SWIG (http://www.swig.org). The gr wlan code
referred to in this paper can be downloaded from the following
link: http://rrsg.ee.uct.ac.za/members/jwamicha/gr-wlan.tar.gz
II. SOFTWARE DEFINED RADIO HARDWARE SETUP
The SDR OFDM implementation presented in this paper was
done using the USRP2 FPGA board working in conjunction
with an RF (Radio Frequency) daughter board [9]. Both boards
are manufactured by Ettus Research, a subsidiary of National
Instruments. The USRP2 consists of an FPGA and mixed
signal circuit section consisting of an ADC (Analog to Digital
Converter) and DAC (Digital to Analog Converter) [8]. Since
the 802.11g WLAN channels fall between a frequency range of
2.412 GHz and 2.484 GHz, the RFX2400 RF daughter board
was used for the OFDM symbol transmission since it is able
to transmit at between 2.25GHz and 2.9GHz.
IEEE Africon 2011 - The Falls Resort and Conference Centre, Livingstone, Zambia, 13 - 15 September 2011
978-1-61284-993-5/11/$26.00 ©2011 IEEE