DESIGN AND IMPLEMENTATION OF A WIRELESS SOFTWARE-

DEFINED RADIO TESTBED

DISEÑO E IMPLEMENTACIÓN DE UN BANCO DE PRUEBAS

INALÁMBRICO CON RADIO DEFINIDO POR SOFTWARE
SERGIO PINO GALLARDO
Facultad de Ingenierías Físico-Mecánicas, Universidad Industrial de Santander, Colombia, [email protected]

HENRY ARGUELLO FUENTES

Ph.D., Facultad de Ingenierías Físico-Mecánicas, Universidad Industrial de Santander, Colombia, [email protected]

Received for review February 15 th, 2013, Accepted May 27 th, 2013, final version July, 3 th, 2013

ABSTRACT: Software-defined radio (SDR) provides a convenient framework for the design and implementation of a communication
system, by separating the signal processing algorithms from the communication hardware. This separation allows researchers to design
testbed systems to validate the gains in performance reported by the theory. Designing and implementing a testbed is a time consuming
and challenging process. This work presents the design and implementation of a flexible testbed system, including the solution to some
synchronization problems. We also present a distributed software architecture that defines the control flow, the subsystem decomposition,
and the mapping to hardware of the testbed. A running example of the testbed for a binary phase-shift keying (BPSK) system is developed
to illustrate the results attained by the proposed testbed in terms of the performance comparison between the testbed measurements, and the
theoretical and simulated ones with a loss of 2 dB.

KEYWORDS: Software-defined radio, Digital communications, Communication testbed.

RESUMEN: El radio definido por software (SDR) proporciona un marco de trabajo conveniente para el diseño e implementación de un
sistema de comunicación, mediante la separación de los algoritmos de procesamiento de señales y el hardware de comunicación. Esta
separación permite a los investigadores diseñar sistemas de banco de pruebas para validar las ganancias en el rendimiento reportados por la
teoría. El diseño e implementación de un banco de pruebas es un proceso que consume tiempo y esfuerzo. Este trabajo presenta el diseño
e implementación de un sistema de banco de pruebas flexible, incluyendo la solución a algunos problemas de sincronización. Además, una
arquitectura distribuida de software que define el flujo de control, la descomposición en subsistemas, y la asignación de hardware de la
plataforma de pruebas es presentada. Se desarrolla un ejemplo del funcionamiento del banco de pruebas para un sistema con modulación
binaria por desplazamiento de fase (BPSK) y se ilustran los resultados obtenidos por el banco de pruebas propuesto en términos de la
comparación de rendimiento entre las medidas obtenidas por del banco de pruebas, y los rendimientos teóricos y simulados con una de
perdida 2 dB.

PALABRAS CLAVE: Radio definido por Software, Comunicaciones digitales, Banco de pruebas de comunicación.

1. INTRODUCTION clearly is overwhelming [1]. On the other hand, the

use of hardware platforms and testbeds in wireless
Conventional radio design traditionally implies the communications takes the role of validating the gains
use of analog and mixed signal design techniques that in performance reported by the theory and simulations.
enable engineers to work on complex radio frequency Unlike the latter, the testbeds operate in real channels
(RF) circuitry. An inherent issue of the former approach under the presence of implementation constraints
lies in the high cost of designing, testing, and building [16, 9, 22]. However, in the early 1990s Joe Mitola
a complete communication system. In addition, the introduced the concept of software-defined radio
impact of a mainly hardware based implementation of (SDR) [3] to refer to the class of reprogrammable or
the communication system consisting of oscillators, reconfigurable radios, or as Jeffrey Reed stated, a radio
filters, mixers, amplifiers, and hardware for source and that is substantially defined in software and whose
channel coding, as well as for modulation; assumes physical layer behavior can be significantly altered
a priori that an upgrade to the resulting hardware through changes to its software [17].

Dyna, year 80, Nro. 180, pp. 67-76. Medellin, August, 2013. ISSN 0012-7353
 68 Pino & Arguello

performance of advanced overtheair technologies, such

as 3GPP LTE type MIMO OFDM systems, has been
developed [4]. Also, middleware software that provides
user access to testbed systems has been proposed [9],
and used for experimental evaluation of alternative
schemes such joint sourcechannel coding [8].

This work presents a complete design and implementation

Figure 1. The frequency shift in SDR includes converting of a flexible wireless testbed system, including the
a digital signal by means of a digital to analog converter solution to some synchronization problems. For this
(DAC), upconverting the analog signal to the desired end, First we explore a set of algorithms that separately
RF center frequency, amplifying the signal to meet the try to reduce the effects on the transmitted signal due to
appropriated power level, and limiting the bandwidth of the following synchronization problems: symbol phase
the signal before its radiation into the medium. synchronization, symbol frequency synchronization,
carrier phase synchronization, carrier frequency
In SDR, conventional frequency shifting of the signal synchronization, and frame synchronization. Also,
at the modulation hardware is replaced by a two steps we propose a distributed software architecture that
process that converts the baseband digital modulated defines the control flow, the subsystem decomposition
signal into a radiated passband analog signal as depicted and the mapping to hardware of the testbed. The
in Fig. 1. The key idea behind the latter process is the testbed is based on the Universal Software Radio
concept of equivalent lowpass signal [21] that suggests Peripheral (USRP 2) which is a low cost, high quality
that, from a mathematical perspective, it is feasible softwaredefined radio (SDR) system designed by
to ignore the center frequency (carrier generation) in Ettus Research [23]. Finally, a running example of the
which the system has to operate and then just perform testbed for a binary phase-shift keying (BPSK) system
all the mathematical treatment of the signal using is implemented in order to show the results attained
the associated lowpass signal. Formally, assume that by the proposed testbed in terms of the performance
s (t ) is the bandpass signal to be transmitted, so its comparison between the testbed measurements; and
spectrum is X ( f )  0 for Fl < f < Fu , where Fl is the the theoretical and simulated ones.
low frequency, and Fu is the high frequency, then it is
possible to represent s (t ) as 2. COMMUNICATION SYSTEM MODEL
j 2 fc t
s(t ) = Re{sl (t )e }, (1) The design of the testbed relies on the underlying
mathematical model that reflect the most important
where sl (t ) = x (t )  jy (t ) is the equivalent lowpass signal characteristics of the transmission environment. Such
or complex envelope of s (t ) . Thus in SDR, while the a model of the channel is used in the design of other
software part has the responsibility of creating the components of the testbed such as the modulator at the
digital representation of the signal sl (t ) , the hardware transmitter, the demodulator at the receiver, the channel
is in control of the generation of the carrier e j 2 f c t , and encoder and decoder. In order to model a wireless
modulate it with sl (t ) . channel, it is common to use the additive noise channel,

There are testbed systems reported in the literature r (t ) = s(t )  n(t ), (2)
that address the design and implementation of such
systems for academic purposes. Rao in [16] presents where is the channel attenuation, and the signals s (t ) , n (t )
a classification scheme for wireless testbeds, including and r (t ) are the transmitted signal, the additive random noise
examples for each case and discussion on the role of the channel, and the received signal, respectively
of such systems in an educational environment. [21]. A particular case is the additive white gaussian
One research focuses on the latency of SDR, and its noise channel AWGN in which the noise n (t ) is normal
impact on throughput in modern wireless protocols distributed, n(t)~N( 0 ,ó 2 ) .
[18]. A flexible SDR system to quantify the real world
 Dyna 180, 2013 69

Here, it is assumed that the communication components

Y = r (t ) ; and d ( x, xˆ ) is the distortion measure [10].
are already defined, such as the target transmission
scheme, and the channel and source encoder/decoder. An important case that enables the expression of OPTA,
Next, the system has to compute the expected in analytic form, is the case of a memoryless Gaussian
performance measurement of the communication, or source and AWGN channel, as depicted in Fig. 2.
the theoretical performance as a function of the signal
to noise ratio (SNR). The SNR is defined as the ratio
of the signal power Ps to the noise power Pn , given by
Ps
SNR = . (3)
Pn

Subsequently, the performance measurement can

be selected based on either its dependency on
bit representation, or its nondependency on bit
representation [10]. First, when the representation is
binary, a common performance measure used is either
the bit error rate BER , or the probability of a bit error
Pb , both of which are theoretically computed using
the posterior probabilities Ps m | r . Furthermore,
r = [r1 r2  rN ] is the observation vector produced either
by the correlation demodulator or the matched filter
demodulator, and sm is the transmitted signal [21].
More specifically, Ps m | r  is given by
P r | s m P s m 
P s m | r  = . (4)
P r 

A straightforward case is the performance of a

modulation system that uses binary phase-shift keying
BPSK as the modulation scheme over an AWGN
channel, Fig. 2.
Figure 2. Top: Theoretical BER vs SNR for a BPSK
On the other hand, in the case when the process modulation scheme over an AWGN channel. Right:
of source and channel coding does not imply an Theoretical OPTA vs SNR for different values of B and W.
intermediate digital representation, it is required that
the performance measure does not depend on a bit Using different values for the source bandwidth B and
representation. For those cases, it is convenient to use the channel bandwidth W , OPTA is defined as
the Optimal Performance Theoretically Attainable, 2  2 
W/B
(OPTA) which is the expression of the limits for s = 1  x  , (6)
2  2 
efficient communications. OPTA is computed by q  n 
equating the expression of channel capacity C and where q2 is the distortion, s2 is the source variance,
2 2
the rate distortion function R (D ) , and solving for the x is the transmit power, and n is the channel noise
signal to noise ratio as [15, 8]. Both the rate distortion function and the channel
capacity for non Gaussian sources and channels, can
be estimated using the Blahut algorithm [2].
R( D) = min I ( X ; Xˆ ) = max I ( X ; Y ) = C , (5)
p( xˆ | x ):E[ d ( x, xˆ )]  D p ( x)
3. MATHEMATICAL TOOLS
where I ( X ; Y ) is the mutual information function of
In the context of a real SDR implementation, there are
the transmitted signal X = s(t ) and the received signal
different non considered sources of error besides the
 70 Pino & Arguello

thermal noise. Thus, the model given by

k m
r (t ) = 0 s (t  T0 )   i si (t  Ti )  n(t )   y j (t j ), (7)
i =1 j =0

extends (2) by considering at the receiver the time

shift or delay T0 of the transmitted signal s (t ) ; the
fading in the channel 0 ; the k reflected signals si (t )
due to multipath and their delay Ti ; the thermal noise
n(t ) in the electronic components of the receiver; and
finally m possible interfere signals y j due to other
communication systems in other frequency bands.
Therefore, from a practical point of view a receiver
has to overcome five basic synchronization problems.
These problems are expressed next as its solution
formulation (Fig. 3 shows some of these problems)
[19], [20],

1. Symbol phase synchronization deals with choosing

when to sample the signal within each symbol time Figure 3. Constellation diagrams for SNR = 7 dB. Top
interval T . Left: no timing or carrier error. Bottom Left: no carrier
error, timing error of T/20 . Top Right: SNR = 15 dB with
2. Symbol frequency synchronization addresses carrier frequency error, no timing error. Bottom Right:
the problem of different oscillator rates at the SNR = 7 dB with constant carrier phase error of 0.4 , no
transmitter and receiver which is common in real timing error.
systems.
3.1. Frame Synchronization
3. Carrier phase synchronization addresses the
aligning of the phase of the carrier at the receiver Given the time delay T0 of the transmitted signal
with the phase of the carrier at the transmitter. it is necessary to perform a time alignment of the
signal at the receiver, which is denominated frame
4. Carrier frequency synchronization deals with synchronization, so that it is feasible to find the
aligning the frequency of the carrier at the receiver optimum sampling time for the beginning of the
to the frequency of the carrier generated at the frame. There are several techniques based on optimum
transmitter. frame synchronization [14, 7, 5]. However, the cross
correlation operation is a straightforward and a well
5. Frame synchronization addresses the problem of studied approach [19, 21, 22]. This correlation is the
finding the initial sample of each message. sequence R [l ] computed by
rw


Therefore, this work explores a set of algorithms (8)
that separately try to reduce the effects on the
Rrw [l ] =  r[n  l ]w[n],
n = 
transmitted signal due to the aforementioned issues. between the received discrete sequence r and a known
However, the aim of the work is to create an initial pilot sequence w , at the time shift parameter l . The
testbed implementation, not to address an exhaustive signal w is supposed to be appended to the transmitted
discussion about the optimum algorithm for solving sequence s so that the maximum value of | Rrw | for
each issue of real communications. l = 0, 1, 2, ,

|R rw [l ] |=|  r[n  l ]w[n] |, (9)
n = 

at the optimum time shift, l = l * , could be interpreted as

 Dyna 180, 2013 71

the time shift which minimizes the difference between 3.3. Estimation of the Signal to Noise Ratio (SNR)
the signals. Once l * is computed the optimum sampling
time for the beginning of the frame, t * , is computed as In order to compute the performance of the system at a
*
t =| r | l , *
(10) given SNR or Channel Signal to Noise Ratio (CSNR),
it is indispensable to estimate the value of the SNR,
where | r | is the number of samples of the signal r . given by (3), at the receiver.
3.2. Channel Attenuation First, the system estimates the statistical characteristics
of the noise in the channel. Hence, the estimation
The channel attenuation introduces a distortion in the is performed in the case when the transmitter is not
amplitude of the signal s (t ) , by scaling it by a factor sending any signal, s (t )  0 . Thus, in theory, the
. Therefore, it is required to estimate the value receiver samples are the noise samples, r (t )  n(t ) .
of the factor at the receiver so that the system An example of the results of the method applied to
can reconstruct a better approximation of the signal samples captured by an USRP 2 SDR hardware, are
s (t ) . For the AWGN channel (2), and assuming shown for both the histogram of the noise and the
that the channel attenuation, , is constant during time and frequency signal representation in Fig. 4.
the transmission time; then, it is suitable to use the Let N be a discrete random variable representing the
Maximum Likelihood Estimate (MLE) in order to discrete version of the AWGN channel noise. Then, the
find the value of the parameter that maximizes the noise power of the channel, 2 , is estimated using the
likelihood function denoted by p(r | ) . That is, find an empirical estimator defined as
that maximizes the correspondence of the selected 2 1 m
 E[ N 2 ]  lim n[ j ]2  Var ( N ), (13)
model with the observation of the received signal r (t ) . m m j =1

The AWGN channel assumes that the noise in the where there are m samples, of the sampled noise of the
channel follows a normal distribution n(t)~N( 0,ó 2 ) , and channel n , available for the estimation.
by letting the transmitted signal be set to a deterministic
signal, such as s (t ) = sin (t ) , the statistics of the received Therefore, given the AWGN channel, (2), and assuming
signal are E[r (t)] = s(t) , and Var[r (t )] = 2 , which means a discrete signal representation of the received signal
r (t ) given by
that the signal r (t ) follows a normal distribution such
as r(t)~N( s (t ) ,ó 2 ) . r = s  n. (14)
E[s 2 ]
Down converting and then sampling the continuous 2
bandpass signal r (t ) , so that there are n i.i.d The SNR E[n ] is computed using the following
(independent and identically distributed samples of methods:
a random variable) samples of the signal, denoted by
r = r1, r2 ,, rn , available for the estimation procedure.
Then, the optimum value of is estimated as
n
ri si
* i =1
= arg max p (r | ) = n
. ( 11 )
 si2
i =1
Now, considering just the samples where the amplitude
of the signal s (t ) is the maximum/minimum, in this case
s max (t ) = 1 and s min (t ) = 1 , then the products ri si  0
Figure 4. Left: Time domain and Frequency domain
. Thus, the estimation of is reduced to
n representations of the sampled noise recorded using
 | ri | an USRP 2 and plotted in Matlab. Right: Histogram of
* i =1
= . ( 1 2) the sampled channel noise recorded and the Gaussian
n
approximation of these samples.
where | . | denotes the absolute value of the sample ri .
 72 Pino & Arguello

The first method uses existing data at the transmitter latter is composed of three types of components
in order to estimate the SNR at the receiver, named Remote Client (RC); Slave Tx (STX) and
 
E s2
Slave Rx (SRX); and the SDR hardware as in Fig.
2
SNR = , 5, where each component is deployed over different
 
E n2 (15) and distributed hardware nodes. The use of the client/
so that the receiver knows a priori the transmitted signal server architectural style, as the basis for the software
s and the value of is estimated as described before architecture, reduces the coupling (a measure of the
in section 3.5. number of dependencies between two subsystems of a
software system) between the baseband digital signal
On the other hand, the second method just considers processing and the hardware of SDR to be used in the
the received signal r , and the SNR is estimated as testbed.
E [r 2 ]
SNR =  1,
E[n 2 ] (16)
so that the estimation of the SNR depends only on the
received signal, which means that the corresponding
values for and the signal s are implicit in r [12].

3.4. Frequency Mismatch

The testbed requires to execute measurements of the

performance of the communication system at different
ranges of SNR, including low values. Thus, in all cases Figure 5. Left: Deployment diagram of the distributed
the system needs to align the frequency of the oscillator architecture of the testbed. Right: Client/Server
at the transmitter with the frequency of the oscillator architectural style used in the software architecture.
at the receiver. Thus, the testbed uses a global GPS
(Global Positioning System) Disciplined Oscillator Conversely, the architecture increases the cohesion
that is shared between the transmitter and the receiver. (a measure of the number of dependencies within a
The main goal of this sharing is to align the frequency subsystem) within each component, in terms of the
of the carrier at the receiver with the frequency of the definition of a well defined set of boundaries and
carrier at the transmitter, this procedure solves the responsibilities of the algorithms (services) that each
carrier frequency synchronization error. The result component has to provide. Thus, while the RC node
is a coarse grained frequency synchronization that performs the baseband digital signal processing, the
avoids the continuous rotation of the constellation of SRX and STX nodes perform the control of the SDR
the received symbols at the receiver. hardware.

Although the former synchronization procedure First, given (1) the testbed implements the signal
solves most problems of frequency mismatch, there is processing and communication algorithms that
a remaining impairment related to a constant carrier generate the digital representation of the signal sl (t ) , as
phase error. The testbed addresses this impairment software to be executed offline in a personal computer
using SISO frequency estimator algorithms, such as (PC). These algorithms are part of the services available
Kay [11], Fitz [6] and Luise & Reggiannini [13]. on the RC node, and could be implemented using
MATLAB, or python with the GNU Radio library.
4. TESTBED DESCRIPTION AND SOFTWARE Currently, software for single-input single-output
ARCHITECTURE (SISO) BPSK and PAM modulation and demodulation
has been developed for this system. Therefore, the
A distributed Software Architecture is in charge of transmission of a sequence of symbols involves the
defining the control flow of the transmission and following operations:
reception operations performed by the testbed. The
 Dyna 180, 2013 73

• At RC the system performs in software the source

coding, channel coding, and modulation processes.
• The modulated discrete time samples in the output,
a discrete time version of the signal sl (t ) described
in (1), are written to a file. In addition, the file
is shared, using a file synchronization software
such as the standard NFS, to all the STX nodes
so that each transmitter has a copy of the discrete
equivalent lowpass representation of the target
transmitted signal s (t ) .
• The testbed uses a sequence of messages (protocol)
between the components, RC and STX/SRX nodes,
in order to perform a transmission/reception with Figure 6. Sequence diagram of the messages passed
the system, Fig. 6. Also, the former process stored between the components of the testbed in the execution of
the received discrete equivalent lowpass version a transmission.
of r (t ) in a file.
The Universal Software Radio Peripheral 2 (USRP 2)
• At RC the system performs the demodulation, hardware platform for SDR is the target platform in this
channel decoding, and source decoding processes work. Thus, when STX either received a sendSignal or
in software. In addition, RC uses the algorithms sendSine message it transfers the content of the related
presented in section 3 in order to solve the five file to the motherboard of the USRP 2 via the Gigabit
basic problems of synchronization. Ethernet interface. Then, the field programmable gate
array (FPGA) in the USRP 2 performs the interpolation
In the protocol operation, the control flow is dictated process and then the discrete time signal is converted
by the RC component that begins the transmission to a continuous time signal using a digital to analog
process. Thus, in a normal scenario of operation the converter (DAC). Finally, the USRP sends the
protocol behaves as follows: continuous time signal to the daughterboard which
• The RC creates a tcp socket connection with both perform the up conversion or frequency shift, in (1) this
STX and SRX nodes. The system uses the input/ process is the product of the signal sl (t ) by e j 2 f c t , as
output streams of the connection to send and depicted in Fig. 7.
received the protocol messages from the RC to the
STX/SRX. Also, these connections are kept alive
for the duration of the transmission.
• The RC assigns the desired signal to transmit s (t )
to the STX node by passing the path of the file that
contains the signal in the “setSignal” message.
• The testbed performs the transmission and
recording procedures which have the following
message order: RC sends the “record” message Figure 7. Block diagram of the SDR hardware that is used
to the SRX node and SRX responds with an in the testbed.
acknowledge message. RC sends the sendSine or
The USRP 2 hardware, table 1, has a motherboard that
sendData message to STX and STX responds with
uses a Xilinx Spartan-3 2000 FPGA for performing
an acknowledge message after the transmission
the digital signal processing, and a gigabit Ethernet
of the signal is performed. Then, the RC sends a
interface for communication with the PC. Moreover,
stopRecord message to SRX in order to finish the
the daughterboard is a RFX2400 high performance, full
recording of the sampled signal at RX.
duplex transceiver designed specifically for operation
in the 2.4 GHz band [23].
 74 Pino & Arguello

Table 1. Hardware parameters. The transmission settings include the baseband

sampling frequency set to f s = 100512 MS / s Samples
sec
; the
Radio architecture Heterodyne pulse shaping and matched filtering operations use a
System setup 1 x 1 (extendible to 2 x 2 an- squared root raised cosine filter with a roll off factor
tennas) of 20% ; the symbol period is set to Ts = 20 Samples
symbol
. The
Carrier frequency 2.41 GHz transmitter and receiver antennas were at a distance of
approximately 2 m. For the evaluation, a binary signal
Signal bandwidth 25 MHz
s following a Bernoulli distribution with parameter
Transmitter power 50 mW p = 1/2 is modulated using BPSK. In order to get
Transmitter D/A Two 400 MS/s, 16 bit, the performance curve the system performs several
converter AD9777, that handle 160 transmissions using different transmit power values to
MSPS without interpolation, get different CSNR values. For each transmission, the
and up to 400 MSPS with 8x receiver uses the algorithms presented in section 3, to
interpolation attempt to reduce the effects on the transmitted signal
Receiver noise fig- 8 dB due to synchronization problems. Also, to control the
ure transmit power a new parameter k  R > 0 is introduced
Receiver dynamic 72.4 dB SNR and 85 dB in (2),
range SFDR (Spurious Free Dy- r (t ) = ks (t )  n(t ),
namic Range) for signals at (17)
the Nyquist frequency
Receiver A/D con- Two 100 MS/s, 14 bit, so that the value of k attenuates, in a controlled fashion,
the transmitted signal s (t ) and as a result it is possible
verter LTC2284
to get different values for CSNR. Thus, in general the
estimated CSNR in dB given by (15) is,
5 . EX PER IM E NTA L WIRELESS
 E s2 
MEASUREMENTS CSNR = 10log10  2
k2 . (18)
 E n2 

Wireless measurements were conducted in an office Then, for calibration, the system has to perform an
setting. The purpose of these measurements is to initial transmission of the signal s (t ) using a fixed
validate the testbed performance versus the expected value k = 1.0 in which performance is denoted as
theoretical performance computed by the channel CSNR =
E s  . Next, solving (18) for k ,
2
2

model. Thus, the transmission of a sequence of

init
 
E n2

symbols involves the creation of frames to wrap

the target message. All frames have a predefined
structure that is composed of a sequence of z zero k=
exp 
ln(10)CSNR
10 ,
 (19)
CSNRinit
amplitude samples used for the noise estimation, p
known BSPK modulated pilot symbols used for frame where the parameter CSNR , expressed in dB, is the
synchronization, and m encoded symbols or payload. target performance of the system. The results presented
in Fig. 9 display the theoretical AWGN curve, the
simulated AWGN curve and the measured curve for
BPSK. These results show that the testbed is capable
of recording the gains in performance within 1 -2dB
from the theoretical curve. However, when the CSNR
is greater than 10dB, the BER curve presents saturation
effects, possibly due to the limitation in bit resolution of
the ADC/DAC and the nonlinearities of the components
of the SDR hardware.
Figure 8. Photograph of the testbed.
 Dyna 180, 2013 75

possibly due to the nonlinearities of the hardware

components and the limitation in bit resolution of the
DAC/ADC.

ACKNOWLEDGMENTS

The authors gratefully acknowledge the contribution

of the Universidad Industrial de Santander and the
University of Delaware.

REFERENCES

[1] Lingaiah, D. Software Radio: RF Engineering’s New

Era, IEEE Software, vol. 20, No. 4, pp. 84–86, 95, 2003.
Figure 9. Experimental wireless measurements for BPSK
[2] Blahut, R. Computation of Channel Capacity and
using BER as the performance measure scheme versus
RateDistortion Functions, IEEE Transactions on Information
SNR.
Theory, vol. 18, No. 4, pp. 460 – 473, 1972.
CONCLUSIONS
[3] Mitola, J. The Software Radio Architecture, IEEE
Communications Magazine, vol. 33, No. 5, pp. 26 –38, 1995.
We have developed a flexible testbed system that takes
advantage of concepts presented in communications [4] Zheng, K., Huang, L., Li, G., Cao, H., Wang, W.,
theory as the underlying basis for its design and and Dohler, M. Beyond 3G Evolution, IEEE Vehicular
implementation. To this end, a distributed software Technology Magazine, vol. 3, No. 2, pp. 30 –36, 2008.
architecture to support user access to an SDR testbed is
presented. The architecture encapsulates the details and [5] Cassaro, T., and Georghiades, C. N. Frame Synchronization
complexities presented in a testbed implementation, for Coded Systems Over AWGN Channels, IEEE Transactions
so that the researchers could concentrate on the on Communications, vol. 52, No. 1, pp. 166, 2004.
validation of the theory without directly addressing the
[6] Fitz, M. P. Further results in the fast estimation of a single
challenges of a real communications system. Also, the
frequency, IEEE Transactions on Communications, vol. 42,
researchers can individually explore new algorithms for
No. 234, pp. 862-864, 1994.
source/channel encoding, modulation/demodulation,
synchronization or error correction without the need [7] Gansman, J. A., Fitz, M. P., and Krogmeier, J. V.
to develop a completely new testbed. Optimum and suboptimum frame synchronization for
pilot-symbol-assisted modulation, IEEE Transactions on
Also, some of the main issues that decrease the Communications, vol. 45, No. 10, pp. 1327–1337, 1997.
performance of the system have been described
along with algorithms to mitigate these impairments. [8] Garcia-Naya, J. A., Fresnedo, O., Vazquez-Araujo, F.
Although, the former algorithms solve to some degree J., Gonzalez-Lopez, M., Castedo, L., and Garcia-Frias, J.
the impairments, it is possible to introduce more Experimental Evaluation of Analog Joint Source-Channel
sophisticated, robust, faster state of the art algorithms Coding in Indoor Environments. In 2011 IEEE International
using the aforementioned architecture. Conference on Communications (ICC). Kyoto, Japan, pp.
1-5, June 2011.
The results obtained from indoor wireless measurements [9] Garcia-Naya, J. A., Perez-Iglesias, H. J., Fernandez-
showed that the resulting system can perform Carames, T. M., Gonzalez-Lopez, M., and Castedo, L. A
within 1-2dB from the theoretical results for CSNR distributed multilayer architecture enabling end-user access
values below 10dB, thus clearly demonstrating the to mimo testbeds. IEEE 19th International Symposium
effectiveness of the proposed system. For CSNR on Personal, Indoor and Mobile Radio Communications
values above 10dB there is a performance degradation, (PIMRC). Cannes, France, pp. 1-5, September 2008.
 76 Pino & Arguello

[10] Hekland, F. On the design and analysis of Shannon- [17] Reed, J., Software radio: a modern approach to radio
Kotel’nikov mappings for joint source-channel coding [PhD engineering, Prentice Hall Press, Upper Saddle River, 2002.
Thesis]. Norwegian University of Science and Technology,
2007. [18] Schmid, T., Sekkat, O., and Srivastava, M. B. An
experimental study of network performance impact
[11] Kay, S. A fast and accurate single frequency estimator, of increased latency in software defined radios”. In
IEEE Transactions on Acoustics, Speech and Signal Proceedings of the second ACM international workshop
Processing, vol. 37, No. 12, pp. 1987-1990, 1989. on Wireless network testbeds, experimental evaluation and
characterization (WinTECH). Montréal, Canada, pp. 59,
[12] Lang, S., Rao, M. R., and Daneshrad, B. Design and September 2007.
development of a 5.25 GHz software defined wireless OFDM
communication platform, IEEE Communications Magazine, [19] Johnson, C. R., and Sethares, W.A., “Telecommunication
vol. 42, No. 6, pp. S6-12, 2004. breakdown: concepts of communication transmitted via
software-defined radio”, Prentice Hall, Upper Saddle River,
[13] Luise, M., and Reggiannini, R. Carrier frequency 2004.
recovery in all-digital modems for burst-mode transmissions,
IEEE Transactions on Communications, vol. 43, No. 234, [20] Linn, Y. Synchronization, phase detection, lock
pp. 1169-1178, 1995. detection, and SNR estimation in coherent M-PSK receivers
[Ph.D Thesis]. Canada: University of British Columbia,
[14] Massey, J. Optimum Frame Synchronization, IEEE 2007.
Transactions on Communications, vol. 20, No. 2, pp. 115-
119, 1972. [21] Proakis, J. G., Digital Communications, McGraw-Hill,
New York, 2000.
[15] Ramstad, T. A. Shannon mappings for robust
communication, Telektronikk, vol. 98, No. 1, pp. 114-128, [22]Rueda H. F., Correa C. V., and Arguello H., “Design and
2002. development of a speech synthesis software for colombian
spanish applied to communication through mobile devices”.
[16] Rao, R. M., Wiejun, Z., Lang, S., Oberli, C., Browne, D., Journal Dyna, No. 173-II, pp. 71-80, 2012.
Bhatia, J., Frigon, J. F., Wang, J., Gupta, P., Heechoon, L., Liu,
D. N., Wong, S. G., Fitz, M., Daneshrad, B., and Takeshita, O. [23] GNU RADIO. Open-source software development
Multi-antenna testbeds for research and education in wireless toolkit that provides signal processing blocks to implement
communications, IEEE Communications Magazine, vol. 42, software radios. Available: http://gnuradio.org, [cited June
No. 12, pp. 72-81, 2004. 2012].