Content uploaded by Hermes Jose Loschi
Author content
All content in this area was uploaded by Hermes Jose Loschi on Apr 18, 2018
Content may be subject to copyright.
Computational Simulation Performance based in
Hybrid Modelling with Discrete Events for
Telecommunication Systems
Reinaldo Padilha1, Yuzo Iano1, Edson Moschim1, Ana Carolina Borges Monteiro1 and Hermes José Loschi1
1- University of Campinas – UNICAMP, Brazil
Abstract — With the objective of improving the transmission of
content in telecommunication systems, in simulation environment
is proposed a pre-coding process of bits based in the application
of discrete events in the signal before of the modulation process.
The proposal brings a different approach of usual technical, in
which the signal transmission on the channel is realized in the
discrete domain with the implementation of discrete entities in
the process of bit generation. In general, the simulation tools for
a model of telecommunication transmission system are based on
continuous and discrete signals. The present work implements a
model based on discrete events applied at a low level of
abstraction in a telecommunication system named hybrid
method, being used the Simulink simulation environment of the
MATLAB® software. In the simulation are considered advanced
modulation format for signal transmission in an AWGN channel.
The results show improvement of 9 to 22% in memory
utilization, as also best computational performance.
Keywords — Methodologies, Discrete Events, Simulation.
I. INTRODUCTION
Computer simulations are strong tools that support the best
knowledge of how a telecommunication system is operating. In
the simulation environment, the designer has the flexibility to
implement different types of system architecture to analyze
different layers, such as physical, transport, transmission and
higher layer, improving and validating the system for different
applications [1] [2] [3] [4] [5] [6].
In simulation, the designer can analyze several system
parameters to validate its operation, such as i.e., the bit error
rate (BER) that is an important parameter to evaluate the
quality of the signal transmission evaluations.
The great majority of processes observed in the world
consist of continuous changes. Where continuous simulation is
suitable for systems in which the variables can change
continuously, being non-linear, with differential equations of
the first order or integrals. Such systems emphasize a
continuous view, that is, looking beyond individual events, but
rather the structure as a whole [1] [2]. Such a concept can be
well understood, in the simulation of a telecommunications
system.
The term Discrete Event is however mainly used in the
sense to denote the modeling that suggests representing the
system being analyzed as a sequence of operations being
performed on entities (transactions) of certain types such as
data packets, bits, etc. The entities are discrete items of interest
in a discrete event simulation, the meaning of an entity depends
on what is being modeled and the type of system, and can have
attributes that affect the way they are handled or may change as
the entity flows through the process [1] [2].
This technique is usually used to model concepts with a
high level of abstraction, such as patients in a hospital, clients
in a queue, emails on a server, flow of vehicles, manufacturing
enterprise, transmission of data packets in telecommunications
systems, among others [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
[12].
In this work, a hybrid model for telecommunication
systems was made and implemented using an AWGN channel
and advanced modulation format DBPSK in simulation
environment, with the objective of to increase the transmission
capacity of information content through of the channel.
Where a bit treatment with discrete events methodology
was modeled in the step of bit generation, being the differential
of this work the use of discrete events applied in a low level of
abstraction. The results show better computational performance
related to memory utilization of the simulation model.
The present work is organized as follows: Section 2
discusses traditional simulation models, showing the modeling
of transmission channel AWGN. Section 3 presents the
proposal of this paper, based on the hybrid model with discrete
event methodology. Section 4 presents the results and, finally,
in Section 5, the conclusions are presented as also the potential
of the search.
II. TELECOMMUNICATION SYSTEM
The communication channel is the medium that provides
the physical connection between transmitters and receivers in a
communication system, be it as a wire, or to a logical
connection over a multiplexed medium such as a radio channel
in telecommunications and computer networking. Carrying
data, using normally two types of media as cable (twisted-pair
wire, cable, and fiber-optic cable) and broadcast (microwave,
satellite, radio, and infrared). And so, for the analysis of
communication systems, it is important to construct
mathematical models that describe the main characteristics of
these means and the changes that the signal undergoes when
transmitted [13] [14] [15] [16] [17] [18].
A model widely used due to its simplicity and mathematical
treatment, and what applies to a large set of physical channels,
is the Additive White Gaussian Noise channel model, AWGN,
which introduces in the transmitted signals a noise modeled,
statistically as a white gaussian additive process [13] [14] [15].
The term differential in modulation format, is related in the
codification of the data, being by the presence of a binary one
or zero, by similarity or difference of the symbols in relation to
the previous signal.
The DBPSK modulation (Differential Binary Phase Shift
Keying) eliminates phase ambiguity and the need for phase
acquisition and tracking, resulting in advantages in the reduced
cost of energy. In this modulation, is employed a non-coherent
way to solve the need for a coherent reference signal at the
receiver. Thus, the input binary sequence is first differentially
encoded and then modulated using a BPSK modulator. And in
demodulation, there is no need to know about the initial state of
the bit, simplifying synchronization [13] [14] [15] [18].
In BPSK (Binary Phase-Shift Keying), one phase
represents the binary 1 and the other phase represents the
binary 0, and as the digital input signal changes state, the phase
of the output signal changes between two angles separated by
180° [13] [14] [15] [18].
The models presented in this section aim to display a
telecommunication system with AWGN channel and DBPSK
modulation. For this, was used the Simulink simulation
environment of the MATLAB® software in its version 8.3 of
64 bits (2014a).
In the model, Figure 01, the signals corresponding as the
bits 0 and 1 are generated, and then modulated in DBPSK,
following for a AWGN channel according to the parameters
specified as sample time of 1 second, power input signal of 1
watt, initial seed in the generator of 37 and in the channel of
67, Eb/No of 0 to 14dB. Then the signal is demodulated in
order to perform the bit error rate (BER) of the channel. The
values obtained referring the BER are sent to the MATLAB®
workspace, for further processing and generating of the signal
BER graph.
Figure 01 – Traditional Model
III. PROPOSAL
The modeling according to proposal implemented with
discrete events is similar to that shown previously,
differentiating that in this model, was added the discrete events
process of pre-coding, consisting of the treatment performed on
the signal corresponding to bit 0, being converted into discrete
entities, and forwarded for a FIFO queue with infinite capacity,
without limit of capacity and retention in the block, storing
entities in the First-In-First-Out sequence, ordering the bits
following really your order of arrival, and thus driving to a
server, which have configuration of service time equal to the
simulation time.
Where the differential of this work is in the use of discrete
events applied in such low level of abstraction, being the bit
generation. After the signal passes through the server, is
converted back to its original format respecting the original
format and data type specified and maintaining the sampling
period respectively. Thus, the signal is modulated in DBPSK
and inserted into the AWGN channel, and then demodulated
for the purposes of calculating the BER of the signal. These
relative values to this BER are also sent to the MATLAB®
workspace, for further processing and generating of the signal
BER graph.
The model presented in Figure 02, incorporates the
traditional modeling with a proposal presented, as well as
highlights the part modeled according to the approach of
discrete events, in blue, as previously described.
Figure 02 – Hybrid Model
And in Figure 03, using 10000 seconds of simulation time
was placed the flows of transmission of the DBPSK signal in
relation to the hybrid model (below) and traditional model
(top) for better viewing and comparison, noting that both
methodologies generated the same result.
Figure 03 – Transmission Flow DBPSK
Was used the Constellation Diagram, to view the
constellation of a modulated digital signal and useful for
comparing the performance of one system with another.
In Figure 04 is shown the results for visualization of the
constellations in 5, 10 e 15 dB, according to the hybrid model
(below) and traditional model (top).
Figure 04 – Simulated DBPSK Constellation
IV. RESULTS
In this section, the results will be presented on the
evaluations of the calculation of the time spent of the
simulations. To obtain the same were performed 5 sequential
simulations with the models presented previously, on physical
machines with different hardware configuration, consisting of
an Intel Core i5 processor and 8GB RAM, and another with an
Intel Core i3 processor and 4GB RAM.
Was used the sldiagnostics function, that displays
diagnostic information about the modeling system in Simulink,
calculating the sum all of the memory consumption processes
used in the model in simulation, by the ProcessMemUsage
parameter, which counts the amount of memory utilized in
each phase of the model, during the entire simulation,
displaying the total amount in MB, according presented in the
Figures 05 and 06.
Figure 05 – time simulation
Figure 06 – Memory Consumption Simulation
Also was analyzed the first simulation of both models,
because it is in the first that the variables are allocated and the
memory reserved for the execution of the model, having a
better performance as shown in Table 1 and related with the
Figures 07 and 08.
Memory Consumption
Machines
i3
i5
Modelo DBPSK
22,01%
9,22%
Table1: Computational Improvement
Figure 07 – first time simulation
Figure 08 – First Memory Consumption Simulation
To analyze the relationship between the simulation
methodology and the impact on the physical layer of the
channel, scripts were made in the MATLAB® for processing of
the graph relative to BER, that allows analyzing the
performance of bit error rate (BER).
In the Figure 09, is displayed the performance of the
models according to simulation methodologies under study,
along with a transmission with noise ranging from 0 to 12 dB.
Figure 09 – Performance BER
V. CONCLUSIONS
In all scenarios analyzed, the simulation model of the
system with discrete event methodology on both different
hardware configurations, evaluated on memory consumption,
obtained better results, when compared with the model with
the traditional methodology, either in its first simulation or
along the sequence of 5 simulations.
Thus, the use of discrete events applied in a low level of
abstraction such as bit generation in telecommunication
system model performed a treatment of the bits prior to the
modulation process, functioning as a pre-coding process
differentiated.
The extension of the results of this work, being the
compression of the information, has a strong impact on the
methods performed in higher layers, like MPEG-4 in a
broadcasting system for example, as well as others, being able
to improve them even more, since this proposal acts on the bits.
The purpose of this research, was the development of
simulation models of telecommunication systems, taking a
different approach from what is normally done and applying a
concept of a methodology at a lower abstraction level than it is
normally used, in which in the transmission in the channel
were created discrete entities in the process of creation of the
bit, as also following the orientation of the modeling of each
technique studied, as also contribute to the study area.
REFERENCES
[1] Digital Modulation in Communications Systems. An Introduction,
Agilent Technologies.
[2] Padilha, R.; Martins, B. I.; Moschim, E.Discrete Event Simulation and
Dynamical Systems: A study of art. BTSym'16, Campinas, SP – Brasil,
December, 2016.
[3] Pereira, F. T.; Takano, A. M.; Leal, F.; Pinho, F. A.Aplicação Da
Simulação A Eventos Discretos Em Um Ambiente Hospitalar Visando A
Melhoria No Processo De Atendimento. XLVSBPO, Natal, RN –
Brasil,2013.
[4] Sharda, B.; Bury, J. S.A Discrete Event Simulation Model For
Reliability Modeling Of A Chemical Plant.Winter Simulation
Conference,2008.
[5] Hu, W.; Sarjoughian, H.S.Discrete-event simulation of network systems
using distributed object computing. SPECTS'05,2005
[6] Sasaki, N. K.; Moschim, E. Simulação de Sistemas de Comunicação
Óptica Baseada em Simulação a Eventos Discretos.Universidade
Estadual de Campinas. Campinas, SP – Brasil. July 2007.
[7] Pissinelli, J. G.; Risso, L. A.; Picanco, S. R. A.;Ignacio, A. S. P.; Silva,
L. A. Modelo De Simulação De Eventos Discretos Para Análise De
Fluxo De Veículos. ENEGEP, Fortaleza, CE – Brasil,2015.
[8] Rangel, J.J.A.; Costa, J.V.S.; Laurindo, Q.M.G.; Peixoto, T.A.; Matias,
I.O.Análise do fluxo de operações em um servidor de e-mail através de
simulação a eventos discretos com o software livre Ururau. Produto &
Produção, vol. 17, n. 1, p. 1-12, mar. 2016.
[9] Gomes, E.N.; Fernandes, M.S.R.; Campos, C.A.V.; Viana, A.C. Um
Mecanismo de Remoção de Mensagens Obsoletas para as Redes
Tolerantes a Atrasos e Interrupções. CSBC, 2012.
[10] Godoy, E.P.; Lopes, W.C.;Sousa, R.V.; Porto, A.J.V. Modelagem E
Simulação De Redes De Comunicação Baseadas No Protocolo Can -
Controller Area Network. Revista SBA: Controle & Automação, Vol.21
no.4, 2010
[11] Helal, M.A Hybrid System Dynamics-Discrete Event Simulation
Approach to Simulating the Manufacturing Enterprise.PhD Thesis,
Department of Industrial Engineering and Management Systems,
College of Engineering and Computer Science, University of Central
Florida,2008
[12] Forrester, J.W., 1968. Industrial Dynamics - After the First Decade.
Management Science, 14(7), pp. 398-415,1968
[13] Freeman, R.L. Fundamentals of Telecommunications, John Wiley &
Sons, 1999.
[14] Freeman, R.L. Telecommunication System Engineering, 4th Edition,
John Wiley & Sons, 2004.
[15] John G. Proakis (2008). Digital Communications, 5rd edition, McGraw-
Hill.
[16] Tozer, E. P., Broadcast Engineer's Reference Book, 1th Edition, FOCAL
PRESS, 2012
[17] Whitaker C. J., Standard Handbook of Broadcast Engineering, 1th
Edition, McGraw-Hill, 2005
[18] L.W. Couch II. Digital and Analog Communication Systems, 8th
Edition, Prentice Hall, 2013
