<?xml version="1.0" encoding="ISO-8859-1"?><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id>2215-3470</journal-id>
<journal-title><![CDATA[Uniciencia]]></journal-title>
<abbrev-journal-title><![CDATA[Uniciencia]]></abbrev-journal-title>
<issn>2215-3470</issn>
<publisher>
<publisher-name><![CDATA[Universidad Nacional, Costa Rica]]></publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id>S2215-34702022000100612</article-id>
<article-id pub-id-type="doi">10.15359/ru.36-1.39</article-id>
<title-group>
<article-title xml:lang="en"><![CDATA[Measuring traffic dynamics at the Edge]]></article-title>
<article-title xml:lang="es"><![CDATA[Midiendo la dinámica de tráfico vehicular en el Edge]]></article-title>
<article-title xml:lang="pt"><![CDATA[Medindo a dinâmica do tráfego de veículos na Edge]]></article-title>
</title-group>
<contrib-group>
<contrib contrib-type="author">
<name>
<surname><![CDATA[León-Vega]]></surname>
<given-names><![CDATA[Luis Gerardo]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
<contrib contrib-type="author">
<name>
<surname><![CDATA[Castro-Godínez]]></surname>
<given-names><![CDATA[Jorge]]></given-names>
</name>
<xref ref-type="aff" rid="Aff"/>
</contrib>
</contrib-group>
<aff id="Af1">
<institution><![CDATA[,Instituto Tecnológico de Costa Rica Electronics Engineering School ]]></institution>
<addr-line><![CDATA[Cartago ]]></addr-line>
<country>Costa Rica</country>
</aff>
<aff id="Af2">
<institution><![CDATA[,Instituto Tecnológico de Costa Rica Electronics Engineering School ]]></institution>
<addr-line><![CDATA[Cartago ]]></addr-line>
<country>Costa Rica</country>
</aff>
<pub-date pub-type="pub">
<day>00</day>
<month>12</month>
<year>2022</year>
</pub-date>
<pub-date pub-type="epub">
<day>00</day>
<month>12</month>
<year>2022</year>
</pub-date>
<volume>36</volume>
<numero>1</numero>
<fpage>612</fpage>
<lpage>620</lpage>
<copyright-statement/>
<copyright-year/>
<self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_arttext&amp;pid=S2215-34702022000100612&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_abstract&amp;pid=S2215-34702022000100612&amp;lng=en&amp;nrm=iso"></self-uri><self-uri xlink:href="http://www.scielo.sa.cr/scielo.php?script=sci_pdf&amp;pid=S2215-34702022000100612&amp;lng=en&amp;nrm=iso"></self-uri><abstract abstract-type="short" xml:lang="en"><p><![CDATA[Abstract This work aims to measure the impact of approximate computing on a case study of traffic dynamics metering on a System-on-Chip edge computing device. Firstly, the study proposes a baseline implementation of the metering system in C++. To analyze the application in detail, study profiled the baseline using a built-in instrumented profiler, presenting the overall performance of each of its parts. During the hotspot analysis, some parts had optimization opportunities exploitable by multi-threading and approximate computing techniques, particularly frame skipping, which is inspired by the loop perforation approximate technique. The first optimization employed was multi-threading, which led to a 1.32x speedup on the application without introducing errors in the metrics. Then, the meter was optimized by using frame skipping. This work demonstrated that adaptatively modifying the number of frames skipped improved the error in the final metrics compared to keeping it fixed. In terms of performance, the frame skipping brought the overall speed up to 1.76x. Approximate computing, in particular, frame skipping, managed to contribute up to 25% of the overall speedup, managing to accelerate the meter from 8.7 frames per second to 15 fps in the most critical case in exchange for some numerical error on the final metrics.]]></p></abstract>
<abstract abstract-type="short" xml:lang="es"><p><![CDATA[Resumen El objetivo de este trabajo es medir el impacto de la aplicación de técnicas de computación aproximada sobre un caso de estudio de implementación de un medidor de la dinámica de tráfico vehicular en una unidad computacional basada en un sistema embebido. El estudio parte de una implementación inicial del medidor hecha en C++. Para el análisis de la aplicación, la implementación inicial se perfila con un perfilador empotrado en el mismo código, que muestra información detallada de cada una de las partes. Durante el análisis de consumo, se encontraron partes optimizables con paralelismo a nivel de hilos y técnicas de computación aproximada como omisión de recuadros, que es una técnica inspirada en la perforación de lazos. La primer optimización realizada fue la implementación multihilo, que logró acelerar la aplicación 1.32 veces sin introducir errores en los resultados. Posteriormente, la implementación fue optimizada con la omisión de recuadros. Durante el desarrollo de este trabajo, se demuestra que modificar el número de recuadros omitidos de forma dinámica en tiempo de ejecución mejora considerablemente el error introducido comparado a mantener constante el número de recuadros omitidos. La combinación de ambas optimizaciones concluyó en una implementación 1.76 veces más rápida, donde la aplicación de computación aproximada mediante omisión de recuadros contribuyó hasta en un 25% sobre el total de la mejora, acelerando el medidor de 8.7 recuadros por segundo a 15 en el escenario más crítico a cambio de la introducción de errores numéricos.]]></p></abstract>
<abstract abstract-type="short" xml:lang="pt"><p><![CDATA[Resumo O objetivo deste trabalho é medir o impacto da aplicação de técnicas de computação aproximada em um estudo de caso da implementação de um medidor de dinâmica de tráfego de veículos em uma unidade computacional baseada em um sistema embarcado. O estudo começa a partir de uma implementação inicial do medidor feita em C++. Para a análise da aplicação, a implementação inicial é perfilada com um perfilador incorporado no próprio código, que exibe informações detalhadas sobre cada uma das partes. Durante a análise de consumo, foram encontradas peças otimizáveis com paralelismo em nível de thread e técnicas de computação aproximada, como o salto de quadros, que é uma técnica inspirada na perfuração de laço. A primeira otimização realizada foi a implementação multithread, que conseguiu acelerar a aplicação em 1,32 vezes sem introduzir erros nos resultados. Posteriormente, a implementação foi otimizada com o salto de quadros. Durante o desenvolvimento deste trabalho, mostra-se que modificar a dinâmica do número de salto de quadros em tempo de execução melhora consideravelmente o erro introduzido em comparação com a manutenção constante do número de salto de quadros. A combinação de ambas as otimizações resultou em uma implementação 1,76 vezes mais rápida, onde a aplicação da computação aproximada por meio do salto de quadros contribuiu com até 25% da melhoria total, acelerando o medidor de 8,7 quadros por segundo para 15 no cenário mais crítico, em troca da introdução de erros numéricos.]]></p></abstract>
<kwd-group>
<kwd lng="en"><![CDATA[traffic dynamics meter]]></kwd>
<kwd lng="en"><![CDATA[edge computing]]></kwd>
<kwd lng="en"><![CDATA[approximate computing]]></kwd>
<kwd lng="es"><![CDATA[Medición de dinámica vehicular]]></kwd>
<kwd lng="es"><![CDATA[computación en el Edge]]></kwd>
<kwd lng="es"><![CDATA[computación aproximada.]]></kwd>
<kwd lng="pt"><![CDATA[Medição da dinâmica de veículos]]></kwd>
<kwd lng="pt"><![CDATA[computação de borda]]></kwd>
<kwd lng="pt"><![CDATA[computação aproximada]]></kwd>
</kwd-group>
</article-meta>
</front><back>
<ref-list>
<ref id="B1">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Bolme]]></surname>
<given-names><![CDATA[D. S.]]></given-names>
</name>
<name>
<surname><![CDATA[Beveridge]]></surname>
<given-names><![CDATA[J. R.]]></given-names>
</name>
<name>
<surname><![CDATA[Draper]]></surname>
<given-names><![CDATA[B. A.]]></given-names>
</name>
<name>
<surname><![CDATA[Lui]]></surname>
<given-names><![CDATA[Y. M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Visual object tracking using adaptive correlation filters.]]></article-title>
<source><![CDATA[Computer Society Conference on Computer Vision and Pattern Recognition,]]></source>
<year>2010</year>
<page-range>2544-50</page-range><publisher-name><![CDATA[IEEE]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B2">
<nlm-citation citation-type="">
<collab>California Department of Transportation (Caltrans)</collab>
<source><![CDATA[https://cruz511.org/drive/traffic-conditions/traffic-cameras/]]></source>
<year></year>
<publisher-loc><![CDATA[Soquel Ave. Santa Cruz, California, United States of America ]]></publisher-loc>
</nlm-citation>
</ref>
<ref id="B3">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[D'Andrea]]></surname>
<given-names><![CDATA[E.]]></given-names>
</name>
<name>
<surname><![CDATA[Ducange]]></surname>
<given-names><![CDATA[P.]]></given-names>
</name>
<name>
<surname><![CDATA[Lazzerini]]></surname>
<given-names><![CDATA[B.]]></given-names>
</name>
<name>
<surname><![CDATA[Marcelloni]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Real-Time Detection of Traffic From Twitter Stream Analysis.]]></article-title>
<source><![CDATA[IEEE Transactions on Intelligent Transportation Systems]]></source>
<year>2015</year>
<volume>16</volume>
<page-range>2269-83</page-range><publisher-name><![CDATA[IEEE]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B4">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Giridharan]]></surname>
<given-names><![CDATA[E. N.]]></given-names>
</name>
<name>
<surname><![CDATA[Kadaieaswaran]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
<name>
<surname><![CDATA[Arunprasath]]></surname>
<given-names><![CDATA[V.]]></given-names>
</name>
<name>
<surname><![CDATA[Karthika]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Big Data Solution for Improving Traffic Management System with Video Processing.]]></article-title>
<source><![CDATA[International Journal of Engineering Science and Computing]]></source>
<year>2017</year>
<volume>7</volume>
<numero>2</numero>
<issue>2</issue>
<page-range>4606-9</page-range></nlm-citation>
</ref>
<ref id="B5">
<nlm-citation citation-type="">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Hall]]></surname>
<given-names><![CDATA[F. D.]]></given-names>
</name>
</person-group>
<source><![CDATA[Traffic Stream Characteristics. Federal Highway Administration.]]></source>
<year>1992</year>
</nlm-citation>
</ref>
<ref id="B6">
<nlm-citation citation-type="">
<collab>Ministerio de Obras Públicas y Transportes</collab>
<source><![CDATA[Cámaras viales CR]]></source>
<year>n. d</year>
<month>.</month>
</nlm-citation>
</ref>
<ref id="B7">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Samie]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
<name>
<surname><![CDATA[Bauer]]></surname>
<given-names><![CDATA[L.]]></given-names>
</name>
<name>
<surname><![CDATA[Henkel]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[From Cloud Down to Things: An Overview of Machine Learning in Internet of Things.]]></article-title>
<source><![CDATA[IEEE Internet of Things Journal]]></source>
<year>2019</year>
<volume>6</volume>
<numero>3</numero>
<issue>3</issue>
<page-range>4921-34</page-range></nlm-citation>
</ref>
<ref id="B8">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Sidiroglou-Douskos]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
<name>
<surname><![CDATA[Misailovic]]></surname>
<given-names><![CDATA[S.]]></given-names>
</name>
<name>
<surname><![CDATA[Hoffmann]]></surname>
<given-names><![CDATA[H.]]></given-names>
</name>
<name>
<surname><![CDATA[Rinard]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Managing Performance vs. Accuracy Trade-Offs with Loop Perforation.]]></article-title>
<source><![CDATA[19th ACM SIGSOFT Symposium and the 13th European Conference on Foundations of Software Engineering]]></source>
<year>2011</year>
<page-range>124-34</page-range></nlm-citation>
</ref>
<ref id="B9">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Viola]]></surname>
<given-names><![CDATA[P.]]></given-names>
</name>
<name>
<surname><![CDATA[Jones]]></surname>
<given-names><![CDATA[M.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Rapid o ject detection using a boosted cascade of simple features.]]></article-title>
<source><![CDATA[IEEE Computer Society Conference on Computer Vision and Pattern Recognition]]></source>
<year>2001</year>
<page-range>511-8</page-range><publisher-name><![CDATA[IEEE]]></publisher-name>
</nlm-citation>
</ref>
<ref id="B10">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Yu]]></surname>
<given-names><![CDATA[W.]]></given-names>
</name>
<name>
<surname><![CDATA[Liang]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
<name>
<surname><![CDATA[He]]></surname>
<given-names><![CDATA[X.]]></given-names>
</name>
<name>
<surname><![CDATA[Hatcher]]></surname>
<given-names><![CDATA[W. G.]]></given-names>
</name>
<name>
<surname><![CDATA[Lu]]></surname>
<given-names><![CDATA[C.]]></given-names>
</name>
<name>
<surname><![CDATA[Lin]]></surname>
<given-names><![CDATA[J.]]></given-names>
</name>
<name>
<surname><![CDATA[Yang]]></surname>
<given-names><![CDATA[X.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[A Survey on the Edge Computing for the Internet of Things.]]></article-title>
<source><![CDATA[IEEE Access]]></source>
<year>2017</year>
<volume>6</volume>
<page-range>6900-19</page-range></nlm-citation>
</ref>
<ref id="B11">
<nlm-citation citation-type="journal">
<person-group person-group-type="author">
<name>
<surname><![CDATA[Zhu]]></surname>
<given-names><![CDATA[L.]]></given-names>
</name>
<name>
<surname><![CDATA[Yu]]></surname>
<given-names><![CDATA[F.]]></given-names>
</name>
<name>
<surname><![CDATA[Wang]]></surname>
<given-names><![CDATA[Y.]]></given-names>
</name>
<name>
<surname><![CDATA[Ning]]></surname>
<given-names><![CDATA[B.]]></given-names>
</name>
<name>
<surname><![CDATA[Tang]]></surname>
<given-names><![CDATA[T.]]></given-names>
</name>
</person-group>
<article-title xml:lang=""><![CDATA[Big Data Analytics in Intelligent Transportation Systems: A Survey.]]></article-title>
<source><![CDATA[IEEE Transactions on Intelligent Transportation Systems]]></source>
<year>2019</year>
<page-range>383-98</page-range></nlm-citation>
</ref>
</ref-list>
</back>
</article>
