Gunawan Ibnu, Noertjahyana Agustinus. Desain

COVER The 1st Conference on Information Technology, Computer, and Electrical Engineering (CITACEE 2013)

INNER COVER The 1st Conference on Information Technology, Computer, and Electrical Engineering (CITACEE 2013)

PREFACE Dear Collegues, On behalf of Technical Committee and Organizing Committee of CITACEE 2013, I am honored to welcome you to The 1st Conference on Information Technology, Computer, and Electrical Engineering (CITACEE). The conference is planned to be conducted annually. This conference program is organized by Computer Engineering Department, Faculty of Engineering, Diponegoro University, Semarang. The main theme of the conference is “Information Technology and its Application towards the Implementation of Green Technology”. The conference aims to provide a forum for researchers, academicians, professionals, and industries to expose and exchange innovative ideas, methods, and experience in information technology, computer engineering, as well as in electrical engineering, and their applications, related with the aspects of green technology. Papers in this conference are presented in both in international or Bahasa Indonesia session. This conference also provides forum for researchers, scientists, and engineers to exchange ideas and their current achievements. In this year we have received 98 paper submissions from various universities, research centers, and as well as from industries. However, after indepth review, the Technical Committee accepted 67 selected papers will be presented in this conference. From this 67 papers, 17 papers will be in International Session (conducted in English). While, 50 papers will be presented in the National Session (conducted in Bahasa Indonesia). The accepted papers are categorized into five groups, there are: Information Technology and System, Signal and Circuit, Power and Control Engineering, General Papers, and Interdisciplinary Papers related to Green Technology. The Proceedings of this conference is highly expected to be used as reference for Academics, Practitioners, as well as for Researchers. We thank all authors and all parties which cannot be mentioned here who have contributed and participated in presenting their works at this conference. We also gratefully acknowledge the important review supports provided by the 10 members of Conference Committee from Indonesia or abroad. Their efforts were crucial to the success of the conference. We are also so blessed by the presence of 3 (three) invited Keynote Speakers from different institutions which will address the important trends relating to information technology and its application towards the implementation of Green Technology. Finally, we wish you all can enjoy one day discussion through this conference and could spend to enjoy the beauty of Semarang City and Undip Campus. We hope to meet you again in the next conference, the 2nd Conference on Information Technology, Computer, and Electrical Engineering (CITACEE 2014).

General Chair of CITACEE 2013 Dr. R. Rizal Isnanto, S.T., M.M., M.T.

SAMBUTAN Dekan Fakultas Teknik Universitas Diponegoro Bapak Rektor Universitas Diponegoro, Ketua APTIKOM, para pembicara kunci, para pemakalah, para peserta conference, para mitra pendukung, para tamu undangan, Ketua Program Studi Sistem Komputer dan panitia conference yang saya hormati. Assalamu’alaikum wa rahmatullahi wa barakatuh Rasa syukur yang paling dalam kita panjatkan kehadirat Allah subhanahu wa ta’ala atas rahmat, taufiq, hidayah, dan inayah-Nya sehingga kita dapat hadir dalam conference di bidang teknologi informasi, komputer, dan Teknik Elektro sebagai ajang silaturahmi ilmiah dalam keadaan sehat wal afiat, semangat dan bahagia. Kegiatan ini merupakan upaya Fakultas Teknik Program Studi Sistem Komputer Universitas Diponegoro untuk melaksanakan Misi UNDIP dalam rangka mewujudkan Visi UNDIP menjadi universitas riset yang unggul di tahun 2020. Kita sadar bahwa peran teknologi sebagai upaya budi daya manusia dalam mewujudkan kesejahteraannya sangatlah penting. Bahkan dalam kenyataannya dengan perkembangan teknologi yang sangat cepat perkembangan dan perubahan diberbagai aspek kehidupan menunjukkan percepatan yang sangat tinggi. Lebih-lebih dengan perkembangan teknologi di bidang komputer dan teknologi informasi lebih besar dan lebih cepat lagi pengaruhnya bahkan terhadap perubahan sosial dan budaya. Netralitas teknologi ibarat sebilah pisau bermata dua, kalau tidak kita gunakan dengan benar dan bijaksana teknologi bisa menjadi bumerang bagi kehidupan umat manusia sendiri. Namun, betapa pun teknologi tidak dapat kita bendung perkembangannya dan tetap sangat kita butuhkan. Sejalan dengan Visi Misi UNDIP, ajang silaturahmi ilmiah seperti ini menjadi penting artinya untuk senantiasa kita lakukan untuk mensinergikan potensi kita untuk mengembangkannya dan mensikapinya secara arif dalam menerapkannya agar harmonisasi terhadap perubahan sosial, budaya, dan bersahabat dengan lingkungan dapat kita wujudkan. Pada kesempatan yang sangat baik ini perkenankanlah kami mengucapkan selamat datang di kampus utama UNDIP ini untuk berdiskusi berbagi pemikiran, pengetahuan, dan pengalaman, yang memberi manfaat tidak saja bagi peserta conference yang hadir pada saat ini, tetapi lebih luas bagi dunia industri dan masyarakat pada umumnya.

The 1st Conference on Information Technology, Computer, and Electrical Engineering (CITACEE 2013) adalah conference di bidang Teknologi Informasi, Komputer, dan Teknik Elektro perdana yang diselenggarakan oleh Program Studi Sistem Komputer Fakultas Teknik UNDIP dengan tema “Information Technology and its Application towards the Implementation of Green Technology”. Conference ini diselenggarakan bertepatan dengan Dies Natalis Fakultas Teknik UNDIP yang ke 55 dan Dies Natalis UNDIP yang ke 56. Conference ini diselenggarakan agar dapat dijadikan sebagai ajang silaturahmi ilmiah bagi para peneliti, akademisi, mahasiswa, profesional, dan dunia industri untuk berdiskusi dan secara lebih luas berbagi pemikiran, ide-ide inovatif, metode-metode, dan pengalaman di bidang Teknologi Informasi, Komputer, dan Teknik Elektro serta aplikasinya terkait dengan penerapan teknologi hijau (green technology). Dengan demikian, temuan-temuan dan simpulan-simpulan baru dari conference ini diharapkan akan diperoleh. Pada gilirannya dapat untuk mengembangkan teknologi dan aplikasinya secara berkelanjutan di masa yang akan datang. Alhamdulillah, conference ini telah mampu menghimpun makalah dalam berbagai topik: teknologi informasi, sistem informasi, sistem ketenagaan, sistem, sinyal dan rangkaian, rekayasa biomedik, sistem komunikasi, dan berbagai bidang yang terkait. Makalah juga mencakup berbagai aspek dari wilayah penelitian yang fundamental, eksperimental, sampai dengan inovasi-inovasi penerapannya. Selamat ber-conference, semoga sukses memberi manfaat bagi kita semua dalam keberkahan. Wassalamu’alaikum wa rahmatullahi wa barakatuh

Dekan Fakultas Teknik Universitas Diponegoro

Ir. Bambang Pudjianto, M.T.

SAMBUTAN Ketua Program Studi Sistem Komputer Universitas Diponegoro Assalamu ‘alaikum Wr. Wb. Segala puji syukur kita panjatkan kehadirat Allah subhanahu wa ta’ala, Tuhan yang Mahakuasa, atas berkat rahmat dan hidayah-Nya, Konferensi tentang Teknologi Informasi, Teknik Komputer dan Teknik Elektro (CITACEE) 2013 yang diselenggarakan oleh Program Studi Sistem Komputer, Fakultas Teknik, Universitas Diponegoro Semarang berhasil diselenggarakan untuk pertama kalinya. Seminar/konferensi ini mengambil tema Information Technology and its Applications towards the Implementation of Green Technology, yang diharapkan menjadi wadah penyebaran keilmuan bagi para akademisi, peneliti, praktisi, serta para pengguna teknologi informasi. Selain itu saya juga berharap konferensi ini menjadi sarana berbagi pengetahuanbagi perkembangan teknologi informasi, khususnya teknologi informasi yang ramah lingkungan. Ucapan terima kasih saya sampaikan kepada para pembicara tamu dan seluruh peserta konferensi, terima kasih atas segala partisipasinya. Semoga konferensi ini memberikan kontribusi dalam pengembangan ilmu dan teknologi khususnya teknologi informasi yang ramah lingkungan. Ucapan terimakasih juga tidak lupa saya sampaikan kepada semua pihak yang ikut membantu terlaksananya konferensi ini. Selamat berkonferensi, semoga kita mampu berkontribusi nyata pada kemajuan bangsa dan negara kita tercinta, utamanya dalam bidang pengembangan teknologi informasi yang ramah lingkungan, sertamampu menangkap berbagai peluang yang sangat menjanjikan di masa yang akan datang, Amiin. Wassalamu ‘alaikum Wr. Wb.

Ketua Program Studi Sistem Komputer Universitas Diponegoro Semarang

Ir. Kodrat Iman Satoto, M.T.

SAMBUTAN KETUA APTIKOM WILAYAH VI JAWA TENGAH Yth. para Profesor, Narasumber, Tamu Undangan, dan para peserta “The 1st Conference on Information Technology, Computer, and Electrical Engineering” (CITACEE 2013) di Universitas Diponegoro. Dengan memanjatkan puji syukur ke hadirat Allah subhanahu wa ta’ala kami merasa berbangga hati atas terselenggaranya kegiatan conference yang pertama dengan bertemakan ”Information Technology and Its Application Toward the Implementation of Green Technology” ini. Kegiatan conference ini semakin melengkapi beberapa kegiatan APTIKOM Wilayah VI Jawa Tengah baik itu conference, workshop, dan ADB. Kegiatan conference dalam waktu dekat antara lain ICISBC 2013 Magister Sistem Informasi UNDIP pada tanggal 5-6 Desember 2013. Kegiatan Workshop seperti Workshop Kurikulum, KKNI, LAM dan PJJ yang diselenggarakan di UMM Magelang 26 Oktober 2013 lalu, adalah merupakan tindaklanjut Hasil Workshop 1 Jakarta 12.12.12 (12 Desember 2012). Seperti diketahui, APTIKOM Jawa Tengah diminta melaksanakan kegiatan ini. Hasil rumusan tersebut ditindaklanjuti pada Rakornas APTIKOM di Samarinda 31 Oktober – 2 November 2013 lalu. Selanjutnya akan diteruskan dengan kegiatan APTIKOM Doctoral Booth Camp VII (ADB VII) di Jawa Tengah tahun 2014. Kegiatan ini merupakan wujud proker APTIKOM dengan melatih peserta untuk menyusun proposal penelitian doktoral di bawah bimbingan Profesor, sehingga mendorong para Dosen APTIKOM lainnya untuk meningkatkan kemampuan akademik agar bergelar Doktor. Kami menyambut baik permintaan panitia CITACEE untuk sekadar memberikan sekapur kata pengantar pada Prosiding Conference ini, dengan mengucapkan selamat datang kepada nara sumber dan tamu undangan, serta selamat ber-conference kepada para peserta. Kami berharap conference ini dapat dilakukan secara kontinu, dapat memacu dan merangsang penelitian dalam rangka sharing informasi, khususnya perkembangan teknologi informasi dan implementasinya, serta merupakan agenda rutin di APTIKOM Wilayah VI Jawa Tengah. Demikian yang dapat kami informasikan, dan sekali lagi selamat ber-conference, semoga Tuhan YME memudahkan dan meringankan langkah kita untuk mendarmabaktikan pengetahuan agar benar-benar bermanfaat dan bermartabat bagi sesama manusia dan “go green” bagi lingkungannya. Terima Kasih.

Semarang, 4 November 2013 Ketua APTIKOM Wilayah VI Jawa Tengah

Drs. Eko Adi Sarwoko, M.Kom. NIDN : 0007116503 / NA : 04-105

CITACEE 2013 CONFERENCE COMMITTEE

General Chair : Dr. R. Rizal Isnanto, S.T., M.M., M.T. (Universitas Diponegoro) Co-Chair : Dr. Oky Dwi Nurhayati, S.T., M.T. (Universitas Diponegoro) Technical Committee: Prof. Drs. Mustafid, M.Eng, Ph.D. (Universitas Diponegoro) Dr. Ir. Hermawan, DEA. (Universitas Diponegoro) Dr. Aris Triwiyatno, S.T., M.T. (Universitas Diponegoro) Dr. Eng. Wahyul Amien Syafei, S.T., M.T. (Universitas Diponegoro) Dr. Munawar Agus Riyadi, S.T., M.T. (Universitas Diponegoro) Dr. Moch. Facta, S.T., M.T. (Universitas Diponegoro) Dr. Ir. Abdul Kadir, M.T., M.M. (Universiti Teknikal Malaysia, Melaka) Dr. Ratna Wardani, S.Si., M.T. (Universitas Negeri Yogyakarta) Dr. Ir. Sri Ratna Sulistyanti, M.T. (Universitas Lampung) Organizing Committee: Ir. Kodrat Iman Satoto, M.T. Adian Fatchur Rochim, S.T., M.T. Abdul Syakur, S.T., M.T. Rinta Kridalukmana, S.Kom, M.T. Kurniawan Teguh Martono, S.T., M.T. Eko Didik Widianto, S.T., M.T. Ike Pertiwi Windasari, S.T., M.T. Adnan Fauzi, S.T. Andi Widiasmoro, S.T. Okta Purnamasari, S.E. Melati Mawas Titi, A.Md.

SCHEDULE OF EVENTS Ruang Sidang SNMPTN, Gedung Widya Puraya Lt. 2 Universitas Diponegoro Semarang, Indonesia Saturday, 16 November 2013 07.30 – 08.00 : Registration 08.00 – 08.05 : Welcome greetings from MC 08.05 – 08.10 : Report from General Chair 08.10 – 08.20 : Opening speech from Dean of Engineering Faculty, Diponegoro University 08.20 – 08.45 : Cisco Networking Academy at a glance 08.45 – 09.00 : Morning coffee break 09.00 – 10.30 : Plenary Session Moderator: Dr. Mohd. Facta, S.T., M.T. Keynote Speaker: 1. Prof. Dr. Richardus Eko Indrajit “Presentation from Prof. Dr. Richardus Eko Indrajit” 2. Prof. Dr. Ir. Tarcisius Haryono, M.Sc. “Protection Characteristic of ZnO Block Lightning Arrester Used in 20 kV Distribution System Against Multiple Lightning Strikes” 3. Prof. Drs. Mustafid, M.Eng., Ph.D. “Green Information System : Innovation for Environmental Sustainability” 10.30 – 11.45 : Parallel Session I 11.45 – 12.45 : Lunch Break 12.45 – 15.45 : Parallel Session II 15.45 – 15.50 : Closing speech from each moderator of Parallel Session II. Certificate for speakers will be given after the Closing Speech 15.50 – 16.05 : Afternoon coffee

Parallel Session Room 1 Moderator: Dr. Ir. Hermawan, DEA

Parallel Session 1 (10.30 – 11.45) 1

2

3

Desain dan Realisasi Antena Mikrostrip untuk Tag dan Reader RFID pada Frekuensi UHF 923 – 925 MHz Dina Angela, Yuyu Wahyu, Vito Andri Lukito, Andi Ilham Syalaby High Performance Low Complexity Interference Canceller for High Throughput WLAN 802.11n System Wahyul Amien Syafei, Anky Setyadewa, Imam Santoso Accurate Channel Estimation in Low SNR Channel for WLAN 802.11n System Wahyul Amien Syafei, Grifina Nuzulia, Sukiswo

4

Optimized Pilot Alocation using Genetic Algorithm for Better Performance of Wi-Fi IEEE 802.11n Wahyul Amien Syafei, Yunda Kumala Nasution, Sukiswo


2 3

4

5

6

7

Iris Feature Extraction using Daubechies Wavelet Transform and its Recognition using Normalized Euclidean Distance R. Rizal Isnanto, Achmad Hidayatno, Antonius Dwi Hartanto High Performance Interference Canceller for 600 Mbps HT WLAN IEEE 802.11n Wahyul Amien Syafei, Sukiswo, Imam Santoso Klasifikasi dan Pengenalan Pola pada Sinyal EKG Berdasarkan Sifat Keacakan (Entropy) dengan 6 Channel Jaenal Arifin, Oyas Wahyunggoro, Rudy Hartanto Rancang Bangun Generator Suara Digital untuk Meningkatkan Produktivitas Hasil Pertanian Didik Widianto, Priguna Septia Putra, Budi Setyawan, Arena Bayu Chandra P Model Sistem Akuisisi Data Multiplatform Menggunakan Aplikasi Antarmuka Pengguna Berbasis Bahasa Pemrograman Processing Arief Hendra Saptadi, Paulus Insap Santosa, Bambang Sutopo Aplikasi Pembelajaran Pengucapan Nama Hewan (Satu Suku Kata) Dalam Bahasa Inggris Ajub Ajulian Zahra M, Achmad Hidayatno Perancangan Sistem SCADA Beban Penerangan pada Prototype Gedung A Teknik Elektro Fakultas Teknik Universitas Diponegoro Rizky Adi Nugraha

Room 2 Moderator: Ir. Kodrat Iman Satoto, M.T.


2 3

4

Optimasi MPPT (Maximum Power Point Tracker) Pada Sistem Photovoltaic menggunakan Algoritma Incremental Conductance Harmini, Titik Nurhayati Optimalisasi Engine Power Mode HD785-7 Adityo Dharma S, Erwin Dermawan Studi Empiris Komponen Dasar Sistem Pemanen Energi Vibrasi dengan Tranduser Piezoelektrik Deddy Susilo, Eka Firmansyah, Litasari Perancangan Sistem Kontrol Auto Tuning PID Menggunakan Fuzzy Logic: Studi Kasus pada Exhaust Gas Recirculation Heavy Duty Diesel Engine Ulinnuha Latifa, Aris Triwiyatno, Sumardi


2 3

4

5

6

7

Perancangan Sistem Kendali Logika Fuzzy Menggunakan Algoritma Genetika: Studi Kasus pada Exhaust Gas Recirculation (EGR) Heavy Duty Diesel Engine Fildzah Imanina, Aris Triwiyatno, dan Sumardi Operating a Four-Leg PWM Converter as a Three-Phase Controlled Current Source Slamet Riyadi Desain Sistem Kontrol Fuzzy Model Reference Learning Control (FMRLC) Studi Kasus: Pengontrolan Ketinggian Air pada Conical Tank M Arif Syukur D, Aris Triwiyatno, dan Wahyudi Desain dan Implementasi Digital Maximum Power Point Tracker Berbasis Mikrokontroller ATMEGA8535 Lukas Aditya M, Leonardus H. Pratomo Estimasi Sudut Orientasi Benda Menggunakan Sensor 6 DoF IMU dan Sensor Magnetometer 3 Aksis Rahadian Nurfansyah, Wahyudi, dan Budi Setiyono Switching Table Based on Space Vector Modulation for Three Phase Inverter using dsPIC Arifin Wibisono, Slamet Riyadi Paralel Dua Inverter Sumber Tegangan Berbasis Mikrokontroler ATmega 8535 Ram Karliutama dan Leonardus. H. Pratomo

Room 3

Moderator: Adian Fatchur Rochim. S.T., M.T. Parallel Session 1 (10.30 – 11.45) 1

2 3

4

Web Based Map Generations of Mobile Robot Movement using Scalable Vector Graphic Harindra Wisnu Pradhana, Achmad Widodo, Suryono Fleksibel Input – Output (Fleksi IO) Berbasis Web Client – Server Tamas Riyo Sesono, Darmawan Utomo, Hartanto Kusuma Wardana Evaluation of Hidden Terminal Problem in Wireless Local Area Network Using Network Simulator Version 3 Alexander William, Antonius Suhartomo Physics Analysis of Erbium Doped Fiber Amplifier (EDFA) Alexander William, Antonius Suhartomo

Parallel Session 2 (12.45 – 15.45) 1 2 3

4

5 6

7

Rancang Bangun Alat Kendali Sistem Keamanan Rumah Wahyu Kusuma, Jamilah, Rizky Satrio Putro Desain Kontroler Fuzzy pada Model Automatic Braking System dengan Antilock Rezki Ahmaliansyah, Aris Triwiyatno, dan Budi Setiyono Miniatur Sistem Palang Perlintasan Otomatis Kereta Api Menggunakan Sensor Infra Merah dan Mikrokontroler AT89S51 Ferry Sudarto, Indrianto, Satriyo Budi Santoso Perbandingan Standar Nasional Indonesia (SNI) terhadap Standar Internasional pada Produk Biomedical Implant Plate Budi Setiyana, Muhammad Khafidh, Rifky Ismail, M. Tauviqirrahman, dan Jamari Environment Monitoring System (EMS) Berbasis Jaringan GPRS Dina Angela, Tunggul A. Nugroho, Sinung Suakanto, Herry I. Sitepu Pengaruh Durasi Paparan Sinar Ultraviolet terhadap Arus Bocor Permukaan Bahan Isolator Resin Epoksi dengan Bahan Pengisi Silane dan Pasir Silika Abdul Syakur, Hermawan, Tommy Perdana Putra Analisis Pengaruh Sudut Kemiringan Panel Surya Tipe Array Tetap terhadap Energi yang Dihasilkan pada Solar Home System (SHS) Pangestuningtyas Diah L., Hermawan, dan Karnoto

Room 4

Moderator: Rinta Kridalukmana, S.Kom., M.T.


2 3 4

Kajian Perancangan Aplikasi Helpdesk and Ticketing Studi Kasus AJB Bumiputra Farham Harvianto, Akbar Muchbarak, Ahmad Pudoli, Sofian Lusa Kendali Mobil Remote Control Menggunakan Handphone Android Ferry Sudarto, Sudaryono, Hendra Kusumah Dynamics System Modeling Approach in Node Mobility on Mobile Ad-hoc Network S.N.M.P. Simamora Analysis of the Application current X2 Interface Handover Process in LTE Technology Uke Kurniawan Usman


2

3 4 5

6 7

Perencanaan Strategis E-Government Pada Pemerintah Daerah Provinsi Maluku Utara Untuk Mewujudkan Good Governance Muhammad Ridha Albaar Desain Sistem Keamanan Pada Infrastruktur Berbasis Jaringan Komputer di Universitas Kristen Petra Ibnu Gunawan, Agustinus Noertjahyana Aplikasi Deteksi Plagiarisme Berbasis Web Menggunakan Framework CodeIgniter Agung Toto Wibowo, Adeva Oktoveri, Ari Moesriami Barmawi, Anditya Arifianto Firewall Session accelerate QoS, NAT and Routing using Decision Tree Budi Dwi Satoto Deteksi Jumlah Kendaraan di Jalan dengan Transceiver SRF02 dan Mikrokontroller ATmega8A Joko Lianto Buliali, Victor Hariadi, Karisma Trinanda Putra, Syahri Muharom Klasifikasi Spermatozoa Pembawa Kromosom X atau Y dengan Metode Naïve Bayes Muhammad Hasan Wahyudi Perancangan dan Implementasi Virtual Hosting Menggunakan Linux Dahlan Abdullah

Room 5 Moderator: Teguh Kurniawan Martono, S.T., M.T.


2

3

4

Estimasi Jarak Berbasis Konektivitas Untuk Penentuan Posisi Node pada Jaringan Sensor Nirkabel Aries Pratiarso, Prima Kristalina, Ninis Ari Fianti Learning Support Application of Gesture Language for Mute and Deaf Case Study in SLB-B YRTRW (Yayasan Rehabilitasi Anak Tuna Rungu Wicara) Surakarta Aris Rakhmadi, Anwar Dwi Harnanto, Fajar Suryawan Audit Kesiapan Organisasi dalam Implementasi Knowledge Management System: Studi Kasus Perusahaan di Bidang Jasa Keuangan Dewi Puspasari, Ovi Novianto, Bayu Kelana Distance Regularized Level Set Evolution for Medical Image Segmentation Indra Rianto, Pranowo


2

3

4

5

6 7

A New Model of Information Processing based on Human Brain Mechanism: Toward a Cognitive Intelligent System Arwin Datumaya Wahyudi Sumari Perancangan Dashboard System untuk Sekolah Musik dengan Menggunakan Key Performance Indicator (Studi Kasus: Sekolah Musik XYZ) Yosi Yonata, Arief Samuel Gunawan, Samuel Deyunior Aplikasi Presensi Kelas Kuliah Dengan Near Field Communication (NFC) Pada Android Andreas Handojo, Julius Wonodihardjo, Justinus Andjarwirawan Pengembangan Simulasi Peternakan Sapi Perah dengan Game Maker Berbasis Windows Prima Widyaningrum, R Rizal Isnanto, Kurniawan Teguh Martono Sistem Cerdas untuk Klasifikasi Kemampuan Kognitif dengan Adaptive Neuro Fuzzy Inference System (ANFIS) Muhamad Afif Effindi, I Nyoman Sukajaya, I Ketut Eddy Purnama, Mauridhi Hery Purnomo Pembangunan E-Commerce UKM Gamelan di Kecamatan Serengan Surakarta Abdul Aziz, Meiyanto Eko Sulistyo Pengembangan Lingkungan Pembelajaran Berbasis Digital Berbasis Kerangka Kerja QoS Adaptif Ratna Wardani, Lukito Edi Nugroho

Room 6 Moderator: Dr. Munawar Agus Riyadi, S.T., M.T.

Parallel Session 1 (10.30 – 11.45) 1 2

3

4

Pengembangan Sistem Pemrograman Acara pada Stasiun TV9 Surabaya Khakim Ghozali, Feby Artwodini Muqtadiroh, Angga Kusumandaru SMS sebagai Perubah Informasi Matriks LED Berbasis AVR ATMega8 pada Perguruan TInggi Raharja Asep Saefullah, Ferry Sudarto, Sigit Maulana Kuncoro Perancangan Permainan Mengasah Daya Ingat “Memory Training” Menggunakan Greenfoot Yessy Kurniasari, R Rizal Isnanto, Oky Dwi Nurhayati Perancangan Permainan Proses Pembuatan Bioetanol Ekstrak Limbah Buah Menggunakan Adobe Flash Cs3 Professional Nurul Arifa, R. Rizal Isnanto, Oky Dwi Nurhayati


2

3

4 5

6

7

Evaluasi Pemeliharaan Sistem Akuntansi Keuangan Berbasis Cobit 4.1. dan Sistem Dinamik Irani Hoeronis, Husni Sastramihardja, Arie Ardiyanti Efﬁcient Work Stealing for Portability of Nested Parallelism and Composability of Multithreaded Program Adnan, Zahir Zainuddin, Wardi Desain Dan Diagnosis Pengembangan Sistem Cerdas Computer Aided Process Planning (Capp) untuk Estimasi Pemanfaatan Limbah Buah Menjadi Bioetanol Sebagai Salah Satu Energi Alternatif Endro Sutrisno, Sri Sumiyati, Oky Dwi Nurhayati Design of Smart Classroom in Collaborative Learning Era : a Personalize Baginda Anggun Nan Cenka and Zainal A. Hasibuan Implementasi Sistem Baitul Mal Wat Tamwil (Bmt) dengan Teknologi Cloud Computing Sebagai Software As a Service (SaaS) Mirzam Muhammad AAN, Abdul Aziz, Wiharto Perancangan Sistem Informasi Perpustakaan di Jurusan Teknik Lingkungan Fakultas Teknik UNDIP Ike Pertiwi Windasari, Dr. Oky Dwi Nurhayati, Sri Sumiyati, Lia Dorothy Implementasi Sistem Pakar Diagnosis Penyakit Diabetes Mellitus Menggunakan Metode Fuzzy Logic Berbasis Web Fauzan Masykur

TABLE OF C ONTENTS

International Session 1 Protection Characteristic of ZnO Block Lightning Arrester Used in 20 kV Distribution System Against Multiple Lightning Strikes Tarcicius Haryono 18 Green Information System: Innovation for Environmental Sustainability Mustafid Mustafid 23 Web Based Map Generations of Mobile Robot Movement using Scalable Vector Graphic Harindra Wisnu Pradhana, Achmad Widodo, Suryono Suryono 28 Evaluation of Hidden Terminal Problem in Wireless Local Area Network Using Network Simulator Version 3 Alexander William, Antonius Suhartomo 32 Physics Analysis of Erbium Doped Fiber Amplifier (EDFA) Alexander William, Antonius Suhartomo 35 Dynamics System Modeling Approach in Node Mobility on Mobile Ad-hoc Network S N M P Simamora 40 Analysis of the Application current X2 Interface handover Process in LTE Technology Uke Kurniawan Usman 45 Learning Support Application of Gesture Language for Mute and Deaf Aris Rakhmadi, Anwar Dwi Harnanto, Fajar Suryawan 49 Distance Regularized Level Set Evolution for Medical Image Segmentation Indra Rianto, Pranowo Pranowo 52 Firewall Session accelerate Qos, NAT and Routing using Decision Tree Budi Dwi Satoto 56 A New Model of Information Processing based on Human Brain Mechanism: Toward a Cognitive Intelligent System Arwin Datumaya Wahyu Sumari, Adang Suwandi Ahmad 62 Efficient Work Stealing for Portability of Nested Parallelism and Composability of Multithreaded Program Adnan Adnan, Zahir Zainuddin, Wardi Wardi 68 Design of Smart Classroom in Collaborative Learning Era Baginda Anggun Nan Cenka, Zainal A. Hasibuan 74 Operating a Four-Leg PWM Converter as a Three-Phase Controlled Current Source Slamet Riyadi 79 Switching Table Based on Space Vector Modulation for Three Phase Inverter using dsPIC Arifin Wibisono, Slamet Riyadi 84 High Performance Low Complexity Interference Canceller for High Throughput WLAN 802.11n System Wahyul Amien Syafei, Anky Setyadewa, Imam Santoso 89 Accurate Channel Estimation in Low SNR Channel For WLAN 802.11n System Wahyul Amien Syafei, Grifina Nuzulia, Sukiswo Sukiswo 94 Iris Feature Extraction using Daubechies Wavelet Transform and its Recognition using Normalized Euclidean Distance R. Rizal Isnanto, Achmad Hidayatno, Antonius Dwi Hartanto 100 High Performance Interference Canceller for 600 Mbps HT WLAN IEEE 802.11n Wahyul Amien Syafei, Sukiswo Sukiswo, Imam Santoso

National Session 105 Rancang Bangun Alat Kendali Sistem Keamanan Rumah Wahyu Kusuma, Jamilah Jamilah, Rizky Satrio Putro 111 Fleksibel Input - Output (Fleksi IO) Berbasis Web Client - Server Tamas Riyo Sesono, Darmawan Utomo, Hartanto Kusuma Wardana 118 Desain Kontroler Fuzzy pada Model Automatic Braking System dengan Antilock Rezki Ahmaliansyah, Aris Triwiyatno, Budi Setiyono 124 Miniatur Sistem Palang Perlintasan Otomatis Kereta Api Menggunakan Sensor Infra Merah dan Mikrokontroler AT89S51 Ferry Sudarto, Indrianto Indrianto, Satriyo Budi Santoso 129 Perbandingan Standar Nasional Indonesia (SNI) terhadap Standar Internasional pada Produk Biomedical Implant Plate Budi Setiyana, Muhammad Khafidh, Rifky Ismail, M. Tauviqirrahman, Jamari Jamari 133 Environment Monitoring System (EMS) Berbasis Jaringan GPRS Dina Angela, Tunggul Arief Nugroho, Sinung Suakanto, Herry Imanta Sitepu 138 Kajian Perancangan Aplikasi Helpdesk and Ticketing Farham Harvianto, Akbar Muchbarak, Achmad Pudoli, Sofian Lusa 143 Kendali Mobil Remote Control Menggunakan Handphone Android Ferry Sudarto, Sudaryono Sudaryono, Hendra Kusumah 147 Estimasi Jarak Berbasis Konektivitas Untuk Penentuan Posisi Node pada Jaringan Sensor Nirkabel Aries Pratiarso, Prima Kristalina, Ninis Ari Fianti 152 Audit Kesiapan Organisasi dalam Implementasi Knowledge Management System: Studi Kasus Perusahaan di Bidang Jasa Keuangan Dewi Puspasari, Ovi Novianto, Bayu Kelana 159 Pengembangan Sistem Pemrograman Acara pada Stasiun TV9 Surabaya Khakim Ghozali, Feby Artwodini Muqtadiroh, Angga Kusumandaru 168 SMS Sebagai Perubah Informasi Matriks LED Berbasis AVR ATMega8 pada Perguruan Tinggi Raharja Asep Saefullah, Ferry Sudarto, Sigit Maulana Kuncoro 174 Perancangan Permainan Mengasah Daya Ingat Memory Training Menggunakan Greenfoot Yessy Kurniasari, R. Rizal Isnanto, Oky Dwi Nurhayati 181 Perancangan Permainan Proses Pembuatan Bioetanol Ekstrak Limbah Buah Menggunakan Adobe Flash CS3 Professional Nurul Arifa, R. Rizal Isnanto, Oky Dwi Nurhayati 188 Perencanaan Strategis E-Government Pada Pemerintah Daerah Provinsi Maluku Utara Untuk Mewujudkan Good Governance Muhammad Ridha Albaar 192 Desain Sistem Keamanan Pada Infrastruktur Berbasis Jaringan Komputer di Universitas Kristen Petra Ibnu Gunawan, Agustinus Noertjahyana 197 Aplikasi Deteksi Plagiarisme Berbasis Web Menggunakan Framework CodeIgniter Agung Toto Wibowo, Adeva Oktoveri, Ari Moesriami Barmawi, Anditya Arifianto 204 Deteksi Jumlah Kendaraan di Jalan dengan Transceiver SRF02 dan Mikrokontroller ATmega8A Joko Lianto Buliali, Victor Hariadi, Karisma Trinanda Putra, Syahri Muharom 210 Perancangan Dashboard System Untuk Sekolah Musik Dengan Menggunakan Key Performance Indicator Yosi Yonata, Arief Samuel Gunawan, Samuel Deyunior 216 Aplikasi Presensi Kelas Kuliah Dengan Near Field Communication (NFC) Pada Android Andreas Handoyo, Julius Wonodihardjo, Justinus Andjarwirawan 222 Pengembangan Simulasi Peternakan Sapi Perah dengan Game Maker Berbasis Windows Prima Widyaningrum, R. Rizal Isnanto, Kurniawan Teguh Martono 229 Sistem Cerdas untuk Klasifikasi Kemampuan Kognitif dengan Adaptive Neuro Fuzzy Inference System (ANFIS) Muhamad Afif Effindi, I Nyoman Sukajaya, I Ketut Eddy Purnama, Mauridhi Hery Purnomo 233 Evaluasi Pemeliharaan Sistem Akuntansi Keuangan Berbasis COBIT 4.1 dan Sistem Dinamik irani Hoeronis, Husni Sastramihardja, Arie Ardiyanti

238 Desain dan Diagnosis Pengembangan Sistem Cerdas Computer Aided Process Planning (CAPP) untuk Estimasi Pemanfaatan Limbah Buah Menjadi Bioetanol Sebagai Salah Satu Energi Alternatif Endro Sutrisno, Sri Sumiyati, Oky Dwi Nurhayati 242 Implementasi Sistem Baitul Mal Wat Tamwil (BMT) dengan Teknologi Cloud Computing sebagai Software as a Service (SaaS) Mirzam Muhammad A A N, Abdul Aziz, Wiharto Wiharto 249 Klasifikasi Spermatozoa Pembawa Kromosom X Atau Y Dengan Metode Naive Bayes Muhammad Hasan Wahyudi 253 Pembangunan E-Commerce UKM Gamelan di Kecamatan Serengan Surakarta Abdul Aziz, Meiyanto Eko Sulistyo 258 Pengembangan Lingkungan Pembelajaran Berbasis Digital Berbasis Kerangka Kerja QoS Adaptif Ratna Wardani, Lukito Edi Nugroho 262 Perancangan Sistem Informasi Perpustakaan di Jurusan Teknik Lingkungan Fakultas Teknik Undip Ike Pertiwi W., Oky Dwi Nurhayati, Sri Sumiyati, Lia Dorothy 270 Implementasi Sistem Pakar Diagnosis Penyakit Diabetes Mellitus Menggunakan Metode Fuzzy Logic Berbasis Web Fauzan Masykur 277 Perancangan dan Implementasi Virtual Hosting Menggunakan Linux Dahlan Abdullah 292 Optimasi MPPT (Maximum Power Point Tracker) Pada Sistem Photovoltaic menggunakan Algoritma Incremental Conductance Harmini Harmini, Titik Nurhayati 297 Optimalisasi Engine Power Mode HD785-7 Adityo Dharma S., Erwin Dermawan 302 Studi Empiris Komponen Dasar Sistem Pemanen Energi Vibrasi dengan Tranduser Piezoelektrik Deddy Susilo, Eka Firmansyah, Litasari Litasari 306 Perancangan Sistem Kontrol Auto Tuning PID Menggunakan Fuzzy Logic: Studi Kasus pada Exhaust Gas Recirculation Heavy Duty Diesel Engine Ulinnuha Latifa, Aris Triwiyatno, Sumardi Sumardi 312 Perancangan Sistem Kendali Logika Fuzzy Menggunakan Algoritma Genetika: Studi Kasus pada Exhaust Gas Recirculation (EGR) Heavy Duty Diesel Engine Fildzah Imanina, Aris Triwiyatno, Sumardi Sumardi 318 Desain Sistem Kontrol Fuzzy Model Reference Learning Control (FMRLC) Studi Kasus: Pengontrolan Ketinggian Air pada Conical Tank M. Arif Syukur D., Aris Triwiyatno, Wahyudi Wahyudi 324 Desain dan Impelementasi Digital Maximum Power Point Tracker Berbasis Mikrokontroller ATMEGA8535 Lukas Aditya Mulyapratama, Leonardus H. Pratomo 329 Estimasi Sudut Orientasi Benda Menggunakan Sensor 6 DoF IMU dan Sensor Magnetometer 3 Aksis Rahadian Nurfansyah, Wahyudi Wahyudi, Budi Setiyono 335 Paralel Dua Inverter Sumber Tegangan Berbasis Mikrokontroler ATMega 8535 Ram Karliutama, Leonardus H. Pratomo 339 Pengaruh Durasi Paparan Sinar Ultraviolet terhadap Arus Bocor Permukaan Bahan Isolator Resin Epoksi dengan Bahan Pengisi Silane dan Pasir Silika Abdul Syakur, Hermawan Hermawan, Tommy Perdana Putra 345 Analisis Pengaruh Sudut Kemiringan Panel Surya Tipe Array Tetap terhadap Energi yang Dihasilkan pada Solar Home System (SHS) Pangestuningtyas Diah L., Hermawan Hermawan, Karnoto Karnoto 350 Desain dan Realisasi Antena Mikrostrip untuk Tag dan Reader RFID pada Frekuensi UHF 923 - 925 MHz Dina Angela, Yuyu Wahyu, Vito Andri Lukito, Andi Ilham Syalaby 356 Optimized Pilot Alocation using Genetic Algoritm for Better Performance of Wi-Fi IEEE 802.11n Wahyul Amien Syafei, Yunda Kumala Nasution, Sukiswo Sukiswo

361 Klasifikasi dan Pengenalan Pola pada Sinyal EKG Berdasarkan Sifat Keacakan (Entropy) dengan 6 Channel Jaenal Arifin, Oyas Wahyunggoro, Rudy Hartanto 366 Rancang Bangun Generator Suara Digital untuk Meningkatkan Produktivitas Hasil Pertanian Eko Didik Widianto, Priguna Septia Putra, Budi Setyawan, Arena Bayu Chandra Permana 370 Model Sistem Akuisisi Data Multiplatform Menggunakan Aplikasi Antarmuka Pengguna Berbasis Bahasa Pemrograman Processing Arief Hendra Saptadi, Paulus Insap Santosa, Bambang Sutopo 375 Aplikasi Pembelajaran Pengucapan Nama Hewan (Satu Suku Kata) dalam Bahasa Inggris Ajub Ajulian Zahra, Achmad Hidayatno 382 Perancangan Sistem SCADA Bebab Penerangan pada Prototype Gedung A Teknik Elektro Fakultas Teknik Universitas Diponegoro Rizky Adi Nugroho

CITACEE 2013

ISSN: 2338-5154

CITACEE 2013, UNDIP, SEMARANG

Protection Characteristic of ZnO Block Lightning Arrester Used in 20 kV Distribution System Against Multiple Lightning Strikes by

T. Haryono Electrical Engineering and Information Technology Department Engineering Faculty Gadjah Mada University

Several kinds of lightning strike (Sirait 2010)

Yogyakarta 16 Nopember 2013

Damaged transformer due to an arrester being not working: • Transformer replacement • cost due to no electricity • cost due to transformer oil cleaning • arrester cost

Eight kinds of lightning strike (sirait,2010)

Lightning Arrester

Lightning Arrester

Transformer

20 kV lightning arrester on 50 kVA transformer at PPKP Yogyakarta

Transfor mer

20 kV lightning arrester on 25 kVA 1 fase transformer at Pogung Dalangan, Yogyakarta

1 Diponegoro University

Semarang, 16 November 2013

1

CITACEE 2013

ISSN: 2338-5154

Transfor mer

Lightning Arrester

20 kV lightning arrester on 25 kVA 1 phase transformer

20 kV lightning arresters protecting underground cable supplying MM UGM 20 kV electrical system in Yogyakarta

at Pogung Dalangan, Yogyakarta

Bantul 120 MVA Substation Equipment protected by lightning arresters in 150 kV incoming feeders from Solo

Llightning arresters protect 3 phase 60 MVA 150 kV Bantul Substation Transformer at Yogyakarta

Lightning Arresters protect load circuit breakers at 1000 MVA 500 kV substation at Pedan

Lightning arrresters protect 500 MVA 3 x 1 fase 500/150 kV transformers at 1000 MVA 500 kV substation at Pedan



2

CITACEE 2013

ISSN: 2338-5154

Lightning arresters protect 1000 MVA 500 kV substation at Pedan at incoming feeders from Cimahi

Introduction

Lightning arresters protect 1000 MVA 500 kV substation at Pedan At incoming feeders from Kediri

– More ZnO arrester is installed • No follow current produced

Background Tropical area spesific characteristics: • Temperature, humidity, and rain debit

A lightning arrester is an equipment to protect other electrical equipment against overvoltage due to lightning surge voltage

high

– Problem :

• Overvoltage protection types : – spark gap : rod gap (Couper, 1983) • Follow current circuit breaker operation – expulsion tube arrester • No follow current – surge diverter / lightning arrester • SiC (silicon carbide) arrester – When electrical current is switched off, electrical spark may happen in series gap due to the dielectric strength of the air gap having not returned back to it’s normal condition.

sea water and industrial polutan cover the housing of a lightning arrester produce unhomogen capasitance between outer surface and inner part of a lightning arrester housing

unhomogen voltage distribution on outer surface of a lightning arrrester housing

– External flashover risk through the outside surface of a lightning arrester housing – Internal partial discharge of the ZnO block to the outside surface of a lightning arrester housing

the ZnO block temperature may increase it’s thermal stability

A lightning arrester



3

CITACEE 2013

ISSN: 2338-5154

Porselin ZnO Arrester system

20 kV lightning arresters

One ZnO Block Four ZnO Blocks connected in series

120 kV Impulse voltage SiC Block Arrester

ZnO blocks Arrester

… … … .

120 kV impulse voltage cut by 20 kV SiC arrester 120 kV impulse voltage cut by 20 kV ZnO arrester



4

CITACEE 2013

ISSN: 2338-5154

Problem Peak voltage, kV

1. The increase of lightning arrester temperature exceeds the maximum limit making the arrester unstabil. 2. Multiple lightning surge current flows through ZnO block causing large change of the ZnO physic characteristic that may result in the damage of the lightning arrester

Isolation characteristic

isolation protection characteristic mikrosecond

Isolation protection

PREVIOUS STUDIES

The Objective of this research •

To investigate the ZnO block performance when it was stroken by multiple lightning strikes The performance of a lightning arrester determines : • The reliability of protected equipment • Life time of equipment



The implementation of MOV (metal oxide varistor) in unearthing distribution system was not proper (Burke James J., dkk, 1991) because in high voltage condition due to fault may create mal operation. Chrzan Krystian, et.al, 1991, reported that for 110 kV metal oxide arrester which was tested in pollution condition using solid layer and salty fog methode created stabil dry tape layer on arrester housing increasing the varistor temperature reaching 70 ºC. In1993, Darveniza found that by applying 5 surge impulse of 5 kA 8/20 microsecs using duration time between impulse of 35 msecs using current of 5 kA up to 9.5 kA ZnO damaging arrester and the damage was accumulative. Japan investigators, Fujiwara, et al, 1996, obtained that by applying pin insolator having metal oxide arrester installed at every two isolator pole reduced 33% lightning strikes to distribution lines. If it is installed at every pole, the reduction in having lightning strike reaching up to 97.5 %

ZnO characteristic

ZnO varistor thermal stability

ZnO arrester current-time characteristic



5

CITACEE 2013

ISSN: 2338-5154

 Aplication

Top

ZnO arrester production process

: Arrester instalation

Lightning Arrester at 420 kV substation

Bottom : Cutting the voltage

Controlling devices

Main equipment • Testing Box made from pleksi glass using the dimension of 0,3 meter x 0,3 meter x 0,3 meter for putting ZnO Block • Multiple High Current Surge Generator : 5 kA - 10 kA • Free induction measurement chamber

Measuring and control devices Measuring tool in testing box

• Electronic controller for testing box temperature • Humidity controller in box testing • Trigger circuit for 5 high impulse current in sequence produced by high impulse current generators • Computer-based control to trigger the high impulse current generator circuit

– Thermometer, barometer, dan hygrometer

Measuring device outside testing box – – – –

Current shunt and pick up coil Digital Osiloscope SEM Infrared camera

Single high current impulse generator circuit

V = R im + L (dim/dt) + (1/C)



t 0

imdt

Usual underdamped circuit then : Switch

L construction

R construction

R/2 0, and α ϵ R are coefficient of the energy functional and …(3)

b) Initial Contour of Level Set Function

Where P is potential energy …(4) …(5) is speed of level set function accelerated. is minimum when level set function at object boundary. V.

RESULT

Distance Regularized Level Set Evolution (DRLSE) method is used in this research for medical image segmentation. Medical image that is used is a MRI Scan of brain and the size of this image is 256x256 pixels. This research shows that DRLSE can be used for image segmentation and the segmentation result is good.

c) Contour at Iteration 100

Figure 2 shows the image segmentation result using DRLSE method. First, select the input image. In this research, we using brain scan image (a). After that, initialize the contour of level set function (b). Then the level set function will evolve, moving the zero level set toward the desired object boundary. Curve evolution process of DRLSE model, it is display the zero level

Diponegoro University


50

CITACEE 2013

ISSN: 2338-5154

computation. Parallel computation can be done in GPU in order to speed-up the computational time. ACKNOWLEDGMENT Thank you Universitas Atma Jaya Yogyakarta for the financial support.

REFERENCES [1]

d) Contour at Iteration 500

[2]

[3]

[4]

[5]

[6]

[7]

e) Contour at Iteration 1000

[8]

[9]

[10]

[11]

f) Contour at Iteration 1600, also the image segmentation result Figure 2. Image Segmentation Result using DRLSE Method

[12]

[13]

VI.

CONCLUSION

DRLSE are one of many method that is used for image segmentation in order to analyze the medical image. DRLSE do not need to re-initialization so the computation time can be faster than level set method. But, if medical image has a big size then time that is need to compute will take longer. To minimize the computational time, maybe in next research, image segmentation with DRLSE method are done in parallel


[14]

[15] [16]

T. McInerney and D. Terzopoulos, "Deformable Models in Medical Image Analysis: A Survey," Medical Image Analysis, vol. 1, no. 2, pp. 91-108, 1996. D. D. Patil and S. G. Deore, "Medical Image Segmentation: A Review," International Journal of Computer Science and Mobile Computing, vol. 2, no. 1, pp. 22-27, 2013. C. Li, C. Xu, C. Gui and M. D. Fox, "Distance Regularized Level Set Evolution and Its Application to Image Segmentation," IEEE Transactions on Image Processing, vol. 19, no. 12, pp. 3243-3254, 2010. U. R. N., P. V. Subbaiah, D. V. Rao and N. K, "Optimal Segmentation of Brain Tumors using DRLSE Levelset," International Journal of Computer Application, vol. 29, no. 9, pp. 6-11, 2011. J.-q. Liu and W.-w. Liu, "Adaptive Medical Image Segmentation Algorithm Combined with DRLSE Model," Procedia Engineering, pp. 20634-2638, 2011. D. K. Lakovidis, M. A. Savelonas, S. A. Karkanis and D. E. Maroulis, "A Genetically Optimized Level Set Approach to Segmentation of Thyroid Ultrasound Images," Appl Intell, pp. 192-203, 2007. C. Li, R. Huang, Z. Ding, J. C. Gatenby and D. N. Mataxas, "A Level Set Method for Image Segmentation in the Presence of Intensity Inhomogeneities With Application to MRI," IEEE Transactions on Image Processing, vol. 20, no. 7, pp. 2007-2016, 2011. S. Yu, Y. Mou, D. Xu, X. You, Zou Long and W. Zeng, "A New Algorithm for Shoreline Extraction from Satellite Imagery with Non-Separable Wavelet and Level Set Method," International Journal of Machine Learning and Computing, vol. 3, no. 1, pp. 158-163, 2013. D. Jayadevappa, S. S. Kumar and D. S. Murty, "A New Deformable Model Based on Level Sets for Medical Image Segmentation," IAENG International Journal of Computer Science, 2009. T. Schoenemann, F. Kahl, S. Masnou and D. Cremers, "A Linear Framework for Region-Based Image Segmentation and Inpainting Involving Curvature Penalization," Int J Comput Vis, pp. 53-68, 2012. S. P. Duraizamy and R. Kayalvizhi, "A New Multilevel Thresholding Method Using Swarm Intelligence Algorithm for Image Segmentation," J. Intelligent Learning System & Applications, vol. 2, pp. 126-138, 2010. P. A. Chochia, "A Pyramidal Image Segmentation Algorithm," Journal of Communications Technology and Electronics, vol. 55, no. 12, pp. 1550-1560, 2010. N. Senthilkumaran and R. Rajesh, "Edge Detection Techniques for Image Segmentation - A Survery of Soft Computing Approaches," International Journal of Recent Trends in Engineering, vol. 1, no. 2, pp. 250-254, 2009. S. W. Yoon, C. Lee, J. K. Kim and L. Myoungho, "Wavelet-based Multi-resolution Deformation for Medical Endoscopic Image Segmentation," J Med Syst, pp. 207-214, 2008. A. McAndrew, An Introduction to Digital Image Processing with MATLAB, Victoria University of Technology, 2004. C. Solomon and T. Breckon, Fundamental of Digital Image Processing: A Practical Approach with Examples in Matlab, Chichester, West Sussex: Wiley Blackwell, 2011.


51

CITACEE 2013

ISSN: 2338-5154

Firewall Session accelerate QoS, NAT and Routing using Decision Tree Budi Dwi Satoto Lecture Informatic Management Trunojoyo University Ngagel Mulyo XII no 39 Surabaya, Indonesia, 60245

[email protected] Abstract— Firewall Security, Network address translation and QoS are precious objectives for network systems to be attained. Unfortunately, they are any conflict, while QoS tries to minimize processing delay, strong security protection requires more processing time and cause packet delay. This article is a step towards resolving this conflict by extending the firewall session table to accelerate NAT, QoS classification, and routing processing time while providing the same level of security protection. Decision Tree helps to reduce syntax in creating filter rule in packet session. Keywords : Packet filtering; Firewall; Session/state table; QoS; NAT; Routing.

I. Introduction Many firewall security mechanisms have evolved tomitigate the ever continuously increasing number of network attacks. Securing open and complex systems have become more and more complicated. If in addition we should take QoS requirements into account, the problem becomes more complicated and necessitates indepth reflexions. Router is a key component in the internet. Its main function is to control data packet flow and determine an optimal path to reach the destination. However, as the networking technology has evolved, much new functionality have been added and implemented in the router. In our context, besides routing function; we are also interested in other functionalities such as NAT, QoS, and stateful packet filtering. Firewall is the primary defense perimeter to protect networks. Firewall technology had evolved from stateless packet filtering toward stateful packet filtering implemented in network routers. In order to be able to trace connection state; SFP builds a session table (also called state table). This session table makes SFP faster and more secure. Some implementations have extended the session table to include NAT mapping information. In this work we find that adding QoS and routing information to the session table is a natural extension that will speedup performance and enhance router scalability and availability. A. Router Main Function. First, Routing is the process of selecting paths in a network along which to send network traffic [1]. Routing directs packet forwarding, the transit of logically addressed packets from their source towards their ultimate destination through intermediate nodes. The


routing process usually directs forwarding on the basis of routing tables which maintain a record of the routes to various network destinations. Thus, constructing routing tables, which are held in the routers’ memory, is very important for efficient routing. Routing table may be configured manually or dynamically utilizing routing algorithm such as OSPF [2]. Fast routing table lookup is an important requirement to implement high performance router. A lot of researches have been conducted to speedup routing table lookup [3], [4]. B. Network Address Translation (NAT). NAT is an IETF [5] standard that enables a local area network (LAN) to modify network IP ddresses and ports numbers in headers of datagram packets (in transit across a traffic routing device) for the purpose of remapping a given address space into another. One of the main objectives of NAT is to solve the scalability problem when the number of IP addresses allowed to access the external network is limited. From the security point of view, NAT more or less hides internal private network addresses from outsiders, enforces control over outbound connections, and restricts incoming traffics [6]. The NAT table is the heart of the whole NAT operation, which takes place within the router as packets arrive and leave its interfaces. Each connection from the internal (private) network to the external (public-Internet) network, and vice versa, is tracked and a special NAT table is created to help the router determine what to do with all incoming packets on all of its interfaces. Again, NAT table lookup plays an important role in enhancing router performance. C. Quality of service (QoS). QoS is a heavily loaded term with many different meanings depending upon the specific context. IETF [7] has defined QoS as nature of the packet delivery service provided, as described by parameters such as achieved bandwidth, packet delay, and packet loss rates. The main goal of QoS is to provide priority treatment. A QoS policy should identify what priority level will be given to each traffic flow. After that, classification algorithms [8] can be used to inspect each packet and mark it with its associated priority level. High priority traffic such as VoIP should beserved before nonpriority one such as e-mail or FTP packets. To achieve this goal packets are placed in queues (waiting for processing) according to its priority levels. Queues


52

CITACEE 2013

ISSN: 2338-5154

represent locations where packets may be held (or dropped). Packet scheduling refers to the decision process used to choose which packets should be serviced or dropped. Buffer management refers to any particular discipline used to regulate the occupancy of a particular queue. Packets will be placed in different queues according to their priority levels. Afterwards, schedulers will pick packet to be served according to their priorities. The most important objectives of scheduling are computational efficiency and fairness [9]. D. Packet Filtering Firewalls are network devices that filter network traffic at one or more of the seven ISO network model most commonly at the network, transport, and application levels [10]. Packet Filtering Firewalls (PFs) were the first generation of firewalls. They are basically screening routers [11] that control the flow of data in and out of a network by looking at certain fields in the packet header: Source IP Address, Destination IP Address, Protocol identifier, Source port number, and Destination port number. The PF inspects all incoming and outgoing packets and applies the specified policy (e.g., drop or accept the packets). PF was considered as an efficient, fast, and cost effective solution since a single ranyouter can protect an entire network. However, PF has a lot of limitations: it is based on IP addresses without any authentication, it depends on port number for identifying communicating applications and this is not a reliable indicator because many current protocols such as network file system (NFS) uses varying port numbers. It cannot defend against man in the middle attacks and forged packets with spoofed IP addresses. But the most important limitation is the difficulty of writing correct filters [12] for complex and permanently evolving systems. Generally, filtering rules are far from providing perfect security against holes in the PF. Stateful Inspection Packet Filtering Firewall (SPF) While PF works by statically inspecting each packet against the rule set, SPF works not only by inspecting the packet headers but also by correlating the incoming traffic to the earlier outgoing requests [13]. Basically, SPF builds dynamic session/state table to record relevant information of each communication to trace the validity of each packet in these connections. SPF dynamically opens and closes ports according to the connection needs, in this way it makes network management easier. The following section describes the structure of the session table. II ARCHITECTURE AND PROCESSING Fig. 1. Shows the general architecture of the session table; the first five fields are used to identify a unique session, it is the same fields used by stateless firewall and QoS to classify traffic; it is called SID (session ID or selector). State field is used to store the state of this session and time field contains session timeout [14]. Fig. 2. Represents casual stateful packet filtering processing in router based implementation. For incoming packet, first, NAT


translation will be performed to make the necessary mapping between external and internal addresses; then the session table will be searched, if an entry is found in the session table the packet will be inspected to ensure its conformance to the session state. If there is no entry in the session table, this means that the packet is the first one in this session so, the packet will be validated against the filtering rules. If it is allowed to pass, an entry in the state table will be added. After that, QoS classification will be performed and the DSCP (differentiated service code point) priority value will be set in the ToS ―Type of Service‖ (also called Traffic Class in IP v6) field in the packet IP header to tell core routers how to treat this packet [15]. Finally routing table will be looked up to determine the next hop and then the packet is transmitted.

Figure. 1. General architecture of session table. Receive packet

NAT

Search Session Table

exist

Check Filtering Rule

Check Session State

Add State Entry

QOS Classification

Routing

Transmit Packet

Figure 2. Casual stateful packet filtering processing.


53

CITACEE 2013

ISSN: 2338-5154

Additional process to check filtering rule Prunning

Partition

Classification

Count Atribut

Gain, Entropy

Figure 3. Check filtering Rule using Decision Tree. Not that all of the previously mentioned operations need tables’ lookup and most of them search multiple fields in the table to find the appropriate entry. Reducing this lookup time is an important goal to enhance performance. For this reasons some firewall implementation such as NetBSD PF merged with NAT and session state information in the session table. III. INTEGRATED SESSION TABLE ARCHITECTURE AND PROCESSING Our goal is to increase security edge router processing capacity and enhance its scalability and availability. In order to achieve this goal, we merge all the needed information to perform SPF, NAT, QoS classification and routing in an integrated session table. This will make all the needed information available in only one search process in short session table, which is a great saving in processing time. Fig. 4. Shows the architecture of our integrated session table; due to column wide space limitation the table is divided into two parts.

address and port number. Finally, extaddress and ext-port are external communicating host IP address and port number. For stateful filtering: < Lan-addr, lan-port, extaddr, ext-port, IP-proto> constitute session ID and the session state and time out are stored in state and time fields. For QoS classification the SID will be used to do classification and the QoS priority value will be stored in the DSCP field. For routing table ext-next-hop will be used to send the packet to a destination outside the protected network. While, Lan-next-hop will be used to send the packet to a destination inside the protected network. Fig. 5. Represents our integrated session table processing. For incoming packet, the session table will be searched, If there is no entry in the session table, this means that the packet is the first one in this session so, first, NAT translation will be performed to make the necessary mapping between external and internal addresses; then the packet will be validated against the filtering rules. If it is allowed to pass, an entry in the state table will be added. After that, QoS classification will be performed and the DSCP will be added to its field in session table. Finally routing table will be searched to obtain the next hop values. If an entry is found in the session table the packet will be inspected to ensure its conformance to the session state and all the needed session processing will be performed in one shot without further research overhead, as all the needed information is available from the single lookup in the short session table. This is clearly a great enhancement which save processing time and increase performance. Receive packet

NAT

Search Session Table

exist

Check Filtering Rule

Add State Entry

Perform session processing

QOS Classification

Figure 4. Our integrated session table architecture. The first part of the table contains NAT information necessary to perform mapping between private and public addresses. Lan-addr and lan-port are the internally private addresses and port number while gwy-addr and gwy-port are the NATed publicly available


Routing

Transmit Packet

Figure 5. integrated Session table processing


54

CITACEE 2013

ISSN: 2338-5154

IV. CONCLUSION In this article, we presented the tree structure to simply packet filtering using decision tree that include in integrated session table which allow security edge router to performer the needed session processing (i.e. stateful filtering, NAT, QoS classification and routing). The use of our integrated session table produce great saving in router processing time and enhance its availability and scalability as it will be able to serve more traffic flows. The implemented integrated session table will be tested and performance test results will be analyzed. The final implementation will be integrated in a QoS cable integrated security gateway which assures high level of security protection and high availability for time critical traffic. REFERENCES [1] Mahmoud MOSTAFA, Anas ABOU EL KALAM, Christian FRABOUL, Extending Firewall Session Table to Accelerate NAT, QoS Classification and Routing, 19th International Conference on Computer Theory and Applications (ICCTA 2009), Alexandria, Egypt (2009)November 2009 [2] J. Moy, OSPF Version 2, IETF RFC 2328, April 1998. [3] Jinxian Lin, Huimin Li, "Optimization of the Routing Table Lookup Algorithm for IPv6,", in: Second International Conference on Genetic and Evolutionary Computing, 2008. WGEC '08. [4] T. Chiueh and P. Pradhan, "High performance IP routing table lookup using CPU caching," in Proc. Of IEEE INFOCOM'99, New York, NY, 1999. [5] Egevang, K. and P. Francis, "The IP Network Address


Translator (NAT)", IETF RFC 1631, May 1994. [6] Smith, M.; Hunt, R.; ―Network security using NAT and NAPT‖ in 10th IEEE International Conference on Networks, 2002. ICON 2002.27-30 Aug. 2002 Page(s):355 - 360 [7] Shenker, S. and J. Wroclawski, Network Element Service Specification Template. RFC 2216, September 1997 [8] David E. Taylor, ―Survey and taxonomy of packet classification techniques‖, ACM Computing Surveys. 2005. [9] John Evans and Clarence Filsfils, ―Deploying IP and MPLS QoS for Multiservice Networks: Theory and Practice‖; Morgan Kaufmann, 2007 [10] Kenneth Ingham and Stephanie Forrest, ―A History and Survey of Network Firewalls‖, Technical Report, TRCS-2002-37, University New Mexico, 2002. [11] Zwicky, E. D.; Cooper S. and Chapman D. B.: "Building Internet Firewalls", Orielly & Associates Inc., 2nd Edition, June 2000 [12] Al-Shaer, E.; Hamed, H.; Boutaba, R. and Hasan, M.:"Conflict Classification and Analysis of Distributed Firewall Policies", In IEEE Journal on Selected Areas in Communications, Volume 23, No. 10, pp. 2069 –2084, October 2005. [13] Siyan, Karanjit and Hare, Chris, "Internet Firewalls and Network Security", Indianapolis: New Riders, Publishing, 1995. [14] Xin Li, Zhenzhou Ji, Mingzeng Hu; ―Session Table Architecture for Defending SYN Flood Attack‖; in 7th International Conference Information and Communications Security, ICICS 2005. [15] Nichols, K., S. Blake, F. Baker, and D. Black, Definition of the Differentiated Services Field in the IPv4 and IPv6 Headers. RFC 2474, December 1998.


55

CITACEE 2013

ISSN: 2338-5154

A New Model of Information Processing based on Human Brain Mechanism: Toward a Cognitive Intelligent System Arwin Datumaya Wahyudi Sumari Asymmetric Warfare Study Program Faculty of Defense Strategy, Indonesia Defense University Jl. Salemba Raya No. 14, Jakarta – 10430, Indonesia

[email protected] Adang Suwandi Ahmad School of Electrical Engineering and Informatics Institut Teknologi Bandung, Jl. Ganesha 10, Bandung – 40132, Indonesia

[email protected] Abstract—To have a system than can think, act, and decide like that of human does has been a challenge for scientists, researchers, and practitioners since the invention of computer and electronic devices which support it. At the heart of a computer-based system is a matter that has task to process information and based on the processed information, the computer will act or decide on that information. At the next level, it is needed a system that is able to process information intelligently. In this paper, we introduce our research on developing a new model of information system extracted from human brain mechanism that we call as information-inferencing fusion. The approach on this model is a part of our research on Cognitive Artificial Intelligence to create a cognitive intelligent system. Keywords-Cognitive Artificial Intelligence; human intelligent; information-inferenceing fusion; information processing model, intelligent system

I.

INTRODUCTION

From one perspective, an intelligent system is defined as a system that brings human intelligent in its daily work in helping humans in their daily activities at home or at works. “Intelligent” can be meant as “having knowledge”. Therefore the term “intelligent” used in intelligent system can be viewed from many perspectives such as intelligent in recognizing faces or fingerprints in biometrics application, intelligent in making air maneuvers in aviation applications, intelligent in avoiding terrain obstacles in ground vehicle applications, and intelligent in making decision and processing information in Executive Support System (ESS) applications. Therefore a system that can be categorized as intelligent system is a system that has knowledge within its “brain”. The most important matter regarding this is to have knowledge the brain has to carry out something from something to produce something regarding that something. Back to our very special natural system given by our creator namely the brain. Let us observe an infant can recognize his/her environment. The process of this recognizing involves learning where within it lies an important mechanism within his/her brain called


information processing. At first gaze, he/she may not recognize who the people around him/her. As the time passes, he/she will recognize them as his/her parent. How can he/she get knowledge? Infant is a very good example for studying how knowledge is grown or constructed within the brain. As stated by [1], the best way to determine what, if anything, humans are born knowing, is to go straight to the source, and consult the recently born. An infant has knowledge from information received by his/her sensory organs. The received information is processed by the brain to obtain knowledge and by using the knowledge he/she can recognize his/her environment. The big question that has been challenging is what kind of information processing taken place within the brain that ends up with knowledge. If this mechanism can be extracted and emulated, the expectation to create a cognitive intelligent system is just a matter of time. One of our researches in Cognitive Artificial Intelligence field has resulted in a model of human thought called Sense-Inference-Decide&Act (SIDA) extracted from human brain mechanisme as depicted in Fig.1

Fig. 1 SIDA human thought model [2].

SIDA human thought model is based on our observation to the knowledge generation taken place on an infant, especially within the brain. In short, this model


56

CITACEE 2013

ISSN: 2338-5154

shows that information from the environment in term of phenomenon is perceived by sensory organs. The perceived information is then delivered to the brain to be processed, where within it the information is fused to obtain inference along with formulation necessary decisions. At the end of the loop, the formulated decisions are used the basis for making decision or action to the pheneomenon in environment. The cycle is then repeated with the same phenomenon or different one. The more sensory organs perceive the same phenomenon, the more information delivered to be processed by brain, and the more knowledge grown. This means that information processing holds a very important role. Based on a brief explanation has been delivered, the aim of the paper is to disseminate our research results in Cognitive Aritificial Intelligence field and the focus of the paper is to show our approach in creating a new model of information processing based on human brain mechanism. The arrangement of the paper is as follow. In Section I we have delivered a brief of why intelligent systems are needed and one of the foundations in building the core of an intelligent system, namely cognitive that taken place within an infant, the most natural human kind. Next, some of existing information processing models will be presented in Section II as the comparison to our model. In Section III the proposed information processing model as well as how it works will be delivered. The paper converges in Section IV with some concluding remarks. II.

MODELS OF INFORMATION PROCESSING

Essentially, information processing model is a theory of human development that uses the computer as a metaphor for explaining thought processes. Similar to computers, humans transform information to solve cognitive problems. Development is viewed in terms of changes in memory-storage capacities and use of different types of cognitive strategies [3]. On the other hand, information processing can be defined as the acquisition, recording, organization, retrieval, display, and dissemination of information [4]. Information processing model can be viewed from diverse disciplines ranges from psychology science to social and sport sciences. Therefore the models proposed are based on its own necessity. In this section we only take a look on three information processing models, namely Wicken’s model, Welford’s model, and Whiting’s model. A. Wickens' Model Wicken models the human information processing by dividing it into seven parts as depicted in Fig. 2 with explanations as follow [4]. • Sensory processing. The input (called stimuli) comes from all the human senses (also known as sensory systems). • Short-term sensory store. Each sensory system has a mechanism which prolongs any stimulus for a short time after the stimulus has occurred.


• Perceptual encoding. The stimuli are assigned to a single perceptual category. This is also known as detection, recognition, identification, categorization, pattern recognition, etcetera.

Fig. 2 Wickens' model of human information processing [5].

• Decision making and response selection. After categorizing the stimulus, human must decide what to do with it. • Response execution. The response (action) is executed. • Feedback and information flow. Human monitors the consequences of his/her action. • Attention. Human selects information sources to be processed, meaning pay more attention to the source(s) that deliver suspicious information. Wicken’s model is used for giving an introduction of some of the basics of cognitive science that apply to Human Factors design, which in this case is given to people who work in aviation industry. B. Welford’s Model Another better known model is Welford’s information processing model as depicted in Fig. 3. As summarized by [6], this model has three processes namely stimulus identification stage, response identification/ selection stage, and response programming stage. The elements of the model are as follow. • Sensory input. The sensory information relevant to the situation is stored in the short term memory. The information is taken in through the senses prior to a decision being made in three main ways. • Short and long term memories. All information gathered from the various sensory inputs is stored for a split second in the short term memory before it is processed. It is suggested that the short term memory can only hold up to seven pieces of information and that it is retained for less than a minute. The long term memory, which appears to have limitless capacity, contains information relating to past experiences. • Decision process. The decision process takes place by comparing the current situation, held in the short term memory, with previous experiences,


57

CITACEE 2013

ISSN: 2338-5154

held in the long term memory, to determine an appropriate action. • Action. The action is performed with reference to the movement pattern stored in long term memory. Once the action is completed, the situation and result are stored in the long term memory for future reference. The decision process is represented by a process named translation from perception to action which is a selection the most appropriate response from alternatives as the resulted of information processing. After the decision is made, the effector control controls which effector will do the response in term of action. This action is done to the phenomenon in the environment which is reperesented by external object.

• Effector mechanism which is the part of the brain which carries out the decisions and sends messages to the limbs and parts of the body via the nervous system. Effector mechanism is also called as Action. • Output which involves the effector mechanism and muscle movement. The nerves send messages to the muscles which move based on the observed phenomenon. • Feedback data that is information which is used during and after an action or movement which enables an agent to adjust or change performance according to this new information.

Fig. 3 Welford's hypothetical model of information processing [7].

Welford’s model is used commonly in sport applications because in sport an athlete has to learn and perform a wide range of perceptual motor skills and to select the appropriate skill action for a given situation [8]. Not different from Wicken’s model, this model takes account the memories to store the perceived information as well as information from decision process. C. Whiting’s Model Whiting’s model of information processing introduced in 1969 as comparison to Welford’s model. This model, as depicted in Fig. 4, has seven elements [9][10] with some relevan modifications as follow. • Display which refers to the range of actions and things that are happening in the surrounding environment of the performer. • Input data from display which involves information from the environment which enters the brain via the sense organs or sensory organs. • Receptor systems which refers to the sense organs which receive information • Perceptual mechanism which is the part of the brain which perceives the surroundings and gives them meaning. Perceptual mechanism is also called as Sensory Input. • Translatory mechanism which is the part of the brain which makes decisions and sorts out and processes the few relevant bits of information from the many inputs from the surroundings or environment. Translatory mechanism is also called Decision Process.


Fig. 4 Whiting’s model of information processing [11].

In this model we can see that the processing of information simply illustrated. It does not need to take account what is done by the memories. The most important matter in this model is the inputs from perceptual mechanism are directly processed by translatory mechanism to become knowledge to be delivered to effector mechanism. III.

A NEW INFORMATION PROCESSING MODEL

The three models have been presented in Section II just describe the mental process within human brain and heavily focused on how brain’s memories work in the existence of information or sensory inputs perceived by sensory organs or sensory system. In all models are not shown how perceived information is processed to obtain knowledge regarding the observed phenomenon or external object or something that causes stimuli (stimulus). Therefore, the intent of the paper is giving a new insight regarding a new information processing model which focuses on the mechanism of growing the knowledge within human brain. This mechanism is then extracted to be emulated on an agent and by using this mechanism it becomes intelligent cognitively or already brings an emulation of human mental proces. The process of the generation of knowledge within human brain is called as knowledge-growing [2].


58

CITACEE 2013

ISSN: 2338-5154

A. Where did We Start? Human brain instead of storing information for just-in time use in short-time memory or for later use in longterm memory, actually the perceived information is processed to generate new knowledge based on the existing knowledge or previous one. The new information from multi-information sources, namely sensory organs which are eyes (sight), ears (hear), nose (smell), tongue (taste), and skin (touch), is combined with an existing knowledge or a priori knowledge by a mechanism called information fusion. The result of the fusion is inferencing that will be used as the basis for making decision or action to the environment, namely the observed phenomenon. How new knowledge is generated in human brain by an information processing called information fusion is illustrated in Fig. 4.

environment from information multi-source, while recalling the information already stored in its brain. The new and existing information is then fused to obtain inferencing of the fused information. The inferencing becomes new knowledge that will be used as the basis for making decision of the most appropriate action to be taken by KGS with its effector to the environment.

Fig. 6 KGS is an intelligent agent that is equipped with knowledgegrowing mechanism based on SIDA cycle [13].

Fig. 5 A simple illustration of a human information fusion system [12].

B. SIDA Cycle Revisited Take a look back to Fig. 1 that depicts SIDA human thought cycle that has three stages with explanation follow. • Sense. This phase is represented by the activities of the sensory organs to sense and to perceive the environment (phenomenon) dynamics.

D. Knowlegde-Growing Mechanism – Cognitive Process Inference and Decision Formulation is the phase where knowledge-growing mechanism occurs through fusion or combination of new information with the existing knowledge in order to obtaining inferencing or new knowledge. The illustration of the knowledgegrowing mechanism simply presented in Table I. Multisensor terminology used in the table represents the five sensory organs belong to human. TABLE I.

THE ILLUSTRATION OF THE KNOWLEDGE-GROWING MECHANISM [14]

• Inference and Decision Formulation. This phase is represented by inference performed by the brain, and it is where knowledge-growing mechanism occurs. New knowledge is grown through fusion or combination of information delivered from Sense phase with the existing knowledge. • Decide&Act. This phase is where the product previous phase is implemented to affect the environment. C. Building a Cognitive Intelligent Agent Now we implement SIDA cycle to the agent in order to build a cognitive intelligent agen called KnowledgeGrowing System (KGS), that is, an agent that is already equipped with knowledge-growing mechanism. As we see in Fig. 1, by using SIDA cycle it is easy to follow how the knowledge is grown in KGS. At the first place, the KGS’ sensors sense and perceive information regarding the


Let us assume that δ = 1, …, i, …, n is the number of sensor or information source of information multi-source or multi-sensor, λ = 1,…, j, …, m is a collection of hypotheses or multi-hypothesis of phenomenon regarding the information supplied by the multi-sensor. At the end of the computation, λ is also functioned as the numbers of


59

CITACEE 2013

ISSN: 2338-5154

fused information from multi-sensor that explain a collection of individual phenomena based on the multihypothesis. Notation P (υi j ) represents the probability hypothesis j is true given information sensed and perceived by sensor i. The Degree of Certainty (DoC) [14], P (ψ 1j ) that hypothesis j is selected based on the fusion of the information delivered from multi-sensor, that is, from P (υ1j ) to P (υδj ) where j = 1,…, λ. The subscript “1” in notation P (ψ 1j ) means that the computation results are

DoC at time 1. The fusion to obtain a collection of DoC is given in (1). δ

∑ P (υ ) j

P (ψ

j 1

)=

i

i =1

δ

(1)

where P (ψ 1j ) ∈ Ψ and it is called as New Knowledge Probability Distribution (NKPD) [1]. The inferencing or new knowledge at this point can be obtained by applying (2).

P (ψ 1j )

estimate

.

= O  P (ψ 1j ) 

(2)

.

where O [...] = max [...] . This is a new notation proposed as a part of A3S (Arwin-Adang-Aciek-Sembiring) information-inferencing fusion method which is the basis for KGS knowledge-growing mechanism [15]. is inferencing of P (ψ 1j ) ∈ Ψ which later P (ψ 1j ) estimate

become new knowledge of KGS. The growing of knowledge over time is obtained by replace the first column of Table I to time parameter. The advancement of A3S method that already involves time parameter gave rise to new method called Observation Multi-time A3S (OMA3S) method and knowledge distribution resulted from the application of this method is called as New Knowledge Probability Distribution over Time (NKPDT) [16]. Back to Table I, the mechanism occurred in the table illustrates that the agent tries to recognize the observed phenomenon or external object by collecting information through its sensory organs (multi-sensor). Actually in the beginning it knows nothing about the phenomenon. Therefore when it uses different sensor, it perceives the same phenomenon as different one. By using information from several sensors simultaneously, it will have more complete knowledge regarding the phenomenon. The knowledge acquisition is carried out on some time interval using OMA3S in order to ensure that it has sufficient knowledge to recognize the phenomenon. The quality of the grown knowledge is measured using (3).


= DoC

P (θ j )

estimate

− φ1j × 100%

(3)

with P (θ j ) is information-inferencing of P (ψ γj ) after several observation time to the phenomenon and φ1j is the knowledge probability of the most correct fusedinformation j which represents the phenomenon at observation time γ 1 . IV.

CONCLUDING REMARKS

As stated in psychology community that at the very heart of cognitive psychology is the idea of information processing and it sees the individual as a processor of information, in much the same way that a computer takes in information and follows a program to produce an output [17]. Moreover, the focus of the research in information processing which results in some information processing models is in how human memories work on the information perceived by the sensory organs. Until the writing of this paper, we have not found any research on this field that study how the brain works to obtain new knowledge from the perceived information. By studying the existing models, we have created a new model of information processing which focused on the process of growing the knowledge within human brain. There three famous information processing models that have been used and studied. The two ones, Wicken’s and Welford’s model, involve memory models while the other one, Whiting’s model, gives more focus on how perceived information is directly translated into knowledge to become a basis of an action. As we can see that Whiting’s model almost similar with SIDA human thought model. Our model tries to extract what is taken place within decision process or translatory mechanism in all models, namely how the perceived information delivered by sensory organs is translated into knowledge. Meaning that the perceived information becomes inferencing or new knowledge because of a mechanism called knowledgegrowing by utilizing the power of information fusion. Knowledge-growing is a cognitive process. Therefore the application of knowledge-growing-based information processing model to an intelligent agent can transform it into a cognitive intelligent agent. We also have presented the mathematical model of knowledge-growing which is called A3S (Arwin-AdangAciek-Sembiring) information-inferencing fusion method along with Observation Multi-time A3S (OMA3S) method, that will be the core of our new information processing model. Our cognitive intelligent agent can be compared with cognitive intelligent agent in [18] and [19]. ACKNOWLEDGEMENT The authors would like to deliver great appreciation to Dr. Ir. Aciek Ida Wuryandari and Dr. Ir. Jaka Sembiring, M.Eng. of the School of Electrical Engineering and Informatics, Institut Teknologi Bandung for great discussions and supports in this challenging research.


60

CITACEE 2013

ISSN: 2338-5154

REFERENCES [1]

N. Angier, “Insights From the Youngest Minds”. [Online]. Available: http://www.nytimes.com/2012/05/01/science/insightsin-human-knowledge-from-the-minds-of-babes.html?pagewanted =all&_r=0. [2] A.D.W. Sumari, A.S. Ahmad, A.I. Wuryandari, and J. Sembiring, “Constructing brain-inspired knowledge-growing system: a review and a design concept”, Proc. of Second Int. Con. on Distributed Framework and Applications 2010 (DFmA2010), 2-3 August 2010, pp. 95-102. [3] “1: Historical Perspectives on Alternative Assessment”. [Online]. http://textbooks.brookespublishing.com/losardo/ Available: chapter1/keyterms.htm. [4] “information processing”. [Online]. Available: http://global.britannica.com/EBchecked/topic/287847/information -processing. [5] “Basics of human-machine interaction”. [Online]. Available: http://www.aerostudents.com/files/humanMachineSystems/basics OfHumanMachineInteraction.pdf. [6] “Wickens' Model”. [Online]. Available: https://www.hf.faa.gov/ webtraining/Cognition/CogFinal008.htm. [7] “Models of information processing”. [Online]. Available: https://sites.google.com/site/jacksrevisionwebsite/pe-revision/aspe/as-pe-theoretical-models/whiting-s-model. [8] A.T. Welford, Skilled Performance: Perceptual and Motor Skills. Palo Alto, CA: Scott, Foresman and Company, 1976 in “Decision-making and Motor Behavior”. [Online]. Available: http://www.kin.sjsu.edu/faculty/ewughalter/UG%20ML%20Decis ion-making%20and%20Motor%20Behavior.ppt. [9] “Decision Making”. [Online]. Available: http://castleschoolpe.wikispaces.com/file/view/Decision+making+ powerpoint.ppt. [10] “Information Processing”. [Online]. Available: http://www.blessedgeorgenapier.co.uk/media/downloads/303information-processing.pdf.

”


[11] “Models of information processing”. [Online]. Available: https://sites.google.com/site/jacksrevisionwebsite/pe-revision/aspe/as-pe-theoretical-models/whiting-s-model. [12] A.D.W. Sumari, “The modeling of knowledge growing system based on multi-time observation A3S (OMA3S)”, Tech. Rep., Institut Teknologi Bandung, 2009. [13] A.D.W. Sumari, “Knowledge-growing system basics: a review on human thought models for building an intelligent agent”, Proc. of Conf. on Information Technology and Electrical Engineering 2010 (CITEE2010), 20 July 2010, pp. 25-31. [14] A.D.W. Sumari, “The modeling of Knowledge-Growing System based on Observation Multi-time Arwin-Adang-Aciek-Sembiring (OMA3S) method,” EL-8095 Technical Report, School of Electrical Engineering and Informatics, Institut Teknologi Bandung, May 2009. [15] A.D.W. Sumari, A.S. Ahmad, A.I. Wuryandari dan J. Sembiring, A Mathematical Model of Knowledge-Growing System: A Novel Perspective in Artificial Intelligence, Proc. of IndoMS Int. Conf. on Mathematics and Its Applications 2009 (IICMA2009), Gadjah Mada University, 12-13 October 2009, pp. 229-240. [16] A.D.W. Sumari, A.S. Ahmad, A.I. Wuryandari, and J. Sembiring, “The application of knowledge-growing system to multiagent collaborative computation for inferring the behavior of genes interaction,” Int. J. of Computer Science & Network Security (IJCSNS), Vol. 9, No. 11, pp. 82–92, November 2009. [17] S. McLeod, Information Processing”. [Online]. Available: http://www.simplypsychology.org/information-processing.html. [18] N. Popescu-Bodorin and V.E. Balas,” From Cognitive Binary Logic to Cognitive Intelligent Agents”. [Online]. Available: http://arxiv.org/pdf/1106.5995v1.pdf. [19] X. Xu, “Biopsychically Inspired Cognitive System for Intelligent Agents in Aerospace Applications”. [Online]. Available: http://www.ohio.edu/people/starzykj/network/Research/papers/Re cent%20conferences/ABiologicallyInspiredCognitiveSystem.pdf.


61

CITACEE 2013

ISSN: 2338-5154

Efficient Work Stealing for Portability of Nested Parallelism and Composability of Multithreaded Program Adnan, Zahir Zainuddin, Wardi Electrical Engineering Dept Universitas Hasanuddin, Indonesia [email protected]

Abstract We present performance evaluations of parallel-for loop with work stealing technique. The parallel-for by work stealing transforms the parallel-loop into a form of binary tree by making use of method of divide-and-conquer. Iterations are distributed in the leaves procedures of the binary tree, and the parallel executions are performed by stealing subtrees from the bottom of the tree. The work stealing and divide-and-conquer are used to address the portability problem in nested parallelism and composability. By work stealing and divide-and-conquer techniques, fine-grained parallel-for can be implemented without contributing large work overhead. Low work overhead is important as the number of processor could be less than expected. Low overhead and fine-grained of work stealing scheduler makes highly parallel processor cores are able to scale the performance. In addition, the approach used in this work makes efficient nested parallelism is possible. Because of a low overhead, we show that the work stealing and divide-and-conquer deliver good scalability in nested parallel Sparse LU factorization. Keywords Composability, nested-parallelism, work-stealing, multicore

1.

Introduction

As many-core processors are available recently, composability is an increasingly important feature of programming languages. Composability refers to the ability to develop a parallel program from its components. The components themselves can be executed by a single thread or multiple threads. Whenever threads make a call to modules where parallelism exists, they may launch new threads as children for that modules. Composability allows modular parallel programming. This means that a programmer will not need to have low-level knowledge about the parallelism that exists within a subroutine. Instead, the programmer simply inserts calls to functions and lets the runtime system schedule the parallelism within the function dynamically. Therefore, composability brings as a consequence the need for nested parallelism. In a program with nested parallelism, there is a possibility that a program will lack parallelism at the outer loop while parallelism is ample in the inner loop. In such cases, threads must be scheduled efficiently for the sake of

the parallelism that exists in the inner loop. Also, it is sometime optimized parallelism for specific number of processor cores is not portable. Therefore, nested parallelism must be performed automatically at runtime. Without efficient implementation of thread scheduling, the nested parallelism that constitutes composability contributes to a large overhead. For instance, an OpenMP thread must launch a set of new threads, which contributes to overhead, whenever it encounters a nested work-sharing construct. In meanwhile, load imbalances may sometime cause performance degradation because thread assignment for nested parallelism is static. Therefore, it is a strong requirement for the programmer to make a careful assignment of a number of threads in such situations. Low overhead and flexibility of work stealing technique are the key factors which support the nested parallelism. Work stealing is an efficient multithreaded programming technique. Parallel execution based on work stealing contributes to small overhead. The overhead is as small as the cost of function call, and threads scheduling of work stealing is flexible. Threads can be scheduled for the parallelism which may exist in anyplace of parallel program. Therefore, we are motivated to make use of work stealing as the base of this work. As the instance of work stealing, StackThreads/MP[12] was used to implement efficient parallel-for which capable of efficient nested parallelism. This paper concerns the portability problem of nested parallelism and composability. In this paper, a form of parallel loop was implemented. The numbers of parallel threads assigned to the loop do not need to be specified. Instead of the numbers of threads, a number of maximum sub iterations assigned for each thread that may run in parallel is specified. A reducer object is featured in the parallel loop. The reducer object can be applied to any data type and size. The contribution of this paper are as follows :

• We demonstrated that parallel-for which is implemented us-

ing divide-and-conquer and work stealing is a solution for the portability of nested parallelism and composability

• We evaluated the performance of nested parallel Sparse LU

factorization using parallel-for.

The remainder of this paper is organized as follows. We discuss some related works of this paper on section 2. We discuss the portability of nested parallelism and composability problems in section 3. We discuss work stealing based execution in section 4. In section 5, we discuss performance evaluation and its result in nested parallel Sparse LU factorization. We conclude this work on section 7.



62

CITACEE 2013

2.

ISSN: 2338-5154

Related Works

Many work stealing[1] techniques work based on lazy task creation, including Cilk[2, 3] and StackThreads/MP[6, 11, 12]. Cilk and StackThreads/MP create logical threads and allow the scheduler to schedule bottommost threads so that any idle worker can steal them. Unlike Cilk and StackThreads/MP, Tascell[5] is logicalthread-free technique that also provides a work stealing capability. However, Tascell only allows work stealing after receiving a steal request while performing polling, which is similar to StackThreads/MP. We implemented parallel-for that built on top the StackThreads/MP. Tanaka[10] implemented nested parallelism for OpenMP via StackThreads/MP for Omni OpenMP[7]. This work also described implementation of a parallel team of threads as logical threads created by workers using divide-and-conquer recursion. Omni OpenMP specifies the number of logical threads that a parent thread creates but our parallel loop implementation specifies a lowerbound and upper-bound of a maximum number of iterations for loop control variable for each logical thread. We do not specify the exact number of threads a parent thread creates. Therefore, a programmer would be allowed to implement a parallel-for that resembles the serial form of the loop-for. Another loop parallelization based on work stealing is the Window algorithm[13]. The Window algorithm differs from divideand-conquer recursion in that divide-and-conquer divides sequences of n elements into n/m chunks of elements that are adjacent. Parallel-for is one natural form of regular parallelism, and the OpenMP [8] is one popular standard that features a parallel-for definition. OpenMP defines that implementations have a scheduler that schedule parallel team of threads to the iteration space. Cilk++ implemented cilk for by means of divide and conquer recursion to provide more parallel slackness and better load balancing. Cilk++ implemented reducers as an C++ hyperobject[4] independently from cilk++ runtime system. As a hyperobject Cilk++ reducer work flexible in both regular and irregular parallelism. We currently implemented reducer as a c structure. We implemented reducer so that it tightly coupled to divide-and-conquer recursion. Together with parallel for,Intel Threaded Building Block[9] also provided a parallel reducer.

3.

Portability of Nested Parallelism and Composability Problem

In this section, we discuss parallel-for and compare two different techniques which are used to implement. The technique are work sharing and work stealing. 3.1

Portability of Nested Parallelism

One common form of parallel construct in many parallel languages is a parallel form of loop-for. For example, one expresses parallelloop using a pragma of omp for in OpenMP. Parallel loop-for of OpenMP is one of construct of OpenMP work sharing. In work sharing, there is a parallel region of threads which are created using a pragma omp parallel. A programmer may use then the pragma omp for within a parallel region or use the pragma omp parallel for. The latter is a combination the both of the pragmas. Runtime systems assigned parallel team of threads separately to the different chunks of iterations. The main task of programmers is to statically decide a number of assigned threads by either num threads() clause or OMP NUM THREADS. For the nested parallel-for of OpenMP, each parallel loop must be written within a parallel region. Therefore, the degree of parallelism on both outer and inner parallel regions are determined by either the OMP NUM THREADS environment variable or the num threads() clause. Code in Fig. 1 shows an example of nested


1. 2. 3. 5. 6. 7 8.

#pragma omp parallel for num_threads(12) for(ii=0;ii