Claim Missing Document
Check
Articles

Found 4 Documents
Search

Peningkatan Efisiensi Penggunaan Bahan Bakar Kendaraan Bermotor Dengan Melakukan Pencarian Jarak Terdekat Menggunakan Jaringan Syaraf Tiruan Hopfield Di Wilayah Purwokerto Agung Mubyarto; Hesti Susilawati
Dinamika Rekayasa Vol 6, No 1 (2010): Dinamika Rekayasa - Februari 2010
Publisher : Jenderal Soedirman University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20884/1.dr.2010.6.1.30

Abstract

This proposed research concern with application of continuous Hopfield neural networks for shortest path routing in a given map, in this case Purwokerto city’s map had choosen as an example. The method that was used in this research based on the ability of continuous Hopfield neural networks for representing their neuron as choosen vertices in the given map. The result shows that the networks can produce optimal route for such source and destination node pairs. As the product of this research, we have knowledge about the shortest way from one to another location at Purwokerto, so we can increase the efficiency on fuel use by using this route
Estimasi Kinerja Kanal MIMO-OFDM Menggunakan Modulasi Adaptif Untuk Teknologi LTE Berdasar Pada Kecepatan Pelanggan Widhiatmoko Hery Purnomo; Hesti Susilawati; Ari Fadli
Dinamika Rekayasa Vol 11, No 1 (2015): Jurnal Ilmiah Dinamika Rekayasa - Februari 2015
Publisher : Jenderal Soedirman University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20884/1.dr.2015.11.1.89

Abstract

OFDM dipakai untuk teknik multiple accesspada downlink LTE. Dalam teknologi LTE, parameter kecepatan user sangat berpengaruh terhadap kualitas jaringan.Pada OFDM, estimasikanalsangatdiperlukan untuk mengatasi fading dan noise.Semakin besar kecepatan user maka semakin besar nilai fading dan noise.Untuk mengatasi masalah tersebut diperlukan estimasi kanal yang digunakan. Metode estimasi kanal yang banyak digunakan ada dua yaitu metode Least Square (LS) dan metode Minimum Mean Square Error (MMSE). Kedua metode tersebut digunakan pada penerima diam dan penerima bergerak dengan modulasi QPSK, 16-QAM, dan 64-QAM. Simulasi dilakukan untuk mengetahui metode estimasi kanal pada beberapa variasi kecepatan user. Hasil analisa menunjukkan bahwa estimasi kanal dengan metode MMSE lebih baik dibandingkan metode LS. Sedangkan pada penerima dengan kecepatan rendah memiliki kualitas jaringan yang lebih baik dibandingkan pada penerima dengan kecepatan tinggi.
Prediksi Penyakit Demam Berdarah Dengue (Dbd) Dan Tifus Menggunakan Jaringan Syaraf Tiruan Perambatan Balik (Back Propagation) Farida Asriani; Hesti Susilawati; Lucyana Harvianti
Dinamika Rekayasa Vol 3, No 1 (2007): Dinamika Rekayasa - Februari 2007
Publisher : Jenderal Soedirman University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20884/1.dr.2007.3.1.122

Abstract

Dengue Haemorrhagic Fever (DHF) and Typhus are diseases which often knock over Indonesian society. Both of these diseases, especially DBD is malignancy that able to result death if losing time to be diagnosed. But civil society, even a doctor which is expert, sometimes also difficult to detect DHF diagnosed earlier, because symptom initially, tending to look like with other acute diseases. One of the diseases that have symptom like DHF is typhus. Back Propagation Artificial Neural Network (ANN) is one of artificial intelligences that can be applicated in the field of health, especially for predicting a disease type. By using input in the form of symptoms the naturalness by a patient, expected this Back Propagation ANN system can help the doctors to diagnose the diseases that happen to their patients, especially DHF and Typhus diseases. ANN system that is designed to predict DHF and typhus diseases, is trained with back propagation algorithm that using adaptive learning rate and momentum. From this research, is got ANN system with network architecture that consist of: one input layer with 18 neurons, one hidden layer with 125 neurons, and one output layer with 1 neuron. And also, value of constanta momentum is 0.95.This ANN system has good performance. From 400 respondents that use for training, result of the memorizing examination reaches percentage equal to 100%, and result of the generalizing examination reaches 96%, from 150 new respondents
Initial Tuning Sebagai Salah Satu Metode Pengoptimalan New Site Pada Jaringan Seluler Gsm (Studi Kasus Pada Area Kerja Pt. Sinergi Telecom Under Ericsson Network ) Hesti Susilawati; Azis Wisnu Widhi Nugraha; Arif Widodo
Dinamika Rekayasa Vol 3, No 1 (2007): Dinamika Rekayasa - Februari 2007
Publisher : Jenderal Soedirman University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20884/1.dr.2007.3.1.124

Abstract

The development of new sites by telecommunication network operator, especially GSM network, purposes for more increase performance and give satisfy for consument. Performance of a site is most affecting for value of network quality totally. New site builded must be monitorized and optimized by its existing in order to can be more realible on serving society with the method that was mentioned “initial tuning activity”. On this method must be concerned some matter as like: application of accurate frequency for avoiding interference problem, creating neighbour relationship between ex isting site and new site, detection for faulth installation for arranging re-use frequency concept, setting accurate coverage, olso changing tilt angle for accurate main beam area. This arrangement must be conditioned on a new site that still “default”, so it can improve quality. On operator side, they have some criteria that must be passed by a new site. Initial tuning activity on Site Simpang RGM had been proved on passing  that site from recommended criteria, olso being totally indicated more increase pe rformance of network