EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Moh. Fandika Aqsa


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Master Theses » Teknik Kes & Keselamatan Kerja
Posted by erna at 08/01/2007 11:33:50  •  5593 Views


KINERJA MULTI-USER DETECTION PADA SISTEM KOMUNIKASI DS-CDMA SINKRON MENGGUNAKAN BACK PROPAGATION -NEURAL NETWORKS

PERFORMANCE OF MULTI-USER DETECTION FOR SYNCHRONOUS DS-CDMA COMMUNICATION SYSTEM USING BACKPROPAGATION - NEURAL NETWORKS

Author :
MILCHAN,MUHAMAD 




ABSTRAK

Pada penelitian ini akan dibandingkan kinerja dari detektor multiuser neural network dengan detektor konvensional matched filter dan detektor multiuser optimal. Ada beberapa skenario simulasi untuk membandingkan kinerja dari masing-masing detektor tersebut meliputi kanal AWGN kanal flat Rayleigh fading kapasitas sistem dan efek near far. Adapun kinerja dari masing-masing detektor diberikan dalam bentuk nilai probabilitas kesalahan bit sebagai fungsi nilai SNR kecuali pada skenario kapasitas sistem sebagai fungsi jumlah pengguna. Detektor multiuser optimal memakai urutan deteksi maximum likelihood yang di-implementasikan dengan algoritma dynamic programming. Sedangkan detektor multiuser neural network menggunakan multilayer perceptron dengan satu layer hidden. Untuk mengimplementasikan neural network dan mentrainingnya digunakan algoritma back propagation. Hasil simulasi pada kanal AWGN memperlihatkan bahwa detektor multiuser optimal mempunyai kinerja yang sangat baik karena tidak terpengaruh oleh multiple access interference MAI. Sayangnya kompleksitas dari detektor multiuser optimal akan berkembang secara eksponensial dengan makin bertambahnya jumlah pengguna. Sedangkan untuk detektor multiuser neural network mempunyai kinerja yang mengalami perbaikan yang signifikan dibanding detektor konvensional matched filter dan mendekati optimal. Pada hasil simulasi. dengan kanal flat Rayleigh Fading dengan kontrol daya sempurna detektor neural-network backpropagation tetap memberikan perbaikan yang signifikan dibanding dengan detektor matched filter konvensional dan juga memberikan perbaikan dibanding detektor multi user optimal. Meskipun kinerja-nya mengalami perbaikan tetapi probabilitas kesalahan bit nya masih belum dapat mencapai standar komunikasi digital untuk voice. Pada skenario efek near far menunjukkan bahwa detektor multiuser neural network lebih resistance terhadap efek near far dibandingkan dengan detektor multi-user optimal dan detektor konvensional matched filter sangat resistance terhadap efek near far.


ABSTRACT

This research will be compared the performance of neural network multiuser with conventional matched filter and optimum multiuser detector. Various scenarios will be investigated. These scenarios include AWGN channel flat Rayleigh fading channel system capacity and the near-far effect. The metric for evaluating performance is the probability of bit error achieved at certain signal-to-noise ratio except the system capacity scenario at certain number of users. Optimum multiuser detector uses maximum likelihood sequence detection that is implemented by a dynamic programming algorithm. Neural network multiuser detector on this research by using multilayer perceptron with one hidden layer. For implementation neural network and training of the network is done with backpropagation algorithm. The simulation results in AWGN channel to show the performance of optimum multiuser detector is excellent because multiple access interference MAI no influence on it. Unfortunately complexity of the optimum multiuser detector is the exponential is the number of users. The performance of neural network multiuser detector has a significant improvements over the conventional matched filter detector and near the optimum. In the flat Rayleigh fading channel neural network multiuser detector has a better performance than the optimum multiuser detector and can improve performance significantly over the conventional matched filter detector. However probability of bit error is not enough to cover digital communication standard for voice. In the near far effect scenario to show that the neural network multiuser more resistant to the near far effect than the optimum multiuser detector and the conventional matched filter detector is most resistant



Keywordsmatched filter ; near far
 
Contributor
  1. Dr. Ir. Mauridhi Hery Purnomo, M.Eng. Ir. Suwadi, MT
Date Create: 08/01/2007
Type: Text
Format: pdf;71 pages
Language: Indonesian
Identifier: ITS-Master-3100003017705
Collection ID: 3100003017705
Call Number: 621.384 56 Mil


Source
Theses Teknik Kes & Keselamatan Kerja RT 621.384 56 Mil, 2002

Coverage
ITS Community

Rights
Copyright @2005 by ITS Library. This publication is protected by copyright and permission should be obtained from the ITS Library prior to any prohibited reproduction, storage in a retrievel system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. For information regarding permission(s), write to ITS Library




[ Download - Summary ]

ITS-Master-3100003017705-866.pdf




 Similar Document...

     No similar subject found !


! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 35.175.248.25
using CCBot/2.0 (https://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2019 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan