Peramalan Isyarat Satelit Menggunakan Kaedah Rangkaian Neural

Satellite communication system is one of the communication methods that are very important and widely used nowadays. Throughout this communication system, we are able to communicate at any location all over the world because of its high capabilities. However there are certain factors that can...

Full description

Saved in:
Bibliographic Details
Main Author: Kahar, Nur Farhan
Format: Monograph
Language:English
Published: Universiti Sains Malaysia 2006
Subjects:
Online Access:http://eprints.usm.my/58788/1/Peramalan%20Isyarat%20Satelit%20Menggunakan%20%20Kaedah%20Rangkaian%20Neural_Nur%20Farhan%20Kahar.pdf
http://eprints.usm.my/58788/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Satellite communication system is one of the communication methods that are very important and widely used nowadays. Throughout this communication system, we are able to communicate at any location all over the world because of its high capabilities. However there are certain factors that can be interfering with the satellite communication systems. One of the causing factors is the volatile weather condition such as rain, high level of humidity, temperature and wind speed. Thus, in this project, research for all causes above has been done in order to produce a smarter system that can solve this problem. The principal of artificial intelligence that has been applied in neural network is the best and practical solution method to be used. The neural network system that has been design in this project is able to make an accurate prediction in the quality of received signal based on weather condition. The objective for designing this system can be explained in the following example. If the system detects rain that causes interference in the wave transmission path, a signal will be sent to the satellite to amplify the transmitting signal. This can prevent communication interference in the satellite communication even in an unpredictable weather condition. This project has produced a neural network system that achieves 91.7% accuracy through the use of Bayesian Regularization algorithm. Thus, we can say that the system is able to make an accurate and convincing estimation. Regression analyses are also done to all the data that has been used. The analysis shows that rain and wind speed are the main factor, and gave the biggest impact to the quality of received satellite signal. As a conclusion, the completed project has achieved its objective that is to produce a smart and effective solution in solving problems for the satellite communication systems.