Document Details

Document Type : Thesis 
Document Title :
A COMPARATIVE STUDY AND IMPLEMENTATION OF LIGHT PATH RESTORATION ALGORITHMS USED IN WAVELENGTH-DIVISION-MULTIPLEXING
دراسة مقارنة و تنفيذ خوارزميات استعادة المسار الضوئي المستعملة في الشبكات الضوئية التي تعدد المقابلات بتقسيم الطول الموجي
 
Subject : A COMPARATIVE STUDY AND IMPLEMENTATION OF LIGHT PATH RESTORATION ALGORITHMS USED IN WAVELENGTH-DIVISION-MULTIPLEXING 
Document Language : Arabic 
Abstract : Optical WDM networks represent the answer to the rise in bandwidth demands. These networks require protection and restoration algorithms in case of failure to provide reliable and continuous data traffic. This thesis is concerned with some Linear Programming and Mixed Integer Programming algorithms that are used in solving restoration and protection problems in WDM optical networks under the assumption of a single link failure. The thesis starts with a brief introduction to optical networks and restorations. Linear programming is also introduced in brief and explained with some solved examples using graphical illustrations and the MATLAB language. Then the work goes on solving a square and a cube network manually for the non-diversity and diversity algorithms. Results are given to demonstrate the cost increments required by the diversity algorithm against the reliability improvement achieved by it. Again, the non-diversity and diversity algorithms are introduced but through the AMPL modeling language, whose name, an abbreviation of “A Mathematical Programming Language”, indicates that it is a particularly suitable and superior language for this kind of applications. Some of the problems that have been solved manually are implemented and solved using AMPL. The implementation is done through two files, the model file which includes the declarations of parameters and constraints and the target function needed to be minimized, and the data file which contains the data and numerical values required to solve the required problem. The code of these files is listed for all used algorithms. Then the run of the programs is showed and the final results are compared and shown to be in total agreement with the results of the manually solved problems. 
Supervisor : Prof. Ali Muhammad Rushdi 
Thesis Type : Master Thesis 
Publishing Year : 1428 AH
2007 AD 
Added Date : Monday, March 2, 2009 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
حسن فايز حكميHAKAMI, HASAN FAYEZResearcherMaster 

Files

File NameTypeDescription
 18760.pdf pdf 

Back To Researches Page