Using Genetic Algorithms to OptimizePreplanned Backup Routes in DWDM Network
| Autor: | Chin-Hung Kuo, 郭志宏 |
|---|---|
| Rok vydání: | 2002 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 90 Any failure of optical node or link in a Wavelength Division Multiplexing (WDM) network may cause it difficultly to transmit data. For maintaining the normal operation of the network, it is important to find the optimal backup restoration path that can solve the problems of the optical node or link failure. Therefore, restoration of WDM network becomes an essential work. The preplanned restoration is one of the common methods we considered to protect or restore the network failures. In this paper, a genetic algorithm is proposed to finding optimal backup lightpath. Simulation results indicate that GA is robust for this problem. Four types of crossover operators (SPC, TPC, FPC, EPC) and four types of mutation operators (SPM, TPM, FPM, EPM) are employed in our genetic algorithm. We also show that parameter setting of Pc and Pm are very important. It will have better performance by suitable parameter setting. We proposed MGA to solve the convergence problem. In MGA, we generate initial populations of GA using most-decent search algorithm. It may have higher probability to avoid convergence. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|