Full Record

New Search | Similar Records

Author
Title The Application of Evolutionary Algorithms for Energy Efficient Grooming of Scheduled Sub-Wavelength Traffic Demands in Optical Networks
URL
Publication Date
Degree MS
Discipline/Department Computer Science
Degree Level masters
University/Publisher University of Windsor
Abstract In recent years there has been a growing recognition of the need for developing energy efficient network design approaches for WDM backbone networks as well. The typical approach has been to switch off some components such as line cards and router ports during low demand periods, and has focussed on traditional static and dynamic traffic models. In this paper, we present a new approach that exploits knowledge of demand holding times to intelligently share resources among non-overlapping demands and reduce the overall power consumption of the network. We consider the fixed-window scheduled traffic model (STM), and present i) a Genetic Algorithm (GA) and ii) a Memetic Algorithm (MA) based strategy that jointly minimizes both power consumption and transceiver cost for the logical topology. Simulation results clearly demonstrate that both of the proposed algorithms outperform traditional holding time unaware (HTU) approaches; the GA leads to additional improvements even compared to the shortest path holding time aware (HTA) heuristic. However, the MA manages to achieve similar results to the GA while taking up 4 to 5 times less computational resources and time to compute.
Subjects/Keywords Applied sciences; Genetic algorithm; Memetic algorithm; Optical networks; Traffic grooming
Contributors Arunita Jaekel
Language en
Rights info:eu-repo/semantics/openAccess
Country of Publication ca
Format application/pdf
Record ID oai:scholar.uwindsor.ca:etd-5878
Other Identifiers TC-OWA-4872
Repository windsor
Date Retrieved
Date Indexed 2020-01-07

Sample Search Hits | Sample Images

…24 2.5.1.2 Classifications . . . . . . . . . . . . . . . . . 26 2.5.1.3 2.6 17 2.4.0.2 2.5 Representation and Fitness . . . . . . . . . . Algorithm . . . . . . . . . . . . . . . . . . . 27 Approaches to Optical Network Optimization…

…39 3.4 Genetic Algorithm . . . . . . . . . . . . . . . . . . . . . . . . 41 3.4.1 GA Based Energy Minimization for Scheduled Traffic 42 3.4.1.1 Chromosome representation . . . . . . . . . . 42 3.4.1.2 Initial population…

…48 3.5.1 Local Search . . . . . . . . . . . . . . . . . . . . . . . 52 3.5.2 Mutation . . . . . . . . . . . . . . . . . . . . . . . . . 54 3.5 4 Experimental Results 59 4.1 Energy consumption within Genetic Algorithm . . . . . . . . 60…

…find its solution. The proposed Genetic Algorithm (GA) and Memetic Algorithm (MA) are able to achieve moderate energy improvements at a much shorter time, and using up less computational resources. On the other hand, our MA consumes…

…The repeaters convert the signal into an electrical signal, and then use a transmitter to send the signal again at a higher density than it was before. These repeaters tend to be very expensive due to the high complexity with modern WavelengthDivision…

…short, variation builds up genetic diversity, while selection reduces it. In terms of evolutionary computing, selection is accomplished with any algorithm that favours data structures with a higher fitness score. There are many possible methods to…

…space of potential 13 solutions to a specific problem. The population evolves towards improving regions of the search space by means of randomized processes of selection, mutation and recombination (sometimes recombination is not used in some…

…evaluation step, and ι : I µ → {true, f alse} us used to denote the termination criterion. Thus, we can represent this with the following algorithm: begin t = 0; − − initializeP (0) = {→ a 1 (0), ..., → a µ (0)…

.