Multi-criteria evaluation and benchmarking for Active Queue Management methods of network congestion control-2
<p>This research aimed to propose a benchmarking decision matrix for the Active Queue Management (AQM) methods of network congestion control based on multi-criteria analysis to aid the developers of AQM methods to make the right decision of selecting the best AQM method. In this study,...
Saved in:
| Main Author: | |
|---|---|
| Format: | thesis |
| Language: | eng |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://ir.upsi.edu.my/detailsg.php?det=8670 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <p>This research aimed to propose a benchmarking decision matrix for the Active Queue Management (AQM) methods of network congestion control based on multi-criteria analysis to aid the developers of AQM methods to make the right decision of selecting the best AQM method. In this study, an experiment was conducted on the basis of several stages. First, decision matrix was proposed for selecting suitable AQM methods based on multi criteria (performance, process overhead and configuration), with each criterion has several sub criteria (Throughput, Mean Queue Length, Drop Rate, Packet Loss, Delay, Time, Space, Estimated Calculation, Sensitivity). In addition, six AQM methods of alternatives were used. Subsequently, the ranking of the AQM methods was utilized by the developed decision matrix using Multi Criteria Decision Making (MCDM) techniques, namely, the Analytic Hierarchy Process (AHP) to weight the evaluation criteria, and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) was used to benchmark and rank the AQM methods. TOPSIS has been applied in two decision-making contexts: individual and group decision making (GDM), as well as in GDM, internal and external group aggregation has applied where internal aggregation is receiving the higher ranked value of 62.50% for RED method, which is ranked first in GDM. Data consisting of three main criteria as the required criteria were collected by developing a Sub-Process that is responsible for implementing the AQM methods to generate the data that used in the constructed Decision Matrix. The research findings showed that the integration of Multi-Layer AHP and Group-TOPSIS was effective in solving the problems associated with the selection of AQM methods, as evidenced by the systematic ranking of six AQM methods. In conclusion, the internal and external aggregations of Group TOPSIS used in different contexts were able to generate the results of AQM method ranking that were similar. The implication of the study is that the AQM developers could use such a novel technique to make the right decision of selecting the best AQM method to prevent the router congestion and improve the performance of the computer networks as a whole</p> |
|---|