Date of Award

2025-12-01

Degree Name

Master of Science

Department

Electrical and Computer Engineering

Advisor(s)

Michael P. McGarry

Abstract

This research investigates the performance of the Network Link Outlier Factor with Most Likely Links (NLOF:MLL), under varying network load conditions. Earlier studies reported that the NLOF:MLL algorithm experienced a noticeable drop in fault-localization accuracy when operating in lightly loaded networks. To further examine this limitation, 240 experiments were carried out to observe how the algorithm responds as overall network load increases. The evaluation focused on the classification performance metrics: precision, recall, and F1-score. The results show that NLOF:MLL’s effectiveness improves as network load increases but that the rate of improvement slows progressively, eventually stabilizing in a pattern consistent with a logarithmic trend. The reduced performance observed under low network load occurs because low throughput range limits the variability within the NLOF:MLL clustering window, resulting in weak flow differentiation and unstable flow outlier detection. Overall, this research provides statistical evidence that NLOF:MLL follows a smooth, saturating performance curve rather than an abrupt transition, thereby clarifying the nature of its “light-load penalty” and offering insight into where refinement is needed to mitigate this penalty.

Language

en

Provenance

Received from ProQuest

File Size

63 p.

File Format

application/pdf

Rights Holder

Sunday Oluwaleke Ogundele

Download

Share

COinS