Development of scalable hybrid switching-based software-defined networking using reinforcement learning by Blose, Max

“…We are therefore of the opinion that the hybrid switching model proposed utilizing machine learning algorithms can address the scalability issues in the design of SDN controller networks, particularly in data centre environments where high switching speeds are of paramount importance.…”
Get full text

Novel motion capture methods for sports analysis : case studies of cycling and rugby goal kicking by Cockroft, Stephen John

Get full text

Development of containerized pipelines for the reproducible analysis of amplicon-, shotgun metagenomic- and metatranscriptomic data. by Madzime, Ruvarashe Joylyne

Get full text

Mathematical optimisation of the suspension system of an off-road vehicle for ride comfort and handling

Get full text
Get full text

Efficient access network selection and data demand prediction for 5G systems by Paul, Udita

“…Firstly, an algorithm is proposed to select the best access network for a subscriber running single or group of applications. …”
Get full text

Analysis of land cover changes of the Hartebeesthoek Radio Astronomy Observatory environment over five decades

Get full text

The analysis of distributed resources on a load sharing reticulation network

Get full text

The stoichiometry and kinetics of biological heat generation in the aerobic stage of dual digestion by Messenger, John

“…In this thesis, the results obtained pertaining to operation and performance of the Milnerton aerobic reactor are presented in detail. …”
Get full text

Aggregated service consolidation by multi-server virtualization by Nkubito, Phillip Uwizeye

Get full text

Transformerless series dip/sag compensation with a multilevel cascaded inverter by Visser, Abraham Johannes

Get full text

Niching strategies for particle swarm optimization

Get full text
Get full text

Hydraulic Data Preprocessing for Anomaly Based Intrusion Detection on SCADA Level of Water Treatment Systems by Mboweni, Ignitious

“…The best performance was achieved when the preprocessed dataset was randomly split into training and testing, yielding a significant improvement in accuracy, F1 score and time to detection for both algorithms used in the study, namely Fine Tree and Boosted Trees Ensemble.…”
Get full text

Improved explainability through uncertainty estimation in automatic target recognition of SAR images

Get full text

Blaze-DEM : a GPU based large scale 3D discrete element particle transport framework

Get full text

Screening for Proteinopathic-related Dementias in Low-resource Clinical Contexts: A Machine Learning Approach by Lewis, Raphaella

“…Results: The C5.0 algorithm reduced the number of clinical variables for screening PRDs from 20 to 9 (Repeatable Battery for the Assessment of Neuropsychological Status [RBANS] Figure Recall, vascular risk factor, phonemic verbal fluency, RBANS List Recall, RBANS List Recognition, Digit Span Backward, CLOX1, CLOX2, and ∆CLOX2-1), and classified the validation sample with an accuracy exceeding chance performance. …”
Get full text

Slotted time based protocol and polling strategies for low speed, narrow band applications by Ragadza, Ntakadzeni Ronald

Get full text

Ubiquitous mesh networking : application to mobile communication and information dissemination in a rural context by Maliwatu, Richard

Get full text

Detecting Electrical Submersible Pump (ESP) Failures and Estimating Run Life Using Artificial Neural Networks by Sobhy, Mostafa Ahmed

“…Concurrently, the failure classification module identifies ten distinct failure types with F1 scores exceeding 0.95, utilizing a suite of complementary algorithms. The datasets were cleaned and preprocessed to address missing values, inconsistencies, and other data quality issues. …”
Get full text
Get full text

Improved decoder metrics for DS-CDMA in practical 3G systems by Elbassiouny, Shady Osama Mahmoud

“…We derive improved log-likelihood ratios (ILLRs) for bit- and chip- asynchronous multipath fading channels. We show that while utilizing knowledge of all users’ code sequences for the ILLR metric is very complicated in chip-asynchronous reception, a simplified expression relying on truncated group delay results in negligible performance loss. …”
Get full text
Get full text