Papers by Clement Nyirenda
arXiv (Cornell University), Sep 3, 2022
This paper compares two deep reinforcement learning approaches for cyber security in software def... more This paper compares two deep reinforcement learning approaches for cyber security in software defined networking. Neural Episodic Control to Deep Q-Network has been implemented and compared with that of Double Deep Q-Networks. The two algorithms are implemented in a format similar to that of a zero-sum game. A two-tailed T-test analysis is done on the two game results containing the amount of turns taken for the defender to win. Another comparison is done on the game scores of the agents in the respective games. The analysis is done to determine which algorithm is the best in game performer and whether there is a significant difference between them, demonstrating if one would have greater preference over the other. It was found that there is no significant statistical difference between the two approaches.
arXiv (Cornell University), Aug 11, 2022
This paper introduces a Modified User Datagram Protocol (UDP) for Federated Learning to ensure ef... more This paper introduces a Modified User Datagram Protocol (UDP) for Federated Learning to ensure efficiency and reliability in the model parameter transport process, maximizing the potential of the Global model in each Federated Learning round. In developing and testing this protocol, the NS3 simulator is utilized to simulate the packet transport over the network and Google TensorFlow is used to create a custom Federated learning environment. In this preliminary implementation, the simulation contains three nodes where two nodes are client nodes, and one is a server node. The results obtained in this paper provide confidence in the capabilities of the protocol in the future of Federated Learning therefore, in future the Modified UDP will be tested on a larger Federated learning system with a TensorFlow model containing more parameters and a comparison between the traditional UDP protocol and the Modified UDP protocol will be simulated. Optimization of the Modified UDP will also be explored to improve efficiency while ensuring reliability.
ICC 2020 - 2020 IEEE International Conference on Communications (ICC)
The Cloud computing system is characterized by large scale servers being utilized by an even larg... more The Cloud computing system is characterized by large scale servers being utilized by an even larger number of users. It is a system where there is the need to frequently and efficiently schedule and manage different application tasks, with varied service requirements. One of the challenges of Cloud computing is managing the quality of service (QoS) rendered to users, specifically scheduling tasks between users and Cloud resources in a timely manner. Cloud users usually have widely diverse QoS requirements and meeting these simultaneously is also a challenge. In this paper, in order to improve on Cloud resource allocation and specifically to tailor it towards meeting varied QoS requirements of users, we proposed a new algorithm which combines Differential Evolution with the Shapley Value economic mode. This combination allows us measure the contribution of each virtual machine (VM), so as to improve the probability of obtaining a better tasks-to-resource allocation thereby improving user satisfaction. From results of conducted experiments, when compared with the traditional DE (Differential Evolution) algorithm and the conventional task-VM binding policy in CloudSim, both for allocations where special QoS requirements are required and in instances of multiple QoS requirements; the modified Shapley value based DE algorithm (SVBDA) shows significant improvement.
Optimisation Algorithms and Swarm Intelligence
The efficacy of the Particle Swarm Optimization (PSO) in determining the optimal locations for ga... more The efficacy of the Particle Swarm Optimization (PSO) in determining the optimal locations for gateways in LoRaWAN networks is investigated. A modified PSO approach, which introduces gateway distancing measures during the initialization phase and flight time, is proposed. For the ease of comparisons and the understanding of the behavior of the algorithms under study, a square LoRaWAN area is used for simulations. Optimization results on a LoRaWAN script, implemented in NS-3, show that the modified PSO converges faster and achieves better results than the traditional PSO, as the number of gateways increases. Results further show that the modified PSO approach achieves similar performance to a deterministic approach, in which gateways are uniformly distributed in the network. This shows that for swarm intelligence techniques such as PSO to be used for gateway placement in LoRaWAN networks, gateway distancing mechanisms must be incorporated in the optimization process. These results fu...
Alcoholism: Clinical and Experimental Research
2021 International Conference on Electrical, Computer and Energy Technologies (ICECET), 2021
The classification of galaxy morphology plays a crucial role in understanding galaxy formation an... more The classification of galaxy morphology plays a crucial role in understanding galaxy formation and evolution. Traditionally, this process is done manually. The emergence of deep learning techniques has given room for the automation of this process. As such, this paper offers a comparison of deep learning architectures to determine which is best suited for optical galaxy morphology classification. Adapting the model training method proposed by Walmsley et al in 2021, the Zoobot Python library is used to train models to predict Galaxy Zoo DECaLS decision tree responses, made by volunteers, using EfficientNet B0, DenseNet121 and ResNet50 as core model architectures. The predicted results are then used to generate accuracy metrics per decision tree question to determine architecture performance. DenseNet121 was found to produce the best results, in terms of accuracy, with a reasonable training time. In future, further testing with more deep learning architectures could prove beneficial.
2018 Open Innovations Conference (OI), 2018
Application areas of wireless body area networks (WBANs) are growing in our everyday lives and th... more Application areas of wireless body area networks (WBANs) are growing in our everyday lives and the main area being the medical field. WBANs are made up of small sensor nodes that monitor activities on the human body and send data to appropriate places. These sensors require energy and they are normally powered by small batteries. It is of utmost importance to make sure that the battery life of these sensors is conserved as much as possible as well as reliability of the network should be as high as possible. In this paper, media access control (MAC) protocols used for communication by the sensor nodes are analyzed using Castalia simulator based on the OMNET++ framework. A network consisting of 5 nodes is set up and mobility is configured on the suitable nodes, packets are then transmitted to the sink node. The IEEE 802.15.4, IEE802.15.6 and TMAC protocols are then used for media access and parameters such as battery usage, reliability, lifetime and latency of the network are observed as simulations are being carried out. It's observed that no protocol has all the desired properties such as high reliability and lower battery usage.
2018 IEEE PES/IAS PowerAfrica, 2018
An enormous percentage of the world's population dwells in energy-underprivileged communities... more An enormous percentage of the world's population dwells in energy-underprivileged communities that are isolated, remote and sparsely populated. The hybrid energy systems (HES) based on renewable energy (RE) resources of increasing interest such as photovoltaic (PV) and wind energy are considered to be an effective option to electrify these communities. This is pertinent for areas with good solar radiation potential and sufficient average wind speed. This paper proposes a DC microgrid made up of a solar-PV/Wind/Diesel hybrid system with a backup battery bank. The proposed DC microgrid was modelled, simulated and optimized for Oluundje village, a remote rural area in the Northern part of Namibia. The load forecasting and system modelling process involved a site survey to collect the load demand, wind resource, and solar radiation data. HOMER software was used for system modelling and cost optimization of the system. In view of the simulation results, it is found that the net present cost (NPC), the cost of energy (CoE) and a payback period of the optimal system are $459 545, $0.248/kWh and 4 years, respectively. In view of the economic and environmental analysis, it is apparent that electrifying energy-deprived communities using DC microgrids based on hybrid systems with numerous RE sources are beneficial due to the lower operating costs and the environmental friendliness associated with these hybrid systems.
2017 IST-Africa Week Conference (IST-Africa), 2017
A method of leakage location detection in Tsumeb east area using Support Vector Machine (SVM) and... more A method of leakage location detection in Tsumeb east area using Support Vector Machine (SVM) and Radial Basis Function (RBF) kernel is presented. The pressure data used to train and test are obtained from EPANET, a computer program that performs extended period simulation of hydraulic and water quality behavior within pressurized pipe networks. Tsumeb East water distribution network was modeled in EPANET. The simulated network consisted of 81 nodes, 96 pipes and 1 tank. 15 nodes were selected deterministically as leaking nodes. Pressure data is divided into 540 training set and 60 testing set. In terms of results, 90% leakage detection accuracy was achieved. This research has provided a base ground for future researchers interested in this area. More other SVM parameters could be used in other researches. Highly sensitive pressure sensor are recommended for applicability of this study.
Swarm Intelligence [Working Title], 2021
The efficacy of the Particle Swarm Optimization (PSO) in determining the optimal locations for ga... more The efficacy of the Particle Swarm Optimization (PSO) in determining the optimal locations for gateways in LoRaWAN networks is investigated. A modified PSO approach, which introduces gateway distancing measures during the initialization phase and flight time, is proposed. For the ease of comparisons and the understanding of the behavior of the algorithms under study, a square LoRaWAN area is used for simulations. Optimization results on a LoRaWAN script, implemented in NS-3, show that the modified PSO converges faster and achieves better results than the traditional PSO, as the number of gateways increases. Results further show that the modified PSO approach achieves similar performance to a deterministic approach, in which gateways are uniformly distributed in the network. This shows that for swarm intelligence techniques such as PSO to be used for gateway placement in LoRaWAN networks, gateway distancing mechanisms must be incorporated in the optimization process. These results fu...
2021 IST-Africa Conference (IST-Africa), 2021
This paper aims to address the problem of Deepfake Detection using a Two-Stream Capsule Network. ... more This paper aims to address the problem of Deepfake Detection using a Two-Stream Capsule Network. First we review methods used to create Deepfake content, as well as methods proposed in the literature to detect such Deepfake content. We then propose a novel architecture to detect Deepfakes, which consists of a two-stream Capsule network running in parallel that takes in both RGB images/frames as well as Error Level Analysis images. Results show that the proposed approach exhibits the detection accuracy of 73.39 % and 57.45 % for the Deepfake Detection Challenge (DFDC) and the Celeb-DF datasets respectively. These results are, however, from a preliminary implementation of the proposed approach. As part of future work, population-based optimization techniques such as Particle Swarm Optimization (PSO) will be used to tune the hyper parameters for better performance.
The absence of electricity in remote and rural areas is one of the major challenges faced by many... more The absence of electricity in remote and rural areas is one of the major challenges faced by many poor and developing countries. Hybrid energy systems (HES) based on photovoltaics (PV) are considered to be an effective option to electrify remote and isolated areas, which are far from conventional grids. This is true for areas that receive high averages of solar radiation annually. This research investigation involves the modelling, simulation and optimization of a PV-Diesel hybrid system for Oluundje village in Namibia. A site survey was conducted in a form of questionnaires and interviews for the purpose of load forecasting and system modelling. HOMER software was used to design and model the proposed hybrid energy system. Costs of different components, hourly solar radiation, and rating parameters are inputs of the simulation program. Sensitivity analysis was carried out using Homer. The optimal PV-Diesel hybrid system and diesel-generator-only system were compared both technically and economically. Based on simulation results, it was found that electrifying a remote village using a PVdiesel hybrid system is more advantageous when compared to the diesel-generator-only system as it has lower operating costs and emissions. This system may be used as a preliminary design to guide in the planning and modelling of similar systems for other remote villages.
A Particle Swarm Optimization (PSO) algorithm for the placement of Data Acquisition Points (DAPs)... more A Particle Swarm Optimization (PSO) algorithm for the placement of Data Acquisition Points (DAPs) in a Smart Water Metering Networks is investigated. The PSO algorithm generates particles, which denote the coordinates of the DAPs and creates the topology file by appending these coordinates to the smart meter topology file. It then invokes the Java LinkLayerModel, which generates the link gain file of the network. Once that is done, the TOSSIM Python script is invoked to simulate the network and the packet delivery ratio (PDR) is calculated and designated as the fitness value for the particle. Updates of global best solution are carried out if necessary. This process continues until 50 iterations are reached. Results show that the PDR for 10 DAPs (0.97) in the PSO placement mechanism is better than that of the meter density based placement for 25 DAPs (0.96). It is, therefore, possible to deploy fewer DAPs while achieving even better PDR values. The PSO mechanism also shows more cons...
2021 IEEE AFRICON, 2021
This paper presents a study on crime classification using two 3D deep learning algorithms, i.e. 3... more This paper presents a study on crime classification using two 3D deep learning algorithms, i.e. 3D Convolutional Neural Network and the 3D Residual Network. The Chicago crime dataset, which has records from 2001 to 2020, with a record count of 7.29 million records, is used for training the models. The models are evaluated by using F1 score, Area Under Receiver Operator Curve (AUROC), and Area Under Curve-Precision Recall (AUCPR). Furthermore, the effectiveness of spatial grid resolutions on the performance of the models is also evaluated. Results show that the 3D ResNet achieved the best performance with a F1 score of 0.9985, whereas the 3D CNN achieved a F1 score of 0.9979, when training on a spatial resolution of 16 pixels. In terms of future work, we would want to test these algorithms on multi label classifaction and regression crime problems, also we want to improve the performance of the 3D CNN by adding RNN layers and evaluate an implementation of 3D ResNeXt for crime prediction and classification.
2021 IEEE AFRICON, 2021
The objective of this study is to identify cephalometric landmarks on 2D cephalograms (X-rays) us... more The objective of this study is to identify cephalometric landmarks on 2D cephalograms (X-rays) using a two-stage Artificial Intelligence (AI) based object detection method. The proposed work implements a Faster Region-based Convolutional Neural Network (Faster R-CNN), a deep-neural network, which consists of a 50 layered Residual Network (ResNet50) with Feature Pyramid Network (FPN) as a backbone network. The algorithm is trained and tested on a dataset presented in the IEEE International Symposium on Biomedical Imaging Challenge (ISBI-2015). The detection was based on the algorithm’s performance, in terms of mean error and the success rate under the clinically accepted accuracy range of 2 mm. The hypothesis behind this work was that Faster R-CNN will have a difficulty in detecting the landmarks due to either fuzzy features and, or low-resolution representations, but with help of FPN, the performance might be better. Results show that the model achieves approximately 90% and 0.9 mm in terms success rate and mean error respectively. In terms of future work, there is still a need to improve Faster R-CNN performance by increasing or modifying the dataset. Furthermore, the use of a more powerful computational platform would lead to faster training time, which would give room to the implementation of optimization algorithms of the hyper parameters by using evolutionary computation methods.
2017 IST-Africa Week Conference (IST-Africa), 2017
This paper seeks to determine the Bit Error Rates (BER) performance for feasible free space optic... more This paper seeks to determine the Bit Error Rates (BER) performance for feasible free space optical communication (FSOC) deployment in Namibian and South African atmospheric weather conditions. In order to achieve this objective, the Pulse Position Modulation (PPM) based BER performance model was invoked to compute various BER values in selected towns of both countries. The results showed that in winter season, BER values were generally lower than in summer season. A comparative analysis for various towns with the lowest and the highest scintillation in both countries showed that BER increases rapidly as photon count increases in towns with high scintillation index values. The BER values realized from this study could be used to simulate internet traffic QoS advantages of the FSOC links when compared to the RF technology networks. The study outcomes are likely to benefit operators, industry, regulators, governments, and researchers in understanding feasible areas of FSOC deployment in both countries.
Sensors, 2019
The internet of things (IoT) and cloud computing are two technologies which have recently changed... more The internet of things (IoT) and cloud computing are two technologies which have recently changed both the academia and industry and impacted our daily lives in different ways. However, despite their impact, both technologies have their shortcomings. Though being cheap and convenient, cloud services consume a huge amount of network bandwidth. Furthermore, the physical distance between data source(s) and the data centre makes delays a frequent problem in cloud computing infrastructures. Fog computing has been proposed as a distributed service computing model that provides a solution to these limitations. It is based on a para-virtualized architecture that fully utilizes the computing functions of terminal devices and the advantages of local proximity processing. This paper proposes a multi-layer IoT-based fog computing model called IoT-FCM, which uses a genetic algorithm for resource allocation between the terminal layer and fog layer and a multi-sink version of the least interferenc...
Uploads
Papers by Clement Nyirenda