Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9571000
F. Mthethwa, C. Gomes, D. Dorrell
Individual remote agro-based micro-grid is prone to reliability and resilient instability issues due to large sudden load or generation fluctuations. Therefore, it is important to integrate several micro-grids to solve the issue of reliability. An interconnected micro-grid system takes advantage of various complementary energy sources and effectively coordinates the energy sharing among the neighbouring micro-grids to improve the stability, reliability, and energy efficiency of the system in case of loss or insufficient power supply from one micro-grid. Control of energy management and communication for inter-micro-grid becomes complex and challenging in these areas due to the excess demand of agro-based consumer loads. In this paper, a decision-making algorithm that provides smart solution in interconnecting several neighbouring micro-grids to optimally share the supply is developed. A case study is presented and HOMER Pro software is used to optimize three proposed micro-grids and ensure optimal energy sharing.
{"title":"Development of Optimal Algorithm to Interconnect Multiple Microgrids in an Agricultural-based Remote Community","authors":"F. Mthethwa, C. Gomes, D. Dorrell","doi":"10.1109/africon51333.2021.9571000","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9571000","url":null,"abstract":"Individual remote agro-based micro-grid is prone to reliability and resilient instability issues due to large sudden load or generation fluctuations. Therefore, it is important to integrate several micro-grids to solve the issue of reliability. An interconnected micro-grid system takes advantage of various complementary energy sources and effectively coordinates the energy sharing among the neighbouring micro-grids to improve the stability, reliability, and energy efficiency of the system in case of loss or insufficient power supply from one micro-grid. Control of energy management and communication for inter-micro-grid becomes complex and challenging in these areas due to the excess demand of agro-based consumer loads. In this paper, a decision-making algorithm that provides smart solution in interconnecting several neighbouring micro-grids to optimally share the supply is developed. A case study is presented and HOMER Pro software is used to optimize three proposed micro-grids and ensure optimal energy sharing.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132692201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570950
Michael Njeru, C. Maina, K. Langat
Monitoring of wild animals has taken different approaches with an aim to provide vital information used in animal protection in their natural habitats. To recognize animal species without human trackers requires machine learning models that extract specie's features from an image. This project proposes a method of counting animals in an image and specifying the species of each animal using Unet and a variant of the SqueezeNet model. To train the Unet model, images and corresponding masks are used as the training data. Different optimizers are applied to each model. During inference, Unet outputs a binary mask with ones where an animal is detected and zeros elsewhere. SqueezeNet model is trained with images corresponding to six classes: bushbuck, impala, llama, warthog, waterbuck, and zebra. Three variants of the SqueezeNet model have been trained. The first contains the original backbone while the other two have the original backbone with an additional fire module. In one model the Fire module is similar to the Fire modules of the original backbone while in the other model, the extra fire module contains batch normalization layers. The trained models show that Unet trained with Nadam optimizer achieves the highest dice coefficient while the SqueezeNet with an extra Fire module containing batch norm layers and RMSprop optimizer achieves the highest training accuracy. The combined system containing the two models takes an image and outputs the image with bounding boxes around each animal and the corresponding animal species. The system achieves both counting and recognition of the species for each image placed at the input.
{"title":"Mammalian Species Detection Using a Cascade of Unet and SqueezeNet","authors":"Michael Njeru, C. Maina, K. Langat","doi":"10.1109/africon51333.2021.9570950","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570950","url":null,"abstract":"Monitoring of wild animals has taken different approaches with an aim to provide vital information used in animal protection in their natural habitats. To recognize animal species without human trackers requires machine learning models that extract specie's features from an image. This project proposes a method of counting animals in an image and specifying the species of each animal using Unet and a variant of the SqueezeNet model. To train the Unet model, images and corresponding masks are used as the training data. Different optimizers are applied to each model. During inference, Unet outputs a binary mask with ones where an animal is detected and zeros elsewhere. SqueezeNet model is trained with images corresponding to six classes: bushbuck, impala, llama, warthog, waterbuck, and zebra. Three variants of the SqueezeNet model have been trained. The first contains the original backbone while the other two have the original backbone with an additional fire module. In one model the Fire module is similar to the Fire modules of the original backbone while in the other model, the extra fire module contains batch normalization layers. The trained models show that Unet trained with Nadam optimizer achieves the highest dice coefficient while the SqueezeNet with an extra Fire module containing batch norm layers and RMSprop optimizer achieves the highest training accuracy. The combined system containing the two models takes an image and outputs the image with bounding boxes around each animal and the corresponding animal species. The system achieves both counting and recognition of the species for each image placed at the input.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134304194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570895
Cameron S Sheehan, T. Green, N. Daina
Electric motorcycles are being introduced in some African countries to combat the negative environmental impacts from the rapid growth in the use of traditional internal combustion engine motorcycle taxis. However, the electricity systems in many of these countries are strained, with generation and/or distribution capacity at their limits, leading to regular power outages that could impact the charging of these e-motorcycles. These fragile grids may be put under further strain by additional e-motorcycle charging. Commercial motorcycle taxi drivers may not be willing to wait for extended periods to charge during their shift. The use of battery swapping stations could mitigate these issues. However, modelling of their system impacts is required to fully understand their potential. This paper presents a hybrid model to simulate the key operational processes of battery swapping stations and their energy systems, allowing various configurations and scenarios to be investigated for the specific context of e-motorcycles in Africa. The configuration parameters include the numbers of batteries and charging slots, the charging power, and the addition of solar PV and static battery energy storage capacity. Power outages can be modelled for various scenarios. A test case of a battery swap station in Nairobi, Kenya, was used to showcase and validate the model. The results demonstrated how the various sub-models performed and interacted with each other, and clearly showed what impact the chosen BSS configuration would have on the grid.
{"title":"A Simulation Approach to Analyse the Impacts of Battery Swap Stations for e-Motorcycles in Africa","authors":"Cameron S Sheehan, T. Green, N. Daina","doi":"10.1109/africon51333.2021.9570895","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570895","url":null,"abstract":"Electric motorcycles are being introduced in some African countries to combat the negative environmental impacts from the rapid growth in the use of traditional internal combustion engine motorcycle taxis. However, the electricity systems in many of these countries are strained, with generation and/or distribution capacity at their limits, leading to regular power outages that could impact the charging of these e-motorcycles. These fragile grids may be put under further strain by additional e-motorcycle charging. Commercial motorcycle taxi drivers may not be willing to wait for extended periods to charge during their shift. The use of battery swapping stations could mitigate these issues. However, modelling of their system impacts is required to fully understand their potential. This paper presents a hybrid model to simulate the key operational processes of battery swapping stations and their energy systems, allowing various configurations and scenarios to be investigated for the specific context of e-motorcycles in Africa. The configuration parameters include the numbers of batteries and charging slots, the charging power, and the addition of solar PV and static battery energy storage capacity. Power outages can be modelled for various scenarios. A test case of a battery swap station in Nairobi, Kenya, was used to showcase and validate the model. The results demonstrated how the various sub-models performed and interacted with each other, and clearly showed what impact the chosen BSS configuration would have on the grid.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133273015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570977
Kaboko Jean-Jacques Monga, M. Grobler, Kabuya Isaac Kamiba
This paper reports the experimental characterization of a ring configuration erbium-doped fibre laser operating in continuous wave (CW). The output power stability and the linewidth as a function of the cavity parameters were investigated. The fibre laser cavity includes two filters namely, a fibre Bragg grating (FBG) and a fibre Fabry-Perot (FP) tunable filter. Without altering the basic cavity configuration, the study focused on the optimization of key parameters of the fibre laser cavity namely, the erbium-doped fibre length, the pump power, the output coupling ratio and the erbium ion concentration. We have demonstrated that the output power, as well as the power stability of the fibre laser, increased as the output coupling ratio. The maximum power fluctuation of the output power was 7.52 %, corresponding to 0.24 dB and was obtained for a 10 % output coupling. Power stability of 0.71 % and 0.8 % were demonstrated for 80 % and 90 % coupling ratios, respectively. To improve the power stability, we introduced a 2 m unpumped erbium-doped fibre laser into the fibre laser cavity. A stable fibre laser with a linewidth of 10 kHz was obtained during a measurement period of 120 min.
{"title":"Optimized, stabilized and narrow linewidth CW-Erbium fibre ring laser","authors":"Kaboko Jean-Jacques Monga, M. Grobler, Kabuya Isaac Kamiba","doi":"10.1109/africon51333.2021.9570977","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570977","url":null,"abstract":"This paper reports the experimental characterization of a ring configuration erbium-doped fibre laser operating in continuous wave (CW). The output power stability and the linewidth as a function of the cavity parameters were investigated. The fibre laser cavity includes two filters namely, a fibre Bragg grating (FBG) and a fibre Fabry-Perot (FP) tunable filter. Without altering the basic cavity configuration, the study focused on the optimization of key parameters of the fibre laser cavity namely, the erbium-doped fibre length, the pump power, the output coupling ratio and the erbium ion concentration. We have demonstrated that the output power, as well as the power stability of the fibre laser, increased as the output coupling ratio. The maximum power fluctuation of the output power was 7.52 %, corresponding to 0.24 dB and was obtained for a 10 % output coupling. Power stability of 0.71 % and 0.8 % were demonstrated for 80 % and 90 % coupling ratios, respectively. To improve the power stability, we introduced a 2 m unpumped erbium-doped fibre laser into the fibre laser cavity. A stable fibre laser with a linewidth of 10 kHz was obtained during a measurement period of 120 min.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126156954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570904
A. Periola, A. Alonge, K. Ogudo
Increasing population and the need to realize sustainable living necessitates finding alternative housing solutions when land resources are insufficient. A suitable solution is ocean based future cities. These cities have subscribers requiring network connectivity to access content like existing terrestrial network subscribers. Therefore, communication networks should be designed for floating city subscribers. This paper presents the network architecture for a floating city network suited for floating city subscribers. The research presents the architecture and computing entities required to realize cloud computing capability in the proposed network architecture. In addition, the presented research proposes new role for networking architecture in the context of the blue economy and the realization of future human living spaces.
{"title":"Network Architecture Design for Floating Cities","authors":"A. Periola, A. Alonge, K. Ogudo","doi":"10.1109/africon51333.2021.9570904","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570904","url":null,"abstract":"Increasing population and the need to realize sustainable living necessitates finding alternative housing solutions when land resources are insufficient. A suitable solution is ocean based future cities. These cities have subscribers requiring network connectivity to access content like existing terrestrial network subscribers. Therefore, communication networks should be designed for floating city subscribers. This paper presents the network architecture for a floating city network suited for floating city subscribers. The research presents the architecture and computing entities required to realize cloud computing capability in the proposed network architecture. In addition, the presented research proposes new role for networking architecture in the context of the blue economy and the realization of future human living spaces.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127788973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570928
H. M. Hussien, K. Katzis, L. Mfupe, T. Ephrem
Cognitive radio-based TV White Space (TVWS) is a promising idea which seeks to increase spectrum utilization by opportunistically accessing spectrum initially licensed to TV transmitters or incumbents. Orthogonal frequency multiple access (OFDM) is thought to be a promising technology for TVWS systems. In this paper, we consider an OFDM based cognitive High-Altitude Platform (HAP) exploiting the TVWS spectrum. We employ dynamic resource allocation for providing wireless access from a HAP at an altitude of 20km, while utilizing the TVWS spectrum. This paper focuses on the resource allocation algorithms, which are designed to increase the system transmission rate of secondary users while keeping the disturbance applied to the incumbent band below a threshold level and the overall power within a range using the Artificial Immune System Algorithm. As per the simulation demonstrated so far, the proposed algorithm outperforms the water-filling algorithm, implying that the system transmission rate is greatly optimized. Moreover, the proposed resource allocation algorithm allocates resources to users in a fair manner, without favoring any particular user, and has good convergence performance.
{"title":"A Novel Resource Allocation for HAP Wireless Networks Exploiting TVWS Spectrum","authors":"H. M. Hussien, K. Katzis, L. Mfupe, T. Ephrem","doi":"10.1109/africon51333.2021.9570928","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570928","url":null,"abstract":"Cognitive radio-based TV White Space (TVWS) is a promising idea which seeks to increase spectrum utilization by opportunistically accessing spectrum initially licensed to TV transmitters or incumbents. Orthogonal frequency multiple access (OFDM) is thought to be a promising technology for TVWS systems. In this paper, we consider an OFDM based cognitive High-Altitude Platform (HAP) exploiting the TVWS spectrum. We employ dynamic resource allocation for providing wireless access from a HAP at an altitude of 20km, while utilizing the TVWS spectrum. This paper focuses on the resource allocation algorithms, which are designed to increase the system transmission rate of secondary users while keeping the disturbance applied to the incumbent band below a threshold level and the overall power within a range using the Artificial Immune System Algorithm. As per the simulation demonstrated so far, the proposed algorithm outperforms the water-filling algorithm, implying that the system transmission rate is greatly optimized. Moreover, the proposed resource allocation algorithm allocates resources to users in a fair manner, without favoring any particular user, and has good convergence performance.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125916109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570942
M. Dimoiu, D. Popescu, L. Ichim
Generative Adversarial Network (GAN) is an algorithmic architecture containing two neural networks, placed against each other to generate new synthetic images and it has been used successfully in image segmentation. The paper analyzes different GAN implementations for segmentation of images acquired by aerial robots in a real context of a rural zone in Romania. To improve the segmentation performance, a new GAN network is proposed by adding a new layer. Data augmentation was done by the following techniques: mirroring, rotation, scaling, gray scaling, blurring, sharpening, etc. Five classes of region of interest are considered: floods, vegetations, buildings, roads, and dry land. GAN implementations were tested on CPU, GPU, and TPU, on individual computing devices and in the cloud. A new layer was added. The performances were analyzed in terms of learning time, operating time, and statistical indicators. The batch size was generally low: batches of 1, 4 or 16 images were used in this paper. The results confirm that the use of batch achieves the best training and generalization performance in terms of computational cost, for a wide range of experiments.
{"title":"Improved Conditional GAN for Aerial Image Segmentation","authors":"M. Dimoiu, D. Popescu, L. Ichim","doi":"10.1109/africon51333.2021.9570942","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570942","url":null,"abstract":"Generative Adversarial Network (GAN) is an algorithmic architecture containing two neural networks, placed against each other to generate new synthetic images and it has been used successfully in image segmentation. The paper analyzes different GAN implementations for segmentation of images acquired by aerial robots in a real context of a rural zone in Romania. To improve the segmentation performance, a new GAN network is proposed by adding a new layer. Data augmentation was done by the following techniques: mirroring, rotation, scaling, gray scaling, blurring, sharpening, etc. Five classes of region of interest are considered: floods, vegetations, buildings, roads, and dry land. GAN implementations were tested on CPU, GPU, and TPU, on individual computing devices and in the cloud. A new layer was added. The performances were analyzed in terms of learning time, operating time, and statistical indicators. The batch size was generally low: batches of 1, 4 or 16 images were used in this paper. The results confirm that the use of batch achieves the best training and generalization performance in terms of computational cost, for a wide range of experiments.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126768404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570866
Mariam Khamis Ali, F. Simba
Advancement in technology has led teaching and learning to improve from traditional to electronic learning (E-Learning). E-learning consists of different multimedia including videos which facilitate learning much easier. Unfortunately, videos accessed on e-learning platform suffer in Quality of Service (QoS) and Quality of Experience (QoE). The aim of this paper is to evaluate QoS in terms of jitter, delay and packet loss, also to evaluate and compare objective and subjective QoE of e-learning video transmitted via Universal Mobile Telecommunication System (UMTS) network. A model of UMTS network was developed by using Network Simulator 2 (NS2) and EvalVid framework. The developed model was used for e-learning video streaming. The streamed video was objectively and subjectively evaluated for its QoE, also jitter, delay and packet loss were used as the parameters for QoS. The obtained results has shown that e-learning video streaming delivered through 3G/UMTS suffers packet losses that exceeds the accepted value of 1%, hence poor video streaming QoS. Subjective QoE turned out to be much worse than objective QoE. Therefore, this paper recommends preference in using subjective QoE, because it gives real feelings of users with regards to QoS. Results suggest that UMTS is not suitable for e-learning video streaming.
{"title":"Performance evaluation of 3G (UMTS Network) for E-Learning Video Streaming","authors":"Mariam Khamis Ali, F. Simba","doi":"10.1109/africon51333.2021.9570866","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570866","url":null,"abstract":"Advancement in technology has led teaching and learning to improve from traditional to electronic learning (E-Learning). E-learning consists of different multimedia including videos which facilitate learning much easier. Unfortunately, videos accessed on e-learning platform suffer in Quality of Service (QoS) and Quality of Experience (QoE). The aim of this paper is to evaluate QoS in terms of jitter, delay and packet loss, also to evaluate and compare objective and subjective QoE of e-learning video transmitted via Universal Mobile Telecommunication System (UMTS) network. A model of UMTS network was developed by using Network Simulator 2 (NS2) and EvalVid framework. The developed model was used for e-learning video streaming. The streamed video was objectively and subjectively evaluated for its QoE, also jitter, delay and packet loss were used as the parameters for QoS. The obtained results has shown that e-learning video streaming delivered through 3G/UMTS suffers packet losses that exceeds the accepted value of 1%, hence poor video streaming QoS. Subjective QoE turned out to be much worse than objective QoE. Therefore, this paper recommends preference in using subjective QoE, because it gives real feelings of users with regards to QoS. Results suggest that UMTS is not suitable for e-learning video streaming.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123162517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-09-13DOI: 10.1109/africon51333.2021.9570951
O. A. Alimi, K. Ouahada, A. Abu-Mahfouz, S. Rimer, Kuburat Oyeranti Adefemi Alimi
The increasing integration of advanced information and communication tools in industrial control systems (ICS) has vastly increased the vulnerabilities and threats of intrusions into the various critical infrastructures which include the water distribution system, electrical power system, etc. that rely on the ICS systems. Currently, providing and ensuring adequate security for these ICS infrastructures are major concerns globally. The quick and accurate detection of any intrusive action into the ICS systems is highly important. Traditional intrusion detection systems (IDS) have exhibited worrying forms of limitations and shortcomings due to the heterogeneity of different cyberattacks and intrusions. Thus, there are needs to devise effective security measures. This paper proposes an IDS model based on the hybridization of particle swarm optimization (PSO) with back-propagation neural network (BPNN) for classifying intrusions in water system infrastructure. The PSO is used to optimize the parameters for the BPNN, thus improving the efficiency of classification. For the validation of the proposed method, the iTrust Lab's secure water treatment dataset was used for experimentation. Using prominent classification metrics, the 97% accuracy and 98.7% precision results achieved using the developed BPNN-PSO model is better compared to other methods including models from related works. Thus, the proposed model can meet the requirements of cyberattacks and intrusions detection in practical water distribution infrastructure.
{"title":"Intrusion Detection for Water Distribution Systems based on an Hybrid Particle Swarm Optimization with Back Propagation Neural Network","authors":"O. A. Alimi, K. Ouahada, A. Abu-Mahfouz, S. Rimer, Kuburat Oyeranti Adefemi Alimi","doi":"10.1109/africon51333.2021.9570951","DOIUrl":"https://doi.org/10.1109/africon51333.2021.9570951","url":null,"abstract":"The increasing integration of advanced information and communication tools in industrial control systems (ICS) has vastly increased the vulnerabilities and threats of intrusions into the various critical infrastructures which include the water distribution system, electrical power system, etc. that rely on the ICS systems. Currently, providing and ensuring adequate security for these ICS infrastructures are major concerns globally. The quick and accurate detection of any intrusive action into the ICS systems is highly important. Traditional intrusion detection systems (IDS) have exhibited worrying forms of limitations and shortcomings due to the heterogeneity of different cyberattacks and intrusions. Thus, there are needs to devise effective security measures. This paper proposes an IDS model based on the hybridization of particle swarm optimization (PSO) with back-propagation neural network (BPNN) for classifying intrusions in water system infrastructure. The PSO is used to optimize the parameters for the BPNN, thus improving the efficiency of classification. For the validation of the proposed method, the iTrust Lab's secure water treatment dataset was used for experimentation. Using prominent classification metrics, the 97% accuracy and 98.7% precision results achieved using the developed BPNN-PSO model is better compared to other methods including models from related works. Thus, the proposed model can meet the requirements of cyberattacks and intrusions detection in practical water distribution infrastructure.","PeriodicalId":170342,"journal":{"name":"2021 IEEE AFRICON","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131457280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：