World Congress on Electrical Engineering and Computer Systems and Science
Rehabilitation of manual dexterity of patients with sensory motor impairments of the upper extrem... more Rehabilitation of manual dexterity of patients with sensory motor impairments of the upper extremity due to stroke requires task-specific and repetitive exercises to facilitate recovery of a function. Given that patients require a considerable amount of time in therapy, it is essential that the rehabilitation devices are engaging and game-based to prevent the boredom of long-term repetitive task practice exercise regimes. In this study, an easy-to-use finger-thumb mechanism is designed that supports the index and middle fingers, and the thumb for patients with hand injuries. The device is connected to a game wirelessly, allowing patients to interact in real time with engaging computer games using goal-directed thumb and finger movements. The connection is established using two Raspberry Pi boards with the aid of a server-client network. The device also provides assistive-resistive forces during the game to assist or challenge patients depending on the state of their motor control. The slight delay of 10 ms for data transfer and 50 ms for game event updates enables the patients to play the game and modify the game events in real time, using the wearable device.
World Congress on Electrical Engineering and Computer Systems and Science
Machine digit recognition from various multi-digit displays is a complex task due to the sheer nu... more Machine digit recognition from various multi-digit displays is a complex task due to the sheer number of unique digit forms, each varying significantly in shape, size, and orientation. Traditional digit recognition libraries may not perform well for all cases, especially when dealing with digital screens that can be highly variable in terms of style, fonts, colour, contrast, intensity, pixel resolution, digit aspect ratio, and spacing. To address these challenges, we present a digit recognition algorithm that is designed to be fast, easy to use, and highly adaptable. Unlike a single fit-all solution, our system can be easily modified to fit different use cases and applications, incorporating additional layers of flexibility and adaptability. This is desirable since different types of displayed digits may have unique features or characteristics that traditional digit recognition libraries do not capture well. To further demonstrate the efficacy of the proposed system, we tested it on a unique pump-flowmeter digits format, which poses significant challenges for digit recognition algorithms due to the complicated shape and layout of the digits. This paper provides a detailed step-by-step account of our system's development and its performance on this challenging dataset. The presented system achieved an accuracy of 80% on test data, is simple and can be used by researchers, developers, and practitioners working in fields such as handwriting recognition, computer vision, machine learning, image processing, pattern recognition, and neural networks.
World Congress on Electrical Engineering and Computer Systems and Science
In this paper a new class of fuzzy systems called scalable fuzzy (SF) systems are proposed. The S... more In this paper a new class of fuzzy systems called scalable fuzzy (SF) systems are proposed. The SF design is built upon the idea of extending the conventional fuzzy logic approach to linguistic variables to all numbers. This leads to a new set of infinite continuous rule-base and membership functions which are located on all rational numbers and are defined based on scale, position, and input variables. The consequent of rules in the Takagi-Sugeno form are then modified, and a mathematical solution based on the convolution theorem is employed for SF modeling purposes. The discrete form of SF systems is developed, and its application is exemplified using several case studies.
Proceedings of the Canadian Engineering Education Association, 2015
In recent years, the emergence of many newdisciplines and the increasing depth of existing discip... more In recent years, the emergence of many newdisciplines and the increasing depth of existing disciplineshave prompted the development of the Body of Knowledge(BoK) concept in order to improve the uniformity of newundergraduate programs in accredited professionalschools, and to provide a requisite standard forcertifications in vocational education. A body of knowledgeis the collected wisdom, experience, processes, and factsthat both inform a profession and provide the solidfoundation from which continuous improvements andinnovative change can occur. This paper provides arguablythe first summary of 32 BoK examples, and proposes a newmodel of a BoK for professional practitioners
2014 IEEE Canada International Humanitarian Technology Conference - (IHTC), 2014
This paper describes an enhanced crowdsourcing that may be the only alternative when everything e... more This paper describes an enhanced crowdsourcing that may be the only alternative when everything else fails in a disaster. This scheme requires radio operators that can work not only together with other communications media (such as the Internet and wireless phones), but they can also work on their own, if the other media either fail or do not exist in the area. In order to increase the number of such radio operators, a scheme for accelerated radio education and training is required. This paper describes the structure of such a course, its delivery, demonstrations, workshops, and examination that lead to a governmentissued Radio Operator's Certificate. The course has evolved over the last 25 years.
2021 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2021
This paper introduces an architecture of a convolutional neural network (CNN) to detect a distrib... more This paper introduces an architecture of a convolutional neural network (CNN) to detect a distributed denial of service (DDoS) attack. The main procedure is the application of an adaptive mother wavelet that is created using a genetic neural network (GNN) to increase the detection rate of the DDoS. In addition, an adaptive-wavelet CNN is trained to extract features from Internet traffic containing DDoS attacks to classify the Internet traffic data (ITD) with DDoS attacks (DDoS ITD) as normal or anomalous. Moreover, a multi-objective optimization based on a genetic algorithm and a weighted cost function based on an information-theoretic measure are used to train and evaluate the adaptive-wavelet CNN. Finally, the adaptive-wavelet CNN's classification efficiency is assessed to find the detection rate of the proposed architecture. The adaptive-wavelet CNN detects the DDoS attack with 95% accuracy.
2015 IEEE 14th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), 2015
The paper deals with the problem of joint spectrum sensing and power control optimization for a m... more The paper deals with the problem of joint spectrum sensing and power control optimization for a multichannel, multiple-user cognitive radio network. In particular, we investigate trade-off factors in designing efficient spectrum sensing techniques to maximize the throughputs and minimize the interferences. To maximize the throughputs of secondary users and minimize the interferences to primary users, it requires for a joint determination of the sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. We propose an analytical approach based on Newton's methods and nonlinear barrier method to solve this large-scale joint multiobjective optimization problem.
2016 IEEE 7th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2016
This paper presents the construction of a low-cost data acquisition system (DAS) prototype based ... more This paper presents the construction of a low-cost data acquisition system (DAS) prototype based on Raspberry Pi-2 microcomputer. The prototype is designed to operate as a standalone system without the need for an additional personal computer (PC). It performs the data acquisition, online plotting and data logging, simultaneously. This is a general-purpose setup, as it is capable of reading any analog sensor giving output in the designed range, or through appropriate signal conditioning. The system is tested in a Mechanical laboratory by collecting data which are compared to a benchmark DAS. Statistical analyses are also performed on the acquired data. It is proved that both signals are identical with only minor differences.
Features of an Internet traffic time series can be estimated using dynamical systems. Dynamical s... more Features of an Internet traffic time series can be estimated using dynamical systems. Dynamical systems may exhibit chaos and strange attractors [1] [2]. Since Internet traffic shows non stationarity and long term dependence among data samples, a cognitive polyscale approach should be taken to analyze the hidden features in a nonlinear data time series. It is necessary to estimate a reasonable window of time series so that the polyscale analysis can be performed without violating the statistical bounds of the analysis. In this work, a feature extraction algorithm is developed using variance fractal dimension trajectory and the statistical parameters of the calculation are validated using an autonomous varying window of data samples. Our analysis shows promising results since the algorithm is able to capture the presence of DNS denial of service attack and has extracted the bursts of data sample accurately. Keywords— Cognitive machine learning, Fractal, Polyscale, DNS DDoS amplificat...
International Journal of Cognitive Informatics and Natural Intelligence, 2020
Distributed denial-of-service (DDoS) attacks are serious threats to the availability of a smart g... more Distributed denial-of-service (DDoS) attacks are serious threats to the availability of a smart grid infrastructure services because they can cause massive blackouts. This study describes an anomaly detection method for improving the detection rate of a DDoS attack in a smart grid. This improvement was achieved by increasing the classification of the training and testing phases in a convolutional neural network (CNN). A full version of the variance fractal dimension trajectory (VFDTv2) was used to extract inherent features from the stochastic fractal input data. A discrete wavelet transform (DWT) was applied to the input data and the VFDTv2 to extract significant distinguishing features during data pre-processing. A support vector machine (SVM) was used for data post-processing. The implementation detected the DDoS attack with 87.35% accuracy.
International Journal of Cognitive Informatics and Natural Intelligence, 2016
In this paper, the authors investigate trade-off factors in designing efficient spectrum sensing ... more In this paper, the authors investigate trade-off factors in designing efficient spectrum sensing and optimal power control techniques for a multichannel, multiple-user cognitive wireless network. They introduce the problem of joint spectrum sensing and power control as a maximization of the network throughput and a minimization of the interference to the network. These two optimization objectives can be achieved by a joint determination of sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. The authors propose an analytical approach based on Newton's methods and nonlinear barrier method to solve this large-scale joint multiobjective optimization problem.
2015 IEEE 14th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), 2015
This paper presents a cognitive feature extraction model based on scaling and multifractal dimens... more This paper presents a cognitive feature extraction model based on scaling and multifractal dimension trajectory to analyze internet traffic time series. DNS (Domain Naming System) traffic time series is considered that contains tagged DNS Denial of Service attacks. The first step of the analysis involves transforming the DNS time series into a multifractal variance dimension trajectory keeping statistical stationarity of data intact. Then features of the trajectory are extracted to remove high variability noise. The extracted set of features indicates the presence of an attack when the denoised trajectory shows increasing variance fractal dimension. This technique is superior in finding changing patterns of a data series due to the presence of noise and denial of service attack because it is not dependent on integer dimensions and mono-scale measurement of variations in data series. Moreover, this technique provides adaptive and locally stationary windows in a highly non stationary data series.
International Journal of Cognitive Informatics and Natural Intelligence, 2013
This paper proposes the application of chaos in large search space problems, and suggests that th... more This paper proposes the application of chaos in large search space problems, and suggests that this represents the next evolutionary step in the development of adaptive and intelligent systems towards cognitive machines and systems. Three different versions of chaotic simulated annealing (XSA) were applied to combinatorial optimization problems in multiprocessor task allocation. Chaotic walks in the solution space were taken to search for the global optimum or “good enough” task-to-processor allocation solutions. Chaotic variables were generated to set the number of perturbations made in each iteration of a XSA algorithm. In addition, parameters of a chaotic variable generator were adjusted to create different chaotic distributions with which to search the solution space. The results show that the convergence rate of the XSA algorithm is faster than simulated annealing when the solutions are far apart in the solution space. In particular, the XSA algorithms found simulated annealing...
International Journal of Cognitive Informatics and Natural Intelligence, 2013
Cognitive Informatics (CI) is a contemporary multidisciplinary field spanning across computer sci... more Cognitive Informatics (CI) is a contemporary multidisciplinary field spanning across computer science, information science, cognitive science, brain science, intelligence science, knowledge science, cognitive linguistics, and cognitive philosophy. CI aims to investigate the internal information processing mechanisms and processes of the brain, the underlying abstract intelligence theories and denotational mathematics, and their engineering applications in cognitive computing and computational intelligence. This paper reports a set of eleven position statements presented in the plenary panel of IEEE ICCI*CC’13 on Cognitive Computers and Knowledge Processors contributed from invited panelists who are part of the world’s renowned researchers and scholars in the field of cognitive informatics and cognitive computing.
World Congress on Electrical Engineering and Computer Systems and Science
Rehabilitation of manual dexterity of patients with sensory motor impairments of the upper extrem... more Rehabilitation of manual dexterity of patients with sensory motor impairments of the upper extremity due to stroke requires task-specific and repetitive exercises to facilitate recovery of a function. Given that patients require a considerable amount of time in therapy, it is essential that the rehabilitation devices are engaging and game-based to prevent the boredom of long-term repetitive task practice exercise regimes. In this study, an easy-to-use finger-thumb mechanism is designed that supports the index and middle fingers, and the thumb for patients with hand injuries. The device is connected to a game wirelessly, allowing patients to interact in real time with engaging computer games using goal-directed thumb and finger movements. The connection is established using two Raspberry Pi boards with the aid of a server-client network. The device also provides assistive-resistive forces during the game to assist or challenge patients depending on the state of their motor control. The slight delay of 10 ms for data transfer and 50 ms for game event updates enables the patients to play the game and modify the game events in real time, using the wearable device.
World Congress on Electrical Engineering and Computer Systems and Science
Machine digit recognition from various multi-digit displays is a complex task due to the sheer nu... more Machine digit recognition from various multi-digit displays is a complex task due to the sheer number of unique digit forms, each varying significantly in shape, size, and orientation. Traditional digit recognition libraries may not perform well for all cases, especially when dealing with digital screens that can be highly variable in terms of style, fonts, colour, contrast, intensity, pixel resolution, digit aspect ratio, and spacing. To address these challenges, we present a digit recognition algorithm that is designed to be fast, easy to use, and highly adaptable. Unlike a single fit-all solution, our system can be easily modified to fit different use cases and applications, incorporating additional layers of flexibility and adaptability. This is desirable since different types of displayed digits may have unique features or characteristics that traditional digit recognition libraries do not capture well. To further demonstrate the efficacy of the proposed system, we tested it on a unique pump-flowmeter digits format, which poses significant challenges for digit recognition algorithms due to the complicated shape and layout of the digits. This paper provides a detailed step-by-step account of our system's development and its performance on this challenging dataset. The presented system achieved an accuracy of 80% on test data, is simple and can be used by researchers, developers, and practitioners working in fields such as handwriting recognition, computer vision, machine learning, image processing, pattern recognition, and neural networks.
World Congress on Electrical Engineering and Computer Systems and Science
In this paper a new class of fuzzy systems called scalable fuzzy (SF) systems are proposed. The S... more In this paper a new class of fuzzy systems called scalable fuzzy (SF) systems are proposed. The SF design is built upon the idea of extending the conventional fuzzy logic approach to linguistic variables to all numbers. This leads to a new set of infinite continuous rule-base and membership functions which are located on all rational numbers and are defined based on scale, position, and input variables. The consequent of rules in the Takagi-Sugeno form are then modified, and a mathematical solution based on the convolution theorem is employed for SF modeling purposes. The discrete form of SF systems is developed, and its application is exemplified using several case studies.
Proceedings of the Canadian Engineering Education Association, 2015
In recent years, the emergence of many newdisciplines and the increasing depth of existing discip... more In recent years, the emergence of many newdisciplines and the increasing depth of existing disciplineshave prompted the development of the Body of Knowledge(BoK) concept in order to improve the uniformity of newundergraduate programs in accredited professionalschools, and to provide a requisite standard forcertifications in vocational education. A body of knowledgeis the collected wisdom, experience, processes, and factsthat both inform a profession and provide the solidfoundation from which continuous improvements andinnovative change can occur. This paper provides arguablythe first summary of 32 BoK examples, and proposes a newmodel of a BoK for professional practitioners
2014 IEEE Canada International Humanitarian Technology Conference - (IHTC), 2014
This paper describes an enhanced crowdsourcing that may be the only alternative when everything e... more This paper describes an enhanced crowdsourcing that may be the only alternative when everything else fails in a disaster. This scheme requires radio operators that can work not only together with other communications media (such as the Internet and wireless phones), but they can also work on their own, if the other media either fail or do not exist in the area. In order to increase the number of such radio operators, a scheme for accelerated radio education and training is required. This paper describes the structure of such a course, its delivery, demonstrations, workshops, and examination that lead to a governmentissued Radio Operator's Certificate. The course has evolved over the last 25 years.
2021 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2021
This paper introduces an architecture of a convolutional neural network (CNN) to detect a distrib... more This paper introduces an architecture of a convolutional neural network (CNN) to detect a distributed denial of service (DDoS) attack. The main procedure is the application of an adaptive mother wavelet that is created using a genetic neural network (GNN) to increase the detection rate of the DDoS. In addition, an adaptive-wavelet CNN is trained to extract features from Internet traffic containing DDoS attacks to classify the Internet traffic data (ITD) with DDoS attacks (DDoS ITD) as normal or anomalous. Moreover, a multi-objective optimization based on a genetic algorithm and a weighted cost function based on an information-theoretic measure are used to train and evaluate the adaptive-wavelet CNN. Finally, the adaptive-wavelet CNN's classification efficiency is assessed to find the detection rate of the proposed architecture. The adaptive-wavelet CNN detects the DDoS attack with 95% accuracy.
2015 IEEE 14th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), 2015
The paper deals with the problem of joint spectrum sensing and power control optimization for a m... more The paper deals with the problem of joint spectrum sensing and power control optimization for a multichannel, multiple-user cognitive radio network. In particular, we investigate trade-off factors in designing efficient spectrum sensing techniques to maximize the throughputs and minimize the interferences. To maximize the throughputs of secondary users and minimize the interferences to primary users, it requires for a joint determination of the sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. We propose an analytical approach based on Newton's methods and nonlinear barrier method to solve this large-scale joint multiobjective optimization problem.
2016 IEEE 7th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2016
This paper presents the construction of a low-cost data acquisition system (DAS) prototype based ... more This paper presents the construction of a low-cost data acquisition system (DAS) prototype based on Raspberry Pi-2 microcomputer. The prototype is designed to operate as a standalone system without the need for an additional personal computer (PC). It performs the data acquisition, online plotting and data logging, simultaneously. This is a general-purpose setup, as it is capable of reading any analog sensor giving output in the designed range, or through appropriate signal conditioning. The system is tested in a Mechanical laboratory by collecting data which are compared to a benchmark DAS. Statistical analyses are also performed on the acquired data. It is proved that both signals are identical with only minor differences.
Features of an Internet traffic time series can be estimated using dynamical systems. Dynamical s... more Features of an Internet traffic time series can be estimated using dynamical systems. Dynamical systems may exhibit chaos and strange attractors [1] [2]. Since Internet traffic shows non stationarity and long term dependence among data samples, a cognitive polyscale approach should be taken to analyze the hidden features in a nonlinear data time series. It is necessary to estimate a reasonable window of time series so that the polyscale analysis can be performed without violating the statistical bounds of the analysis. In this work, a feature extraction algorithm is developed using variance fractal dimension trajectory and the statistical parameters of the calculation are validated using an autonomous varying window of data samples. Our analysis shows promising results since the algorithm is able to capture the presence of DNS denial of service attack and has extracted the bursts of data sample accurately. Keywords— Cognitive machine learning, Fractal, Polyscale, DNS DDoS amplificat...
International Journal of Cognitive Informatics and Natural Intelligence, 2020
Distributed denial-of-service (DDoS) attacks are serious threats to the availability of a smart g... more Distributed denial-of-service (DDoS) attacks are serious threats to the availability of a smart grid infrastructure services because they can cause massive blackouts. This study describes an anomaly detection method for improving the detection rate of a DDoS attack in a smart grid. This improvement was achieved by increasing the classification of the training and testing phases in a convolutional neural network (CNN). A full version of the variance fractal dimension trajectory (VFDTv2) was used to extract inherent features from the stochastic fractal input data. A discrete wavelet transform (DWT) was applied to the input data and the VFDTv2 to extract significant distinguishing features during data pre-processing. A support vector machine (SVM) was used for data post-processing. The implementation detected the DDoS attack with 87.35% accuracy.
International Journal of Cognitive Informatics and Natural Intelligence, 2016
In this paper, the authors investigate trade-off factors in designing efficient spectrum sensing ... more In this paper, the authors investigate trade-off factors in designing efficient spectrum sensing and optimal power control techniques for a multichannel, multiple-user cognitive wireless network. They introduce the problem of joint spectrum sensing and power control as a maximization of the network throughput and a minimization of the interference to the network. These two optimization objectives can be achieved by a joint determination of sensing and transmission parameters of the secondary users, such as sensing times, decision threshold vectors, and power allocation vectors. There is a conflict between these two objectives, thus a multiobjective optimization problem is introduced. The authors propose an analytical approach based on Newton's methods and nonlinear barrier method to solve this large-scale joint multiobjective optimization problem.
2015 IEEE 14th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), 2015
This paper presents a cognitive feature extraction model based on scaling and multifractal dimens... more This paper presents a cognitive feature extraction model based on scaling and multifractal dimension trajectory to analyze internet traffic time series. DNS (Domain Naming System) traffic time series is considered that contains tagged DNS Denial of Service attacks. The first step of the analysis involves transforming the DNS time series into a multifractal variance dimension trajectory keeping statistical stationarity of data intact. Then features of the trajectory are extracted to remove high variability noise. The extracted set of features indicates the presence of an attack when the denoised trajectory shows increasing variance fractal dimension. This technique is superior in finding changing patterns of a data series due to the presence of noise and denial of service attack because it is not dependent on integer dimensions and mono-scale measurement of variations in data series. Moreover, this technique provides adaptive and locally stationary windows in a highly non stationary data series.
International Journal of Cognitive Informatics and Natural Intelligence, 2013
This paper proposes the application of chaos in large search space problems, and suggests that th... more This paper proposes the application of chaos in large search space problems, and suggests that this represents the next evolutionary step in the development of adaptive and intelligent systems towards cognitive machines and systems. Three different versions of chaotic simulated annealing (XSA) were applied to combinatorial optimization problems in multiprocessor task allocation. Chaotic walks in the solution space were taken to search for the global optimum or “good enough” task-to-processor allocation solutions. Chaotic variables were generated to set the number of perturbations made in each iteration of a XSA algorithm. In addition, parameters of a chaotic variable generator were adjusted to create different chaotic distributions with which to search the solution space. The results show that the convergence rate of the XSA algorithm is faster than simulated annealing when the solutions are far apart in the solution space. In particular, the XSA algorithms found simulated annealing...
International Journal of Cognitive Informatics and Natural Intelligence, 2013
Cognitive Informatics (CI) is a contemporary multidisciplinary field spanning across computer sci... more Cognitive Informatics (CI) is a contemporary multidisciplinary field spanning across computer science, information science, cognitive science, brain science, intelligence science, knowledge science, cognitive linguistics, and cognitive philosophy. CI aims to investigate the internal information processing mechanisms and processes of the brain, the underlying abstract intelligence theories and denotational mathematics, and their engineering applications in cognitive computing and computational intelligence. This paper reports a set of eleven position statements presented in the plenary panel of IEEE ICCI*CC’13 on Cognitive Computers and Knowledge Processors contributed from invited panelists who are part of the world’s renowned researchers and scholars in the field of cognitive informatics and cognitive computing.
Uploads
Papers by Witold Kinsner