godfrey mills - Academia.edu (original) (raw)

Papers by godfrey mills

2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS)

Efficient use of electric energy by users is one of the critical ways for increasing energy effic... more Efficient use of electric energy by users is one of the critical ways for increasing energy efficiency, reducing cost of power generation and supply, and environmental concerns. As part of remedial measures to address issues of energy use, many technologies for demand reduction ranging from energy saving devices through a variety of energy management systems have emerged to help users manage and control their consumption. In this paper, we present the design and implementation of a home energy and power management system that enables users to monitor, regulate, and manage their demand response dynamically through scheduling and controlling of their appliances using a mobile application. At the crux of the power management system is an embedded microcontroller which is encoded with schemes for monitoring, managing, scheduling and controlling operations. The user interface on the mobile device directly communicates with the power management system via integrated Raspberry Pi server that uses Wi-Fi communication protocol while communication between the Raspberry Pi and the embedded microcontroller is based on Bluetooth communication. A numerical simulation of the proposed design was first developed and a prototype of the design was implemented and tested. The test results show that the system was able to effectively monitor, control and regulate the power usage in the home. The flexibility offered by the system allows the user to self-regulate the amount of power usage, which translates into cost savings with the overall effect of flattening of the demand peaks.

IEEE Industry Applications Magazine

A novel algorithm using computer vision and machine learning techniques has been developed in thi... more A novel algorithm using computer vision and machine learning techniques has been developed in this research and applied to automate the reading of analog meters. This approach does not rely on any prior information about the meter being read or any human intervention during the process. High-level features of the meter, including the graduation values and angles, are extracted using a cascade of image contour filters with a series of digit classifiers. The features are refined and used to train regression models that return the reading of the analog meter automatically.

2021 IEEE 8th International Conference on Adaptive Science and Technology (ICAST), 2021

Traditionally, electric power utilities acquire the energy consumption information of users for b... more Traditionally, electric power utilities acquire the energy consumption information of users for billing through manual meter reading. With the advent of smart digital energy meters coupled with IoT solutions, meter reading services have improved considerably where user energy data could be collected remotely through telemetry. In many developing countries where most electric energy meters are still post-paid or non-smart devices, the utilities continue to rely on physical inspection and recording the user energy consumption for billing. This method is tedious and prone to error and delays in customer bill preparations. This paper proposes a mobile application solution that involves taking real-time pictures of energy meter readings using a mobile device and transmitting the data to a central server to process and extract the user consumption information using an artificial intelligence engine. The mobile application allows users to enter details of the meter being read. The optical character recognition technology was used as the intelligence engine at the central server to extract the meter readings from the images. The character recognition engine was trained and tested using the open-source MNIST database, which has 60,000 samples for training and 10,000 samples for testing. The meter reading system was first tested using an existing database of recorded images of energy meters, then tested at selected residences in different communities. Results revealed that the application could extract the different customer energy consumption records from the image data with an accuracy of 99.09%. An average time of 1.52 s was recorded to extract the customer energy consumption data from the images and 0.34 s to transmit the data to the billing server. The transmission time is, however, dependent on the communication service provider used for the data transmission. This software-based solution will provide enormous benefits to electric utilities that use post-paid energy meters and rely on the manual recording of the user data. The utility company will make reading faster, easier, and more accurate by automating the meter reading process.

2021 IEEE 8th International Conference on Adaptive Science and Technology (ICAST), 2021

Power substations are the most critical units of any electricity supply system. These substations... more Power substations are the most critical units of any electricity supply system. These substations either step down or step up voltages from one level to another, or maintain the same level of voltage for interconnections. Substations that operate at medium to high voltage levels have critical components such as transformers, circuit breakers, protection relays, DC batteries among others, that require regular monitoring and analysis for maintenance operations. These components also detect potential anomalies that could lead to unplanned outages and system failure. In this paper, we present an automated monitoring system that enables authorized substation operators to remotely access and receive critical substation information on trips and alarms necessary for rapid maintenance response and intervention of faults. At the crux of the monitoring system is a hardware controller system which has a microcontroller, relays, and shift registers, for acquiring and processing signals from the transformer control panel at the substation. The controller system, which also functions as a data logger, communicates the substation data to a cloud server using the GSM/GPRS (Global Service Mobile/General Packet Radio Service) communication protocol. A user interface system on a mobile device interacts and access prevailing conditions at the substation from the controller system through the server system. A prototype of the substation monitoring system design was implemented and tested through numerical simulation and field tests at a 33/11kV substation of the ECG (Electricity Company of Ghana). The results show that authorized substation operators are able to remotely monitor and receive in real-time, critical substation information such as alarms, trips, warnings, and the winding loads on transformers necessary to facilitate rapid maintenance to avoid system failures and shut-down.

Journal of Network and Systems Management, 2022

Peer-to-Peer (P2P) technology is a popular tool for sharing files and multimedia services on netw... more Peer-to-Peer (P2P) technology is a popular tool for sharing files and multimedia services on networks. While the technology has been serving a good purpose of facilitating sharing of large volumes of data on networks, in other aspects, it has also become a potential source through which attackers could ride on to launch various malicious attacks on the networks. In networks with limited bandwidth resources, uncontrolled P2P activities may also come with problems of congestion in such networks. As P2P continues to evolve on the internet in more complex forms, the need for dynamic mechanisms with the ability to learn the evolving P2P behavior will be essential for accurate monitoring and detection of the P2P traffic to minimize its effects on networks. Supervised machine learning classifiers have been used in recent times, as potential tools for monitoring and detection of the P2P traffic. Incidentally, the capabilities of such classifiers decline over time due to the changing dynamics of the P2P features, making it necessary for the classifiers to undergo continuous retraining in order to maintain their capability of providing effective detection of new P2P traffic features in real-time operations. This paper presents a hybrid machine-learning framework that combines the capabilities of self-organizing map (SOM) model with a multilayer perceptron (MLP) network to achieve real-time detection of P2P traffic in networks. The SOM model generates sets of clustered features contained in the traffic flows and organizes the features into P2P and non-P2P, which are used for training the MLP model for subsequent detection and control of the P2P traffic. The proposed P2P detection framework was tested using real traffic data from the University of Ghana campus network. The test results revealed an average detection rate of 99.89% of the observed instances of P2P traffic in the experimental data. The good detection rate from the detection framework suggests its capability to serve as a potential tool for dynamic monitoring, detection, and control of P2P traffic to manage bandwidth resources and isolation of undesirable P2P-driven traffic in networks.

Satellite dishes are used to receive beams of signals from satellites and other broadcasting sour... more Satellite dishes are used to receive beams of signals from satellites and other broadcasting sources which arethen focused onto an antenna. The dish needs to be adjusted to get the desired azimuth and elevation formaximum signal reception. To overcome the difficulty of adjusting it manually, it would be beneficial to have asystem that aligns the satellite receiver by mechanical means while allowing the user to interact with the systemremotely to achieve a line of sight communication with the satellite source of interest. This paper proposesthe design and development of a system which receives user specifications from an Android application viaBluetooth by either specifying the direction of orientation of the dish or selecting a satellite of interest. Acontrol system interacting with the developed user interface achieves this. It employs a microcontroller, a GPSdevice, a compass and two servo motors to manage the orientation of the dish on its horizontal and verticalaxes. The Smartph...

2018 IEEE Industry Applications Society Annual Meeting (IAS), 2018

A novel algorithm using computer vision and machine learning techniques have been developed in th... more A novel algorithm using computer vision and machine learning techniques have been developed in this research and applied to automate the reading of analog meters. This approach does not rely on any prior information about the meter being read or any human intervention during the process. High-level features of the meter including the graduation values and angles are extracted using a cascade of image contour filters with a series of digit classifiers. The features are refined and used to train regression models that return the reading of the analogue meter automatically. The proposed approach was tested to read a variety of offline and live-feed images of analog pointer meters automatically without any prior information about the meters.

ACM Transactions on Knowledge Discovery from Data, 2022

Class imbalance problem is prevalent in many real-world domains. It has become an active area of ... more Class imbalance problem is prevalent in many real-world domains. It has become an active area of research. In binary classification problems, imbalance learning refers to learning from a dataset with a high degree of skewness to the negative class. This phenomenon causes classification algorithms to perform woefully when predicting positive classes with new examples. Data resampling, which involves manipulating the training data before applying standard classification techniques, is among the most commonly used techniques to deal with the class imbalance problem. This article presents a new hybrid sampling technique that improves the overall performance of classification algorithms for solving the class imbalance problem significantly. The proposed method called the Hybrid Cluster-Based Undersampling Technique (HCBST) uses a combination of the cluster undersampling technique to under-sample the majority instances and an oversampling technique derived from Sigma Nearest Oversampling ...

International Journal of Telemedicine and Applications, 2021

Obesity is a major global health challenge and a risk factor for the leading causes of death, inc... more Obesity is a major global health challenge and a risk factor for the leading causes of death, including heart disease, stroke, diabetes, and several types of cancer. Attempts to manage and regulate obesity have led to the implementation of various dietary regulatory initiatives to provide information on the calorie contents of meals. Although knowledge of the calorie content is useful for meal planning, it is not sufficient as other factors, including health status (diabetes, hypertension, etc.) and level of physical activity, are essential in the decision process for obesity management. In this work, we present an artificial intelligence- (AI-) based application that is driven by a genetic algorithm (GA) as a potential tool for tracking a user’s energy balance and predicting possible calorie intake required to meet daily calorie needs for obesity management. The algorithm takes the users’ input information on desired foods which are selected from a database and extracted records of...

Journal of Sensors, 2020

There is rapid interest growing in the use of smart, connected devices. The developing world mark... more There is rapid interest growing in the use of smart, connected devices. The developing world market for smart technology is evolving to adopt and adapt to the interconnected world of devices leading to the Internet of Things (IoT) everywhere. This research paper presents the design, development, and deployment of a prototype for the secure wireless home automation system with OpenHAB 2. We employed the use of two (2) high-performance microcontrollers, namely, the Arduino Mega 2560, interfaced with a 16-channel relay, and Raspberry Pi Model B, running the OpenHAB software. The Raspberry Pi functioned as the server to develop a prototype of an automated smart home that is remotely controllable from both a web application and an Android mobile app. In designing a wireless controlled switch for home appliances, two security procedures were implemented, namely, the token-based JSON Web Token (JWT) interface and Advanced Encryption Standard (AES) procedures for authentication and data enc...

Biomedical Engineering: Applications, Basis and Communications, 2018

Acoustic stethoscopes are used to monitor signals from patients. Incidentally, the connecting tub... more Acoustic stethoscopes are used to monitor signals from patients. Incidentally, the connecting tube between the chest-piece and the ear-piece of common stethoscopes is known to serve as a medium for transmitting pathogens from patients to physicians or from one patient to another patient. This work presents a wireless stethoscope design with mobile integration that transmits heart sounds to mobile devices for evaluation and analysis, thus, eliminates the connecting tube. This is an extension of the previous work that presented the proof of concept of a wireless stethoscope with Bluetooth transmission. In this work, however, the chest-piece of the traditional stethoscope is integrated with microcontroller unit and Bluetooth communication device. Captured signals are processed and transmitted wirelessly to a mobile device with interface application software for recording, listening and visual display of waveforms. Following numerical simulation, a prototype was developed. Testing condu...

Journal of Computer Networks and Communications, 2019

A Mobile Ad-Hoc Network (MANET) is a convenient wireless infrastructure which presents many advan... more A Mobile Ad-Hoc Network (MANET) is a convenient wireless infrastructure which presents many advantages in network settings. With Mobile Ad-Hoc Network, there are many challenges. These networks are more susceptible to attacks such as black hole and man-in-the-middle (MITM) than their corresponding wired networks. This is due to the decentralized nature of their overall architecture. In this paper, ANN classification methods in intrusion detection for MANETs were developed and used with NS2 simulation platform for attack detection, identification, blacklisting, and node reconfiguration for control of nodes attacked. The ANN classification algorithm for intrusion detection was evaluated using several metrics. The performance of the ANN as a predictive technique for attack detection, isolation, and reconfiguration was measured on a dataset with network-varied traffic conditions and mobility patterns for multiple attacks. With a final detection rate of 88.235%, this work not only offere...

IEEE Industry Applications Magazine, 2019

International Journal of Machine Learning and Computing, 2016

Adequately learning from datasets that are highly imbalance has become one of the most challengin... more Adequately learning from datasets that are highly imbalance has become one of the most challenging tasks in Data Mining and Machine Learning disciplines. Most datasets from high risk application areas are often adversely affected by the class imbalance problem due to the limited occurrence of positive examples. This paper presents a new undersampling technique, called Cluster Undersampling Technique (CUST) that has the capability of further improving the performance of classification algorithms when learning from imbalance datasets. The performance of CUST is evaluated by using it to undersample 16 real world class imbalance datasets prior to building classification models using C4.5 decision tree and OneR algorithms. The performance of the models are compared to the performance of random undersampling and oversampling, synthetic minority oversampling, one-sided selection, and cluster-based undersampling. The experimental results using area under receiver-operating characteristic curve and geometric mean showed that CUST resulted in higher performance and is statistically better compared to well-known techniques.

Biomedical Engineering: Applications, Basis and Communications, 2016

A need assessment exercise at various resource-limited hospitals in Ghana revealed that a convent... more A need assessment exercise at various resource-limited hospitals in Ghana revealed that a conventional method of monitoring uterine contractions is employed. This method is time consuming and ineffective with a likelihood of misrepresenting data on uterine contractions. There is therefore a need for a system that can potentially overcome the identified challenges. In this paper, the authors present the proof of concept for development of an automated uterine contraction monitoring system designed for use in resource-limited settings. Following the engineering design process, data were gathered to draft product specifications. Various concepts were evaluated and a mathematical model of chosen concept was built and simulated. A functional prototype was constructed to test the system’s ability to measure the frequency and average duration of muscle contractions over a specified interval. The results indicate the capability of the chosen concept to meet design specifications. The design...

2016 IEEE Industry Applications Society Annual Meeting, 2016

The immense benefits of fire in road transport cannot be overemphasized. However more than two th... more The immense benefits of fire in road transport cannot be overemphasized. However more than two thousand vehicles are damaged by unwanted fire on a daily basis. On a global scale, incendiary losses to the automobile and insurance industries have ran into billions of dollars over the last decade. A not-so-distant contributory factor is the lack of a sophisticated fire safety system on the automobile. This has been addressed by designing and implementing fuzzy logic control system with feedback over an Arduino micro-controller system. The automatic system consisting of flame sensors, temperature sensors, smoke sensors and a re-engineered mobile carbon dioxide air-conditioning unit was tested on a medium sized physical car. Results show that the automobile fire detection and control system devoid of false alarms, detects and extinguishes fire under 20 seconds. An innovative, very promising solution module for hardware implementation in fire detection and control for automobiles has been developed by using new algorithms and fuzzy logic.

2015 IEEE Industry Applications Society Annual Meeting, 2015

This paper presents a rotational energy harvester using a brushless dc (BLdc) generator to harves... more This paper presents a rotational energy harvester using a brushless dc (BLdc) generator to harvest ambient energy for quadcopter in order to prolong it flight duration. For a quadcopter, its endurance is essential in order to achieve operational goals such as scientific research, security, surveillance, and reconnaissance. Because quadcopters have a limitation on size and mass, they cannot carry a large mass of on-board energy thereby having short flight time. In this paper, BLdc generators are coupled with the propellers of the quadcopter to transfer kinetic energy from the propellers to the generator. Taking into consideration the power requirement of quadcopter, the output of the generator is amplified using dc-dc boost, and is regulated to power and charge the on-board battery. The BLdc generator is simulated in MATLAB/Simulink. A final prototype of the rotational energy harvesting system is built, and this comprises a quadcopter, power management system, and a battery charging system.

Stethoscopes are used to listen to acoustic signals from the internal organs of the human body. A... more Stethoscopes are used to listen to acoustic signals from the internal organs of the human body. Although stethoscopes play a very important role in the diagnosis process, the chest piece and the connecting cable are known to facilitate transmission of pathogens from patient to patient and from patient to the user. Replacing the connecting cable with a wireless system may help reduce the potential risk and further allow broadcasting of the signals to multi-users for examination. This work reports on the design of a two-piece Bluetooth-based wireless system that eliminates the connecting cables in electronic stethoscopes. The design consists of a Bluetooth based integrated chest-piece module for captured acoustic sound transmission and a microcontroller-based (MSP430) head-piece receiver module for decoding the data for the three operational modes of the stethoscope. The design was first tested using a chirp signal source with frequency of 10 Hz – 5 kHz. Results obtained for the three...

This paper presents design reconfiguration of analog water meter to provide remote access to user... more This paper presents design reconfiguration of analog water meter to provide remote access to user water consumption and billing records, payments, and meter device monitoring using photo-encoding as the detecting method for water consumption, a PIC18F2423 microcontroller for data processing, and SMS (short message service) technology for data transportation. To validate the system design, an analog water meter was converted into a digital equivalent and interfaced to the cellular network to transmit parameters of the meter to-and-from a consolidation server (CS). The prototype was tested for operations such as user consumption and billing records, payments, device control, and tamper notification. Test results conducted within a 12km radius gave an average response time of 16 seconds spanning from time of SMS request to the CS to read data from meter through data processing and submission to user. To manage system congestion from multi-users, a multi-threaded algorithm was implement...

2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS)

Efficient use of electric energy by users is one of the critical ways for increasing energy effic... more Efficient use of electric energy by users is one of the critical ways for increasing energy efficiency, reducing cost of power generation and supply, and environmental concerns. As part of remedial measures to address issues of energy use, many technologies for demand reduction ranging from energy saving devices through a variety of energy management systems have emerged to help users manage and control their consumption. In this paper, we present the design and implementation of a home energy and power management system that enables users to monitor, regulate, and manage their demand response dynamically through scheduling and controlling of their appliances using a mobile application. At the crux of the power management system is an embedded microcontroller which is encoded with schemes for monitoring, managing, scheduling and controlling operations. The user interface on the mobile device directly communicates with the power management system via integrated Raspberry Pi server that uses Wi-Fi communication protocol while communication between the Raspberry Pi and the embedded microcontroller is based on Bluetooth communication. A numerical simulation of the proposed design was first developed and a prototype of the design was implemented and tested. The test results show that the system was able to effectively monitor, control and regulate the power usage in the home. The flexibility offered by the system allows the user to self-regulate the amount of power usage, which translates into cost savings with the overall effect of flattening of the demand peaks.

IEEE Industry Applications Magazine

A novel algorithm using computer vision and machine learning techniques has been developed in thi... more A novel algorithm using computer vision and machine learning techniques has been developed in this research and applied to automate the reading of analog meters. This approach does not rely on any prior information about the meter being read or any human intervention during the process. High-level features of the meter, including the graduation values and angles, are extracted using a cascade of image contour filters with a series of digit classifiers. The features are refined and used to train regression models that return the reading of the analog meter automatically.

2021 IEEE 8th International Conference on Adaptive Science and Technology (ICAST), 2021

Traditionally, electric power utilities acquire the energy consumption information of users for b... more Traditionally, electric power utilities acquire the energy consumption information of users for billing through manual meter reading. With the advent of smart digital energy meters coupled with IoT solutions, meter reading services have improved considerably where user energy data could be collected remotely through telemetry. In many developing countries where most electric energy meters are still post-paid or non-smart devices, the utilities continue to rely on physical inspection and recording the user energy consumption for billing. This method is tedious and prone to error and delays in customer bill preparations. This paper proposes a mobile application solution that involves taking real-time pictures of energy meter readings using a mobile device and transmitting the data to a central server to process and extract the user consumption information using an artificial intelligence engine. The mobile application allows users to enter details of the meter being read. The optical character recognition technology was used as the intelligence engine at the central server to extract the meter readings from the images. The character recognition engine was trained and tested using the open-source MNIST database, which has 60,000 samples for training and 10,000 samples for testing. The meter reading system was first tested using an existing database of recorded images of energy meters, then tested at selected residences in different communities. Results revealed that the application could extract the different customer energy consumption records from the image data with an accuracy of 99.09%. An average time of 1.52 s was recorded to extract the customer energy consumption data from the images and 0.34 s to transmit the data to the billing server. The transmission time is, however, dependent on the communication service provider used for the data transmission. This software-based solution will provide enormous benefits to electric utilities that use post-paid energy meters and rely on the manual recording of the user data. The utility company will make reading faster, easier, and more accurate by automating the meter reading process.

2021 IEEE 8th International Conference on Adaptive Science and Technology (ICAST), 2021

Power substations are the most critical units of any electricity supply system. These substations... more Power substations are the most critical units of any electricity supply system. These substations either step down or step up voltages from one level to another, or maintain the same level of voltage for interconnections. Substations that operate at medium to high voltage levels have critical components such as transformers, circuit breakers, protection relays, DC batteries among others, that require regular monitoring and analysis for maintenance operations. These components also detect potential anomalies that could lead to unplanned outages and system failure. In this paper, we present an automated monitoring system that enables authorized substation operators to remotely access and receive critical substation information on trips and alarms necessary for rapid maintenance response and intervention of faults. At the crux of the monitoring system is a hardware controller system which has a microcontroller, relays, and shift registers, for acquiring and processing signals from the transformer control panel at the substation. The controller system, which also functions as a data logger, communicates the substation data to a cloud server using the GSM/GPRS (Global Service Mobile/General Packet Radio Service) communication protocol. A user interface system on a mobile device interacts and access prevailing conditions at the substation from the controller system through the server system. A prototype of the substation monitoring system design was implemented and tested through numerical simulation and field tests at a 33/11kV substation of the ECG (Electricity Company of Ghana). The results show that authorized substation operators are able to remotely monitor and receive in real-time, critical substation information such as alarms, trips, warnings, and the winding loads on transformers necessary to facilitate rapid maintenance to avoid system failures and shut-down.

Journal of Network and Systems Management, 2022

Peer-to-Peer (P2P) technology is a popular tool for sharing files and multimedia services on netw... more Peer-to-Peer (P2P) technology is a popular tool for sharing files and multimedia services on networks. While the technology has been serving a good purpose of facilitating sharing of large volumes of data on networks, in other aspects, it has also become a potential source through which attackers could ride on to launch various malicious attacks on the networks. In networks with limited bandwidth resources, uncontrolled P2P activities may also come with problems of congestion in such networks. As P2P continues to evolve on the internet in more complex forms, the need for dynamic mechanisms with the ability to learn the evolving P2P behavior will be essential for accurate monitoring and detection of the P2P traffic to minimize its effects on networks. Supervised machine learning classifiers have been used in recent times, as potential tools for monitoring and detection of the P2P traffic. Incidentally, the capabilities of such classifiers decline over time due to the changing dynamics of the P2P features, making it necessary for the classifiers to undergo continuous retraining in order to maintain their capability of providing effective detection of new P2P traffic features in real-time operations. This paper presents a hybrid machine-learning framework that combines the capabilities of self-organizing map (SOM) model with a multilayer perceptron (MLP) network to achieve real-time detection of P2P traffic in networks. The SOM model generates sets of clustered features contained in the traffic flows and organizes the features into P2P and non-P2P, which are used for training the MLP model for subsequent detection and control of the P2P traffic. The proposed P2P detection framework was tested using real traffic data from the University of Ghana campus network. The test results revealed an average detection rate of 99.89% of the observed instances of P2P traffic in the experimental data. The good detection rate from the detection framework suggests its capability to serve as a potential tool for dynamic monitoring, detection, and control of P2P traffic to manage bandwidth resources and isolation of undesirable P2P-driven traffic in networks.

Satellite dishes are used to receive beams of signals from satellites and other broadcasting sour... more Satellite dishes are used to receive beams of signals from satellites and other broadcasting sources which arethen focused onto an antenna. The dish needs to be adjusted to get the desired azimuth and elevation formaximum signal reception. To overcome the difficulty of adjusting it manually, it would be beneficial to have asystem that aligns the satellite receiver by mechanical means while allowing the user to interact with the systemremotely to achieve a line of sight communication with the satellite source of interest. This paper proposesthe design and development of a system which receives user specifications from an Android application viaBluetooth by either specifying the direction of orientation of the dish or selecting a satellite of interest. Acontrol system interacting with the developed user interface achieves this. It employs a microcontroller, a GPSdevice, a compass and two servo motors to manage the orientation of the dish on its horizontal and verticalaxes. The Smartph...

2018 IEEE Industry Applications Society Annual Meeting (IAS), 2018

A novel algorithm using computer vision and machine learning techniques have been developed in th... more A novel algorithm using computer vision and machine learning techniques have been developed in this research and applied to automate the reading of analog meters. This approach does not rely on any prior information about the meter being read or any human intervention during the process. High-level features of the meter including the graduation values and angles are extracted using a cascade of image contour filters with a series of digit classifiers. The features are refined and used to train regression models that return the reading of the analogue meter automatically. The proposed approach was tested to read a variety of offline and live-feed images of analog pointer meters automatically without any prior information about the meters.

ACM Transactions on Knowledge Discovery from Data, 2022

Class imbalance problem is prevalent in many real-world domains. It has become an active area of ... more Class imbalance problem is prevalent in many real-world domains. It has become an active area of research. In binary classification problems, imbalance learning refers to learning from a dataset with a high degree of skewness to the negative class. This phenomenon causes classification algorithms to perform woefully when predicting positive classes with new examples. Data resampling, which involves manipulating the training data before applying standard classification techniques, is among the most commonly used techniques to deal with the class imbalance problem. This article presents a new hybrid sampling technique that improves the overall performance of classification algorithms for solving the class imbalance problem significantly. The proposed method called the Hybrid Cluster-Based Undersampling Technique (HCBST) uses a combination of the cluster undersampling technique to under-sample the majority instances and an oversampling technique derived from Sigma Nearest Oversampling ...

International Journal of Telemedicine and Applications, 2021

Obesity is a major global health challenge and a risk factor for the leading causes of death, inc... more Obesity is a major global health challenge and a risk factor for the leading causes of death, including heart disease, stroke, diabetes, and several types of cancer. Attempts to manage and regulate obesity have led to the implementation of various dietary regulatory initiatives to provide information on the calorie contents of meals. Although knowledge of the calorie content is useful for meal planning, it is not sufficient as other factors, including health status (diabetes, hypertension, etc.) and level of physical activity, are essential in the decision process for obesity management. In this work, we present an artificial intelligence- (AI-) based application that is driven by a genetic algorithm (GA) as a potential tool for tracking a user’s energy balance and predicting possible calorie intake required to meet daily calorie needs for obesity management. The algorithm takes the users’ input information on desired foods which are selected from a database and extracted records of...

Journal of Sensors, 2020

There is rapid interest growing in the use of smart, connected devices. The developing world mark... more There is rapid interest growing in the use of smart, connected devices. The developing world market for smart technology is evolving to adopt and adapt to the interconnected world of devices leading to the Internet of Things (IoT) everywhere. This research paper presents the design, development, and deployment of a prototype for the secure wireless home automation system with OpenHAB 2. We employed the use of two (2) high-performance microcontrollers, namely, the Arduino Mega 2560, interfaced with a 16-channel relay, and Raspberry Pi Model B, running the OpenHAB software. The Raspberry Pi functioned as the server to develop a prototype of an automated smart home that is remotely controllable from both a web application and an Android mobile app. In designing a wireless controlled switch for home appliances, two security procedures were implemented, namely, the token-based JSON Web Token (JWT) interface and Advanced Encryption Standard (AES) procedures for authentication and data enc...

Biomedical Engineering: Applications, Basis and Communications, 2018

Acoustic stethoscopes are used to monitor signals from patients. Incidentally, the connecting tub... more Acoustic stethoscopes are used to monitor signals from patients. Incidentally, the connecting tube between the chest-piece and the ear-piece of common stethoscopes is known to serve as a medium for transmitting pathogens from patients to physicians or from one patient to another patient. This work presents a wireless stethoscope design with mobile integration that transmits heart sounds to mobile devices for evaluation and analysis, thus, eliminates the connecting tube. This is an extension of the previous work that presented the proof of concept of a wireless stethoscope with Bluetooth transmission. In this work, however, the chest-piece of the traditional stethoscope is integrated with microcontroller unit and Bluetooth communication device. Captured signals are processed and transmitted wirelessly to a mobile device with interface application software for recording, listening and visual display of waveforms. Following numerical simulation, a prototype was developed. Testing condu...

Journal of Computer Networks and Communications, 2019

A Mobile Ad-Hoc Network (MANET) is a convenient wireless infrastructure which presents many advan... more A Mobile Ad-Hoc Network (MANET) is a convenient wireless infrastructure which presents many advantages in network settings. With Mobile Ad-Hoc Network, there are many challenges. These networks are more susceptible to attacks such as black hole and man-in-the-middle (MITM) than their corresponding wired networks. This is due to the decentralized nature of their overall architecture. In this paper, ANN classification methods in intrusion detection for MANETs were developed and used with NS2 simulation platform for attack detection, identification, blacklisting, and node reconfiguration for control of nodes attacked. The ANN classification algorithm for intrusion detection was evaluated using several metrics. The performance of the ANN as a predictive technique for attack detection, isolation, and reconfiguration was measured on a dataset with network-varied traffic conditions and mobility patterns for multiple attacks. With a final detection rate of 88.235%, this work not only offere...

IEEE Industry Applications Magazine, 2019

International Journal of Machine Learning and Computing, 2016

Adequately learning from datasets that are highly imbalance has become one of the most challengin... more Adequately learning from datasets that are highly imbalance has become one of the most challenging tasks in Data Mining and Machine Learning disciplines. Most datasets from high risk application areas are often adversely affected by the class imbalance problem due to the limited occurrence of positive examples. This paper presents a new undersampling technique, called Cluster Undersampling Technique (CUST) that has the capability of further improving the performance of classification algorithms when learning from imbalance datasets. The performance of CUST is evaluated by using it to undersample 16 real world class imbalance datasets prior to building classification models using C4.5 decision tree and OneR algorithms. The performance of the models are compared to the performance of random undersampling and oversampling, synthetic minority oversampling, one-sided selection, and cluster-based undersampling. The experimental results using area under receiver-operating characteristic curve and geometric mean showed that CUST resulted in higher performance and is statistically better compared to well-known techniques.

Biomedical Engineering: Applications, Basis and Communications, 2016

A need assessment exercise at various resource-limited hospitals in Ghana revealed that a convent... more A need assessment exercise at various resource-limited hospitals in Ghana revealed that a conventional method of monitoring uterine contractions is employed. This method is time consuming and ineffective with a likelihood of misrepresenting data on uterine contractions. There is therefore a need for a system that can potentially overcome the identified challenges. In this paper, the authors present the proof of concept for development of an automated uterine contraction monitoring system designed for use in resource-limited settings. Following the engineering design process, data were gathered to draft product specifications. Various concepts were evaluated and a mathematical model of chosen concept was built and simulated. A functional prototype was constructed to test the system’s ability to measure the frequency and average duration of muscle contractions over a specified interval. The results indicate the capability of the chosen concept to meet design specifications. The design...

2016 IEEE Industry Applications Society Annual Meeting, 2016

The immense benefits of fire in road transport cannot be overemphasized. However more than two th... more The immense benefits of fire in road transport cannot be overemphasized. However more than two thousand vehicles are damaged by unwanted fire on a daily basis. On a global scale, incendiary losses to the automobile and insurance industries have ran into billions of dollars over the last decade. A not-so-distant contributory factor is the lack of a sophisticated fire safety system on the automobile. This has been addressed by designing and implementing fuzzy logic control system with feedback over an Arduino micro-controller system. The automatic system consisting of flame sensors, temperature sensors, smoke sensors and a re-engineered mobile carbon dioxide air-conditioning unit was tested on a medium sized physical car. Results show that the automobile fire detection and control system devoid of false alarms, detects and extinguishes fire under 20 seconds. An innovative, very promising solution module for hardware implementation in fire detection and control for automobiles has been developed by using new algorithms and fuzzy logic.

2015 IEEE Industry Applications Society Annual Meeting, 2015

This paper presents a rotational energy harvester using a brushless dc (BLdc) generator to harves... more This paper presents a rotational energy harvester using a brushless dc (BLdc) generator to harvest ambient energy for quadcopter in order to prolong it flight duration. For a quadcopter, its endurance is essential in order to achieve operational goals such as scientific research, security, surveillance, and reconnaissance. Because quadcopters have a limitation on size and mass, they cannot carry a large mass of on-board energy thereby having short flight time. In this paper, BLdc generators are coupled with the propellers of the quadcopter to transfer kinetic energy from the propellers to the generator. Taking into consideration the power requirement of quadcopter, the output of the generator is amplified using dc-dc boost, and is regulated to power and charge the on-board battery. The BLdc generator is simulated in MATLAB/Simulink. A final prototype of the rotational energy harvesting system is built, and this comprises a quadcopter, power management system, and a battery charging system.

Stethoscopes are used to listen to acoustic signals from the internal organs of the human body. A... more Stethoscopes are used to listen to acoustic signals from the internal organs of the human body. Although stethoscopes play a very important role in the diagnosis process, the chest piece and the connecting cable are known to facilitate transmission of pathogens from patient to patient and from patient to the user. Replacing the connecting cable with a wireless system may help reduce the potential risk and further allow broadcasting of the signals to multi-users for examination. This work reports on the design of a two-piece Bluetooth-based wireless system that eliminates the connecting cables in electronic stethoscopes. The design consists of a Bluetooth based integrated chest-piece module for captured acoustic sound transmission and a microcontroller-based (MSP430) head-piece receiver module for decoding the data for the three operational modes of the stethoscope. The design was first tested using a chirp signal source with frequency of 10 Hz – 5 kHz. Results obtained for the three...

This paper presents design reconfiguration of analog water meter to provide remote access to user... more This paper presents design reconfiguration of analog water meter to provide remote access to user water consumption and billing records, payments, and meter device monitoring using photo-encoding as the detecting method for water consumption, a PIC18F2423 microcontroller for data processing, and SMS (short message service) technology for data transportation. To validate the system design, an analog water meter was converted into a digital equivalent and interfaced to the cellular network to transmit parameters of the meter to-and-from a consolidation server (CS). The prototype was tested for operations such as user consumption and billing records, payments, device control, and tamper notification. Test results conducted within a 12km radius gave an average response time of 16 seconds spanning from time of SMS request to the CS to read data from meter through data processing and submission to user. To manage system congestion from multi-users, a multi-threaded algorithm was implement...