Sufian Yousef
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sufian Yousef has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Akram Qashou
Anglia Ruskin University, Cambridge, UK
3 shared publications Abdallah A. Smadi
University of Idaho, Moscow, England
1 shared publicationAkram Qashou
Anglia Ruskin University
1 shared publicationAmaechi Stanley Okoro
Anglia Ruskin University
1 shared publicationAmani A. AlOmari
University of Idaho, Moscow, England
1 shared publication- Erika Sanchez-Velazquez
School of Computing and Information Science, Anglia Ruskin University, Cambridge CB1 1PT, UK
1 shared publication
Latest Publications
- Open Access
- 21 July 2025
- Published by Mesopotamian Academic Press in Babylonian Journal of Networking
- Vol. 2025, 106-115
Applying Keras-Based Deep Learning for Intelligent Analysis in Network Security and Monitoring Systems
With the advent of digital age, network access and protection of sensitive data from unauthorized access or use has been a great challenge. Face detection and recognition is becoming a prevalent method in network security system by utilising the biometric principles. In this survey, we use Convolutional Neural Networks (CNNs) and the Keras deep learning framework to improve network security by building efficient face detection systems. A high-level and user-friendly API implemented by Keras (over TensorFlow), which makes it very easy to use deep learning models for tasks such as face detection. Being multi-GPU ready and distributed training friendly alongside compatibility with OpenCV and TensorFlow enables it to be used in developing reliable, secure, real-time face authentication systems. In this paper, we review the statistics of CNN models for face detection, compare performance of the Keras models on multiple datasets, and show applications of securing a network such as login authentication and access control, as well as real-world uses for surveillance system. In addition, the survey details applications of face detection where the possibility of face spoofing is an issue and current research directions on the topic. Utilizing CNNs with Keras can enable the development of more flexible, performative, and accurate biometric authentication systems, and these systems play a critical role in the global cybersecurity ecosystem.
- 21 July 2025
- Published by Mesopotamian Academic Press in Babylonian Journal of Networking
- Vol. 2025, 106-115
Open AccessApplying Keras-Based Deep Learning for Intelligent Analysis in Network Security and Monitoring Systems
With the advent of digital age, network access and protection of sensitive data from unauthorized access or use has been a great challenge. Face detection and recognition is becoming a prevalent method in network security system by utilising the biometric principles. In this survey, we use Convolutional Neural Networks (CNNs) and the Keras deep learning framework to improve network security by building efficient face detection systems. A high-level and user-friendly API implemented by Keras (over TensorFlow), which makes it very easy to use deep learning models for tasks such as face detection. Being multi-GPU ready and distributed training friendly alongside compatibility with OpenCV and TensorFlow enables it to be used in developing reliable, secure, real-time face authentication systems. In this paper, we review the statistics of CNN models for face detection, compare performance of the Keras models on multiple datasets, and show applications of securing a network such as login authentication and access control, as well as real-world uses for surveillance system. In addition, the survey details applications of face detection where the possibility of face spoofing is an issue and current research directions on the topic. Utilizing CNNs with Keras can enable the development of more flexible, performative, and accurate biometric authentication systems, and these systems play a critical role in the global cybersecurity ecosystem.
- Open Access
- 22 August 2024
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 15 (9) , 4426-4442
Temporal forecasting by converting stochastic behaviour into a stable pattern in electric grid
The malfunction variables of power stations are related to the areas of weather, physical structure, control, and load behavior. To predict temporal power failure is difficult due to their unpredictable characteristics. As high accuracy is normally required, the estimation of failures of short-term temporal prediction is highly difficult. This study presents a method for converting stochastic behavior into a stable pattern, which can subsequently be used in a short-term estimator. For this conversion, K-means clustering is employed, followed by long-short-term memory and gated recurrent unit algorithms are used to perform the short-term estimation. The environment, the operation, and the generated signal factors are all simulated using mathematical models. Weather parameters and load samples have been collected as part of a dataset. Monte-Carlo simulation using MATLAB programming has been used to conduct experimental estimation of failures. The estimated failures of the experiment are then compared with the actual system temporal failures and found to be in good match. Therefore, to address the gap in knowledge for any future power grid estimated failures, the achieved results in this paper form good basis for a testbed to estimate any grid future failures.
- 22 August 2024
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 15 (9) , 4426-4442
Open AccessTemporal forecasting by converting stochastic behaviour into a stable pattern in electric grid
The malfunction variables of power stations are related to the areas of weather, physical structure, control, and load behavior. To predict temporal power failure is difficult due to their unpredictable characteristics. As high accuracy is normally required, the estimation of failures of short-term temporal prediction is highly difficult. This study presents a method for converting stochastic behavior into a stable pattern, which can subsequently be used in a short-term estimator. For this conversion, K-means clustering is employed, followed by long-short-term memory and gated recurrent unit algorithms are used to perform the short-term estimation. The environment, the operation, and the generated signal factors are all simulated using mathematical models. Weather parameters and load samples have been collected as part of a dataset. Monte-Carlo simulation using MATLAB programming has been used to conduct experimental estimation of failures. The estimated failures of the experiment are then compared with the actual system temporal failures and found to be in good match. Therefore, to address the gap in knowledge for any future power grid estimated failures, the achieved results in this paper form good basis for a testbed to estimate any grid future failures.
- 24 August 2023
- Published by Springer Nature
- p. 71-101
Investigating Gesture Control of Robotic Arm via Lora Technology for Smart Cities
- 24 August 2023
- Published by Springer Nature
- p. 71-101
Investigating Gesture Control of Robotic Arm via Lora Technology for Smart Cities
- Open Access
- 24 March 2022
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 13 (5) , 2371-2390
Mining sensor data in a smart environment: a study of control algorithms and microgrid testbed for temporal forecasting and patterns of failure
The generation of active power in renewable energy is dependent on several factors. These variables are related to the areas of weather, physical structure, control, and load behavior. Estimating the future value of the active power to be generated is difficult due to their unpredictable character. However, because of the higher precision required of the estimation, this problem becomes more complex if we examine a short-term temporal prediction. This study presents a method for converting stochastic behavior into a stable pattern, which can subsequently be used in a short-term estimator. For this conversion, K-means clustering is employed, followed by Long-Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) algorithms to perform the Short-term estimate. The environment, the operation, and the generated (normal or faulty) signal are all simulated using mathematical models. Weather parameters and load samples have been collected as part of a dataset. Monte-Carlo simulation using MATLAB programming has been realized to conduct an experiment. In addition, the LSTM and the GRU are compared to see how well they perform in this system. The proposed method's end findings outperform the current state-of-the-art.
- 24 March 2022
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 13 (5) , 2371-2390
Open AccessMining sensor data in a smart environment: a study of control algorithms and microgrid testbed for temporal forecasting and patterns of failure
The generation of active power in renewable energy is dependent on several factors. These variables are related to the areas of weather, physical structure, control, and load behavior. Estimating the future value of the active power to be generated is difficult due to their unpredictable character. However, because of the higher precision required of the estimation, this problem becomes more complex if we examine a short-term temporal prediction. This study presents a method for converting stochastic behavior into a stable pattern, which can subsequently be used in a short-term estimator. For this conversion, K-means clustering is employed, followed by Long-Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) algorithms to perform the Short-term estimate. The environment, the operation, and the generated (normal or faulty) signal are all simulated using mathematical models. Weather parameters and load samples have been collected as part of a dataset. Monte-Carlo simulation using MATLAB programming has been realized to conduct an experiment. In addition, the LSTM and the GRU are compared to see how well they perform in this system. The proposed method's end findings outperform the current state-of-the-art.
- Open Access
- 24 July 2021
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 12 (5) , 976-989
Distribution system power quality compensation using a HSeAPF based on SRF and SMC features
The purpose of this paper is to describe the design of a Hybrid Series Active Power Filter (HSeAPF) system to improve the quality of power on three-phase power distribution grids. The system controls are comprise of Pulse Width Modulation (PWM) based on the Synchronous Reference Frame (SRF) theory, and supported by Phase Locked Loop (PLL) for generating the switching pulses to control a Voltage Source Converter (VSC). The DC link voltage is controlled by Non-Linear Sliding Mode Control (SMC) for faster response and to ensure that it is maintained at a constant value. When this voltage is compared with Proportional Integral (PI), then the improvements made can be shown. The function of HSeAPF control is to eliminate voltage fluctuations, voltage swell/sag, and prevent voltage/current harmonics are produced by both non-linear loads and small inverters connected to the distribution network. A digital Phase Locked Loop that generates frequencies and an oscillating phase-locked output signal controls the voltage. The results from the simulation indicate that the HSeAPF can effectively suppress the dynamic and harmonic reactive power compensation system. Also, the distribution network has a low Total Harmonic Distortion (< 5%), demonstrating that the designed system is efficient, which is an essential requirement when it comes to the IEEE-519 and IEC 61,000–3-6 standards.
- 24 July 2021
- Published by Springer Nature in International Journal of System Assurance Engineering and Management
- Vol. 12 (5) , 976-989
Open AccessDistribution system power quality compensation using a HSeAPF based on SRF and SMC features
The purpose of this paper is to describe the design of a Hybrid Series Active Power Filter (HSeAPF) system to improve the quality of power on three-phase power distribution grids. The system controls are comprise of Pulse Width Modulation (PWM) based on the Synchronous Reference Frame (SRF) theory, and supported by Phase Locked Loop (PLL) for generating the switching pulses to control a Voltage Source Converter (VSC). The DC link voltage is controlled by Non-Linear Sliding Mode Control (SMC) for faster response and to ensure that it is maintained at a constant value. When this voltage is compared with Proportional Integral (PI), then the improvements made can be shown. The function of HSeAPF control is to eliminate voltage fluctuations, voltage swell/sag, and prevent voltage/current harmonics are produced by both non-linear loads and small inverters connected to the distribution network. A digital Phase Locked Loop that generates frequencies and an oscillating phase-locked output signal controls the voltage. The results from the simulation indicate that the HSeAPF can effectively suppress the dynamic and harmonic reactive power compensation system. Also, the distribution network has a low Total Harmonic Distortion (< 5%), demonstrating that the designed system is efficient, which is an essential requirement when it comes to the IEEE-519 and IEC 61,000–3-6 standards.