Mohamed Hefeida
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mohamed Hefeida has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Sameh Sorour
School of Computing, Queens University, Kingston, ON, Canada
5 shared publications - Umair Mohammad
Florida International University
3 shared publications - Amr S. El-Wakeel
Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USA
2 shared publications - Obinna Agbo
Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, WV, USA
2 shared publications - Umair Mohammad
School of Computing and Information Sciences Florida International University Miami FL USA
2 shared publications - Aboelmagd Noureldin
Royal Military College of Canada
1 shared publication
Latest Publications
- 23 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (17) , 36353-36369
Hybrid-CNN Intrusion Detection Framework for CAN Networks in Connected and Autonomous Vehicles
The increasing cyber–physical integration in connected and autonomous vehicles (CAVs) has amplified the need for robust intrusion detection systems (IDSs) to secure controller area network (CAN) communications. Existing deep learning approaches, such as CNNs and LSTMs, have demonstrated strong potential but often struggle to detect stealthy and sophisticated attacks, primarily due to inadequate temporal modeling and the use of datasets with limited fidelity that do not reflect real-world attack complexities. To address these limitations, we propose an optimized hybrid IDS model that integrates 1-D convolutional neural networks (1-D-CNN) with bidirectional long short-term memory (BiLSTM) layers. The architecture is specifically designed to extract spatio-temporal features from CAN traffic with minimal complexity, enabling efficient detection of subtle attack patterns. Our model incorporates a time-aware design that allows it to detect attacks as they evolve, even when they occur in localized segments of the payload. It was trained and evaluated using the high-fidelity ORNL ROAD dataset, which includes physically validated fuzzing and targeted CAN ID attacks. The proposed model achieved an accuracy of 99.73%, a Matthews Correlation Coefficient (MCC) of 0.9401, and maintained average low false positive of (≤ 0.00045) and false negative of (≤ 0.00047) rates. It also demonstrated a low average prediction latency of 0.535 ms per CAN frame, highlighting its architectural efficiency. This performance reflects a deliberate design balance between lightweight complexity and high detection accuracy, advancing the development of practical and robust deep learning-based IDS solutions for modern vehicular networks.
- 23 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (17) , 36353-36369
Hybrid-CNN Intrusion Detection Framework for CAN Networks in Connected and Autonomous Vehicles
The increasing cyber–physical integration in connected and autonomous vehicles (CAVs) has amplified the need for robust intrusion detection systems (IDSs) to secure controller area network (CAN) communications. Existing deep learning approaches, such as CNNs and LSTMs, have demonstrated strong potential but often struggle to detect stealthy and sophisticated attacks, primarily due to inadequate temporal modeling and the use of datasets with limited fidelity that do not reflect real-world attack complexities. To address these limitations, we propose an optimized hybrid IDS model that integrates 1-D convolutional neural networks (1-D-CNN) with bidirectional long short-term memory (BiLSTM) layers. The architecture is specifically designed to extract spatio-temporal features from CAN traffic with minimal complexity, enabling efficient detection of subtle attack patterns. Our model incorporates a time-aware design that allows it to detect attacks as they evolve, even when they occur in localized segments of the payload. It was trained and evaluated using the high-fidelity ORNL ROAD dataset, which includes physically validated fuzzing and targeted CAN ID attacks. The proposed model achieved an accuracy of 99.73%, a Matthews Correlation Coefficient (MCC) of 0.9401, and maintained average low false positive of (≤ 0.00045) and false negative of (≤ 0.00047) rates. It also demonstrated a low average prediction latency of 0.535 ms per CAN frame, highlighting its architectural efficiency. This performance reflects a deliberate design balance between lightweight complexity and high detection accuracy, advancing the development of practical and robust deep learning-based IDS solutions for modern vehicular networks.
- Open Access
- 4 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 23-31
Optimized Deep Learning-based Intrusion Detection System for Connected Vehicles
- 4 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 23-31
Open AccessOptimized Deep Learning-based Intrusion Detection System for Connected Vehicles
- Open Access
- 4 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 118122-118134
AutoTA: A Dynamic Intent-Based Virtual Teaching Assistant for Students Using Open Source LLMs
Large Language Models (LLMs) are explored for their potential to transform education by serving as virtual teaching assistants, offering personalized support through human-like responses to tasks such as content-related questions and coursework guidance. In this study, we present a novel framework that leverages intent classification to enhance the effectiveness of LLMs in this role. Our framework, AutoTA, categorizes student queries into distinct topics— lecture discussions, homework assistance, and syllabus questions—triggering specific conversation chains tailored to each intent. Additionally, we incorporate a custom vector-space filter that refines responses based on filename tracking after intent identification. To evaluate the framework, we used course materials from the undergraduate-level CS course, Computer Incident Response, and compared the performance of several open-source LLMs, including Llama 3.1. Our results show that the framework accurately classifies intent and provides appropriate guidance, measured through quantitative and qualitative metrics. These findings highlight the potential of the proposed framework to enhance personalized learning and improve student engagement. While tested in a computer science course, the framework incorporates diverse assessment types that suggest potential for broader application.
- 4 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 118122-118134
Open AccessAutoTA: A Dynamic Intent-Based Virtual Teaching Assistant for Students Using Open Source LLMs
Large Language Models (LLMs) are explored for their potential to transform education by serving as virtual teaching assistants, offering personalized support through human-like responses to tasks such as content-related questions and coursework guidance. In this study, we present a novel framework that leverages intent classification to enhance the effectiveness of LLMs in this role. Our framework, AutoTA, categorizes student queries into distinct topics— lecture discussions, homework assistance, and syllabus questions—triggering specific conversation chains tailored to each intent. Additionally, we incorporate a custom vector-space filter that refines responses based on filename tracking after intent identification. To evaluate the framework, we used course materials from the undergraduate-level CS course, Computer Incident Response, and compared the performance of several open-source LLMs, including Llama 3.1. Our results show that the framework accurately classifies intent and provides appropriate guidance, measured through quantitative and qualitative metrics. These findings highlight the potential of the proposed framework to enhance personalized learning and improve student engagement. While tested in a computer science course, the framework incorporates diverse assessment types that suggest potential for broader application.
- 12 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (8) , 10978-10989
STC: Spatial and Temporal Clustering for Cooperative Perception System
Transportation safety is a very important concern for many decision-makers. Not only does it aim at preventing vehicle accidents, but also saves the lives of individuals who may be distracted while crossing the streets. Various traffic conditions can hinder the visibility of pedestrians and other vehicles when only relying on the local sensors attached to these AVs. Therefore, Cooperative Perception (CP) among Connected Autonomous Vehicles (CAV) is suggested to overcome this problem by utilizing inter-vehicle communication. However, communication network limitations, including limited bandwidth, packet loss, operator compatibilities, or even lack of coverage, can significantly impede the performance of these cooperative perception solutions. To this end, we propose a spatial safety-aware clustering algorithm. This algorithm clusters the perceived objects across AVs. This effective idea allows a decrease in communication payload by approximately 20% and efficiently increases the information reception over an infrastructure-less Vehicle-to-Vehicle (V2V) network by 10% compared to ETSI. Furthermore, we suggest integration of the spatial clustering algorithm with existing baselines, showing improvements of up to 18% for road object perception. Lastly, we propose a Spatial and Temporal Clustering (STC) approach that performs a clustering for information sent over the time domain in addition to the spatial clustering. This further decreases the payload by 41% and increases the perception up to 37% while showing more than a 12X increase in the reception of safety-relevant information compared to ETSI.
- 12 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 73 (8) , 10978-10989
STC: Spatial and Temporal Clustering for Cooperative Perception System
Transportation safety is a very important concern for many decision-makers. Not only does it aim at preventing vehicle accidents, but also saves the lives of individuals who may be distracted while crossing the streets. Various traffic conditions can hinder the visibility of pedestrians and other vehicles when only relying on the local sensors attached to these AVs. Therefore, Cooperative Perception (CP) among Connected Autonomous Vehicles (CAV) is suggested to overcome this problem by utilizing inter-vehicle communication. However, communication network limitations, including limited bandwidth, packet loss, operator compatibilities, or even lack of coverage, can significantly impede the performance of these cooperative perception solutions. To this end, we propose a spatial safety-aware clustering algorithm. This algorithm clusters the perceived objects across AVs. This effective idea allows a decrease in communication payload by approximately 20% and efficiently increases the information reception over an infrastructure-less Vehicle-to-Vehicle (V2V) network by 10% compared to ETSI. Furthermore, we suggest integration of the spatial clustering algorithm with existing baselines, showing improvements of up to 18% for road object perception. Lastly, we propose a Spatial and Temporal Clustering (STC) approach that performs a clustering for information sent over the time domain in addition to the spatial clustering. This further decreases the payload by 41% and increases the perception up to 37% while showing more than a 12X increase in the reception of safety-relevant information compared to ETSI.
- 14 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Green Communications and Networking
- Vol. 7 (4) , 1766-1777
Energy Aware Task Allocation for Semi-Asynchronous Mobile Edge Learning
This paper extends the paradigm of "mobile edge learning (MEL)" by designing an energy-aware optimal task allocation scheme for training a machine learning (ML) model in a semi-asynchronous manner across multiple learners connected via the resource-constrained wireless edge network. The tasks are allocated such that the local dataset size selected at each learner ensures completion within a given global delay constraint and a local maximum energy consumption limit. Hence, the designed method is heterogeneity aware (HA) because it offers a trade-off between resource consumption and MEL performance by directly relating the time and energy consumption to the heterogeneous communication/computational capabilities of learners. Because the resulting optimization is an NP-hard quadratically-constrained integer linear program (QCILP), a two-step suggest-and-improve (SAI) solution is proposed. The proposed HA semi-asynchronous (HA-Asyn) approach is compared against the HA synchronous (HA-Sync) scheme and the heterogeneity unaware (HU) synchronous/asynchronous (HU-Sync/Asyn) equal batch allocation schemes. Results from a system of 20 learners tested for various completion time and energy consumption constraints show that the proposed HA-Asyn method works better than the HU-Sync/Asyn approaches and can even provide gains of up-to 25% compared to the HA-Sync scheme.
- 14 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Green Communications and Networking
- Vol. 7 (4) , 1766-1777
Energy Aware Task Allocation for Semi-Asynchronous Mobile Edge Learning
This paper extends the paradigm of "mobile edge learning (MEL)" by designing an energy-aware optimal task allocation scheme for training a machine learning (ML) model in a semi-asynchronous manner across multiple learners connected via the resource-constrained wireless edge network. The tasks are allocated such that the local dataset size selected at each learner ensures completion within a given global delay constraint and a local maximum energy consumption limit. Hence, the designed method is heterogeneity aware (HA) because it offers a trade-off between resource consumption and MEL performance by directly relating the time and energy consumption to the heterogeneous communication/computational capabilities of learners. Because the resulting optimization is an NP-hard quadratically-constrained integer linear program (QCILP), a two-step suggest-and-improve (SAI) solution is proposed. The proposed HA semi-asynchronous (HA-Asyn) approach is compared against the HA synchronous (HA-Sync) scheme and the heterogeneity unaware (HU) synchronous/asynchronous (HU-Sync/Asyn) equal batch allocation schemes. Results from a system of 20 learners tested for various completion time and energy consumption constraints show that the proposed HA-Asyn method works better than the HU-Sync/Asyn approaches and can even provide gains of up-to 25% compared to the HA-Sync scheme.
- 21 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1
Dynamic Task Allocation for Mobile Edge Learning
This paper introduces the new paradigm of Mobile Edge Learning "MEL" that enables the implementation of realistic distributed machine learning (DML) tasks on wireless edge nodes while taking into consideration the heterogeneous computing and networking environments. Therefore, a heterogeneity aware (HA) scheme is designed to solve the problem of dynamic task allocation for MEL in a way that maximizes the DML accuracy over wireless heterogeneous nodes or 'learners' while respecting the time constraints. The problem is first formulated as a quadratically-constrained integer linear program (QCILP). Being NP-hard, it is relaxed into a non-convex problem over real variables which can be solved using commercially available numerical solvers. The relaxation also allows us to propose a solution based on deriving the analytical upper bounds of the optimal solution using Lagrangian analysis and Karush-Kuhn-Tucker (KKT) conditions. The merits of the proposed analytical solution are demonstrated by comparing its performance to the numerical approaches and comparing the validation accuracy of the proposed HA scheme to the baseline heterogeneity unaware (HU) equal task allocation approach. Simulation results show that the HA schemes decrease convergence time up-to 56% and increase the final validation accuracy up-to 8%.
- 21 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1
Dynamic Task Allocation for Mobile Edge Learning
This paper introduces the new paradigm of Mobile Edge Learning "MEL" that enables the implementation of realistic distributed machine learning (DML) tasks on wireless edge nodes while taking into consideration the heterogeneous computing and networking environments. Therefore, a heterogeneity aware (HA) scheme is designed to solve the problem of dynamic task allocation for MEL in a way that maximizes the DML accuracy over wireless heterogeneous nodes or 'learners' while respecting the time constraints. The problem is first formulated as a quadratically-constrained integer linear program (QCILP). Being NP-hard, it is relaxed into a non-convex problem over real variables which can be solved using commercially available numerical solvers. The relaxation also allows us to propose a solution based on deriving the analytical upper bounds of the optimal solution using Lagrangian analysis and Karush-Kuhn-Tucker (KKT) conditions. The merits of the proposed analytical solution are demonstrated by comparing its performance to the numerical approaches and comparing the validation accuracy of the proposed HA scheme to the baseline heterogeneity unaware (HU) equal task allocation approach. Simulation results show that the HA schemes decrease convergence time up-to 56% and increase the final validation accuracy up-to 8%.
- Open Access
Jointly Optimizing Dataset Size and Local Updates in Heterogeneous Mobile Edge Learning
This paper proposes to maximize the accuracy of a distributed machine learning (ML) model trained on learners connected via the resource-constrained wireless edge. We jointly optimize the number of local/global updates and the task size allocation to minimize the loss while taking into account heterogeneous communication and computation capabilities of each learner. By leveraging existing bounds on the difference between the training loss at any given iteration and the theoretically optimal loss, we derive an expression for the objective function in terms of the number of local updates. The resulting convex program is solved to obtain the optimal number of local updates which is used to obtain the total updates and batch sizes for each learner. The merits of the proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performance to the heterogeneity unaware (HU) approach.
Open AccessJointly Optimizing Dataset Size and Local Updates in Heterogeneous Mobile Edge Learning
This paper proposes to maximize the accuracy of a distributed machine learning (ML) model trained on learners connected via the resource-constrained wireless edge. We jointly optimize the number of local/global updates and the task size allocation to minimize the loss while taking into account heterogeneous communication and computation capabilities of each learner. By leveraging existing bounds on the difference between the training loss at any given iteration and the theoretically optimal loss, we derive an expression for the objective function in terms of the number of local updates. The resulting convex program is solved to obtain the optimal number of local updates which is used to obtain the total updates and batch sizes for each learner. The merits of the proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performance to the heterogeneity unaware (HU) approach.
- 9 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Optimal Task Allocation for Mobile Edge Learning with Global Training Time Constraints
This paper proposes to maximize the accuracy of a distributed machine learning (ML) model trained on learners connected via the resource-constrained wireless edge. We jointly optimize the number of local/global updates and the task size allocation to minimize the loss while taking into account heterogeneous communication and computation capabilities of each learner. By leveraging existing bounds on the difference between the optimal and actual training loss, we derive an expression for the objective function in terms of the local updates. The resulting convex program is solved to obtain the optimal number of local updates which is used to obtain the total updates and batch sizes for each learner. The merits of the proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performance to the heterogeneity unaware (HU) approach.
- 9 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Optimal Task Allocation for Mobile Edge Learning with Global Training Time Constraints
This paper proposes to maximize the accuracy of a distributed machine learning (ML) model trained on learners connected via the resource-constrained wireless edge. We jointly optimize the number of local/global updates and the task size allocation to minimize the loss while taking into account heterogeneous communication and computation capabilities of each learner. By leveraging existing bounds on the difference between the optimal and actual training loss, we derive an expression for the objective function in terms of the local updates. The resulting convex program is solved to obtain the optimal number of local updates which is used to obtain the total updates and batch sizes for each learner. The merits of the proposed solution, which is heterogeneity aware (HA), are exhibited by comparing its performance to the heterogeneity unaware (HU) approach.
- Open Access
Task Allocation for Asynchronous Mobile Edge Learning with Delay and Energy Constraints
This paper extends the paradigm of "mobile edge learning (MEL)" by designing an optimal task allocation scheme for training a machine learning model in an asynchronous manner across mutiple edge nodes or learners connected via a resource-constrained wireless edge network. The optimization is done such that the portion of the task allotted to each learner is completed within a given global delay constraint and a local maximum energy consumption limit. The time and energy consumed are related directly to the heterogeneous communication and computational capabilities of the learners; i.e. the proposed model is heterogeneity aware (HA). Because the resulting optimization is an NP-hard quadratically-constrained integer linear program (QCILP), a two-step suggest-and-improve (SAI) solution is proposed based on using the solution of the relaxed synchronous problem to obtain the solution to the asynchronous problem. The proposed HA asynchronous (HA-Asyn) approach is compared against the HA synchronous (HA-Sync) scheme and the heterogeneity unaware (HU) equal batch allocation scheme. Results from a system of 20 learners tested for various completion time and energy consumption constraints show that the proposed HA-Asyn method works better than the HU synchronous/asynchronous (HU-Sync/Asyn) approach and can provide gains of up-to 25\% compared to the HA-Sync scheme.
Open AccessTask Allocation for Asynchronous Mobile Edge Learning with Delay and Energy Constraints
This paper extends the paradigm of "mobile edge learning (MEL)" by designing an optimal task allocation scheme for training a machine learning model in an asynchronous manner across mutiple edge nodes or learners connected via a resource-constrained wireless edge network. The optimization is done such that the portion of the task allotted to each learner is completed within a given global delay constraint and a local maximum energy consumption limit. The time and energy consumed are related directly to the heterogeneous communication and computational capabilities of the learners; i.e. the proposed model is heterogeneity aware (HA). Because the resulting optimization is an NP-hard quadratically-constrained integer linear program (QCILP), a two-step suggest-and-improve (SAI) solution is proposed based on using the solution of the relaxed synchronous problem to obtain the solution to the asynchronous problem. The proposed HA asynchronous (HA-Asyn) approach is compared against the HA synchronous (HA-Sync) scheme and the heterogeneity unaware (HU) equal batch allocation scheme. Results from a system of 20 learners tested for various completion time and energy consumption constraints show that the proposed HA-Asyn method works better than the HU synchronous/asynchronous (HU-Sync/Asyn) approach and can provide gains of up-to 25\% compared to the HA-Sync scheme.