Muhammad Furqan Zia
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
Muhammad Furqan Zia has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
Adlen Ksentini
2 shared publicationsDaniel Massicotte
2 shared publications- Jehad M. Hamamreh
Antalya Bilim University
2 shared publications - Maria Siddiqua
Department of Artificial Intelligence and Data Science, The National University of Computer and Emerging Sciences (NUCES), Karachi, Pakistan
2 shared publications Messaoud Ahmed Ouameur
2 shared publicationsMiloud Bagaa
2 shared publications
Latest Publications
- Open Access
- 13 October 2026
- Published by MDPI AG in Preprints.org
Confidence-Aware Explainable AI for Reliable Large Language Model Decision Support
Large language models (LLMs) are increasingly used in decision-support systems because of their ability to interpret complex queries, synthesize information, and generate contextually relevant responses. However, their use in reliability sensitive settings remains challenging because LLMs can produce incorrect or unsupported outputs while expressing a high degree of linguistic confidence. This paper presents a confidence aware explainable AI framework for improving the reliability of LLM-based decision support. The proposed approach combines model confidence, semantic consistency across multiple stochastic generations, and explanation support to construct a composite reliability score. A selective decision module then uses this score to determine whether a generated response should be accepted, flagged for review, or withheld through abstention. The framework is evaluated across factual, commonsense, and multi-step reasoning tasks using multiple open-weight LLMs. Performance is assessed in terms of answer accuracy, expected calibration error, failure-detection AUROC, selective accuracy, and response coverage. The evaluation investigates whether combining complementary reliability signals provides a more informative assessment of response correctness than self-reported confidence, model confidence, or semantic consistency alone. The proposed framework provides a practical reliability layer for LLM-based decision-support systems by jointly communicating confidence, explanation support, and an actionable response decision, thereby helping distinguish sufficiently reliable outputs from those requiring further verification or abstention.
- 13 October 2026
- Published by MDPI AG in Preprints.org
Open AccessConfidence-Aware Explainable AI for Reliable Large Language Model Decision Support
Large language models (LLMs) are increasingly used in decision-support systems because of their ability to interpret complex queries, synthesize information, and generate contextually relevant responses. However, their use in reliability sensitive settings remains challenging because LLMs can produce incorrect or unsupported outputs while expressing a high degree of linguistic confidence. This paper presents a confidence aware explainable AI framework for improving the reliability of LLM-based decision support. The proposed approach combines model confidence, semantic consistency across multiple stochastic generations, and explanation support to construct a composite reliability score. A selective decision module then uses this score to determine whether a generated response should be accepted, flagged for review, or withheld through abstention. The framework is evaluated across factual, commonsense, and multi-step reasoning tasks using multiple open-weight LLMs. Performance is assessed in terms of answer accuracy, expected calibration error, failure-detection AUROC, selective accuracy, and response coverage. The evaluation investigates whether combining complementary reliability signals provides a more informative assessment of response correctness than self-reported confidence, model confidence, or semantic consistency alone. The proposed framework provides a practical reliability layer for LLM-based decision-support systems by jointly communicating confidence, explanation support, and an actionable response decision, thereby helping distinguish sufficiently reliable outputs from those requiring further verification or abstention.
- Open Access
- 25 February 2026
- Published by Elsevier in Computer Networks
- Vol. 280
A survey on explainable AI for semantic communication: Architecture, challenges, and future opportunities
- 25 February 2026
- Published by Elsevier in Computer Networks
- Vol. 280
Open AccessA survey on explainable AI for semantic communication: Architecture, challenges, and future opportunities
- Open Access
- 24 September 2025
- Published by Elsevier in Physical Communication
- Vol. 73
A survey of domain generalization in AI-enabled semantic communication: Architecture, challenges and future opportunities
- 24 September 2025
- Published by Elsevier in Physical Communication
- Vol. 73
Open AccessA survey of domain generalization in AI-enabled semantic communication: Architecture, challenges and future opportunities
- Open Access
- 21 March 2025
- Published by Science Publishing Group in Advances in Networks
- Vol. 12 (1) , 1-18
Securing the Future: A Survey on Smart Home Security in IoT-Integrated Smart Cities
The rapid growth of urbanization and technological advancements have led to the rise of smart cities and smart homes, where the Internet of Things (IoT) plays a pivotal role. Smart homes enhance energy efficiency, security, and convenience through automated systems and interconnected devices. This survey provides a comprehensive review of smart home architectures, communication technologies, and applications, emphasizing their integration within smart city infrastructures. It explores key components such as sensors, controllers, and cloud-based platforms that enable seamless automation. Additionally, this paper discusses major challenges in smart home security, including privacy risks, cyber threats, and interoperability issues among IoT devices. Security concerns such as unauthorized access, data breaches, and denial-of-service attacks are analyzed, alongside strategies to mitigate these risks. The study also highlights the importance of secure communication protocols, authentication mechanisms, and encryption techniques to ensure the resilience of smart home systems. Furthermore, this survey examines emerging research directions in smart home technology, including AI-driven automation, energy-efficient systems, and blockchain-based security solutions. As smart homes continue to evolve, addressing these challenges will be crucial for their widespread adoption. This paper aims to serve as a valuable resource for researchers, developers, and policymakers seeking to enhance the security and functionality of smart homes within the broader framework of smart cities.
- 21 March 2025
- Published by Science Publishing Group in Advances in Networks
- Vol. 12 (1) , 1-18
Open AccessSecuring the Future: A Survey on Smart Home Security in IoT-Integrated Smart Cities
The rapid growth of urbanization and technological advancements have led to the rise of smart cities and smart homes, where the Internet of Things (IoT) plays a pivotal role. Smart homes enhance energy efficiency, security, and convenience through automated systems and interconnected devices. This survey provides a comprehensive review of smart home architectures, communication technologies, and applications, emphasizing their integration within smart city infrastructures. It explores key components such as sensors, controllers, and cloud-based platforms that enable seamless automation. Additionally, this paper discusses major challenges in smart home security, including privacy risks, cyber threats, and interoperability issues among IoT devices. Security concerns such as unauthorized access, data breaches, and denial-of-service attacks are analyzed, alongside strategies to mitigate these risks. The study also highlights the importance of secure communication protocols, authentication mechanisms, and encryption techniques to ensure the resilience of smart home systems. Furthermore, this survey examines emerging research directions in smart home technology, including AI-driven automation, energy-efficient systems, and blockchain-based security solutions. As smart homes continue to evolve, addressing these challenges will be crucial for their widespread adoption. This paper aims to serve as a valuable resource for researchers, developers, and policymakers seeking to enhance the security and functionality of smart homes within the broader framework of smart cities.
- Open Access
- 30 December 2024
- Published by Pakistan Association for the Advancement of Science in International Journal of Emerging Engineering and Technology
- Vol. 3 (2) , 13-18
A Robust Framework for 2D Human Face Reconstruction from Half-Frontal Views in Low-Quality Surveillance Footage
This paper proposes a robust framework for reconstructing 2D human facial images from half-frontal views, primarily captured under low-quality surveillance conditions. A custom MATLAB-based Graphical User Interface (GUI) is developed to support the complete pipeline, including frame extraction, enhancement, and face reconstruction. Representative frames are extracted and enhanced for video inputs using one of three techniques: histogram equalization, contrast stretching, or logarithmic transformation. Reconstruction involves detecting a single eye from the half-frontal image, followed by horizontal flipping and concatenation to generate a symmetric full-frontal face. The reconstructed faces are validated using the Viola-Jones object detection algorithm to confirm the presence and alignment of facial features. Quantitative evaluation uses the Structural Similarity Index (SSIM) and Jaccard Index (JI) to measure image quality and geometric accuracy. The proposed method is tested on publicly available datasets and a custom-designed dataset reflecting real-world surveillance challenges such as low resolution and poor illumination. Experimental results demonstrate that the framework delivers accurate and visually coherent reconstructions with low computational overhead, making it suitable for real-time surveillance and facial analysis applications.
- 30 December 2024
- Published by Pakistan Association for the Advancement of Science in International Journal of Emerging Engineering and Technology
- Vol. 3 (2) , 13-18
Open AccessA Robust Framework for 2D Human Face Reconstruction from Half-Frontal Views in Low-Quality Surveillance Footage
This paper proposes a robust framework for reconstructing 2D human facial images from half-frontal views, primarily captured under low-quality surveillance conditions. A custom MATLAB-based Graphical User Interface (GUI) is developed to support the complete pipeline, including frame extraction, enhancement, and face reconstruction. Representative frames are extracted and enhanced for video inputs using one of three techniques: histogram equalization, contrast stretching, or logarithmic transformation. Reconstruction involves detecting a single eye from the half-frontal image, followed by horizontal flipping and concatenation to generate a symmetric full-frontal face. The reconstructed faces are validated using the Viola-Jones object detection algorithm to confirm the presence and alignment of facial features. Quantitative evaluation uses the Structural Similarity Index (SSIM) and Jaccard Index (JI) to measure image quality and geometric accuracy. The proposed method is tested on publicly available datasets and a custom-designed dataset reflecting real-world surveillance challenges such as low resolution and poor illumination. Experimental results demonstrate that the framework delivers accurate and visually coherent reconstructions with low computational overhead, making it suitable for real-time surveillance and facial analysis applications.
- 13 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Secrecy and Resilience in Next-Gen Wi-Fi: Exploring a Multi-User Down-Link Non-Orthogonal Transmission Framework
Next-generation Wi-Fi communication necessitates improved spectral efficiency, low latency, expanded coverage, robustness, and enhanced communication security. A major challenge in current and future Wi-Fi communication is preventing eavesdroppers from intercepting live transmissions and ensuring enhanced communication secrecy. The non-orthogonal multiple access (NOMA) scheme has emerged as a potential candidate for future Wi-Fi communication. This is attributed to its notable characteristic of serving multiple users simultaneously on the same time/frequency resources. In this work, my motivation is to propose a novel NOMA transmission scheme that can serve multiple users without any information leakage to an internal or external eavesdropper. My proposed NOMA transmission scheme utilizes physical layer security (PHY) and is less complex, more efficient, and more robust than the conventional power domain NOMA (PD-NOMA) transmission. PD-NOMA suffers from various drawbacks, such as the use of successive interference cancellation (SIC), which demands power sharing among users and has a high probability of information leakage
- 13 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Secrecy and Resilience in Next-Gen Wi-Fi: Exploring a Multi-User Down-Link Non-Orthogonal Transmission Framework
Next-generation Wi-Fi communication necessitates improved spectral efficiency, low latency, expanded coverage, robustness, and enhanced communication security. A major challenge in current and future Wi-Fi communication is preventing eavesdroppers from intercepting live transmissions and ensuring enhanced communication secrecy. The non-orthogonal multiple access (NOMA) scheme has emerged as a potential candidate for future Wi-Fi communication. This is attributed to its notable characteristic of serving multiple users simultaneously on the same time/frequency resources. In this work, my motivation is to propose a novel NOMA transmission scheme that can serve multiple users without any information leakage to an internal or external eavesdropper. My proposed NOMA transmission scheme utilizes physical layer security (PHY) and is less complex, more efficient, and more robust than the conventional power domain NOMA (PD-NOMA) transmission. PD-NOMA suffers from various drawbacks, such as the use of successive interference cancellation (SIC), which demands power sharing among users and has a high probability of information leakage
- 30 June 2023
- Published by IGI Global Scientific Publishing
- p. 213-246
An Overview of Security Issues in Cognitive Radio Ad Hoc Networks
Cognitive radio ad hoc network (CRAHN) is an emerging discipline of network computing. It combines the advantages of cognitive radio networks and mobile ad hoc networks. This chapter starts with an overview of various research issues of CRAHN along with representative solutions for these research issues. Among the various research issues presented, security is discussed in detail due to its prime importance. A review of existing literature reveals that not much work on security has been reported for cognitive radio networks. Specifically, an overview of security issues in CRAHN, presented in this article, is a novel work of this kind. A major part of this article highlights the importance of security in CRAHN and presents an overview of major security issues and the solutions proposed to address these issues in CRAHN.
- 30 June 2023
- Published by IGI Global Scientific Publishing
- p. 213-246
An Overview of Security Issues in Cognitive Radio Ad Hoc Networks
Cognitive radio ad hoc network (CRAHN) is an emerging discipline of network computing. It combines the advantages of cognitive radio networks and mobile ad hoc networks. This chapter starts with an overview of various research issues of CRAHN along with representative solutions for these research issues. Among the various research issues presented, security is discussed in detail due to its prime importance. A review of existing literature reveals that not much work on security has been reported for cognitive radio networks. Specifically, an overview of security issues in CRAHN, presented in this article, is a novel work of this kind. A major part of this article highlights the importance of security in CRAHN and presents an overview of major security issues and the solutions proposed to address these issues in CRAHN.
- Open Access
- 6 April 2021
- Published by Researcherstore
Multi-cell, Multi-user, and Multi-carrier Secure Communication Using Non-Orthogonal Signals’ Superposition with Dual-Transmission for IoT in 6G and Beyond
Considering the advancements of the internet of things (IoT) in 6G and beyond communications, data transmission security in IoT devices has received extensive interest because of their significant features, such as low computational complexity, led by low power requirements. In such devices, the conventional cryptographic techniques may fail to provide secure communication. To fight this drawback, physical layer security (PLS) has remarkable potential to provide security solutions suitable for such applications. In this work, a highly effective PLS technique is proposed for providing secure communication against external and internal eavesdroppers in a downlink multi-cell, multi-user, and multi-carrier IoT communication system. In our proposed system, we considered two base stations, where each base station uses a single radio frequency (RF) chain to link two antennas that are used for the transmission of data. Further, we transmit the data in two rounds, and each round of transmission occurs through a single active antenna of each base station. A different antenna is used for each round of transmission to communicate with two single antenna IoT devices/users in the presence of a passive eavesdropper. In the proposed algorithm, frequency selective channel-based pre-coder matrices and the dual transmission approach are jointly employed. The dual-transmission is performed simultaneously from two base stations to provide security against internal and external eavesdroppers. The proposed system is suitable for IoT-based applications. Also, the potential capabilities of our proposed algorithm are proved by extensive mathematical and simulation analysis.
- 6 April 2021
- Published by Researcherstore
Open AccessMulti-cell, Multi-user, and Multi-carrier Secure Communication Using Non-Orthogonal Signals’ Superposition with Dual-Transmission for IoT in 6G and Beyond
Considering the advancements of the internet of things (IoT) in 6G and beyond communications, data transmission security in IoT devices has received extensive interest because of their significant features, such as low computational complexity, led by low power requirements. In such devices, the conventional cryptographic techniques may fail to provide secure communication. To fight this drawback, physical layer security (PLS) has remarkable potential to provide security solutions suitable for such applications. In this work, a highly effective PLS technique is proposed for providing secure communication against external and internal eavesdroppers in a downlink multi-cell, multi-user, and multi-carrier IoT communication system. In our proposed system, we considered two base stations, where each base station uses a single radio frequency (RF) chain to link two antennas that are used for the transmission of data. Further, we transmit the data in two rounds, and each round of transmission occurs through a single active antenna of each base station. A different antenna is used for each round of transmission to communicate with two single antenna IoT devices/users in the presence of a passive eavesdropper. In the proposed algorithm, frequency selective channel-based pre-coder matrices and the dual transmission approach are jointly employed. The dual-transmission is performed simultaneously from two base stations to provide security against internal and external eavesdroppers. The proposed system is suitable for IoT-based applications. Also, the potential capabilities of our proposed algorithm are proved by extensive mathematical and simulation analysis.
- Open Access
- 3 December 2020
- Published by Researcherstore
An Advanced Non-Orthogonal Multiple Access Security Technique for Future Wireless Communication Networks
The future wireless communication systems demand much more enhanced security and reliability compared to currently deployed systems. In this work, we propose a much simpler yet more efficient physical layer security (PLS) technique for achieving reliable and secure communication in the multiple-input single-output non-orthogonal multiple access (MISO-NOMA) systems. This system is capable of providing enhanced confidential communication as well as inter-user interference cancellation without using the successive interference cancellation (SIC) method. The conventional NOMA was previously adopted under the name of multi-user superposition transmission (MUST) in release 13 of 3GPP but recently excluded from 3GPP-release 17 due to its performance degradation. In this work, we analyze the drawbacks in conventional NOMA and present a new kind of NOMA with more improved performance metrics. The proposed algorithm combines the benefit of pre-coder matrices with simultaneous transmission using antenna diversity to provide simple, reliable, and secure communication without complex processing at the receivers in downlink scenarios. The effectiveness of the proposed algorithm is verified and proven by extensive analysis and numerical simulations.
- 3 December 2020
- Published by Researcherstore
Open AccessAn Advanced Non-Orthogonal Multiple Access Security Technique for Future Wireless Communication Networks
The future wireless communication systems demand much more enhanced security and reliability compared to currently deployed systems. In this work, we propose a much simpler yet more efficient physical layer security (PLS) technique for achieving reliable and secure communication in the multiple-input single-output non-orthogonal multiple access (MISO-NOMA) systems. This system is capable of providing enhanced confidential communication as well as inter-user interference cancellation without using the successive interference cancellation (SIC) method. The conventional NOMA was previously adopted under the name of multi-user superposition transmission (MUST) in release 13 of 3GPP but recently excluded from 3GPP-release 17 due to its performance degradation. In this work, we analyze the drawbacks in conventional NOMA and present a new kind of NOMA with more improved performance metrics. The proposed algorithm combines the benefit of pre-coder matrices with simultaneous transmission using antenna diversity to provide simple, reliable, and secure communication without complex processing at the receivers in downlink scenarios. The effectiveness of the proposed algorithm is verified and proven by extensive analysis and numerical simulations.