Mahdi Nangir
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
Mahdi Nangir has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
Abdolrasoul Sakhaei Gharagezlou
Faculty of Electrical and Computer Engineering University of Tabriz Tabriz Iran
6 shared publications- Javad Musevi Niya
Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
5 shared publications - Nima Imani
Department of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
5 shared publications - Abdolrasoul Sakhaei Gharagezlou
Department of Electrical and Computer Engineering, University of Tabriz, Tabriz, Iran
4 shared publications Nima Imani
Faculty of Electrical and Computer Engineering University of Tabriz Tabriz Iran
4 shared publications- Reza Asvadi
Shahid Beheshti University
4 shared publications
Latest Publications
- 13 February 2026
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 21 (4) , 4105-4121
MM Wave (4 × 1) MIMO Planar Printed Antenna with High Isolation Using a Hashtag-Like Decoupling Technique for 5G Applications
- 13 February 2026
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 21 (4) , 4105-4121
MM Wave (4 × 1) MIMO Planar Printed Antenna with High Isolation Using a Hashtag-Like Decoupling Technique for 5G Applications
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Image Processing
- Vol. 20 (1)
Architecture‐Aware Block‐Level Pruning Across CNN Models for Breast Cancer Histopathology Image Classification
Convolutional neural networks (CNNs) have achieved strong performance in medical image analysis; however, their deployment in edge computing environments remains challenging. This paper introduces an architecture‐aware block‐level pruning framework that progressively removes implementation‐level blocks from sequential (VGG16, VGG19) and branched (ResNet50, InceptionV3) CNN architectures. Each pruning step is followed by training under one of three paradigms—training from scratch, feature extraction, or fine‐tuning—and performance is evaluated through the accuracy–efficiency trade‐off and complementary metrics. To exploit the diverse and architecture‐dependent representations learned by the best‐performing models in each regime, a stacking ensemble strategy is employed. Experiments on the PatchCamelyon dataset reveal a distinctly architecture‐ and regime‐dependent pruning behavior: sequential models trained from scratch and fine‐tuned exhibit monotonic accuracy–efficiency degradation, while under feature extraction they show non‐monotonic patterns; branched architectures display non‐monotonic behavior across all regimes due to their multi‐path or residual block redundancy. Under moderate pruning, pruned Inception‐V3 model outperforms prior state‐of‐the‐art (SOTA) with a 5.47% accuracy improvement; under strong pruning, pruned Inception‐V3 and ResNet‐50 models exceed the literature with 4.37% and 5.38% accuracy enhancement, respectively. In aggressive pruning, two different pruned configuration of Inception‐V3 achieve superior performance relative to SOTA with 1.21% and 10% accuracy improvements. The stacking ensembles further yield 1.78%–6.61% accuracy gains, 2%–5% AUC improvements, and 4%–6% increases in F1‐score over individual CNN backbones. These findings demonstrate that the notable improvements obtained via ensemble learning reflect the strong capability of the pruned networks, which maintain excellent predictive behavior even with their compressed architectural design.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in IET Image Processing
- Vol. 20 (1)
Open AccessArchitecture‐Aware Block‐Level Pruning Across CNN Models for Breast Cancer Histopathology Image Classification
Convolutional neural networks (CNNs) have achieved strong performance in medical image analysis; however, their deployment in edge computing environments remains challenging. This paper introduces an architecture‐aware block‐level pruning framework that progressively removes implementation‐level blocks from sequential (VGG16, VGG19) and branched (ResNet50, InceptionV3) CNN architectures. Each pruning step is followed by training under one of three paradigms—training from scratch, feature extraction, or fine‐tuning—and performance is evaluated through the accuracy–efficiency trade‐off and complementary metrics. To exploit the diverse and architecture‐dependent representations learned by the best‐performing models in each regime, a stacking ensemble strategy is employed. Experiments on the PatchCamelyon dataset reveal a distinctly architecture‐ and regime‐dependent pruning behavior: sequential models trained from scratch and fine‐tuned exhibit monotonic accuracy–efficiency degradation, while under feature extraction they show non‐monotonic patterns; branched architectures display non‐monotonic behavior across all regimes due to their multi‐path or residual block redundancy. Under moderate pruning, pruned Inception‐V3 model outperforms prior state‐of‐the‐art (SOTA) with a 5.47% accuracy improvement; under strong pruning, pruned Inception‐V3 and ResNet‐50 models exceed the literature with 4.37% and 5.38% accuracy enhancement, respectively. In aggressive pruning, two different pruned configuration of Inception‐V3 achieve superior performance relative to SOTA with 1.21% and 10% accuracy improvements. The stacking ensembles further yield 1.78%–6.61% accuracy gains, 2%–5% AUC improvements, and 4%–6% increases in F1‐score over individual CNN backbones. These findings demonstrate that the notable improvements obtained via ensemble learning reflect the strong capability of the pruned networks, which maintain excellent predictive behavior even with their compressed architectural design.
- 15 November 2025
- Published by Elsevier in Measurement
- Vol. 259
Non-destructive measurement of rebar location and depth in concrete based on antenna reflection coefficient analysis
- 15 November 2025
- Published by Elsevier in Measurement
- Vol. 259
Non-destructive measurement of rebar location and depth in concrete based on antenna reflection coefficient analysis
- Open Access
Utilizing Model-Free Reinforcement Learning for Optimizing Secure Multi-Party Computation Protocols
In this manuscript, we explore the application of model-free reinforcement learning in optimizing secure multiparty computation (SMPC) protocols. SMPC is a crucial tool for performing computations on private data without the need to disclose it, holding significant importance in various domains, including information security and privacy. However, the efficiency of current protocols is often suboptimal due to computational and communicational complexities. Our proposed approach leverages model-free reinforcement learning algorithms to enhance the performance of these protocols. We have designed a reinforcement learning model capable of dynamically learning and adapting optimal strategies for secure computations. Our experimental results demonstrate that employing this method leads to a substantial reduction in execution time and communication costs of the protocols. These achievements highlight the high potential of reinforcement learning in improving the efficiency of secure multiparty computation protocols, providing an effective solution to the existing challenges in this field.
Open AccessUtilizing Model-Free Reinforcement Learning for Optimizing Secure Multi-Party Computation Protocols
In this manuscript, we explore the application of model-free reinforcement learning in optimizing secure multiparty computation (SMPC) protocols. SMPC is a crucial tool for performing computations on private data without the need to disclose it, holding significant importance in various domains, including information security and privacy. However, the efficiency of current protocols is often suboptimal due to computational and communicational complexities. Our proposed approach leverages model-free reinforcement learning algorithms to enhance the performance of these protocols. We have designed a reinforcement learning model capable of dynamically learning and adapting optimal strategies for secure computations. Our experimental results demonstrate that employing this method leads to a substantial reduction in execution time and communication costs of the protocols. These achievements highlight the high potential of reinforcement learning in improving the efficiency of secure multiparty computation protocols, providing an effective solution to the existing challenges in this field.
- 8 September 2025
- Published by Springer Nature in Telecommunication Systems
- Vol. 88 (4) , 1-10
Joint Power Allocation and Optimal Antenna Selection for the Massive MIMO Systems with Imperfect CSI and the MRT Precoding
- 8 September 2025
- Published by Springer Nature in Telecommunication Systems
- Vol. 88 (4) , 1-10
Joint Power Allocation and Optimal Antenna Selection for the Massive MIMO Systems with Imperfect CSI and the MRT Precoding
- 18 August 2025
- Published by Polish Academy of Sciences Chancellery in Opto-Electronics Review
- p. 155677
Developing threat detection and weather impact techniques by AI algorithms to enhance the reliability of FSO/RF system
Free space optical (FSO) and radio frequency (RF) communication systems need artificial intelligence (AI) to increase their reliability against cyber threats, as well as the vagaries of bad weather. This paper presents a new AI-decision layer of operation of a hybrid FSO/RF system what dynamically ensures its security and operational stability in case of environmental (fog/dust) and security (eavesdropping/jamming) threats. The authors’ technique fundamentally juxtaposes fuzzy logic rule-based classification with multi-algorithm machine learning (ML) validation (54 actionable rules k-nearest neighbours (KNN), support vector machine (SVM), artificial neural networks (ANN)) towards 99.9% real-time response optimization, vastly superior to conventional threshold-based applications. To the authors’ knowledge, this is the first architecture to accommodate adaptive channel switching/encryption in the < 0.1 ms latency regime while maintaining the high-speed benefits of FSO. Experimental results show that in terms of accuracy, error rate, and the balance between precision and recall, ANN is superior to KNN and SVM. ANN achieves the highest classification accuracy with the fewest false positive rates. The significance of the results lies in their ability to improve the security and efficiency of hybrid FSO/RF systems in a way that requires minimal human intervention.
- 18 August 2025
- Published by Polish Academy of Sciences Chancellery in Opto-Electronics Review
- p. 155677
Developing threat detection and weather impact techniques by AI algorithms to enhance the reliability of FSO/RF system
Free space optical (FSO) and radio frequency (RF) communication systems need artificial intelligence (AI) to increase their reliability against cyber threats, as well as the vagaries of bad weather. This paper presents a new AI-decision layer of operation of a hybrid FSO/RF system what dynamically ensures its security and operational stability in case of environmental (fog/dust) and security (eavesdropping/jamming) threats. The authors’ technique fundamentally juxtaposes fuzzy logic rule-based classification with multi-algorithm machine learning (ML) validation (54 actionable rules k-nearest neighbours (KNN), support vector machine (SVM), artificial neural networks (ANN)) towards 99.9% real-time response optimization, vastly superior to conventional threshold-based applications. To the authors’ knowledge, this is the first architecture to accommodate adaptive channel switching/encryption in the < 0.1 ms latency regime while maintaining the high-speed benefits of FSO. Experimental results show that in terms of accuracy, error rate, and the balance between precision and recall, ANN is superior to KNN and SVM. ANN achieves the highest classification accuracy with the fewest false positive rates. The significance of the results lies in their ability to improve the security and efficiency of hybrid FSO/RF systems in a way that requires minimal human intervention.
- 17 June 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 167
A general MMSE joint detection and estimation of speech signal based on generalized gamma priors
- 17 June 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 167
A general MMSE joint detection and estimation of speech signal based on generalized gamma priors