Megha Mudholkar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Megha Mudholkar has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Pankaj Mudholkar
Marwadi University, Rajkot, India
4 shared publications - Arun Raj S. R
University B.D.T College of Engineering
2 shared publications Budesh Kanwer
Poornima Institute of Engineering and Technology
2 shared publicationsC Madan Kumar
School of CS & AI, SR University
2 shared publications- Subarno Bhattacharyya
O.P. Jindal Global University, India
2 shared publications Vivek V Jog
Symbiosis Skills and Professional University
2 shared publications
Latest Publications
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 303-330
IoT-Enabled Threat Detection for Large-Scale Distributed Networks in Smart Policing
The purpose chapter objectives to design a scalable and intelligent threat detection system to smart policing networks that leverages the IoT technology at scale. It will deal with the problems of timely processing of data of various types of devices and security issues related to such devices. It is suggested to use a multi-layer architecture based on multi-modal (different forms of media) deep learning, near the source (edge) processing of data as well as federated (distributed) learning to provide/network intelligence to the threat detection system. The system is a combination of CNN-BLSTM-based feature learning and adaptive decision mechanisms to allow real-time detection and response. The hybrid dataset of benchmark intrusion data, real IoT traffic, and simulated attack scenarios are used to validate this experiment. The proposed framework has a high detection accuracy (96.7%) and low latency (~280 ms), which is better than the traditional, machine learning, and state-of-the-art deep learning models.
- 25 September 2026
- Published by IGI Global Scientific Publishing
- p. 303-330
IoT-Enabled Threat Detection for Large-Scale Distributed Networks in Smart Policing
The purpose chapter objectives to design a scalable and intelligent threat detection system to smart policing networks that leverages the IoT technology at scale. It will deal with the problems of timely processing of data of various types of devices and security issues related to such devices. It is suggested to use a multi-layer architecture based on multi-modal (different forms of media) deep learning, near the source (edge) processing of data as well as federated (distributed) learning to provide/network intelligence to the threat detection system. The system is a combination of CNN-BLSTM-based feature learning and adaptive decision mechanisms to allow real-time detection and response. The hybrid dataset of benchmark intrusion data, real IoT traffic, and simulated attack scenarios are used to validate this experiment. The proposed framework has a high detection accuracy (96.7%) and low latency (~280 ms), which is better than the traditional, machine learning, and state-of-the-art deep learning models.
- Open Access
- 14 March 2026
- Published by Elsevier in SSRN Electronic Journal
Predicting Soil Suppressiveness: An Integrated IoT and Machine Learning Framework for Pre-Planting Fusarium Wilt Risk Assessment
Soil-borne pathogens such as Fusarium oxysporum are responsible for substantial yield losses in high-value crops, yet most existing digital plant disease detection systems remain reactive, relying on post-infection visual symptoms that emerge too late for effective intervention. While Internet of Things (IoT) technologies and Machine Learning (ML) have been widely adopted for image-based disease diagnosis, their application to pre-planting, soil-level risk assessment remains limited. This study proposes an integrated IoT–ML framework for pre-planting Fusarium wilt risk assessment, leveraging a hybrid feature set that fuses real-time soil sensor data with static farmer-provided soil characteristics. An ESP32-based sensing platform was deployed to collect soil pH, temperature, and moisture, which were combined with manually entered soil texture (sand and clay fractions) and nutrient indicators (NPK). Fifteen supervised Machine Learning models were evaluated to assess predictive reliability under noisy, field-relevant conditions. Among all models, the Bagging Classifier achieved the highest performance, with an accuracy of 92.1% and a Cohen’s Kappa of 0.88, indicating strong agreement beyond chance. In contrast, gradient boosting methods such as XGBoost achieved only 71% accuracy, highlighting their sensitivity to sensor noise and domain variability. Results demonstrate that variance-reduction ensembles, combined with hybrid data fusion, are particularly well-suited for soil-based disease risk modelling. By enabling actionable risk assessment before sowing, the proposed system supports proactive soil management strategies, allowing farmers to implement mitigation measures prior to crop establishment. This work advances the shift from reactive disease detection toward soil-centric, decision-oriented precision agriculture.
- 14 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPredicting Soil Suppressiveness: An Integrated IoT and Machine Learning Framework for Pre-Planting Fusarium Wilt Risk Assessment
Soil-borne pathogens such as Fusarium oxysporum are responsible for substantial yield losses in high-value crops, yet most existing digital plant disease detection systems remain reactive, relying on post-infection visual symptoms that emerge too late for effective intervention. While Internet of Things (IoT) technologies and Machine Learning (ML) have been widely adopted for image-based disease diagnosis, their application to pre-planting, soil-level risk assessment remains limited. This study proposes an integrated IoT–ML framework for pre-planting Fusarium wilt risk assessment, leveraging a hybrid feature set that fuses real-time soil sensor data with static farmer-provided soil characteristics. An ESP32-based sensing platform was deployed to collect soil pH, temperature, and moisture, which were combined with manually entered soil texture (sand and clay fractions) and nutrient indicators (NPK). Fifteen supervised Machine Learning models were evaluated to assess predictive reliability under noisy, field-relevant conditions. Among all models, the Bagging Classifier achieved the highest performance, with an accuracy of 92.1% and a Cohen’s Kappa of 0.88, indicating strong agreement beyond chance. In contrast, gradient boosting methods such as XGBoost achieved only 71% accuracy, highlighting their sensitivity to sensor noise and domain variability. Results demonstrate that variance-reduction ensembles, combined with hybrid data fusion, are particularly well-suited for soil-based disease risk modelling. By enabling actionable risk assessment before sowing, the proposed system supports proactive soil management strategies, allowing farmers to implement mitigation measures prior to crop establishment. This work advances the shift from reactive disease detection toward soil-centric, decision-oriented precision agriculture.
- 20 February 2026
- Published by IGI Global Scientific Publishing
- p. 267-302
Synergizing Quantum and Neuromorphic Architectures for Scalable, Energy-Efficient, and Adaptive Artificial Intelligence Models in IoT Systems
As artificial intelligence is increasingly employed in edge and IoT environments, the need for architectures that combine high fidelity with low energy consumption is urgent. Though neuromorphic and quantum computing methods are powerful in their own right, they typically require tradeoffs between scalability, latency and control granularity. This research explores the nexus of that tradeoff, which discusses a Hybrid Quantum–Neuromorphic Framework (HQNF) that models dynamic conditional redistribution of computational load with adaptive control and reinforcement learning. Existing frameworks tend to address the quantum and neuromorphic layers as a siloed module, with no inherent intelligent decision-making layer to balance uncertainty (reasoning), energy consumption, and output accuracy in real-time. HQNF overcomes these separated operations with a proposed three-tier architecture, enabling reasoning support for high and continuous uncertain states and mechanics to intelligently decide based on observational feedback from knowledge projection.
- 20 February 2026
- Published by IGI Global Scientific Publishing
- p. 267-302
Synergizing Quantum and Neuromorphic Architectures for Scalable, Energy-Efficient, and Adaptive Artificial Intelligence Models in IoT Systems
As artificial intelligence is increasingly employed in edge and IoT environments, the need for architectures that combine high fidelity with low energy consumption is urgent. Though neuromorphic and quantum computing methods are powerful in their own right, they typically require tradeoffs between scalability, latency and control granularity. This research explores the nexus of that tradeoff, which discusses a Hybrid Quantum–Neuromorphic Framework (HQNF) that models dynamic conditional redistribution of computational load with adaptive control and reinforcement learning. Existing frameworks tend to address the quantum and neuromorphic layers as a siloed module, with no inherent intelligent decision-making layer to balance uncertainty (reasoning), energy consumption, and output accuracy in real-time. HQNF overcomes these separated operations with a proposed three-tier architecture, enabling reasoning support for high and continuous uncertain states and mechanics to intelligently decide based on observational feedback from knowledge projection.
- Retraction
- 1 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1
Retraction Notice: Lightweight Security Protocols for IoT Devices in 5G Environments
- 1 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1
RetractionRetraction Notice: Lightweight Security Protocols for IoT Devices in 5G Environments
- Retracted
- 1 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Retracted: Lightweight Security Protocols for IoT Devices in 5G Environments
Contemporary information technology has gradually been used to preserve and deliver vast amounts of health-related data to cut expenses and enhance healthcare institutions. In this setting, the rise of e-health clouds creates new options, such as easy and remote access to healthcare information. Nevertheless, this ability introduces numerous additional hazards and concerns, such as ways to ensure the reliability, safety, and privacy of extremely vulnerable e-Health statistics. Among these problems, security is a significant concern, as it ensures that sensitive clinical statistics in the cloud are not accessible to unauthorized parties. The Internet of Things (IoT) refers to integrated smart gadgets and items over the internet. The explosive rise of IoT and 5G technology presents opportunities for advancement, particularly in medical settings. The smart card, login credentials, and fingerprints are three types of identification that meet the criterion for extreme safety. Several 3-factor ECC-based verification systems for e-healthcare clouds have been proposed currently. Nevertheless, the majority of the protocols have fundamental safety weaknesses and incur significant computational and communication expenses. As a result, this study presents a unique protocol for the e-healthcare cloud that protects against several important vulnerabilities, including anonymous usage, offline login speculation, imitation, and recovered smart card threats. Furthermore, this study performs thorough safety assessments on the protocol employing the Random Oracle System (ROS). The evaluation demonstrates The suggested protocol works better than many of the existing ones because of computing and communication expenses. Therefore, the suggested protocol is shown to be more effective, stronger, and safer.
- 1 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
RetractedRetracted: Lightweight Security Protocols for IoT Devices in 5G Environments
Contemporary information technology has gradually been used to preserve and deliver vast amounts of health-related data to cut expenses and enhance healthcare institutions. In this setting, the rise of e-health clouds creates new options, such as easy and remote access to healthcare information. Nevertheless, this ability introduces numerous additional hazards and concerns, such as ways to ensure the reliability, safety, and privacy of extremely vulnerable e-Health statistics. Among these problems, security is a significant concern, as it ensures that sensitive clinical statistics in the cloud are not accessible to unauthorized parties. The Internet of Things (IoT) refers to integrated smart gadgets and items over the internet. The explosive rise of IoT and 5G technology presents opportunities for advancement, particularly in medical settings. The smart card, login credentials, and fingerprints are three types of identification that meet the criterion for extreme safety. Several 3-factor ECC-based verification systems for e-healthcare clouds have been proposed currently. Nevertheless, the majority of the protocols have fundamental safety weaknesses and incur significant computational and communication expenses. As a result, this study presents a unique protocol for the e-healthcare cloud that protects against several important vulnerabilities, including anonymous usage, offline login speculation, imitation, and recovered smart card threats. Furthermore, this study performs thorough safety assessments on the protocol employing the Random Oracle System (ROS). The evaluation demonstrates The suggested protocol works better than many of the existing ones because of computing and communication expenses. Therefore, the suggested protocol is shown to be more effective, stronger, and safer.