T. V. Padmavathy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
T. V. Padmavathy has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Indra Priyadharshini Sundar
School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India
4 shared publications - Abdul Quadir
School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India.
1 shared publication Abdul Quadir Md
School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India
1 shared publication- Brindha Subburaj
School of Computer Science and Engineering, Vellore Institute of Technology
1 shared publication - D. Shiny Irene
Department of Computing Technologies, School of Computing, SRM Institute of Science and Technology, Kattankulathur, India
1 shared publication - Dayang Norhayati Abang Jawawi
Universiti Teknologi Malaysia
1 shared publication
Latest Publications
- Open Access
- 21 September 2026
- Published by Springer Nature in Scientific Reports
Quantum game-theoretic reinforcement learning for autonomous vehicle navigation and communication in vehicular networks
Autonomous vehicle (AV) navigation and communication in highly dynamic vehicular networks require fast decision-making, reliable coordination, and energy-efficient connectivity. However, traditional Reinforcement Learning (RL) and game-theoretic approaches often suffer from slow convergence and limited scalability in dense multi-agent environments. To address these challenges, this paper proposes a novel Quantum Game-Theoretic Reinforcement Learning (QGTRL) framework that integrates quantum-inspired policy encoding using Parameterized Quantum Circuits (PQCs) with game-theoretic multi-agent coordination. The proposed approach uses quantum superposition and complex probability amplitudes to enhance exploration efficiency and improve policy convergence stability. In addition, a communication-aware coordination mechanism is incorporated to support robust interaction in dynamic Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) environments. The QGTRL framework is benchmarked against state-of-the-art methods, including Deep Q-Network (DQN), Multi-Agent Deep Deterministic Policy Gradient (MADDPG), Quantum Long Short-Term Memory (QLSTM) Deep Recurrent Q-Network (QLSTM-DRQN), and Quantum Deep Reinforcement Learning (QDRL), under realistic simulation conditions. Experimental results demonstrate that QGTRL improves path-planning accuracy by 18.7%, reduces communication latency by 15.3%, and decreases energy consumption by 22.4% compared to baseline models, while maintaining higher mission success rates in dense and high-mobility scenarios. Furthermore, the framework operates on quantum-inspired classical computation, eliminating the need for specialized quantum hardware and enabling practical deployment. These results highlight QGTRL as a scalable, efficient, and robust solution for next-generation cooperative autonomous vehicle systems.
- 21 September 2026
- Published by Springer Nature in Scientific Reports
Open AccessQuantum game-theoretic reinforcement learning for autonomous vehicle navigation and communication in vehicular networks
Autonomous vehicle (AV) navigation and communication in highly dynamic vehicular networks require fast decision-making, reliable coordination, and energy-efficient connectivity. However, traditional Reinforcement Learning (RL) and game-theoretic approaches often suffer from slow convergence and limited scalability in dense multi-agent environments. To address these challenges, this paper proposes a novel Quantum Game-Theoretic Reinforcement Learning (QGTRL) framework that integrates quantum-inspired policy encoding using Parameterized Quantum Circuits (PQCs) with game-theoretic multi-agent coordination. The proposed approach uses quantum superposition and complex probability amplitudes to enhance exploration efficiency and improve policy convergence stability. In addition, a communication-aware coordination mechanism is incorporated to support robust interaction in dynamic Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) environments. The QGTRL framework is benchmarked against state-of-the-art methods, including Deep Q-Network (DQN), Multi-Agent Deep Deterministic Policy Gradient (MADDPG), Quantum Long Short-Term Memory (QLSTM) Deep Recurrent Q-Network (QLSTM-DRQN), and Quantum Deep Reinforcement Learning (QDRL), under realistic simulation conditions. Experimental results demonstrate that QGTRL improves path-planning accuracy by 18.7%, reduces communication latency by 15.3%, and decreases energy consumption by 22.4% compared to baseline models, while maintaining higher mission success rates in dense and high-mobility scenarios. Furthermore, the framework operates on quantum-inspired classical computation, eliminating the need for specialized quantum hardware and enabling practical deployment. These results highlight QGTRL as a scalable, efficient, and robust solution for next-generation cooperative autonomous vehicle systems.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 7, 9352-9374
Explainable AI-Driven Intrusion Detection and Blockchain-Based Trust Evolution for Secure Routing in Underwater Sensor Networks
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Communications Society
- Vol. 7, 9352-9374
Open AccessExplainable AI-Driven Intrusion Detection and Blockchain-Based Trust Evolution for Secure Routing in Underwater Sensor Networks
- Open Access
- 8 December 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (6) , 30520-30528
Energy Harvesting from Vehicle-Induced Wind for Enhanced Communication Efficiency in Autonomous Electric Vehicles
The increasing demand for sustainable transportation solutions presents a challenge in minimizing Electric Vehicles’ (EVs) reliance on external power sources while ensuring effective communication and control. This study addresses the issue by integrating wind energy harvesting and optimizing internal communication systems to support autonomous EV operation. The proposed framework utilizes vehicle-mobility-induced wind energy, supplemented by solar Photovoltaic (PV) cells, to power essential vehicle functions, thus reducing dependence on traditional charging methods and infrastructure. Wind turbine sensors attached to a 3D-printed dynamo assembly convert ambient kinetic energy into electrical energy, which is subsequently stored in a battery pack. This power is intelligently managed and distributed through a system coordinated by an ESP32 controller and Arduino UNO, enabling seamless communication across subsystems. Autonomous control is executed via a joystick interface that transmits steering, acceleration, and braking commands to actuators, with real-time adjustments enabled by motor drivers. Automated transceivers facilitate continuous data exchange among energy-harvesting units, controllers, and actuators, ensuring synchronization and effective power management. The integration of Free Real-Time Operating System (FreeRTOS) enables real-time scheduling and dynamic energy management, supporting responsive operation under changing environmental conditions. By integrating IoT communication with renewable energy sources, this approach provides an energy-efficient, adaptable solution for autonomous EVs, supporting the goal of self-sustaining, eco-friendly transportation systems.
- 8 December 2025
- Published by Engineering, Technology & Applied Science Research in Engineering, Technology & Applied Science Research
- Vol. 15 (6) , 30520-30528
Open AccessEnergy Harvesting from Vehicle-Induced Wind for Enhanced Communication Efficiency in Autonomous Electric Vehicles
The increasing demand for sustainable transportation solutions presents a challenge in minimizing Electric Vehicles’ (EVs) reliance on external power sources while ensuring effective communication and control. This study addresses the issue by integrating wind energy harvesting and optimizing internal communication systems to support autonomous EV operation. The proposed framework utilizes vehicle-mobility-induced wind energy, supplemented by solar Photovoltaic (PV) cells, to power essential vehicle functions, thus reducing dependence on traditional charging methods and infrastructure. Wind turbine sensors attached to a 3D-printed dynamo assembly convert ambient kinetic energy into electrical energy, which is subsequently stored in a battery pack. This power is intelligently managed and distributed through a system coordinated by an ESP32 controller and Arduino UNO, enabling seamless communication across subsystems. Autonomous control is executed via a joystick interface that transmits steering, acceleration, and braking commands to actuators, with real-time adjustments enabled by motor drivers. Automated transceivers facilitate continuous data exchange among energy-harvesting units, controllers, and actuators, ensuring synchronization and effective power management. The integration of Free Real-Time Operating System (FreeRTOS) enables real-time scheduling and dynamic energy management, supporting responsive operation under changing environmental conditions. By integrating IoT communication with renewable energy sources, this approach provides an energy-efficient, adaptable solution for autonomous EVs, supporting the goal of self-sustaining, eco-friendly transportation systems.
- Open Access
- 19 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 37352-37369
Maximizing Wireless Sensor Networks Coverage and Connectivity in Irregular Terrains Using Grover’s Quantum Algorithm
The complicated topographical limitations and NP-hard nature of the sensor placement problem make it difficult to achieve optimal coverage and connectivity in Wireless Sensor Networks (WSNs) deployed over irregular terrains. WSNs encounter critical challenges in uneven regions due to irregular obstacles, elevation variations, and dynamic terrain profiles, all of which influence coverage and connectivity. Classical methods, such as greedy or evolutionary strategies, often struggle to scale efficiently as terrain complexity increases and may converge to suboptimal solutions due to their heuristic nature. This paper applies Grover’s quantum search algorithm to enhance sensor placement, the primary focus of this study. Unlike classical techniques that traverse solution spaces step by step, the Grover algorithm enhances the search process with a quadratic speedup. This enhancement enables the efficient identification of near-optimal sensor configurations in complicated contexts. This quantum-assisted method not only reduces the number of necessary sensors but also enhances performance in coverage and connectivity metrics, along with a substantial decrease in computation burden. The integration of terrain-awareness and quantum search diverges significantly from traditional approaches, providing an intelligent, scalable, and computationally promising methodology for WSN deployment in the real-world terrains.
- 19 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 37352-37369
Open AccessMaximizing Wireless Sensor Networks Coverage and Connectivity in Irregular Terrains Using Grover’s Quantum Algorithm
The complicated topographical limitations and NP-hard nature of the sensor placement problem make it difficult to achieve optimal coverage and connectivity in Wireless Sensor Networks (WSNs) deployed over irregular terrains. WSNs encounter critical challenges in uneven regions due to irregular obstacles, elevation variations, and dynamic terrain profiles, all of which influence coverage and connectivity. Classical methods, such as greedy or evolutionary strategies, often struggle to scale efficiently as terrain complexity increases and may converge to suboptimal solutions due to their heuristic nature. This paper applies Grover’s quantum search algorithm to enhance sensor placement, the primary focus of this study. Unlike classical techniques that traverse solution spaces step by step, the Grover algorithm enhances the search process with a quadratic speedup. This enhancement enables the efficient identification of near-optimal sensor configurations in complicated contexts. This quantum-assisted method not only reduces the number of necessary sensors but also enhances performance in coverage and connectivity metrics, along with a substantial decrease in computation burden. The integration of terrain-awareness and quantum search diverges significantly from traditional approaches, providing an intelligent, scalable, and computationally promising methodology for WSN deployment in the real-world terrains.
- 7 May 2025
- Published by IGI Global Scientific Publishing
- p. 127-144
A Case Study on Cyber Attack Detection Using Machine Learning
This chapter focusses on leveraging machine learning (ML) for cyber-attack prevention, addressing a range of threats and assaults. Machine learning is a crucial component of modern cybersecurity, offering a flexible approach to defend information systems against the constantly evolving tactics of malicious actors. By training both supervised and unsupervised ML algorithms on diverse datasets, we tackle issues such as hostile assaults and class imbalance. A key aspect of our work is prioritizing the interpretability of ML models to effectively manage and reduce false positives and false negatives. Additionally, we explore the challenges of integrating ML findings with existing cybersecurity frameworks, aiming for seamless collaboration between traditional security measures and ML-driven solutions. Our goal is to provide valuable insights on utilizing ML to prevent cyberattacks, highlighting its benefits, limitations, and future potential. Ultimately, we aim to enhance cybersecurity defenses in dynamic threat landscapes by clarifying the role of ML in cybersecurity.
- 7 May 2025
- Published by IGI Global Scientific Publishing
- p. 127-144
A Case Study on Cyber Attack Detection Using Machine Learning
This chapter focusses on leveraging machine learning (ML) for cyber-attack prevention, addressing a range of threats and assaults. Machine learning is a crucial component of modern cybersecurity, offering a flexible approach to defend information systems against the constantly evolving tactics of malicious actors. By training both supervised and unsupervised ML algorithms on diverse datasets, we tackle issues such as hostile assaults and class imbalance. A key aspect of our work is prioritizing the interpretability of ML models to effectively manage and reduce false positives and false negatives. Additionally, we explore the challenges of integrating ML findings with existing cybersecurity frameworks, aiming for seamless collaboration between traditional security measures and ML-driven solutions. Our goal is to provide valuable insights on utilizing ML to prevent cyberattacks, highlighting its benefits, limitations, and future potential. Ultimately, we aim to enhance cybersecurity defenses in dynamic threat landscapes by clarifying the role of ML in cybersecurity.
- 9 April 2025
- Published by IGI Global Scientific Publishing
- p. 303-324
Improving UAV Fault Detection Through a Hybrid CNN-LSTM-GRU Framework for Sensor Data Analysis
Unmanned Aerial Vehicles (UAVs) are essential in numerous applications, such as surveillance and delivery services, where operational reliability and safety are paramount.This research aims to enhance fault detection techniques by utilizing deep learning models capable of interpreting complex sensor data and identifying potential issues. A hybrid deep learning model, combining Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Gated Recurrent Units (GRU), was developed to analyze sequential sensor data. The hybrid model leverages CNN layers to extract spatial features, while LSTM and GRU layers capture temporal dependencies. Experimental results demonstrate that the hybrid CNN-LSTM-GRU model outperforms the individual models, achieving a test accuracy of 85.66% and a validation accuracy of 88.34%. These findings underscore the hybrid model's superior capability in fault detection for UAVs, highlighting its potential to improve the reliability and safety of UAV operations through advanced deep learning frameworks.
- 9 April 2025
- Published by IGI Global Scientific Publishing
- p. 303-324
Improving UAV Fault Detection Through a Hybrid CNN-LSTM-GRU Framework for Sensor Data Analysis
Unmanned Aerial Vehicles (UAVs) are essential in numerous applications, such as surveillance and delivery services, where operational reliability and safety are paramount.This research aims to enhance fault detection techniques by utilizing deep learning models capable of interpreting complex sensor data and identifying potential issues. A hybrid deep learning model, combining Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Gated Recurrent Units (GRU), was developed to analyze sequential sensor data. The hybrid model leverages CNN layers to extract spatial features, while LSTM and GRU layers capture temporal dependencies. Experimental results demonstrate that the hybrid CNN-LSTM-GRU model outperforms the individual models, achieving a test accuracy of 85.66% and a validation accuracy of 88.34%. These findings underscore the hybrid model's superior capability in fault detection for UAVs, highlighting its potential to improve the reliability and safety of UAV operations through advanced deep learning frameworks.
- 24 April 2024
- Published by Springer Nature
- p. 365-377
Cybersecurity Threat Detection in SDN Clouds Using Attention Mechanism in RNN
- 24 April 2024
- Published by Springer Nature
- p. 365-377
Cybersecurity Threat Detection in SDN Clouds Using Attention Mechanism in RNN
- Open Access
- 22 November 2022
- Published by MDPI AG in Sustainability
- Vol. 14 (23) , 15547
Design and Validation of Lifetime Extension Low Latency MAC Protocol (LELLMAC) for Wireless Sensor Networks Using a Hybrid Algorithm
As the battery-operated power source of wireless sensor networks, energy consumption is a major concern. The medium-access protocol design solves the energy usage of sensor nodes while transmitting and receiving data, thereby improving the sensor network’s lifetime. The suggested work employs a hybrid algorithm to improve the energy efficiency of sensor networks with nodes that are regularly placed. Every node in this protocol has three operating modes, which are sleep mode, receive mode, and send mode. Every node enters a periodic sleep state in order to conserve energy, and after waking up, it waits for communication. During the sleep mode, the nodes turn off their radios in order to reduce the amount of energy they consume while not in use. As an added feature, this article offers a channel access mechanism in which the sensors can send data based on the Logical Link Decision (LLD) algorithm and receive data based on the adaptive reception method. It is meant to select acceptable intermediary nodes in order to identify the path from the source to the destination and to minimize data transmission delays among the nodes in the network scenario. Aside from that, both simulation and analytical findings are used to examine the activity of the suggested MAC, and the created models are evaluated depending on their performance. With regard to energy consumption, latency, throughput, and power efficiency, the result demonstrates that the suggested MAC protocol outperforms the corresponding set of rules. The extensive simulation and analytical analysis showed that the energy consumption of the proposed LELLMAC protocol is reduced by 22% and 76.9% the end-to-end latency is 84.7% and 87.4% percent shorter, and the throughput is 60.3% and 70.5% higher than the existing techniques when the number of node is 10 and 100 respectively.
- 22 November 2022
- Published by MDPI AG in Sustainability
- Vol. 14 (23) , 15547
Open AccessDesign and Validation of Lifetime Extension Low Latency MAC Protocol (LELLMAC) for Wireless Sensor Networks Using a Hybrid Algorithm
As the battery-operated power source of wireless sensor networks, energy consumption is a major concern. The medium-access protocol design solves the energy usage of sensor nodes while transmitting and receiving data, thereby improving the sensor network’s lifetime. The suggested work employs a hybrid algorithm to improve the energy efficiency of sensor networks with nodes that are regularly placed. Every node in this protocol has three operating modes, which are sleep mode, receive mode, and send mode. Every node enters a periodic sleep state in order to conserve energy, and after waking up, it waits for communication. During the sleep mode, the nodes turn off their radios in order to reduce the amount of energy they consume while not in use. As an added feature, this article offers a channel access mechanism in which the sensors can send data based on the Logical Link Decision (LLD) algorithm and receive data based on the adaptive reception method. It is meant to select acceptable intermediary nodes in order to identify the path from the source to the destination and to minimize data transmission delays among the nodes in the network scenario. Aside from that, both simulation and analytical findings are used to examine the activity of the suggested MAC, and the created models are evaluated depending on their performance. With regard to energy consumption, latency, throughput, and power efficiency, the result demonstrates that the suggested MAC protocol outperforms the corresponding set of rules. The extensive simulation and analytical analysis showed that the energy consumption of the proposed LELLMAC protocol is reduced by 22% and 76.9% the end-to-end latency is 84.7% and 87.4% percent shorter, and the throughput is 60.3% and 70.5% higher than the existing techniques when the number of node is 10 and 100 respectively.