Purushothaman Ramaiah
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
Purushothaman Ramaiah has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Kathiresan Kokulavani
J.J.College of Engineering and Technology
2 shared publications - Nagaiyanallur Lakshminarayanan Venkatara
J.J.College of Engineering and Technology
2 shared publications - Subramani Suresh Kumar
J.J.College of Engineering and Technology
2 shared publications - Velankanni Gowri
J.J.College of Engineering and Technology
2 shared publications - Adlin Sheeba
Professor, Department of Computer Science and Engineering, St. Joseph's Institute of Technology, Chennai, Tamil Nadu, India
1 shared publication - Arun Anthonisamy
Department of Computer Science and Business System, Panimalar Engineering College, Chennai 600123, TN, India
1 shared publication
Latest Publications
- Open Access
- 25 March 2025
- Published by EDP Sciences in ITM Web of Conferences
- Vol. 76, 01004
Deep Learning Models for Image Classification Advances in Convolutional Neural Network Architectures
Deep learning has improved image classification tasks dramatically, where Convolutional Neural Networks (CNNs) have prevailed as the most successful architecture. But there are challenges posed by current CNN models, either due to computational expense, limited explainability, or poor generalization across domains. To overcome this, the research proposes an optimized CNN architecture that improves efficiency and scalability while enabling knowledge transfer and self-supervised learning for smaller datasets, resulting in improved accuracy. Moreover, it uses some hybrid CNN-Transformer techniques to make use of the advantages of two models and guarantee sufficient feature extraction and enhanced generalization. We will incorporate explainable AI (XAI) methods like Grad-CAM and SHAP to solve interpretability challenges, ensuring that the model is appropriate for high-stakes domains like autonomous systems and medical imaging. In addition, the outlined framework is tailored to practical applications, as it optimizes the CNN for buildings at the edge and in real time, in the speed-accuracy exchange. In designing the next generation of CNN architectures that can build upon computational efficiency, and generalization, making CNN more useful for real world applications in various fields such as health, surveillance and autonomous systems.
- 25 March 2025
- Published by EDP Sciences in ITM Web of Conferences
- Vol. 76, 01004
Open AccessDeep Learning Models for Image Classification Advances in Convolutional Neural Network Architectures
Deep learning has improved image classification tasks dramatically, where Convolutional Neural Networks (CNNs) have prevailed as the most successful architecture. But there are challenges posed by current CNN models, either due to computational expense, limited explainability, or poor generalization across domains. To overcome this, the research proposes an optimized CNN architecture that improves efficiency and scalability while enabling knowledge transfer and self-supervised learning for smaller datasets, resulting in improved accuracy. Moreover, it uses some hybrid CNN-Transformer techniques to make use of the advantages of two models and guarantee sufficient feature extraction and enhanced generalization. We will incorporate explainable AI (XAI) methods like Grad-CAM and SHAP to solve interpretability challenges, ensuring that the model is appropriate for high-stakes domains like autonomous systems and medical imaging. In addition, the outlined framework is tailored to practical applications, as it optimizes the CNN for buildings at the edge and in real time, in the speed-accuracy exchange. In designing the next generation of CNN architectures that can build upon computational efficiency, and generalization, making CNN more useful for real world applications in various fields such as health, surveillance and autonomous systems.
- Open Access
- 25 November 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 82 (1) , 113
Investigation of Incorporation of Internet of Things with Wireless Sensor Networks Based on Path Vector Hop Count and Limited Bandwidth Channel IoT Mechanism
A wireless sensor network (WSN) consists of sensors with wireless transceivers that link autonomously over many hops. It offers various advantages, including less traffic, more stability, extended wireless communication distances, and broader coverage regions at lower cost. Combining emerging Limited Bandwidth Channel Internet of Things (LBC-IoT) technologies with wireless sensor networks offers interesting applications in the defense, medical, smart conveyance, and marketable sectors. This study initially analyzes WSN and LBC-IoT technologies independently before combining them to look into the networking framework of LBC-IoT and WSN, as well as the associated technologies resulting from the fusion. This article describes the typical network node redeployment strategy for wireless sensors, which can lead to poor node connection and inadequate coverage due to a lack of confined subgroup node exploration. The suggested method for localizing WSN nodes, based on the Hop Count Path Vector (HOP-PV) algorithm, enhances the process of calculating the average hop distance and the number of node hops, resulting in the PVHOP-LBCIOT mechanism. Simulation results indicate that the improved PVHOP-LBCIOT algorithm’s three deployment methods (square, central uniform, and cross) outperform the two approaches of HOP-PV (random deployment) and PVHOP-LBCIOT (border uniform deployment) for an equal number of unknown moving anchor positions (11), a disparate number of unspecified nodes (30-13), and a fixed communication radius (6), with a reduced average error rate of 32.79%, from 38%, and improved accuracy for obtaining unknown node locations. The suggested method for localizing WSN nodes using a single node acting as a mobile anchor point, known as the PVHOP-LBCIOT mechanism, enhances and optimizes the process of computing the average hop distance and the number of node hops. A comparison experiment demonstrates that this hopping algorithm has much greater coverage, node power, linkage, and resilience compared to existing methods.
- 25 November 2024
- Published by MDPI AG in Engineering Proceedings
- Vol. 82 (1) , 113
Open AccessInvestigation of Incorporation of Internet of Things with Wireless Sensor Networks Based on Path Vector Hop Count and Limited Bandwidth Channel IoT Mechanism
A wireless sensor network (WSN) consists of sensors with wireless transceivers that link autonomously over many hops. It offers various advantages, including less traffic, more stability, extended wireless communication distances, and broader coverage regions at lower cost. Combining emerging Limited Bandwidth Channel Internet of Things (LBC-IoT) technologies with wireless sensor networks offers interesting applications in the defense, medical, smart conveyance, and marketable sectors. This study initially analyzes WSN and LBC-IoT technologies independently before combining them to look into the networking framework of LBC-IoT and WSN, as well as the associated technologies resulting from the fusion. This article describes the typical network node redeployment strategy for wireless sensors, which can lead to poor node connection and inadequate coverage due to a lack of confined subgroup node exploration. The suggested method for localizing WSN nodes, based on the Hop Count Path Vector (HOP-PV) algorithm, enhances the process of calculating the average hop distance and the number of node hops, resulting in the PVHOP-LBCIOT mechanism. Simulation results indicate that the improved PVHOP-LBCIOT algorithm’s three deployment methods (square, central uniform, and cross) outperform the two approaches of HOP-PV (random deployment) and PVHOP-LBCIOT (border uniform deployment) for an equal number of unknown moving anchor positions (11), a disparate number of unspecified nodes (30-13), and a fixed communication radius (6), with a reduced average error rate of 32.79%, from 38%, and improved accuracy for obtaining unknown node locations. The suggested method for localizing WSN nodes using a single node acting as a mobile anchor point, known as the PVHOP-LBCIOT mechanism, enhances and optimizes the process of computing the average hop distance and the number of node hops. A comparison experiment demonstrates that this hopping algorithm has much greater coverage, node power, linkage, and resilience compared to existing methods.
- Open Access
- 1 January 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (1) , 179-189
Design analysis of moth-flame optimized fault tolerant technique for minimally buffered network-on-chip router
A network on a chip is a solitary silicon chip utilized to perform the communication characteristics of large-scale (LSI) to very large-scale integration (VLSI) systems. Network-on-chip (NoC) architecture includes links, network interfaces (NI), and routers to unite with external memories or processors. NoC is designed to flow messages from the source module to the destination module through several links involving routing decisions. The design of NoC is complex and the buffer section’s expensiveness creates problems while providing secured data service. Moreover, routers and links in NoC setups are liable to faults. This work introduces a minimal buffered router, and the faults in the network are optimized using moth flame optimized (MFO) fault-tolerant technique. The software named Xilinx ISE design suite 14.5 is employed for the minimum buffered router model. The suggested scheme is operated with less area, low power (0.241 mW), and high speed (965.261 Megahertz (MHz)) when matched with previous works.
- 1 January 2024
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 33 (1) , 179-189
Open AccessDesign analysis of moth-flame optimized fault tolerant technique for minimally buffered network-on-chip router
A network on a chip is a solitary silicon chip utilized to perform the communication characteristics of large-scale (LSI) to very large-scale integration (VLSI) systems. Network-on-chip (NoC) architecture includes links, network interfaces (NI), and routers to unite with external memories or processors. NoC is designed to flow messages from the source module to the destination module through several links involving routing decisions. The design of NoC is complex and the buffer section’s expensiveness creates problems while providing secured data service. Moreover, routers and links in NoC setups are liable to faults. This work introduces a minimal buffered router, and the faults in the network are optimized using moth flame optimized (MFO) fault-tolerant technique. The software named Xilinx ISE design suite 14.5 is employed for the minimum buffered router model. The suggested scheme is operated with less area, low power (0.241 mW), and high speed (965.261 Megahertz (MHz)) when matched with previous works.
- Open Access
- 1 December 2023
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 32 (3) , 1337-1345
Design of matrix, distributive round robin, ping pong and enhanced ping lock arbiter for shared resources systems
Arbiter is one of the main core elements in the network scheduler. The significant goal of this work is to design a high-speed and low execution-time arbiter with lock free and fair arbitration scheme. In this work, four types of arbiters such as matrix arbiter (MA), ping pong arbiter (PPA), distributive round-robin arbiter (DRRA) and enhanced ping lock arbiter (EPLA) are designed and analyzed area, delay, and speed of arbiters. MA is worked in square matrix format and matrix transition is performed for effective routing. The DRRA is designed by using a multiplexer and counter. Hence an, effective scheduling is carried out in DRRA. Binary tree format is used in PPA. The PPA provides low chip size and high speed than existing MA and DRRA. The PPA limits fair arbitration during uniformly distributed active request patterns. To overcome this problem, PPA is improved with some lock systems to create an EPLA. A new ping lock arbiter (PLA) leaf and PLA inter structure is proposed at the gate level to reduce the execution delay, improve the speed and achieve fair arbitration over all other existing arbiters. Resource allocation, execution delay, and speed are analyzed using the Xilinx Integrated Software Environment (ISE) tool.
- 1 December 2023
- Published by Institute of Advanced Engineering and Science in Indonesian Journal of Electrical Engineering and Computer Science
- Vol. 32 (3) , 1337-1345
Open AccessDesign of matrix, distributive round robin, ping pong and enhanced ping lock arbiter for shared resources systems
Arbiter is one of the main core elements in the network scheduler. The significant goal of this work is to design a high-speed and low execution-time arbiter with lock free and fair arbitration scheme. In this work, four types of arbiters such as matrix arbiter (MA), ping pong arbiter (PPA), distributive round-robin arbiter (DRRA) and enhanced ping lock arbiter (EPLA) are designed and analyzed area, delay, and speed of arbiters. MA is worked in square matrix format and matrix transition is performed for effective routing. The DRRA is designed by using a multiplexer and counter. Hence an, effective scheduling is carried out in DRRA. Binary tree format is used in PPA. The PPA provides low chip size and high speed than existing MA and DRRA. The PPA limits fair arbitration during uniformly distributed active request patterns. To overcome this problem, PPA is improved with some lock systems to create an EPLA. A new ping lock arbiter (PLA) leaf and PLA inter structure is proposed at the gate level to reduce the execution delay, improve the speed and achieve fair arbitration over all other existing arbiters. Resource allocation, execution delay, and speed are analyzed using the Xilinx Integrated Software Environment (ISE) tool.
- Open Access
- 17 May 2023
- Published by MDPI AG in Engineering Proceedings
- Vol. 37 (1) , 81
Energy-Efficient Control Methods in Heterogeneous Wireless Sensor Networks: A Survey
Due to the advancement of sensor gadgets and telecommunication technology, wireless sensor networks (WSNs) have drawn a lot of observations in the recent period. Inaccessible terrain, disaster zones, or polluted conditions are typically where they are deployed at random, making battery replacement or recharge challenging or even impossible. Network lifespan is therefore extremely important to a WSN. An abundance of power-effective strategies in a diverse wireless sensor network are surveyed in this paper. We first provide an overview of the fundamental network radio representation and how it may be utilized to analyze different trade-offs between network deployment costs and an energy-efficient clustering approach. We also highlight a few protocols that can be utilized in heterogeneous networks that are energy efficient.
- 17 May 2023
- Published by MDPI AG in Engineering Proceedings
- Vol. 37 (1) , 81
Open AccessEnergy-Efficient Control Methods in Heterogeneous Wireless Sensor Networks: A Survey
Due to the advancement of sensor gadgets and telecommunication technology, wireless sensor networks (WSNs) have drawn a lot of observations in the recent period. Inaccessible terrain, disaster zones, or polluted conditions are typically where they are deployed at random, making battery replacement or recharge challenging or even impossible. Network lifespan is therefore extremely important to a WSN. An abundance of power-effective strategies in a diverse wireless sensor network are surveyed in this paper. We first provide an overview of the fundamental network radio representation and how it may be utilized to analyze different trade-offs between network deployment costs and an energy-efficient clustering approach. We also highlight a few protocols that can be utilized in heterogeneous networks that are energy efficient.