AARAV KANNAN JAYAKUMAR

Mr. AARAV KANNAN JAYAKUMAR

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 May 2025 in Sensors
Verified
Journal Article
Open access
PDF file
NMNalini Manogaran
Nalini Manogaran
MMMercy Theresa M
Mercy Theresa M
RRRajalakshmi Raja
Rajalakshmi Raja
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
NMNandagopal Malarvizhi
Nandagopal Malarvizhi
Gheorghita GhineaGheorghita Ghinea
Gheorghita Ghinea
The working of the Internet of Things (IoT) ecosystem indeed depends extensively on the mechanisms of real-time data collection, sharing, and automatic operation. Among these fundamentals, wireless sensor networks (WSNs) are important for maintaining a countenance with their many distributed Sensor Nodes (SNs), which can sense and transmit environmental data wirelessly. Because WSNs possess advantages for remote data collection, they are severely hampered by constraints imposed by the limited energy capacity of SNs; hence, energy-efficient routing is a pertinent challenge. Therefore, in the case of clustering and routing mechanisms, these two play important roles where clustering is performed to reduce energy consumption and prolong the lifetime of the network, while routing refers to the actual paths for transmission of data. Addressing the limitations witnessed in the conventional IoT-based routing of data, this proposal presents an FL-oriented framework that presents a new energy-efficient routing scheme. Such routing is facilitated by the ADDQL model, which creates smart high-speed routing across changing scenarios in WSNs. The proposed ADDQL-IRHO model has been compared to other existing state-of-the-art algorithms according to multiple performance metrics such as energy consumption, communication delay, temporal complexity, data sum rate, message overhead, and scalability, with extensive experimental evaluation reporting superior performance. This also substantiates the applicability and competitiveness of the framework in variable-serviced IoT-oriented WSNs for next-gen intelligent routing solutions.

21 May 2025
Verified
Conference Paper
Closed access
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
RRR. Rajalakshmi
R. Rajalakshmi
GRGarugu Rajasekharachari
Garugu Rajasekharachari
GVGarapati Vikas
Garapati Vikas
RAR. S. Amshavalli
R. S. Amshavalli
Heart disease remains a leading cause of mortality worldwide, necessitating rapid and accurate diagnostic systems to enhance patient outcomes. Traditional electrocardiogram (ECG) analysis methods are prone to errors, particularly when dealing with complex patterns or large datasets. This paper proposes an advanced deep learning approach using Convolutional Neural Networks (CNNs) for multi-class ECG image classification, aimed at improving diagnostic accuracy and efficiency. The methodology incorporates techniques such as transfer learning, data augmentation, and Generative Adversarial Networks (GANs) to enhance model robustness and performance. A user-friendly interface is developed for real-time diagnostic application, promoting clinical integration. Comparative analysis reveals that the proposed model achieves an overall accuracy of 94.7%, significantly outperforming traditional methods such as Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), and Decision Tree. These findings demonstrate the potential of deep learning-based systems to revolutionize cardiac diagnosis, providing scalable, accurate, and efficient diagnostic solutions applicable to diverse clinical environments.

18 February 2025
Verified
Conference Paper
Closed access
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
RAR.S. Amshavalli
R.S. Amshavalli
VDVenkateswara Reddy D
Venkateswara Reddy D
VVVegesna Surya Vinay Varma
Vegesna Surya Vinay Varma
GKG. Kavitha
G. Kavitha
The rapid expansion of cloud computing has led to increased energy consumption, raising concerns about its environmental impact. Despite advancements in scheduling algorithms, existing methods face critical challenges, including suboptimal energy efficiency, limited adaptability to dynamic cloud environments, and insufficient context-aware decision-making. Recent techniques, such as Meta Reinforcement Learning, Genetic Algorithms, and Energy-Aware Scheduling, show potential but often struggle with high computational complexity, static resource allocation, or scalability issues. These shortcomings result in inefficient resource utilization, elevated energy usage, and difficulties in meeting Service Level Agreement (SLA) requirements. To address these challenges, this study introduces the Energy Efficient Semantic Scheduling Framework (ESSF), a novel approach combining semantic reasoning and energy profiling to optimize task scheduling in dynamic cloud environments. The framework integrates three key components: (1) Dynamic Energy Profiling (DEP) for real-time energy monitoring, (2) Semantic Resource Matching (SRM) utilizing knowledge graphs for context-aware allocation, and (3) Adaptive Workload Balancing (AWB) to dynamically redistribute tasks for improved resource utilization and energy efficiency. Experiments using the Google Cluster Trace Dataset demonstrate that ESSF reduces energy consumption by up to 25% compared to traditional methods while achieving 90% resource utilization and 97% SLA compliance. These results highlight the potential of the proposed framework to overcome existing limitations, offering a scalable, adaptive, and environmentally sustainable solution for task scheduling in cloud computing.

7 April 2025
Verified
Conference Paper
Closed access
RAR. S. Amshavalli
R. S. Amshavalli
ASArun Shankar
Arun Shankar
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
JKJ. Kalaivani
J. Kalaivani
Efficient public transportation systems are critical to sustainable urban development, yet they often face challenges such as delayed operations, inadequate fleet management, and limited real-time visibility for passengers. This research addresses these issues by proposing a real-time public transport information system utilizing GPRS technology, which combines GPS and GSM modules to improve the operational efficiency of public transport systems. The system provides real-time vehicle location tracking, SMS notifications when a vehicle starts, and allows passengers to request vehicle location updates or remotely control the vehicle via SMS commands. By integrating affordable components such as Arduino Uno, SIM800L GSM modules, and NEO-6M GPS modules, this system is cost-effective, scalable, and operates without the need for continuous internet connectivity. This feature makes it particularly suitable for areas with limited network infrastructure. The outcome of this research demonstrates that the proposed system offers a feasible, reliable, and affordable solution to enhance passenger convenience, improve operational control, and facilitate scalable fleet management, making public transport more efficient and safer.

2 May 2025
Verified
Conference Paper
Closed access
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
RRR. Rajalakshmi
R. Rajalakshmi
MGM Gowtham
M Gowtham
RGR Geetheshwar
R Geetheshwar
SSS R Srividhya
S R Srividhya

1 August 2025
Verified
Conference Paper
Closed access
AARAV KANNAN JAYAKUMARAARAV KANNAN JAYAKUMAR
AARAV KANNAN JAYAKUMAR
RRR. Rajalakshmi
R. Rajalakshmi
The primary key to the modern innovation found in Enhanced Neurocomputing architectures is the harmonization of digital disparate data modalities into a single uni-analytical pipeline. The combination of renocentric disease diagnosis with cognitive neurocomputing architectures helps with accurate diagnosis and effective timely treatment. This proposed research represents a novel approach for enhancing diagnostic accuracy for the patients suffering from kidney-related issues of renocentric complications using AI-driven image processing methodologies. By leveraging the use of deep learning (DL), we introduce a novel model system that combines convolutional neural networks (CNNs) for interpreting clinical diagnostic data, that are specifically focussed on renal scans and similar related diagnostic images. Our novel approach utilizes an authentic dataset of comprehensive kidney images to train a deep learning (DL) model, for enabling the detection of sequential pathological features that indicates the diabetes induced renal damage. The model's architecture includes advanced techniques like the transfer learning and the data augmentation for improving its generalization robustness and capabilities. By utilizing AI-enhanced image processing methods, our model system aims to provide more accurate and early detection of renal complications, thus facilitating better patient administration and management.