Hybrid Flying Foxes and Enhanced Whale Algorithm-Based Cluster Optimization Method for Efficient Stable Routing in Vehicular Ad Hoc Networks (VANETs) N. Gopinath, A. Chinnasamy, T. Sathies Kumar International Journal of Communication Systems, 2025 Vehicular ad hoc network (VANET) is an indispensable entity to diversified number of intelligent transportation system (ITS)–enabled technologies. But network scalability, frequent topology changes, and high mobility are the major problems due to the sparse distribution of vehicles especially in highways and constantly changing vehicular network topology. Maintenance of stable route in the network between the vehicles is a herculean task as its failure increases the probability of instability. This establishment of stable routes is essential in VANETs for efficiently utilizing the computational resources such that desirable degree of quality of service (QoS) can be achieved. This stable route determination can be attained by addressing the factors of energy balancing, coverage, connectivity, and load balancing for the purpose of guaranteeing the sensed data from all the points of target to the base stations in a reliable manner. In this paper, hybrid flying foxes and enhanced whale algorithm (HFFEWA)–based cluster optimization method is proposed for attaining sustained routing that establishes stable cluster construction during the routing process. This HFFEWA adopted the factors of route along the highway, velocity, number of nodes, and communication range into the fitness function for minimizing the degree of randomness. It specifically used flying fox optimization algorithm (FFOA) for exploring the search space more eminently such that global clusters could be constructed with maximized diversity. On the other hand, enhanced whale algorithm (EWA) is adopted for preventing the issue of premature convergence. It is also proposed with the capability of well‐balanced exploration and exploitation that explores and exploits the search space such that it can be used in generating optimal number of cluster heads (CHs). The simulation results of this HFFEWA conducted different vehicular density confirmed an improved network lifetime of 19.42% with the stabilized cluster construction of 32.18%, better than the competitive approaches. The evaluation of HFFEWA under different network size confirmed better performance in packet delivery rate, end‐to‐end delay, and packet loss.
Metaverse security monitoring based on virtual environment analysis using machine learning techniques Sheryl Oliver, A. Chinnasamy, P. Varun, N. Manikandan, S. Magesh, R. Manikandan Navigating AI and the Metaverse in Scientific Research, 2025 A NextG Internet platform allows users to participate in many kinds of virtual events and communicate with avatars in a 3D virtual environment to perform many activities in this environment. Building an intrusion detection system is computationally challenging in the Metaverse because of its interactive nature as well as large number of user interactions that take place within virtual settings. This research proposes novel techniques in a Metaverse-based virtual environment in security monitoring using machine learning techniques. Here, security monitoring was carried out using reinforcement-federated regressive Gaussian neural networks. The metaverse virtual environment has been deployed, and its analysis is carried out using a cloud edge network with virtual software-defined infrastructure. Experimental analysis is carried out in terms of scalability, quality of service, latency, accuracy, and network integrity. The proposed model attained a scalability of 94%, a quality of service of 95%, an accuracy of 97%, a latency of 96%, and a network integrity of 93%.
Intelligent decision-making framework for big data using enhanced honey badger-based adaptive hybrid deep learning network D. Kavitha, A. Chinnasamy, P. Selvakumari International Journal of Data Mining and Bioinformatics, 2025 By utilising the conventional models, it is also consuming more time for processing. Hence, there is a crucial requirement for real-world application over big data procedures to perform a scalable and effective solution. For the experimentation, input data is gathered from different application-oriented datasets. Initially, the input data is congregated and undergoes for data cleaning stage and then the cleaned data is given as input for optimal feature extraction, in which the enhanced map-reduce model is applied for extracting the optimal features. These obtained optimal features are fed into adaptive cascaded long short-term memory and auto-encoder-based long short-term memory (ACLALSTM), in which the parameters are optimised by using enhanced HBA for effective decision-making in proposed big data analysis. The experimental analysis shows, that the proposed big data-based decision-making model shows the tendency to provide rapid decisions that help to analyse the big data effectively.
Optimized Video Streaming in Mobile Environments: A Novel Multihead Autoencoder and Red-Tailed Hawk Optimization Framework Duraimurugan Samiayya, Chinnasamy Ambayiram, Hariprasad Natarajan, Rajakumar Muthusamy Palani IETE Journal of Research, 2025 In recent years, adaptive video streaming over wireless multimedia sensor networks has gained significant attention due to the growing demand for real-time video delivery in mobile environments. However, traditional video transmission methods tend to face challenges in maintaining high video quality while adjusting to changing network conditions. To address this limitation, we propose a novel Interaction Crossover Red-tailed Hawk Optimized Multihead Autoencoder Matrix-factorization model for super-resolution transmission in adaptive video streaming over wireless multimedia sensor networks. Multihead autoencoder matrix-factorization architecture and an optimization method inspired by red-tailed hawks are two advanced optimization algorithms that are combined to create the proposed model. This combined technique not only enhances video resolution but also modifies transmission parameters dynamically, effectively optimizing video quality, bandwidth usage, and latency, regardless of challenging network conditions. The selection of the proposed model is based on its effective approach to managing video quality and network resources in real-time. This model adapts seamlessly to various network conditions, including bandwidth shifts due to user movement, through an interactive crossover strategy. This innovative approach overcomes the static quality limitations of previous systems, leading to substantial improvements in playback stability, throughput, and clarity of the video content. The experimental results demonstrated that the proposed model surpassed the existing models, achieving impressive metrics: a Peak Signal-to-Noise Ratio of 42.5 dB, a throughput of 900 kbps, and a playback stability of 98%. Overall, the proposed framework exhibits significant potential for adaptive video streaming systems in dynamic and mobile environments.
Wearable Medical Device Security and Network Protection for Reliable Healthcare Wearable Medical Device Security and Network Protection for Reliable Healthcare, 2025
Geotaggingofplantationinthecatchment Area of Hydro Project S Mathupriya, A Chinnasamy, B Benisha, K Kaviya, U Kiruthika 2025 International Conference on Computing and Communication Technologies Iccct 2025, 2025 Geo-tagging of plantations within the catchment area of a hydro project is an essential process for effective environmental management and monitoring. This technique involves capturing the precise geographical coordinates of planted vegetation, coupled with detailed metadata, to support the assessment and management of reforestation or afforestation efforts. The process includes field surveys using GPS technology, data integration into Geographic Information Systems (GIS), and regular updates to track growth and ecological impact. By providing accurate spatial data, geo-tagging facilitates visualization of plantation distribution, supports compliance with environmental regulations, and aids in evaluating the effectiveness of conservation strategies. This approachnotonlyenhancesprojectmanagementbut also contributes to the sustainable development of hydro projectsby ensuring thehealth and impactof catchment area vegetation are systematically monitored and managed.
Deep Learning in Healthcare and Clinical Studies V. Pandimurugan, S. Angayarkanni Annamalai, G. Saranya, V. Rajaram, A. Chinnasamy Deep Learning for Smart Healthcare Trends Challenges and Applications, 2024
Preventing Health Records Risk Analysis with Explainable AI V Pandimurugan, B Balakiruthiga, J Umamageswaran, S A Angayarkanni, V Rajaram, A Chinnasamy 2nd International Conference on Self Sustainable Artificial Intelligence Systems Icssas 2024 Proceedings, 2024
Text Extraction and Audio Generation from Images using NLP R Rakesh, Dwarakanath, Malathi Eswaran, Chinnasamy A, Chetaly Mawal Proceedings of the 2023 2nd International Conference on Augmented Intelligence and Sustainable Systems Icaiss 2023, 2023
Machine Learning Development in Solving Critical Medical Problems Shaista Fatima, G Sangeetha, P Ponmurugan, A. N. Arularasan, A. Prabhu Chakkaravarthy, R Denis, A. Chinnasamy 6th International Conference on Inventive Computation Technologies Icict 2023 Proceedings, 2023
Smart Plant Watering System 13th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2022, 2022
Covid Patient Health Monitoring System 13th International Conference on Advances in Computing Control and Telecommunication Technologies Act 2022, 2022
A Study On Varicose Vein into Ulcer V. Rajathi, A. Chinnasamy, S. Abarna, J. Sunanthinii, S. Bharathy 2022 1st International Conference on Computational Science and Technology Iccst 2022 Proceedings, 2022
