KANIMOZHI R

Dr. KANIMOZHI R

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
5 November 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
PRP. S. Ramesh
P. S. Ramesh
A robust Deep Reinforcement Learning-based Intrusion Detection Scheme (DRL-IDS) for Software-Defined Networking (SDN) which combines the Long-Short Term Sequence Recurrent Neural Network (LFTS-RNN) with the Particle Cloud-Integrated Joint Time- and Feature-Optimization Algorithm (PC-JTFOA). The hybrid model aims to enhance the security of SDN through the detection and mitigation of a wide array of Distributed Denial of Service attacks and network misbehaviors across different SDN planes. The LFTS-RNN is used for accurate attack detection and misbehavior identification. Meanwhile, the PC-JTFOA optimizes feature selection, load balancing, and energy-efficient routing, thus ensuring fast and reliable network traffic management. The deep reinforcement learning approach further enables continuous adaptation to changing network behaviors, thus making the model dynamically adapt to known as well as emerging attack vectors. The proposed DRL-IDS scheme obtains superior performance in experimental results based on the NSL-KDD and WPPD datasets. The LFTS-RNN model indicates a highly impressive sensitivity of 98.67% and specificity of 97.42%, while the DRL-IDS model presents an detection accuracy of 99.85%. The PC-JTFOA further improves the solution by exhibiting a low response time of 1423 ms, which indicates tremendous improvement in computational efficiency. A comparative analysis with the existent intrusion detection methods pointed out that the scheme proposed not only outperforms other models in terms of detection accuracy as well as adaptability, but it also reduces complexity.

9 December 2025 in Scientific Reports
Verified
Journal Article
Open access
VPV. Padmavathi
V. Padmavathi
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
RSR. Saminathan
R. Saminathan
Thanks to Industry 4.0, digital twin technology now leads the way in smart manufacturing, enabling the real-time replication of devices to facilitate maintenance, process improvements, and smooth operation. Still, many conventional Digital Twin approaches heavily rely on cloud solutions, which result in high latency, limited scaling options, and increased risks to data privacy. However, real-time, physics-enabled simulation remains very challenging for making fast and accurate choices in busy factories due to difficulties with computer speed and network capacity. We introduce a new method that combines Digital Twin, Edge AI, and FL to make co-simulation in smart factories quick, efficient, and secure. Moving the learning and simulation steps to edge devices near the factory floor enables real-time data processing and inference. Federated Learning enables these edge devices to share and learn from one another, ensuring that important production data remains within the factories. This engine, based on FMI, enables the real-time co-simulation of physical events with their virtual counterparts, allowing the system's behavior to be tested under various operating modes. The framework achieved a latency reduction of up to 35%, a 28% decrease in cloud usage, and a 13.2% throughput gain compared to cloud-only architectures, highlighting its scalability and industrial applicability. Additional ways to enhance this area include the use of blockchain, automated learning updates, and allowing machines to transfer knowledge from plant to plant.

30 July 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
VPV. Padmavathi
V. Padmavathi
PRP. S. Ramesh
P. S. Ramesh
The widespread use of mobile phones among seniors (i.e., those over the age of 50) is still not sufficiently acknowledged, especially concerning security and online threats. The present research investigates the Perceived Danger (PD) of attacks on computers as a possible obstacle to the acceptance and utilization of cell phones among older people. This study examines the differences between smartphone owners and non-users, and older people with and without mobile phone knowledge. We conducted a qualitative cross-sectional poll among elderly people in TamilNadu The results obtained from multilevel modelling indicate solid evidence of support for the associations between the Motivation to Use (MTU), Learning Adaptability (LA), perceived usefulness (UF), and Technology Anxiety (TA). Additionally, there is a notable association between Avoidance Behaviour [AB], adoption readiness [AR], perceived Ease of Use [EoU], social connectivity impact (SC), self-efficacy [SE], and Digital Engagement (DE).Though Perceived Threat (PT) negatively correlates with general smartphone adoption, the association between PT and both Attitude toward Use (AtU) and Perceived Usefulness (PU) seems surprisingly positive. The findings indicate that MTU has a negative relationship with TA and AB but a beneficial relationship with LA and UF. AR shows poor relationships with TA and AB while having positive relationships with LA and UF. By their impacts on LA, UF, TA, and AB, EoU, SC, SE, and DE have an impact on MTU. These patterns indicate that AR, SE, and SC could indirectly impact PT’s impact on adoption. The research discovered that advertising must reduce TA and AB, which negatively impacts mobile technology use, and boosts smartphone adoption among elderly individuals. LA and UF, together, enhance MTU and have to be increased. Enhancing Learning Adaptability (LA) and Perceived Usefulness (PU), while reducing Technology Anxiety (TA) and Avoidance Behaviour (AB), can be achieved through improved attitude preparedness, perceived ease of use, and strengthened social connectivity. Instructional materials should be easily implemented, develop trust, and encourage DE with self-efficacy.

16 April 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
VPV. Padmavathi
V. Padmavathi
In the time when the communication mode is changing towards the digital aspect, such as; computing devices and the internet, it is natural that there is a need to safeguard the safety as well as the privacy of the data being shared, which can also compromise the integrity. Image steganography is a technique that conceals messages in pictures taken on a digital camera, and serves as a means of secure data trans-mission. In this study, we provide a novel method for image watermarks which is highly secure and robust through the application of Recur-rent Neural Systems and fuzzy logic. By merging fuzzy logic with the flexibility of Recurrent Neural networks, the proposed system enhances the protection and rate of hidden information even under noise, compression and other attacks. The model utilizes the learning capabilities of Recurrent Neural networks for optimization of the embedding process, while the Recurrent Neural model is useful for managing uncertainty and improving decision-making processes. Such two-layered systems ensure the inaudibility and invisibility of steganographic data while limiting the embedding of large amounts of visible data, which is highly effective. Robustness and security of the method is tested by using various images types within different environments. To enhance the efficiency and robustness of steganography of images, this work proposes an innovative two-layer method that combines fuzzy logic with Recurrent Neural Networks (RNNs). By enhancing embedding, strengthening attack resistance, and offering an adaptable, intelligent framework for safe data concealing, it advances the field. It was observed that compared to other techniques used in steganography approaches, it is pleasing to note that the whole embedding efficiency and resistance to attacks are far and above enhancement candidates. The results demonstrate that the seamless integration of fuzzy logic and Recurrent Neural Systems provides an efficient and scalable framework for secure image steganography.

1 January 2019 in International Journal of Cloud Computing
Verified
Journal Article
Closed access
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
Mobile phone cloud application users are hesitating to move their information to the cloud service supplier due to security. When People try to store lot of files, they need to check the storage limit. In the proposed technique user's original data is taken, watermarking is applied and then the data is compressed to reduce the storage size of the data. Then some secrete message is embedded with the image for the authentication and these authenticated data is involved in security processes of Specter encryption. This encrypted data is divided into n pieces and each piece of data is going to be stored in a different storage location of the cloud. When intruder wants to take the data it is impossible because different location of the data pieces and encryption technology. When downloading the reversed process is applied and original data is received. The proposed method uses J-bit encoding (JBE) for compression.

3 March 2026 in Concurrency and Computation: Practice and Experience
Verified
Journal Article
Closed access
KANIMOZHI RKANIMOZHI R
KANIMOZHI R
VPV. Padmavathi
V. Padmavathi