@tce.edu
Assistant Professor, Computer Science and Engineering
Thiagarajar College of Engineering
B.E- Computer Science and Engineering, Noorul Islam College of Engineering / M.S.University, Tirunelveli,2000
M.E- Computer Science and Engineering, Engineering College / Anna University, Chennai,2004
Information and Communication Engineering, Anna University, Chennai,2018
Data Management
Computer Networks
Machine Learning
Cloud Computing
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
S. Karthikeyani, S. Sasipriya, and M. Ramkumar
Elsevier BV
D. Karthika Priya, B. Sathya Bama, M. P. Ramkumar, and S. Mohamed Mansoor Roomi
Springer Science and Business Media LLC
M. Ramkumar, S. Syed Jamaesha, M. S. Gowtham, and C. Santhosh Kumar
Springer Science and Business Media LLC
Sakthi Ulaganathan, M. P. Ramkumar, G. S. R. Emil Selvan, and C. Priya
Springer Science and Business Media LLC
M Ramkumar, C Ganesh Babu, R Sarath Kumar, Akash S, Dhanu Sri T K, and Priyanka G S
IEEE
Many Deep Learning (DL) applications have been able to attain the best possible outcomes. The DL model is used for categorizing electrical brain signals for the interface between the brain and computer. Fast Fourier Transform converts time-domain Electroencephalogram (EEG) data to frequency-domain data. The Deep Metric Learning (DML) model using the nearest neighbor is used to train a spectrogram composed of 72 concatenated EEG channels. There were three sets of triplet pairings of epochs labeled positive, negative, and anchor in the training batch. The Recurrent Neural Network (RNN) is combined with a triplet loss function for minimizing the Euclidean distance in embeddings of the identical class. RNN can also increase the distance among the embeddings of various labeled images to train an embedding feature. An EEG dataset of 22 untrained people was used to test the suggested approach. It was demonstrated that the DML method using the nearest neighbor can be used for training samples in a way that 100 EEG trials for a single subject. This model is used in the modification of the well-known problem of the substantial inter-individual inconsistency of human Motor-Imagery Brain-Computer Interfaces (MI-BCI) EEG with an accuracy of 0.75 percent.
J. Lece Elizabeth Rani, M. P. Ramkumar, and G. S. R. Emil Selvan
Springer Nature Singapore
Harshini Sivakami V, Nivedhidha M, Ramkumar M P, and Emil Selvan G S R
IEEE
The growing dependence on technology in healthcare has resulted in the creation of sophisticated hospital networks that are highly linked and vulnerable to cyber threats. A reliable Network Intrusion Detection System (NIDS) is required to identify and prevent such cyberattacks. The network intrusion detection is vital for safeguarding hospital networks and guaranteeing data security. The CICIDS2017 dataset contains a comprehensive set of network traffic characteristics for assessing network intrusion detection systems. Besides that, class imbalance is a prevalent difficulty in intrusion detection and it may have a considerable impact on the effectiveness of classification algorithms. The suggested solution employs a Machine Learning (ML) based NIDS for hospital networks that utilizes CopulaGAN (Generative Adversarial Network) to address the challenges due to imbalanced class ratio. The synthetic samples of minority classes were created to balance the dataset and improve detection accuracy. The Random Forest (RF) algorithm is used to discover the most defining features in the dataset and its hyperparameters are tuned to improve classification performance. Overall, the CopulaGAN boosted Random Forest based NIDS described here is a valuable solution for detecting network intrusions in hospital networks.
Prasanna S S, G. S. R. Emil Selvan, and M. P. Ramkumar
IEEE
Industrial Control Systems (ICS) become a crucial target for hackers as these devices are unsupported in terms of storage, complex computations, and security. On the other hand, providing security for these devices are very difficult. Because small downtime of these devices will lead to financial loss and reputation loss for the industry. Sometimes, even it may lead to a disaster. We should also ensure that there are no heavy workloads present in this Operational Technology (OT) network. Hence, the intrusion detection systems for ICS should operate with high levels of accuracy utilizing resources as low as possible. In this paper, an anomaly-based ICS intrusion detection system is proposed. The proposed model uses Pearson Correlation for feature selection and Deep Neural Network for detection. The suggested solution is put to the test using an HIL-based Augmented ICS security dataset. The results show that the proposed model has a good fit and achieves a higher accuracy rate.
R. Jeyarohini, K. R. Aravind Britto, and M. P. Ramkumar
IEEE
The recent advances in the semiconductor nano technologies increase the complexity of very large-scale integration circuits. With the fabrication technology entering the deep Nanoscale era more demand at greater huge complication of integrated chip design arises. The development of eccentric floor plan without over-lapping is the intention of VLSI floor planning. The major objective of VLSI is concerned with floor plan area minimization and wire length optimization. Floor planning is the important step that creates an optimal layout solution for the VLSI circuits. This article attempted to collect the necessary information in reducing the floor plan layout by means of reducing the wire length, white space area with the employment of computer aided metaheuristic algorithms. In order to ease the process of execution the best local search algorithms like a Particle Swarm Optimization and Simulated Annealing algorithms is compared.
D. Karthika Priya, M.P. Ramkumar, and D. Menaka
IEEE
Railway networks are crucial for transportation infrastructure, and ensuring their safety and efficiency is of utmost importance. This paper proposes an efficient computer vision-based railway detection system that incorporates state-of-the-art techniques for fastener fault detection. By utilizing an improved version of the YOLOv7 algorithm and the novel Self Improved-Owl Optimization algorithm, the system accurately locates and recognizes the states of rail fasteners, enabling efficient fault detection. Experimental evaluations demonstrate significant improvements in detection accuracy and computational efficiency compared to traditional methods. The system's ability to provide precise localization and state recognition of fasteners contributes to the overall safety and efficiency of railway networks, reducing maintenance costs and enhancing transportation reliability. The proposed model achieves an impressive accuracy of 97%, indicating its effectiveness in accurately detecting and recognizing rail fastener faults. This high level of accuracy demonstrates the robustness and reliability of the model, making it a valuable tool for railway maintenance teams.
Subash A, Shane Rex S, Vijay G, G S R Emil Selvan, and M P Ramkumar
IEEE
This work proposes a method for detecting Android malware by leveraging static permissions and machine learning algorithms. A dataset of 398 Android applications was compiled, with potential malware behaviour identified through analysis of Android API usage. After preprocessing the dataset, three machine learning algorithms―Naive Bayes, Decision Tree, and K-Neighbours―were trained and their accuracies compared. Naive Bayes demonstrated the highest accuracy, making it the most suitable algorithm for detecting Android malware using static permissions. The proposed method offers an effective approach to identifying potential malware behaviour in Android applications, ultimately contributing to a more secure Android ecosystem.
Arun Karthick S, Rahul Shiva Konar, Harshavarthini V S, Ramkumar M P, and Emil Selvan G S R
IEEE
In medical imaging and diagnostics, identifying brain tumors is crucial. The objective of this study is the automated diagnosis of brain tumors using Magnetic Resonance Imaging (MRI) pictures, and the suggestion of a deep learning approach utilizing the YOLOv7 neural network model is presented. Initially a dataset had been collected which consists of 300 images. The image datasets are labelled in such a way that the area of tumor is correctly annotated. The dataset is then expanded in size as a result of the data augmentation approaches that are used. The model is trained on 80% of the photos, then tested on the remaining 20%. On a sizable dataset of brain MRI images, the suggested approach is assessed, and its performance is compared to that of earlier YOLO algorithm versions. The results demonstrate that the proposed approach of using YOLO7 achieves high accuracy in brain tumor detection, outperforming previous versions by a significant margin. The proposed model of YOLOv7 achieves a mAP score of 94%. These results prove that the YOLOv7 model is capable to detect the brain tumor along with its location which proves, that the model has potential in finding tumors using MRI.
Venkatesan C, Thamaraimanalan T, Ramkumar M, Sivaramakrishnan A, and Marimuthu M
IEEE
In a variety of medical applications, electrocardiogram (ECG) records are essential for the identification of cardiovascular disorders. Cardiovascular issues such cardiac arrhythmias and coronary heart disease (CHD) are among the most prevalent and can result in cardiac arrest or sudden cardiac death. This study employs ECG signal preprocessing and feature extraction to identify cardiac arrhythmias and evaluate CHD risk. This research stresses the use of a Support Vector Machine (SVM) classifier for cardiac arrhythmia identification after ECG signal preprocessing. The preprocessed ECG signal is then subjected to arrhythmic beat classification to find anomalies. Extracted R-peaks from the ECG signal are divided into normal and arrhythmic risk subjects using the SVM classification-based approach for abnormality identification. When compared to other similar classifiers, the K-Nearest Neighbor (KNN) classifier offers the highest classification accuracy of 97.5%.
Vadamodula Prasad, Emil Selvan G. S. R., and Ramkumar M. P.
Informa UK Limited
M. P. Ramkumar, G. S. R. Emil Selvan, M. Mahalakshmi, and R. Jeyarohini
Springer Nature Singapore
S. Sams Aafiya Banu, B. Gopika, E. Esakki Rajan, M. P. Ramkumar, M. Mahalakshmi, and G. S. R. Emil Selvan
Springer Nature Singapore
I. Gokul Ganesh, A. Selva Sugan, S. Hariharan, M. P. Ramkumar, M. Mahalakshmi, and G. S. R. Emil Selvan
Springer Nature Singapore
Muthuperumal Periyaperumal Ramkumar, Pauliah David Mano Paul, Balajee Maram, and John Patrick Ananth
Wiley
The Internet of Things (IoT) has appreciably influenced the technology world in the context of interconnectivity, interoperability, and connectivity using smart objects, connected sensors, devices, data, and appliances. The IoT technology has mainly impacted the global economy, and it extends from industry to different application scenarios, like the healthcare system. This research designed anti‐corona virus‐Henry gas solubility optimization‐based deep maxout network (ACV‐HGSO based deep maxout network) for lung cancer detection with medical data in a smart IoT environment. The proposed algorithm ACV‐HGSO is designed by incorporating anti‐corona virus optimization (ACVO) and Henry gas solubility optimization (HGSO). The nodes simulated in the smart IoT framework can transfer the patient medical information to sink through optimal routing in such a way that the best path is selected using a multi‐objective fractional artificial bee colony algorithm with the help of fitness measure. The routing process is deployed for transferring the medical data collected from the nodes to the sink, where detection of disease is done using the proposed method. The noise exists in medical data is removed and processed effectively for increasing the detection performance. The dimension‐reduced features are more probable in reducing the complexity issues. The created approach achieves improved testing accuracy, sensitivity, and specificity as 0.910, 0.914, and 0.912, respectively.
Ramkumar M.P., T. Daniya, P. Mano Paul, and S. Rajakumar
Elsevier BV
Ramkumar M. P, Suresh Ponnan, Siddharth Shelly, Md. Zair Hussain, Mohd Ashraf, and Anandakumar Haldorai
Elsevier BV
Ramkumar Muthuperumal Periyaperumal, Ganeshan Ramasamy, Maria Azees, and Ulaganathan Sakthi
Wiley
Cloud computing is an emerging standard in modern days for the purpose of sharing huge data, as it affords numerous user friendly behaviors. Cloud computing services offer an extensive range of resource pool in order to maintain huge scale data. Although, cloud computing model is disposed to several cyber‐attacks and security problems regarding cloud structure, because of the dynamic and distribute character and exposures in virtualization implementation. Distributed denial‐of‐service (DDoS) attack is a type of cyber‐attack, which disturbs the usual traffic of targeted cloud server. Moreover, DDoS produces malicious traffic in cloud structure, and thus consumes cloud resources. In this paper, an effective DDoS attack detection model, named fractional anti corona virus student psychology optimization‐based deep residual network (FACVSPO‐based DRN) is implemented using spark architecture. The devised FACVSPO approach is newly designed by incorporating anti coronavirus optimization (ACVO) algorithm, fractional calculus (FC) and student psychology based optimization (SPBO) model. Moreover, the hybrid correlative scheme is designed for extracting significant features for attack detection. The DRN structure is utilized for performing attack recognition, which categorizes the data as normal or attack. In addition, the DRN classifier is trained by the developed FACVSPO approach. The developed attack detection model outperformed other existing techniques in terms of testing accuracy, true negative rate (TNR), true positive rate (TPR) of 0.9236, 0.9141, and 0.9412, respectively. The testing accuracy of the implemented model is 12.02%, 8.92%, 7.27%, 6.30%, 5.68%, and 1.20% better than the existing methods, such as Taylor‐elephant herd optimisation based deep belief network (TEHO‐DBN), deep learning, deep neural network (DNN), multiple kernel learning, Fuzzy Taylor elephant herd optimisation (EHO)‐based DBN, fractional anti corona virus optimization‐deep neuro fuzzy network (FACVO‐based DNFN), respectively. Similarly, the TNR is 10.14%, 6.88%, 5.94%, 5.46%, 4.25%, and 3.28% and TPR is 12.33%, 9.46%, 8.05%, 7.41%, 6.02%, and 3.04% better than the existing methods.
Ramkumar M .P . , P.V. Bhaskar Reddy, J.T. Thirukrishna, and Ch. Vidyadhari
Elsevier BV
Subasri I, Emil Selvan G S R, and Ramkumar M P
IEEE
With the growing recognition that current Internet protocols have significant security flaws; several ongoing research projects are attempting to design potential next-generation Internet architectures to eliminate flaws made in the past. These projects are attempting to address privacy and security as their essential parameters. NDN (Named Data Networking) is a new networking paradigm that is being investigated as a potential alternative for the present host-centric IP-based Internet architecture. It concentrates on content delivery, which is probably underserved by IP, and it prioritizes security and privacy. NDN must be resistant to present and upcoming threats in order to become a feasible Internet framework. DDoS (Distributed Denial of Service) attacks are serious attacks that have the potential to interrupt servers, systems, or application layers. Due to the probability of this attack, the network security environment is made susceptible. The resilience of any new architecture against the DDoS attacks which afflict today's Internet is a critical concern that demands comprehensive consideration. As a result, research on feature selection approaches was conducted in order to use machine learning techniques to identify DDoS attacks in NDN. In this research, features were chosen using the Information Gain and Data Reduction approach with the aid of the WEKA machine learning tool to identify DDoS attacks. The dataset was tested using K-Nearest Neighbor (KNN), Decision Table, and Artificial Neural Network (ANN) algorithms to categorize the selected features. Experimental results shows that Decision Table classifier outperforms well when compared to other classification algorithms with the with the accuracy of 85.42% and obtained highest precision and recall score with 0.876 and 0.854 respectively when compared to the other classification techniques.
Deepa S, Nivetha S, Thulasi Lakshmi K, Emil Selvan G S R, and Ramkumar M P
IEEE
Named Data network consists of three data structures for forwarding the incoming Interest packets. Whenever the interest packet arrives, it has to search in the pending interest table and the content store. When a large amount of interest packet arrives, the time required to search the matching content in the pending Interest table can be increased. And also, as the size of Content Store is limited, it will be difficult to satisfy the incoming interest packets’ needs. So the interest packets in the content store have to be replaced. In this paper, Optimal Content Store with tree based approach is proposed and is compared with Optimal Content Store without tree based approach. The comparison shows that packet drop rate and delay is comparatively reduced in Optimal Content store with tree based approach.
Mahalakshmi M, Ramkumar M P, and Emil Selvan G S R
IEEE
Emerging technologies throughout the world work with the baseline of Cyber Physical Systems (CPS) to provide remote accessibility. Large scale industries make use of CPS infrastructure, known as Industrial Control Systems (ICS). The ICS using the Supervisory Control And Data Acquisition systems (SCADA) provides remote accessibility, which are highly prone to cyber-attacks. Here arises the need of Intrusion Detection System (IDS). Any Machine Learning (ML) based IDS heavily depends on the historical data, in that case, the class imbalance problem plays a crucial determinant factor of the IDS performance. Thus, the re-search is to address the adverse effects of the class imbalance problem in ML based IDS using a hybrid approach. To do so, the SCADA dataset of the gas pipeline critical infrastructure, is used. To handle the imbalanced data problem, the Synthetic Minority Oversampling Technique with Support Vector Machine (SMOTE_SVM) is adopted as a data level solution combined with an algorithmic level solution, Cost-Sensitive Machine Learning (CSL) is adopted. Different evaluation metrics are used to evaluate the performance of the ML model with and without data balancing techniques. From the results, the hybrid approach of SMOTE-SVM with CSL is proven to be an efficient method to deal with the effects of the imbalanced dataset. The hybrid approach exhibits the following met-rics on using Logistic Regression tuned as CSL; the highest precision and recall scores with 69% and 66% respectively and a zero False Alarm Rate (FAR).