@ssmiet.ac.in
Associate Professor / ECE
SSM Institute of Engineering and Technology
Wireless Sensor Networks, Network Security, Antennas and Signal Processing
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Thangavel Yuvaraja, Winston Gnanathika Rajan, Salem Jeyaseelan, Rengasamy Ashokkumar, Magudeeswaran Premkumar, PhD W. R. Salem JEYASEELAN and PhD S. R. ASHOKKUMAR
University of Slavonski Brod, Mechanical Engineering Faculty in Slavonski Brod
: This paper outlines a method for identifying and counteracting distributed denial of service (DDoS) and low-rate denial of service (DoS) attacks. These impair significant threats to network security and can disrupt the accessibility and efficacy of systems under attack. The proposed method combines Time-Frequency Analysis (TFA) using Short-Time Fourier Transform (STFT) and a Deep Learning model (DLM), namely Recurrent Neural Network (RNN), to enhance network security. By leveraging the strengths of STFT and RNN, the approach achieves improved detection capabilities and enables timely response and effective mitigation. The CICDDoS2019 dataset has been employed to conduct the evaluation, which provides a diverse set of realistic attack traffic scenarios. The results show that the proposed approach is effective, with an impressive accuracy rate of 99.1%. Compared to traditional methods, the integrated achieves higher accuracy and lower false positive rates. This research highlights the potential of Multimodal Fusion method, for addressing the growing need for advanced defense mechanisms in today's evolving threat landscape.
Dhamodharan Srinivasan, M Premkumar, S Deepa Nivethika, P Dhilipkumar, S Parameswari, and M Kalpana Chowdary
IEEE
In this article, the potential of low-terahertz (THz) technology is discussed to present high data rates in future biomedical systems, and also in the 6G mobile system. However, due to the loss, the design of high-gain antennas is crucial to overcome power limitations and compensate for the additional path loss. This article highlights recent developments in wideband and high-gain sub-mm-wave and low-THz antennas and their fabrication technologies. The advancements in recent technologies have enabled the manufacture of antennas with complicated structures with precise accuracy and less expensive. Overall, the paper emphasizes the importance of high-gain antennas in overcoming the challenges associated with low-THz technology and highlights the potential of AM technology in the development of low-cost, high-performance THz antennas.
M. Premkumar, S. R. Ashokkumar, V. Jeevanantham, G. Mohanbabu, and S. AnuPallavi
Springer Science and Business Media LLC
S. R. Ashokkumar, M. Premkumar, S. Anupallavi, V. Jeevanantham, G. Mohanbabu, and A. Selvapandian
Springer Science and Business Media LLC
S Jayakumar, M Premkumar, G Mohanbabu, R Sangeetha, Josephine Pon Gloria Jeyaraj, and S Deepika
IEEE
This article presents a hexagonal Multiple Input-Multiple Output (MIMO) antenna designed for 5G at 27.5 GHz in the mm-wave spectrum. With soaring data demands, the 28 GHz mm-wave band is vital for high data rates and minimal air attenuation. The hexagonal MIMO antenna efficiently mitigates interference while maintaining a compact design. Utilizing MIMO technology to its fullest, the antenna delivers substantial enhancements in system performance. It achieves exceptional isolation, surpassing -29dB across the entire operating bandwidth, and boasts an impressive peak gain of 6.06dBi. The proposed antenna has robust performance with 89% of overall efficiency and minimal return loss. Beyond 5G, this antenna holds promise in AI applications, serving as a wireless signal collector for critical feature extraction. Notably, in radar data collection, it aids object identification and tracking, critical in autonomous vehicles, aviation, and defense. This article quantifies the antenna’s advantages, emphasizing its role in advancing 5G and its potential in enhancing AI systems. In summary, the hexagonal MIMO antenna is a high-performing solution poised to revolutionize 5G communication and open new vistas in AI signal processing.
Shunmughavel V, Jeevanantham V, Dhamodharan Srinivasan, Premkumar M, S Deepa Nivethika, and Ashokkumar N
IEEE
Nowadays, document retrieval is one of the major tasks in Biomedical Research. Due to the size, a lot of time gets wasted for the user in searching the information they need while retrieving the documents, the problem that arises is the retrieval of non-related documents. In order to overcome this, a concept based ontology mapping method is proposed in this paper. Initially, a query document is given to the retrieval system for retrieving the matched documents based on the concepts in the document. A Stanford parser is used for making the document into sentences and then into Parts-of-speech tagger and Context-free phrase structure grammar representation. After that, the resultant noun phrases and verb phrases are considered as concepts and linking phrases and from these the concept maps are generated for each document. Then the concepts from the concept map of query document are matched with the concept maps of the target document collection using Word Net Ontology. If the measure for the matching is less than the threshold value means, the alternate documents are checked for the matching. Otherwise, the relevant documents for the matched query is retrieved based on the concept maps from the document storage. The experimentation was carried out with the help of medical documents. The proposed retrieval system shows that the documents are retrieved effectively based on concepts and the system performs with higher accuracy value.
Premkumar M, Muthukrishnan A, Ashokkumar S. R, Nagakumararaj S, Sathesh Raaj R, and Dhamodharan Srinivasan
IEEE
In this work presents a lightweight ultra wide band (UWB) antenna with a high fidelity factor (FF) for healthcare applications. In order to obtain the greatest degree of FF in all the field, the design strategy examines the return loss, antenna gain, and group delay throughout the UWB spectrum in each design stage. The final design is an elliptic ground plane with a size of 1620 mm2 and a circular antenna having six rings in the radiating component. In terms of S11and FF, simulations are made in free space and on the body show that it performs admirably within the bandwidth of 3.1 to 10.6 GHz. The results demonstrate that the antenna is capable of identifying malignant tumours and benign tumours.
Ashokkumar S.R, Premkumar M, Dhilipkumar P, P Manikandan, Naveen P, and M. Saravanan
IEEE
An automated epilepsy detection method has been proposed by exploiting the multi-domain features with a few learning algorithms. EEG signals are initially preprocessed to remove the redundant data. Then they are divided into 5-second segments, with each segment containing the extraction of multi-domain information from the frequency domain, temporal domain, connectivity, and graph analysis measurements. From the obtained features, most significant features are selected by the Multi Objective Evolutionary (MOE) method. The correlation matrix is obtained through connectivity calculation, and it is converted into binary undirected and weighted graphs through graph theory analysis. For classification, Support Vector Machine (SVM), Quadratic Discriminant Analysis (QDA), Linear Discriminant Analysis (LDA) has been implemented. Also, the Bayesian optimization (BaO) algorithm has been utilized to optimize SVM parameters. This proposed work is analyzed for EEG signals obtained from the CHB-MIT dataset. The proposed approach resulted in 98.09%, 81.49% and 80.90% accuracy rate for SVM, LDA, and QDA respectively. Conclusively, the SVM classifier outperformed other classifiers in terms of accuracy, Area under the Curve (AUC), sensitivity and specificity with 98.1%, 99.7%, 98.1%, and 98.1% respectively.
M. Premkumar, S.R. Ashokkumar, G. Mohanbabu, V. Jeevanantham, and S. Jayakumar
Elsevier BV
M. Premkumar, S.R. Ashokkumar, V. Jeevanantham, S. Anu Pallavi, G. Mohanbabu, and R. Sathesh Raaj
Elsevier BV
A Puviarasu, M Balaji, R Thirukkumaran, A Siva Kumar, and M Premkumar
Elsevier BV
N Senthilkumar, M Manimegalai, S Karpakam, S.R Ashokkumar, and M Premkumar
Elsevier BV
Magudeeswaran Premkumar, Tharai Vinay Param and G. MohanBabu
Mechanical Engineering Faculty in Slavonski Brod
Security in wireless frameworks is a significant and difficult task because of the open environment. The Denial of Service (DoS) is as yet significant endeavour to make an online assistance inaccessible. The objective of this attack is to keep the authentic nodes from getting to the administrations. Intrusion detection systems assume an essential job in identifying DoS attacks that improve the performance of the system. However massive information from the system presents huge difficulties to the discovery of DoS attack, as the identification framework needs adaptable techniques for gathering, storing and processing a lot of information. In order to defeat these difficulties, this paper proposes Hybrid Intrusion Detection System (HIDS) framework dependent on different MLP strategies. In this article HIDS utilizes Naive Bayes (NB), irregular random forest (RF), decision tree (DT), multilayer perceptron (MLP), K-nearest neighbours (K-NN) and support vector machine (SVM) for better outcomes. The NSL-KDD dataset and UNSW-NB15 dataset are taken to examine the detection accuracy. The experiment results show that the proposed defence system is accomplished with high accuracy, high detection rate and low false alarm rate in both the datasets.
S. R. Ashokkumar, S. Anupallavi, G. MohanBabu, M. Premkumar, and V. Jeevanantham
Wiley
Emotions are biologically based psychological states brought on by neurophysiologic changes, variously associated with thoughts, feelings, behavioral responses, and a degree of pleasure or displeasure. In order to assure active collaboration and trigger appropriate emotional input, accurate identification of human emotions is important. Poor generalization capacity induced by individual variations in emotion perceptions is indeed a concern in the current methods of emotion recognition. This work proposes a new dynamic pattern learning system based on entropy to allow electroencephalogram (EEG) signals for subject‐independent emotion recognition with strong generalization and classification through Recurrent Neural Network and Ensemble learning. First, dynamic entropy measurements are used to derive consecutive entropy values over time from EEG signals in quantitative EEG calculations. Experiment findings indicate that in order to distinguish negative and positive emotions, the highest average accuracy of 94.67% is achieved. In addition, the findings have completely shown that this approach produces outstanding performance for emotion detection across individuals relative to recent studies.
M. Premkumar and T. V. P. Sundararajan
Springer Science and Business Media LLC
S. R. Ashokkumar, S. Anupallavi, M. Premkumar, and V. Jeevanantham
Wiley
Epilepsy is one of the most common neurological diseases of the human brain. It affects the nervous system of brain which shows the impact on an individual's life because of its repetitious occurrences of seizure. Epileptic detection using automatic learning is essential to reduce the substantial work on reviewing continuous electroencephalogram (EEG) signal in spatial and temporal dimensions. A novel methodology is implemented on EEG signals for the detection of epileptic seizure with the combination of fractional S‐transform (FST) and entropies along with deep convolutional neural networks (CNN). The original EEG signals are preprocessed with discrete wavelet transform to generate Daubechies‐4 (Db4) wavelets. FST is enacted on every segment of the preprocessed signal for time‐frequency representation and the features are obtained through entropies. Afterwards, a 15‐layer deep CNN with dropout layer and soft‐max is used for classification. The experimental results showed that the singular value decomposition entropy are more stable and deep CNN models always performed better for this entropy. A specificity of 98.70%, sensitivity of 97.71%, and accuracy of 99.70% are achieved for the multichannel segment.
Ashokkumar S. R, Premkumar M, Selvapandian. A, Jeevanantham V, and Anupallavi S
IEEE
Recently the importance of the Internet of Things (IoT) has opened up the possibility of developing new applications related to mobile-Health. These include real-time monitoring of various physiological data such as EEG and ECG. Due to the complexity of remote monitoring applications, the need for continuous sensing is often overwhelming. This causes data processing to consume a large amount of energy. This research work introduces a data specific transceiver design, which takes the advantage of physical layer data collection characteristics. The proposed design shows excellent performance in terms of low complexity and signal distortion with high data gain.
M. Premkumar and T.V.P. Sundararajan
Elsevier BV