SHARATH KUMAR S

EDUCATION

B.E in Information Science and Engineering
M.Tech in Computer Network Engineering
( pursuing in the area of Software defined Networks

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Networks and Communications, Multidisciplinary, Artificial Intelligence, Computer Science Applications

Scopus Publications

Efficient and interactive fault tolerant, distributed SDN controller for a secured communication in a software defined network
S. Sharathkumar, N. Sreenath
Multimedia Tools and Applications, 2025
HSPC-SDN: Heuristic Driven Self-Configuring Proactive Controller for QoS-Centric Software Defined Network
S Sharathkumar, N Sreenath
International Journal of Computing and Digital Systems, 2023
The exponential rise in software computing, low-cost hardware and allied application demands has broadened the horizon for wireless technologies to serve different purposes.Wireless communication systems being central to the modern innovation and industrial growth have given rise to the different communication ecosystems including internet of things, machine to machine communication, wireless local area network, Ad-hoc networks etc.However, coping with non-negotiable service level agreements have forced industries to ensure quality of service (QoS) and quality of experience demands.To meet such demands, software defined network (SDN) has gained widespread attention.The ability to enable higher programmability, flexibility and scalability makes SDN-based system viable; yet, guaranteeing their robustness towards dynamic network, link-failure and adaptive QoS-centric recovery has remained a challenge.In sync with this motive, in this paper a robust Heuristic Driven Self-Configuring Proactive Controller is designed for QoS-centric SDN network (HSPC-SDN).Unlike classical data-plane SDN controllers or allied routing solutions, HSPC-SDN performs multi-constraints risk assessment followed by heuristic driven disjoint multiple path selection to support proactive network failure-recovery.HSPC-SDN applies dynamic link-quality information, cumulative congestion degree, probability of successful transmission and link quality change index to perform best forwarding device selection to alleviate any malicious behaviour or malfunction during transmission.Subsequently, it applies genetic algorithm to perform disjoint multiple forwarding cum failure recovery path selection that in conjunction with AND logic function enables self-configuring route recovery to meet fault-tolerant QoS-centric communication.The proposed heuristic model exploits network availability information amalgamated with minimal distance and strictly no-shared component criteria to perform multiple disjoint forwarding-path cum recovery-path selection.Simulation based results revealed that HSPC-SDN, which can be implemented as a standalone single data-plane controller as well as a middleware routing concept achieves superior average packet delivery rate of 98.03%, packet loss rate of 1.97%, recovery time of 1.66ms and energy consumption of 77.14mJ over other disjoint forwarding path based SDN controllers.
Performance Evaluation of a standard reliable, fault tolerant Software Defined Wireless Sensor Network with an extended early inactive problem
Sharathkumar. S, N Sreenath
2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering Icacite 2022, 2022
Nowadays, in this modern era, wireless communication is gaining considerable increase in importance over the existing traditional networks and has become the main part of any network architecture. Among them, Software Defined Network (SDN) has gained more priority alongside wireless sensor networks. This results to form a type of typical network i.e. Software defined wireless sensor network (SDWSN). In this scenario, Energy efficiency, increased throughput and data delivery rate are the main goals to be achieved and evaluated to measure the performance of the network. Further, to achieve this transmission reliability in the network, we need to employ fault resilient, reliable and fault tolerant wireless data transfer in the network. Additionally, to achieve transmission reliability, traditional fault tolerant-MAC (FT-MAC) and Reliable timeout-MAC (RT-MAC) protocols in WSN [1] [2] is proposed to satisfy the requirements for the appropriate schedule of routine work and to achieve the final output. In this paper, we have investigated the optimization process in the form of the extension of passive inactiveness problem in certain topologies to evaluate the delay in packet delivery and to measure the reliability of the network. In order to achieve the solution for this problem, here we propose a fault tolerant, reliable extended timeout MAC (FTETMAC) protocol that guides the forwarded connecting nodes to final output state and reschedules synchronization in the form of time management. According to the simulation results our proposed FTETMAC protocol achieves better reliability, high throughput and energy efficiency compared to the other traditional MAC protocols along with their components.
Design of Reliable Fault Tolerant Architecture and Analysis of a Software Defined Network to recover from DDoS Attack
S Sharathkumar, A John Paul, N Sreenath
7th IEEE International Conference on Recent Advances and Innovations in Engineering Icraie 2022 Proceedings, 2022
A new centralized networking architecture is proposed here which mainly separates the control traffic with that of data traffic i.e Software Defined Network (SDN). The main network connecting devices like routers, switches are used for the separation of the network traffic to achieve free movement of information in the network. This technology allows organizations to rapidly manage and reconfigure network resource. In this modern world, Internet has gained more interesting as well as challenging medium, as a result a good fault tolerant and recovery process is required to detect any defects in the network. In this paper, the Fault Tolerance (FT) is used as a critical component to achieve proper stability for any communication medium. To achieve high availability and dependability on systems, reliable fault tolerance technique is required. Fault tolerant SDN is used in this architecture to solve the problem. A mathematical model known as the Shared Risk Link Group (SRLG) is used in this architecture. The performance metrics like Time delays, packet loss, and recovery time are measured under normal traffic and during Distributed Denial of Service (DDOS) attack. As a result, difference between pre-attacked as well as post-attacked network states are analyzed.
Automated Sleep Staging Analysis using Sleep EEG signal: A Machine Learning based Model
Santosh Kumar Satapathy, D Loganathan, Sharathkumar S, Praveena Narayanan
2021 International Conference on Advance Computing and Innovative Technologies in Engineering Icacite 2021, 2021
Sleep is essential for people health and well-being. However, numerous individuals face sleep problems. These problems can lead to several neurological and physical disorder diseases, and therefore, decrease their overall life quality. Artificial intelligence methods for automated sleep stage classification (ASSC) are a fundamental approach to evaluate and treat this public health challenge. The main contribution of this paper is to present the design and development of an ASSC. This study supports the recognition of sleep stages and provides relevant information on the sleep process according to American Academy of Sleep Medicine manuals. The proposed method includes a two-step execution process. On the one hand, the sleep records are extracted through electroencephalogram signals. Three different health condition subjects of distinct gender and different age groups have been analyzed. On the other hand, different session recordings of sleep processes from two additional nights are considered. The proposed work uses a single channel for two-state sleep stage classification. This study uses a public dataset and incorporates data pre-processing, data extraction, and feature selection. The entire experiment was executed on different medical conditioned subjects using a support vector machine (SVM). The reported results signify that the proposed model achieved the best classification accuracy of 97.73% with subgroup-I subject using SVM classification models, respectively.
Automated Sleep Staging Using Convolution Neural Network Based on Single-Channel EEG Signal
Santosh Kumar Satapathy, S. Sharathkumar, D. Loganathan
Lecture Notes in Networks and Systems, 2021
A Machine Learning Model for Automated Classification of Sleep Stages Using Polysomnography Signals
Santosh Kumar Satapathy, Hari Kishan Kondaveeti, D. Loganathan, S. Sharathkumar
Lecture Notes in Electrical Engineering, 2021
Convolutional neural network for classification of multiple sleep stages from dual-channel EEG signals
Santosh Kumar Satapathy, D Loganathan, Praveena Narayanan, S Sharathkumar
4th IEEE Conference on Information and Communication Technology Cict 2020, 2020
Backgrounds and Objectives: Sleep-related disorders are critical diseases and they need to be proper diagnosed as early as possible. The major difficulty is that fewer medical experiments are available in remote locations to diagnose different types of sleep disorders. This research work presents an automated classification of sleep stages using deep learning techniques to diagnosis multiple sleep diseases from single-channel and different combinations of channels of EEG signals.Methods: In this proposed work, we involve three different forms of signal inputs for the automatic classification of sleep stages from EEG signals. The proposed research work presents a one-dimensional convolution neural network (1D-CNN) model for multiple sleep stages because of its high robustness for automatically classifying the sleep stages from brain signals without involving any types of feature extraction/selection, which is one of the challenging processes in the earlier literature. The proposed model contained seven convolution layers followed by two fully connected layers. The main objective for designing such a custom deep neural network is to improve classification accuracy performance with less number of learnable parameters.Results: The proposed model has used two subgroups of the ISRUC-Sleep dataset. We also obtain a k-cross-fold validation approach over the subjects, which ensure that there is no possible contamination in between training and testing. The experimental results for this proposed model for classification of five classes of sleep stages (wake, non-rapid eye movement N1-N3, and rapid eye movement). The proposed model was evaluated by classification accuracy, precision, sensitivity, F1score, and Cohen’s Kappa score. The proposed 1D-CNN architecture achieved the highest classification accuracy of 95.85%(C3-A2),94.11%(O1-A2), and 97.22%(C3-A2+O1-A2) using the SG-I dataset, similarly, the same model reported accuracy using the SG-II dataset of 95.73%(C3-A2),94.02%(O1-A2), and 95.06% (C3-A2+O1-A2)Conclusions: Our proposed methodology is efficient and effective for multiple sleep staging. The proposed 1D-CNN model is ready for clinical purposes and can be managed with a huge number of polysomnography data.
Adaptive content-aware access control of EPR resource in a healthcare system
s. Sharathkumar, G. Jagadamba
2017 International Conference on Advances in Computing Communications and Informatics Icacci 2017, 2017
In this modern era of the digital world, the patient's medical records are electronically maintained and updated for the maintenance of the good health. The electronically stored patients' health status is termed as the Electronic Patient Record (EPR). The EPR service includes the accessing, storage, and maintenance by the authenticated users. However, in the emergency situation, the EPR can be required by the unauthenticated users/physician. In these situations, the primary challenge is to prove the identity of an unknown physician/user. In these cases, there is a need for an adaptive security mechanism, where the adaptive mechanism includes intelligence to identify the emergency situation and adapt subsequently. In the health care, the privacy, integrity and access control over the generated data places an important role. To address these issues in this paper, we use a structural design that involves procedure based and distributed management in conviction to afford an elastic method to access the electronically stored records depending on the users' digital identification.

Publications

HSPC-SDN: Heuristic Driven Self-Configuring Proactive Controller for QoS-Centric Software Defined Network
2022-08-25 | Preprint
DOI: 10.21203/
CONTRIBUTORS: Sharathkumar S; Sreenath N
Show less detail
URL

Contributors
Sharathkumar S (Author)
Sreenath N (Author)

Recommender System Using Machine Learning
1st IEEE International Conference on Advances in Information Technology, ICAIT 2019 - Proceedings
2019 | Conference paper
DOI: 10.1109/
EID: 2-
Part of ISBN: 9781728132419
CONTRIBUTORS: Priyanka, M.B.; Ankit Karan, S.; Pruthvi, M.; Manushree, M.R.; Kumar, S.

Adaptive content-aware access control of EPR resource in a healthcare system
2017 International Conference on Advances in Computing, Communications and Informatics, ICACCI 2017
2017 | Conference paper
DOI: 10.1109/
EID: 2-
CONTRIBUTORS: Sharathkumar, S.; Jagadamba, G.