PRABAKARAN D
@ifet.ac.in
ASSOCIATE PROFESSOR, DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING
IFET COLLEGE OF ENGINEERING, VILLUPURAM
B.E. in ECE
M.Tech in Applied Electronics
Ph.D. pursuing in Network Security
Published more than 10 research articles in SCI, Scopus indexed journals.
EDUCATION
B.E. in ECE
M.Tech in Applied Electronics
Ph.D. pursuing in Network Security
RESEARCH INTERESTS
Network Security, Cloud computing, Antenna
FUTURE PROJECTS
CLOUD COMPUTING SECURITY
TO ENHANCE SECURITY IN CLOUD COMPUTING AND ITS SERVICES
Applications Invited
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Prabakaran. D, Praveena N.G, Beulah Jackson, S. Varalakshmi, and G.Uma Maheswari
IEEE
Virtual Private Network (VPN) is the emerging technology facilitates the secure cloud connectivity and remote access, motivates the migration of physical resource of IT applications to the virtual resource. Despite of tremendous applications of virtual private network, the security remains as a major concern, and is highly prune to the varieties of attacks. The existing works attempts to address the security concern, resulting in unchanged status of the security in terms of enhancement. To overcome this concern, this manuscript introduces a Multi-Factor Authentication (MFA) using the low entropy password along with the graphical password authentication system. The security of the graphical password is enhanced using the novel ElGamal bakers map function to and a cipher graphical is generated to communicate over the insecure network. The proposed work is treated under various attack analysis using black box testing process and the robustness is measured.
Shyamala Ramachandran and Prabakaran D
IEEE
The growing prevalence of Internet of Things (IoT) devices being connected to the internet has resulted in an increase in security concerns, particularly concerning evasion attacks. Reinforcement learning is a category of machine learning in which an agent is taught to make decisions within an environment to maximize a reward signal. In this context, the agent is educated to differentiate network traffic data as either normal or evasive. An evasion attack falls into the category of adversarial attacks, where the attacker seeks to elude detection by exploiting vulnerabilities within the system. This research introduces a novel deep reinforcement learning method for identifying evasion attacks in IoT devices. The proposed design incorporates a neural network and a combination of machine learning classifiers to analyze network traffic data. The features has been extracted using Wireshark, and these features are then utilized as input for a reinforcement learning agent. This agent is trained to distinguish between normal and evasive traffic by combining the outputs of five different classifiers. To train our model, we use two publicly available datasets—one containing benign data and the other containing malicious data. The performance of our approach is evaluated using real-time data, and the results demonstrate that our proposed method surpasses other techniques in terms of accuracy, precision, recall, and F1-score. In conclusion, our deep reinforcement learning approach offers an effective means of detecting and mitigating the risks associated with evasion attacks on IoT devices.
S. Kannan, Prabakaran. D, Dhenesh Kumar. S, and Sivaram. S
IEEE
The demand for power is increasing daily due to an increase in industrial and household energy requirements and consumption. The non-renewable energy generation is considered to be dirty power as it requires oil, gas, coal and reactive substance like uranium for power generation. The non-renewable power generation pollutes the environment and is considered to be more harmful to people making the government and researchers to concentrate on the renewable method of power generation. Wind energy can be harvested round the clock as the wind will be available all time than the availability of sunlight which controls solar power. This paper proposed a novel deep learning-based convolution neural networks (CNN) for forecasting wind power generation one day ahead. The forecasting efficiency is compared with the benchmarking methodologies of Machine learning algorithms and the Root Mean Square Error (RMSE) of the proposed method is determined to be 0.629 which is considered to be efficient for the wind power generation process.
K. Suresh Kumar, D. Prabakaran, R. Senthil Kumaran, and I. Yamuna
Wiley
P Harineeshwari. and D Prabakaran.
IEEE
A wireless devices has an consecutive generation renovation to provide a higher and impressive service. The current 5G generation is the only one with the ability and power to communicate at extremely high speeds. Because they are now required for reliable 5G Wi-Fi connectivity, wide-band antennas with high gain have displaced traditional narrow bandwidth antennas with low gain. The next era, 6G, is projected to assist a large number of information site visitors in managing the increase in connection of Wi-Fi device. As a result, a significant amount of bandwidth will be aimed, and a greater benefit will be required. In order to satisfy the demand of information quotes and bandwidth utilization of conversation systems, maximum of the research studies results in the invent of new designs that fulfill the necessities of the future era verbal exchange. In this paper H-Shaped antenna that contains microstrip feed line is proposed. Among various substrates Polyimide dielectric substrate provides improved performance. The loss tangent and relative permittivity of the polyimide substrate are 3.5 and 0.0027, respectively. Computer software technology (CST) software is used to model the proposed system. The suggested H-shape antenna is shown, together with simulated results for gain, VSWR, return loss, directivity, and radiation pattern.
Gayathri Baskaran, K Saundariya, D Prabakaran, and R Senthilkumaran
IEEE
Web Development is evolving exponentially today at business purpose. Nowadays, there is an increasing need for handyman services and requires a simpler and more accessible way of storing and managing the required information about the Handyman website in the Admin panel and while using PHP/Python/Ajax, there is a blocking of data from the server-side and it takes multiple loading which makes the website with bad user interaction. This project proposed an administration panel for Household services. The purpose of this web design is to provide a non-blocking environment using Node JS to improve the quality of its working process. It is designed with React JS in front-end development, which makes faster response to the browser, SEO friendly. For security purposes, JWT is proposed in order to secure the data from unknown members by setting the Token. Also, a work analysis of the designed web application is presented and demonstrated in this paper.
K Saundariya, M Abirami, Kumaran R Senthil, D Prabakaran, B Srimathi, and G Nagarajan
IEEE
In recent days, there is a rapid increase in the need for handyman services around the world. If any issue is unfortunately encountered in the home, some issues may be hectic and people can’t be repaired on their own, people are busy with their schedule, hence they need workers to maintain and repair their homes. It is tough to find workers offline at the correct time and cost. Therefore, this online website makes it easier to book your own workers at the correct time and cost, it makes the workers available in just one click at your doorstep. Handyman workers have a separate login to showcase themselves by adding the works and skills they have. It also helps the professionals to gain opportunities and money based on their work. There are several categories and services, on the time of users’ login for the need for the services, the workers are listed based on location and cost with their name and contact. Creating a website using React JS makes it faster, boosts productivity, and is SEO friendly. MongoDB is a schema-less database and with ease of scale-out, hence it is easier to manage data. With this, a user can avoid delay and difficulty.
D. Prabakaran and S. Sriuppili
IOP Publishing
Abstract In this digital world, there are many applications to secure and legalize their data and all of these emissions by various techniques and there are many algorithms and methods to process their data. Some extensive method used is biometric authentication and voice recognition is better, since it paves the convenient manner to the user and it merely acquires the voice from the user. Also the background noise is in crisis with Mel Frequency Cepstrum Coefficient (MFCC) which is recognition algorithm where overcome by other tools such as smoothening filter etc. The main focus of this project is to investigate the feature extraction scheme.
Prabakaran. D and Shyamala Ramachandran
IEEE
Authentication is the crucial process of verifying and permitting the right user to access cloud services. The cloud service security proves to be fragile by the brute force efforts of hackers. To revamp the durability of authentication process, integrating the security factors has been exhorted. One such factor is the secure key extracted from the authenticated user’s voiceprint. This paper proposes a model for enhancing the signal to noise ratio of the MFCC algorithm to extract the secure key from the user voiceprint. The proposed method is highly rigid against the voice duplication attacks preventing unauthorized user to access by duplicating the authorized user’s voice print.
Prabakaran D and Sheela K
IEEE
The wireless sensor network is the emerging technology with spatially placed electronic sensors to measure the vital parameters. The necessity of monitoring the patient”s critical parameters of physical parameter., psychological activities drives to the development of wireless health care sensor network. The evolution of Wireless Healthcare Sensor Network (WHSN) had eased the monitoring process of patient round the clock. Despite of benefits leading the Wireless Healthcare Sensor Network to the peak of performance., certain vital parameters like network agility and security of the network still remains as a concern. The security aspect of the Wireless Healthcare Sensor Network (WHSN) is a biggest challenge., as the attackers tends to access the details of the patients illegally and may sometimes leads to masquerade the data being transferred. To overcome this concern and to provide a high level of security to the sensitive health care data., this paper proposes a novel methodology to provide rigid authentication using Elliptical Curve Cryptography (ECC). The ECC-based CPABE secured framework for WBAN without the operation of bilinear pairing. The result shows that there is reductions in cipher text size, reduces the computational cost, and also secure transferring of information.
D. Prabakaran and Shyamala Ramachandran
Computers, Materials and Continua (Tech Science Press)
The rise of the digital economy and the comfort of accessing by way of usermobile devices expedite human endeavors in financial transactions over the Virtual Private Network (VPN) backbone. This prominent application of VPN evades the hurdles involved in physical money exchange. The VPN acts as a gateway for the authorized user in accessing the banking server to provide mutual authentication between the user and the server. The security in the cloud authentication server remains vulnerable to the results of threat in JP Morgan Data breach in 2014, Capital One Data Breach in 2019, and manymore cloud server attacks over and over again. These attacks necessitate the demand for a strong framework for authentication to secure from any class of threat. This research paper, propose a framework with a base of Elliptical Curve Cryptography (ECC) to perform secure financial transactions throughVirtual PrivateNetwork (VPN) by implementing strongMulti-Factor Authentication (MFA)using authentication credentials and biometric identity. The research results prove that the proposed model is to be an ideal scheme for real-time implementation. The security analysis reports that the proposed model exhibits high level of security with a minimal response time of 12 s on an average of 1000 users.
D. Prabakaran and R. Shyamala
IEEE
In the digital era, the computing applications are to be secured from anonymous attacks by strengthening the authentication credentials. Numerous methodologies and algorithms have been proposed implementing human biometric as unique identity and one such identity is human voice print. The human voice print is a unique characteristic of the individual and has a wide variety of techniques in representing and extracting the features from the digital speech signals. The voice recognition techniques were executed on different platforms and exploit different mathematical tools in voice feature extraction, leading to dissimilarity in performance and results. In this paper, we investigate, analyze and present a review on performance of numerous voice recognition techniques.