Dr.J. Premalatha
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Dr.J. Premalatha has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Vani Rajasekar
Vellore Institute of Technology
6 shared publications K. Sathya
5 shared publicationsK Sathya
Kongu Engineering College,Department of CT-UG,Erode,India
2 shared publications- K. Loheswaran
Sasurie College of Engineering, Tirupur District, Pallagoundenpalayam, India
2 shared publications Aswini Balasundaram
UG student, Department of Information Technology, Kongu Engineering College, Erode-638060, Tamil Nadu, India
1 shared publicationKavi Priya Dharshini Seturamalingam
UG student, Department of Information Technology, Kongu Engineering College, Erode-638060, Tamil Nadu, India
1 shared publication
Latest Publications
- 1 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 0766-0769
Reinforcing Cyber World Security with Deep Learning Approaches
In the past decade, the Machine Learning (ML) and Deep learning (DL) has produced much research interest in the society and attracted them. Now-a-days, the Internet and social life make a lead in most of their life but it has serious social threats. It is a challenging thing to protect the sensitive information, data network and the computers which are in unauthorized cyber-attacks. For protecting the data’s we need the cyber security. For these problems, the recent technologies of Deep learning and Machine Learning are integrated with the cyber-attacks to provide the solution for the problems. This paper gives a synopsis of utilizing deep learning to enhance the security of cyber world and various challenges in integrating deep learning into cyber security are analyzed.
- 1 July 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 0766-0769
Reinforcing Cyber World Security with Deep Learning Approaches
In the past decade, the Machine Learning (ML) and Deep learning (DL) has produced much research interest in the society and attracted them. Now-a-days, the Internet and social life make a lead in most of their life but it has serious social threats. It is a challenging thing to protect the sensitive information, data network and the computers which are in unauthorized cyber-attacks. For protecting the data’s we need the cyber security. For these problems, the recent technologies of Deep learning and Machine Learning are integrated with the cyber-attacks to provide the solution for the problems. This paper gives a synopsis of utilizing deep learning to enhance the security of cyber world and various challenges in integrating deep learning into cyber security are analyzed.
- 22 May 2020
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2240 (1) , 140006
Performance analysis of machine learning techniques for the prediction of breast cancer in big data environment
Every year more than a million women are diagnosed with breast cancer that results in the death of more than half of them, due to the delay in diagnosis of the disease. Machine Learning (ML) offers a better approach for the prediction of breast cancer. A feature selection technique, INTERACT is applied to select relevant features for breast cancer diagnosis. The tool used in this project is WEKA 2.3. The classification algorithms, namely Decision tree, Random forest and Support Vector Machine (SVM) are incorporated along with INTERACT technique. The relevant and important features for breast cancer diagnosis are chosen. The dataset used in this project is Wisconsin Diagnostic Breast Cancer (WDBC) subdirectory. The experimental result shows that the SVM outperforms the other classifiers. It improved accuracy of the diagnostic model by using feature selection method.
- 22 May 2020
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2240 (1) , 140006
Performance analysis of machine learning techniques for the prediction of breast cancer in big data environment
Every year more than a million women are diagnosed with breast cancer that results in the death of more than half of them, due to the delay in diagnosis of the disease. Machine Learning (ML) offers a better approach for the prediction of breast cancer. A feature selection technique, INTERACT is applied to select relevant features for breast cancer diagnosis. The tool used in this project is WEKA 2.3. The classification algorithms, namely Decision tree, Random forest and Support Vector Machine (SVM) are incorporated along with INTERACT technique. The relevant and important features for breast cancer diagnosis are chosen. The dataset used in this project is Wisconsin Diagnostic Breast Cancer (WDBC) subdirectory. The experimental result shows that the SVM outperforms the other classifiers. It improved accuracy of the diagnostic model by using feature selection method.
- 30 September 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (3) , 1854-1857
Sensor-Seeded Cryptographically Secure Random Number Generation
Random numbers are essential to generate secret keys, initialization vector, one-time pads, sequence number for packets in network and many other applications. Though there are many Pseudo Random Number Generators available they are not suitable for highly secure applications that require high quality randomness. This paper proposes a cryptographically secure pseudorandom number generator with its entropy source from sensor housed on mobile devices. The sensor data are processed in 3-step approach to generate random sequence which in turn fed to Advanced Encryption Standard algorithm as random key to generate cryptographically secure random numbers.
- 30 September 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (3) , 1854-1857
Sensor-Seeded Cryptographically Secure Random Number Generation
Random numbers are essential to generate secret keys, initialization vector, one-time pads, sequence number for packets in network and many other applications. Though there are many Pseudo Random Number Generators available they are not suitable for highly secure applications that require high quality randomness. This paper proposes a cryptographically secure pseudorandom number generator with its entropy source from sensor housed on mobile devices. The sensor data are processed in 3-step approach to generate random sequence which in turn fed to Advanced Encryption Standard algorithm as random key to generate cryptographically secure random numbers.
- 30 September 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (3) , 1593-1598
An Efficient Signcryption Scheme for Secure Authentication using Hyper Elliptic Curve Cryptography and Keccak Hashing
The need for security is a challenging task nowadays due to the transition from wired to wireless networks, emergence of Internet of Things (IoT), Mobile Networks, Wireless Sensor Networks (WSN) and Radio Frequency Identification System (RFID). Generally wireless systems are prone to insecurity and resource (power) constraint, to deal with these challenges many solutions has been proposed in cryptography. One such important development is light weight cryptography particularly signcryption.Signcryption is a logical combination of digital signature and encryption in a single step therefore the cost of communication and computation is very less compared to the existing signature then encryption scheme. Till date many signcryption techniques were raised based on El-Gamal, RSA and Elliptic Curve Cryptography (ECC). The proposed research work highlights the limitations of existing signcryption based on ECC and it proposes the efficient light weight cryptographic scheme of signcryption based on Hyper Elliptic Curve Cryptography (HECC) and Keccak hashing. Further the proposed research work achieves all the security metrics such as confidentiality, integrity, non- repudiation, forward secrecy and public verifiability.
- 30 September 2019
- Published by Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication (BEIESP) in International Journal of Recent Technology and Engineering (IJRTE)
- Vol. 8 (3) , 1593-1598
An Efficient Signcryption Scheme for Secure Authentication using Hyper Elliptic Curve Cryptography and Keccak Hashing
The need for security is a challenging task nowadays due to the transition from wired to wireless networks, emergence of Internet of Things (IoT), Mobile Networks, Wireless Sensor Networks (WSN) and Radio Frequency Identification System (RFID). Generally wireless systems are prone to insecurity and resource (power) constraint, to deal with these challenges many solutions has been proposed in cryptography. One such important development is light weight cryptography particularly signcryption.Signcryption is a logical combination of digital signature and encryption in a single step therefore the cost of communication and computation is very less compared to the existing signature then encryption scheme. Till date many signcryption techniques were raised based on El-Gamal, RSA and Elliptic Curve Cryptography (ECC). The proposed research work highlights the limitations of existing signcryption based on ECC and it proposes the efficient light weight cryptographic scheme of signcryption based on Hyper Elliptic Curve Cryptography (HECC) and Keccak hashing. Further the proposed research work achieves all the security metrics such as confidentiality, integrity, non- repudiation, forward secrecy and public verifiability.
- 4 September 2017
- Published by Springer Nature in Cluster Computing
- Vol. 22 (S5) , 10615-10623
Intrusion detection of distributed denial of service attack in cloud
- 4 September 2017
- Published by Springer Nature in Cluster Computing
- Vol. 22 (S5) , 10615-10623
Intrusion detection of distributed denial of service attack in cloud
- 1 February 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
A hybrid GA-IWO scheduling algorithm
Cloud computing is the latest need of the hour network which is convenient. It is an on-demand network accessed over a shared group of computing resources. Only minimal management effort is needed to be released with no communication with a service-provider. Optimization problem such as Invasive Weed Optimization is used in cloud computing but the challenges in this technique is its premature convergence and cannot achieve global optimum, specifically for multimodal issues. In the case of Genetic Algorithm (GA), it does not always come with global optimum always, mainly the population is varied when overall solution is required. GA is a technique which is complex to be understood. To overcome these drawbacks Genetic Algorithm-Invasive Weed Optimization (GA-IWO) method is proposed. Less computation time is required in GA-IWO and it is easy to implement on embedded systems and this makes the proposed techniques beneficial in the case of real-time decision making situations. Results show that better performance can be achieved through GA-IWO.
- 1 February 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
A hybrid GA-IWO scheduling algorithm
Cloud computing is the latest need of the hour network which is convenient. It is an on-demand network accessed over a shared group of computing resources. Only minimal management effort is needed to be released with no communication with a service-provider. Optimization problem such as Invasive Weed Optimization is used in cloud computing but the challenges in this technique is its premature convergence and cannot achieve global optimum, specifically for multimodal issues. In the case of Genetic Algorithm (GA), it does not always come with global optimum always, mainly the population is varied when overall solution is required. GA is a technique which is complex to be understood. To overcome these drawbacks Genetic Algorithm-Invasive Weed Optimization (GA-IWO) method is proposed. Less computation time is required in GA-IWO and it is easy to implement on embedded systems and this makes the proposed techniques beneficial in the case of real-time decision making situations. Results show that better performance can be achieved through GA-IWO.
- 1 January 2017
- Published by Inderscience Publishers in International Journal of Business Intelligence and Data Mining
- Vol. 12 (2/3)
Performance Amelioration of Manets Using cooperative Routing with Cross-Layer Design
International publishers of academic, scientific and professional journals since 1979.
- 1 January 2017
- Published by Inderscience Publishers in International Journal of Business Intelligence and Data Mining
- Vol. 12 (2/3)
Performance Amelioration of Manets Using cooperative Routing with Cross-Layer Design
International publishers of academic, scientific and professional journals since 1979.
- 19 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Biometric signcryption using hyperelliptic curve and cryptographically secure random number
The biometric system stores physiological traits of user such as finger print, iris, palm, patterns of veins etc. Iris signcryption is the most successful way to improve the information security because iris consists of most accurate features compared to other authentication systems. In order to provide better security and less computational time lightweight cryptographic scheme called signcryption has introduced. Signcryption is a logical combination of encryption and signature. Cryptographically secure random numbers are used as private key in encryption segment and as key in generating signature of SHA-1 segment. Generally the random numbers passing the NIST Test suite fails to satisfy cryptographically secure requirements. An approach called twining technique along with AES in counter mode is used to generate cryptographically random numbers from the seed values of sensors. Twining technique is a novel method to minimize co linearity that occurs while extracting seed from multiple sensors housed in same environment. One of the most recent researches in the area of public key cryptography is HyperElliptic Curve Cryptography (HECC) because HECC uses smaller key size and less computational time. In the proposed biometric signcryption, biometric data, HECC and cryptographically secure random numbers are used to derive the signcrypted value of iris.
- 19 September 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Biometric signcryption using hyperelliptic curve and cryptographically secure random number
The biometric system stores physiological traits of user such as finger print, iris, palm, patterns of veins etc. Iris signcryption is the most successful way to improve the information security because iris consists of most accurate features compared to other authentication systems. In order to provide better security and less computational time lightweight cryptographic scheme called signcryption has introduced. Signcryption is a logical combination of encryption and signature. Cryptographically secure random numbers are used as private key in encryption segment and as key in generating signature of SHA-1 segment. Generally the random numbers passing the NIST Test suite fails to satisfy cryptographically secure requirements. An approach called twining technique along with AES in counter mode is used to generate cryptographically random numbers from the seed values of sensors. Twining technique is a novel method to minimize co linearity that occurs while extracting seed from multiple sensors housed in same environment. One of the most recent researches in the area of public key cryptography is HyperElliptic Curve Cryptography (HECC) because HECC uses smaller key size and less computational time. In the proposed biometric signcryption, biometric data, HECC and cryptographically secure random numbers are used to derive the signcrypted value of iris.
- 30 August 2016
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 23 (3) , 501-507
Region based find and spray scheme for co-operative data communication in vehicular cyber-physical systems
- 30 August 2016
- Published by Tech Science Press in Intelligent Automation and Soft Computing
- Vol. 23 (3) , 501-507
Region based find and spray scheme for co-operative data communication in vehicular cyber-physical systems