Browsing by Author "Padmavathi, G"
Now showing 1 - 20 of 99
Results Per Page
Sort Options
- ItemAn Accurate Method for Deteetion of Cyber Attaeks(2013) Padmavathi, G; Uma, MDue to the rapid growth of wireless local area networks in day-to-day applications detection of cyber attacks have been increasing. In 2012, a survey stated that an organization experiences with an average of 102 attacks every week, whereas 72 attacks and 50 attacks were detected during 2011 and 2010, approxmiately 42% of attacks have increased in a year. In this paper, the dimensionality reduction techniques are used to detect the cyber attacks. The objective of this research work is to extract the meaningful class labels in the data set using linear dimensionality reduction techniques like Principal Components Analysis (PCA), Independent Component Analysis (ICA) and Linear Discriminant analysis (LDA). The experimental study shows that PCA performs better than LDA and ICA. To improve the obtained results the PCA is combined with SVM, where an accuracy rate increases in the detection of cyber attacks.
- ItemAdvanced Womens’ Safety Monitoring System on Android Mobile Platform(2015-03) Aparna, V S; Padmavathi, G
- ItemANODR-ECC Key Management protocol with TELNET to secure Application and Network layer for Mobile Adhoc Networks(2012) Padmavathi, G; Subashini, PA mobile ad hoc network (MANETs) is a self-organizing network that consists of mobile nodes that are connected through wireless media. A number of unique features, such as lack o f infrastructural or central administrative supports, dynamic network topologies, open communication channels, and limited device capabilities and bandwidths, have made secure, reliable and efficient routing operations in MANET a challenging task. The ultimate goal o f the security solutions for MANET is to provide security services, such as authentication, confidentiality, integrity, anonymity, and availability to mobile users. To achieve the goals, the security solution need for entire protocol stack.. The proposed protocol ANODRECC with Telnet provide application layer security and it ensures route anonymity and location privacy and is robust against eavesdropping attack. For route anonymity, it prevents strong adversaries from tracing a packet flow back 'to its source or destination; for location privacy, it ensures that adversaries cannot discover the real identities of local transmitters. The simulation is done using network simulator qualnet 5.0 for different number o f mobile nodes. The proposed model has exposed improved results in terms o f Average throughput. Average end to end delay. Average packet delivery ratio and Average jitter.
- ItemApplication of Machine learning in Detecting Insider Threat-State of Art and Survey(2013) Padmavathi, GResearch in computer security is more focused to prevent unauthorized and illegitimate access to systems and information. But, many times, the most damaging malicious activity is the result of internal misuse within an organization, which has not drawn much attention. Data Exfiltration refers to illegitimate transfer of data out of a given organization or network. Organizations employ security solutions like IDS, IPS and firewalls at the perimeter level to safe guard their network from external attacks. Insider attacks in the recent decade poses serious impact on the organization in terms of confidentiality and reputation. Machine learning algorithms and techniques has provided solutions to many of the complex real time problems in diversified fields and of great help in decision making and taking preventive and corrective measures. Many of soft computing techniques have been applied in intrusion detection in the recent years to detect and to prevent network intrusions by both external and internal attackers. This paper presents overview of insider threat, its current state of art in research, research challenges, data exfiltration steps, detailed on the machine learning approaches applied to address this problem.
- ItemApplying Computational Intelligence Techniques for Personal Authentication using Keystroke Dynamics(2013) Padmavathi, GABSTRACTUsername with password is the commonly used authentication mechanism for securing huge data transactions being carried out every day via Internet. Most of the text based authentication methods are vulnerable to many attacks as they depend on text and can be strengthened more by combining password with key typing manner of the user. Keystroke Dynamics is one of the inexpensive;and strong behavioral biometric technologies, which identifies the authenticity of a user when the user is working via a keyboard. The paper uses computational intelligence techniques such as Genetic algorithm. Ant Colony Optimization and Particle Swarm Optimization for feature subset selection and Back propagation Neural Network for classification. From the results, it is observed that ACO wrapped with BPN outperforms the other methods.
- ItemAn appraisal of Artificial Immune System(2013) Padmavathi, GArtificial Immune system is an important field of computational Intelligence, which models various algorithms and mechanisms. One of the Computational tools is Artificial Immune System, which emulates with the Biological Immune system to process and function. In this paper, a comparison is made between Natural Immune System and Artificial Immune System. Different Artificial Immune system algorithms are discussed along with recent applications in Artificial Immune System.
- ItemBiometric System Penetration in Resource Constrained Mobile Device(2013) Padmavathi, GOver the past few years, the usage of mobile devices to access data has becoming more frequent, and the usage of mobile devices in the applications such as web-browsing, email, multimedia, entertainment applications (games, videos, and audios), navigation, trading stocks, electronic purchase, banking and health care are increased, therefore data security is essential and also it becomes a challenge in securing the data in the mobile device. User authentication schemes such as Password or Personal Identllication Number (PIN) based authentication in mobile device is a difficult process for providing safe access to precious, private information or personalized services. To address these problems in the mobile devices, biometric system can be developed which are more secure, affordable and memorable authentication scheme based on graphical assistance, images and audio. We believe that biometric authentication is the most secure approach among other authenUcaUon mechanism. This paper discusses the various features of biometrics and mobile device security threats.
- ItemCharacteristics based Detection of Internet worms using Combined Machine Learning Methods and Worm Containment(2015-06) Divya, S; Padmavathi, G
- ItemClassification of Oil Spills using Computational Intelligence(2013) Padmavathi, GRough set theory is one of the useful cohiputati^ni^Ulhfelligence tools which can be used to deal with vague information and to take important decision in uheeftain domain. Rough set theory is useful for decision making in situation where indiscernibility is present. Rough sets do not require experienced knowledge engineers to provide dditional information about the niembership functigns for being processed. In this paper Rough Set method is used ar dimensionality reduction of oil spill feature space in SAR imagery. Reducing the number of features results in reduced feature space which improves the prediction, accuracy and minimizes the classification time.
- ItemClone Detection Using Enhanced EDD (EEDD) with Danger Theory in Mobile Wireless Sensor Network(2015) Padmavathi, GIn the recent years, Wireless Sensor Networks (WSN) are widely used to reap the benefits o f sensing capacity o f mobile nodes and its wireless mode o f communication. Sensor network architectures can be used for various applications that include intrusion detection, military patrols, border monitoring, etc. The security o f nodes in a sensor network is a challenging issue as they are easy targets for attacks. The node replication attack is one among those attacks. The adversary can capture security credentials from a node by compromising the node. The adversary can replicate the node using the same ID. To rectify this problem in a sensor network, various replica node detection techniques have been proposed. These techniques do not work well when multiple replicas are introduced in the neh\>ork against genuine nodes. Subsequently, the detection capability of the nehvork is degi-aded. To overcome these problems, this paper proposes Enhanced Efficient Distributed Detection (EEDD) algorithm. It combines the best features o f EDD and danger theory o f artificial immune system. This hybrid approach EEDD detects the clones. This method prevents multiple replicas in the WSN. The advantages o f the proposed method include (i) increased detection rate, (ii) decreased overheads, (Hi) high Packet Delivery Ratio (PDR) and (iv) low energy consumption. The proposed method is tested in a simulated environment.
- ItemComparison of RSA-Threshold Cryptography and ECC-Threshold Cryptography for Small Mobile Adhoc Networks(2012) Padmavathi, GA mobile ad hoc netw o rk is a special type o f wireless network in which a collection o f mobile hosts with wireless network interfaces may form a tem p o ra ry network. Without the aid o f p ro p e r fixed in fra stru c tu re , providing secure communications is a big challenge. The strength of the security solutions very much depends on the cryptographic keys used for communication. Efficient key management is an im p o rtan t req u irem en t of such networks. For networks like MANET which a re basically constrained networks with minimum resources, identification of suitable asymmetric cryptosystem is a vital one. Hence an a ttem p t has been made in this p ap e r to identity a suitable asymmetric-threshold based cryptosystems for small MANETs. The study focuses on the comparison o f Rivest Sh amir Adelman-Threshold Cry p to g rap h y and Elliptic Curve C ry p to g rap h y Threshold C ry p to g rap h y in terms o f the performance pa ram e te rs like key generation time. Encryption time. Decryption time and communication cost. Different small network scenarios with variable node sizes and key sizes a re experimented and the results show th a t ECC-TC is the most desirable asymmetric-threshold cryptosystem for small MANET.
- ItemA Comprehensive Approach to Mobile Device and Data Security using Iris Authentication, Malware Detection, Secured Cloud Storage and Retrieval(2016-06) Sujithra, M; Padmavathi, G
- ItemA COMPREHENSIVE STUDY ON CLASSIFICATION OF PASSIVE INTRUSION AND EXTRUSION DETECTION SYSTEM(2013) Padmavathi, GCyber criminals compromise Integriry. Availability and Confidentiality of network resources in cyber space and cause remote class intrusions such as U2R, R2L, DoS and probe/scan system attacks .To handle these intrusions, Cyber Security uses three audit and monitoring systems namely Intrusion Prevention Sy.stems (IPS), Intrusion Detection Systems (IDS). Intrusion Detection System (IDS) monitors only inbound traffic which is insufficient to prevent botnet systems. A system to monitor outbound traffic is named as Extrusion Detection System (EDS). Therefore a hybrid system should he designed to handle both inbound and outbound traffic. Due to the increased false alarms preventive systems do not suite to an organizational network. The goal o f this paper is to devise a taxonomy for cyber security and study the existing methods of Intrusion and Extrusion Detection systems based on three primary characteristics. The metrics used to evaluate IDS and EDS are also presented.
- ItemComputational Intelligence (Cl) - Techniques and Applications(2013) Padmavathi, Gomputational Intelligence is the most promising technique which has been introduced in every Field for optimized results. Computational Intelligence is a set of processing technique which is capable of handling the real world problems. The aim of this review paper is to analyze computational intelligence feature, various branches of it and detailed study about its wide range of applications is analyzed. Applications of Computational Intelligence discussed in this paper itself shows versatility of computational intelligence in this modern society. Computational intelligence plays major role in the field of research because of its capability of adopting the real world problems and which produces the optimized results.
- ItemComputational Intelligence (Cl) - Techniques and Applications(2013) Uma, M; Padmavathi, GComputational Intelligence is the most promising technique which has been introduced in every field for optimized results. Computational Intelligence is a set of processing technique which is capable of handling the real world problems. The aim of this review paper is to analyze computational intelligence feature, various branches of it and detailed study about its wide range of applications is analyzed. Applications of Computational Intelligence discussed in this paper itself shows versatility of computational intelligence in this modern society. Computational intelligence plays major role in the field of research because of its capability of adopting the real world problems and which produces the optimized results.
- ItemComputational Intelligence Techniques in Intrusion Detection System for Critical Information Infrastructure Protection(2013) Padmavathi, GCritical National Infrastructures (CNI) are important resources of the nation to be proteeted. Since all the operations are eomputer based the security of CNI are wholly dependent on security of Information systems. Many protection systems such as firewallsl, cryptographic systems, Intrusion Detection Systems exists to provide Critical Information Infrastructure Protection (CUP). Among them IDS play a signifieant role in analyzing and protecting huge volume of network traffic. Computational Intelligence Techniques can be implemented in IDS systems to handle complex real world threats to CNI. Some of the Computational Intelligence Techniques that ean be applied to each layer of IDS are discussed.
- ItemComputer Network Worms Propagation and its Defence Mechanisms: A Survey(2014) Padmavathi, GInformation security is one of the major concerns for military, government, civil and commercial organizations and security risk has been immensely raised on the internet access. Self-duplicating, self-propagating malicious codes knoivn as worms spread themselves without any human interaction and launch the most destructive attacks against networks and cause high security risks. Increasing threats from worms in the network continue to be a challenging task to detect and handle. Various worms exist in network affecting the communication security and different worm handling mechanisms exists to handle the threat created by them. Considering the features and metrics of the existing mechanisms in the detection and contaiiiineiit of worms in the network, this paper surveys the classincation of worms and several existing defence mechanism and metrics to detect those worm attacks in the network.
- ItemCredit Card Fraud Detection Using Specific Machine Learning Methods(2018-04) Akshaya, S; Padmavathi, G
- ItemDecentralized Group Key Management Scheme For Secure Multicast Communication(2016-04) Arthi, K; Padmavathi, G
- ItemDeploying Simulink Models Into System-On-Chip Structures For Object Detection Using Image Processing Techniques(2011-04) Karunambika, S; Padmavathi, G