U. Mahalakshmi ¹, V. S. Shankar Sriram ²

Affiliations
1 School of Computing (CSE), SASTRA University, 613401, Thanjavur, Tamilnadu, IN
2 School of Computing (IT), SASTRA University, 613401, Thanjavur, Tamilnadu, IN

Abstract

Authentication schemes employing multibiometric system begin with several identification and verification based security schemes. Current systems implementing multi biometric scheme remain unsatisfactory facing to the growing security needs. The proposed system provides secured authentication to integrate multi biometric system with Elliptic Curve Cryptography that employs two modalities (i) Fusion of fingerprint, face, iris and signature. (ii) Generating elliptic curve and key using Elliptic Curve Cryptography. In the proposed system, selected portion of multi biometric features are fused into a single image and a curve is generated using ECC based technique that employs secured domain parameters generated through Genetic Algorithm. A one time password is also appended to the system to afford high authentication. The proposed algorithm is highly efficient against False Acceptance Rate and False Rejection Rate.

Keywords

Multibiometric, Elliptic Curve Cryptography, One Time Password, Authentication

Full Text

   

Marc del Valle ¹, B. Manikandan ², V. S. Shankar Sriram ²

Affiliations
1 ETSETB, Universitat Politècnica de Catalunya, Barcelona, ES
2 School of Computing, SASTRA University, Thanjavur, IN

Abstract

Femtocells have been proposed as solution for high-speed mobile network (i.e. LTE) bandwidth requirements. Research work in the field of securing femtocells is still in its infant stage. In this research contribution an attempt has been done to identify and mitigate a possible attack in the femtocells where locational information about a femtocell user is disclosed. The proposed mechanism notifies the femto entity under threat about the attack. Also a novel Multi-hop algorithm has been proposed to hide the details of the communicating parties from the attacker.

Keywords

Femtocell, Mobile Cellular Network, Mobile User Location, Multi-hops Algorithm, Security, Small Cells

Full Text

   

Nivethitha Somu ¹, A. Gangaa ¹, V. S. Shankar Sriram ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur, Tamil Nadu, IN

Abstract

“Big Data” - voluminous and variety of data from different sources, which demands innovative processing and analysis for decision - making analysis. The data can be either in form of structured or unstructured data. Processing big data with the traditional processing tools and the present relational database management systems tends to be a difficult task. Parallel execution environment, like Hadoop is needed for processing voluminous data. For processing the data in a open framework like Hadoop we need a highly secure authentication system for restricting the access to the confidential business data that are processed. In this paper, a novel and a simple authentication model using one time pad algorithm that removes the communication of passwords between the servers is proposed. This model tends to enhance the security in Hadoop environment.

Keywords

Big Data, Hadoop And one Time Padhow

Full Text

   

A. Gangaa ¹, Nivethitha Somu ¹, V. S. Shankar Sriram ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613401, Tamil Nadu, IN

Abstract

The importance of electronic data and its rapid growth leads to gradual increase to the data usage over the network servers and on the other side high priority is needed for the data security and privacy. The existing keyword search schemes for data retrieval from the centralized storage uses keyword as a valid parameter. Even though, keywords are encrypted and then used, intruders are able to guess keywords and can verify it using offline dictionary attack. The ability of guessing keywords by intruders reveals the stored data which affects data security and privacy. We need an additional parameter to the keyword used for search which will make the keyword guessing attack impossible. The additional parameter we used in our encryption framework is a signature file whose content needs to be independent of data. The content of signature file can be any data such as image, table, text, etc. which depends on the choice of user. This signature is hashed with keyword which makes the adversary difficult to guess the content of signature of the user. The existing approaches used keyword along with private key as search parameters, the guessing of appropriate keyword reveals the private key of user. In our framework, addition of signature along with keyword ensures both authentication as well as data privacy.

Keywords

Data Privacy, Data Retrieval, Keyword Guessing Attack, Keyword Privacy, Keyword Search Scheme, Signature Hashing

Full Text

   

J. Bharath ¹, V. S. Shankar Sriram ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613401, Tamil Nadu, IN

Abstract

Objectives: Cloud computing is a virtualized and scalable platform which helps the users to reduce the cost incurred for setting up and maintaining an IT infrastructure. Despite the various benefits offered by cloud, it faces stringent challenges with respect to security and trust management. Trust covers the security aspects of cloud. The proposed system focuses on selection of optimal parameters used to ensure trust value. Methods: The hybridized techniques for cloud Trust Management works with a pre defined set of parameter values. Hence the trust value computed using these optimal parameters has a great impact on the overall accuracy of the trust score. We intend to use Genetically Modified Ant Colony Optimization GM-ACO technique to identify best Trust Metric Parameters (TMPs) with respect to Cloud Service Providers and it significantly outperforms compared with existing techniques. Findings: The genetically modified ACO algorithm will reduce the complexity of calculating the trust score of many service providers in the cloud environment. Applications: Managing trust in peer-to-peer systems, social network based systems, recommendation based systems, policy based systems, reputation based trust systems, trust mining systems etc.

Keywords

Ant Colony Optimization, Cloud Computing, Genetic Algorithm, Trust Evaluation, Trust Management.

Full Text

   

Arivazhagan Nagarajan ¹, V. S. Shankar Sriram ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613401, Tamil Nadu, IN

Abstract

Objectives: An imposter identification system allows authenticated user to request for regeneration of password while restricting users who possess imposter properties. Existing system in the literature either allows the users to request for regeneration of password or denies access to the users. So the imposter acceptance and genuine rejection percentage is high in the existing system. Methods: The proposed system improves the performance of imposter identification system by adding three levels of non biometric factors to it. Fuzzy score is calculated on each level and the scores are accumulated with its own weight at last level. Using weighted score the threshold is compared and evaluated to make further decisions. Based on the evaluations made with respect to the threshold, the user will be allowed or denied to regenerate the password. Neural Network is applied to find out the threshold value whereas initially the threshold value is randomly assigned. The neural system works on two modes; one is learning mode in which the developer has to validate the system using training and testing dataset. Findings: With the validation, the threshold value is set to the most accurate value. Secondly, the system works on production mode where the end user uses the system. The proposed system increases the performance of imposter identification algorithm by increasing the rate of genuine user acceptance and imposter rejection. Applications: For all, ID-Password based authentication applications which supports Forget password as a sub module for the authentication module.

Keywords

Bio-metric, Imposter, Neural System.

Full Text