A Traitor Identification Technique for Numeric Relational Databases with Distortion Minimization and Collusion Avoidance

A Traitor Identification Technique for Numeric Relational Databases with Distortion Minimization and Collusion Avoidance

Arti Arun Mohanpurkar (Dr. D Y Patil School of Engineering &Technology, Lohgaon, Pune, Maharashtra, India) and Madhuri Satish Joshi (MGM's Jawaharlal Nehru Engineering College, Aurangabad, Maharashtra, India)
Copyright: © 2016 |Pages: 24
DOI: 10.4018/IJACI.2016070106
OnDemand PDF Download:
No Current Special Offers


An enormous growth in internet usage has resulted into great amounts of digital data to handle. Data sharing has become significant and unavoidable. Data owners want the data to be secured and perennially available. Data protection and any violations thereby become crucial. This work proposes a traitor identification system which securely embeds the fingerprint to provide protection for numeric relational databases. Digital data of numeric nature calls for preservation of usability. It needs to be done so by achieving minimum distortion. The proposed insertion technique with reduced time complexity ensures that the fingerprint inserted in the form of an optimized error leads to minimum distortion. Collusion attack is an integral part of fingerprinting and a provision to mitigate by avoiding the same is suggested. Robustness of the system against several attacks like tuple insertion, tuple deletion etc. is also depicted.
Article Preview

1. Introduction

“The great Indian sage Vyasa claims to own the copyright on world’s wisdom; that makes all of us traitors”. Then who else can understand a traitor better than a traitor! With the developments in internet, database applications and remote access techniques, the demand that lots of databases on the internet to be permitted to remote query and access, for authorized users has become common, and the problem to be able to protect the copyright of relational databases has arisen. Although this trend is a blessing to end users, it exposes the data owners to the threat of loss of ownership. Providers are therefore keen about demanding technology which facilitates identification of piracy and the traitors of their databases.

In this era of ambient computing, huge datasets are found to be continuously generated by the sensors. The mobile devices or smart systems are supposed to show intelligence and respond based on the values of the parameters sensed by the sensors. Sensors continuously generate data which is numeric in nature which may be further notified in different ways like alarms or automatic switch on/off or by display of some messages/warnings etc. The behaviour of parameters being monitored through sensors may have to be analysed for correct decision making to provide smart solutions to the real time problems recognized by the smart systems. To be able to analyse the sensor generated data, it may have to be shared with several data mining experts who may mine the data, provide an analysis and further an optimal solution to the problem. The data mining tools are extensively employed to extract useful patterns from these datasets, and this requires sharing of datasets on the internet by data owners with the mining experts.

Another area where proving ownership and traitor identification is required is the Electronic Medical Record (EMR) Systems being employed in the hospitals. EMRs when outsourced or shared for opinion of others (experts, researchers, co-workers) have a threat of a traitor trying to prove his ownership or selling it illicitly to a third party. Thus use of EMRs has challenged the security issues. Copyright laws exist for such kind of digital assets which act against the culprits, but beforehand it is important to be able to prove ownership on that database and also be able to find which of the buyers (accuse at least one) has fraudulently sold it to the innocent buyer. Thus arises the need for some techniques like watermarking (Agrawal, Haas, & Kiernan, 2003) (ownership protection) and fingerprinting (for traitor tracing) to identify the owner of the data as well as the malicious owner/s (traitor/s) who illicitly redistributes the data, respectively.

Watermarking is a class of information hiding technique where owner specific marks are inserted with an intention to be able to prove ownership. The identification of such a traitor and proof of ownership can be obtained by a popularly known technique called fingerprinting. Fingerprinting is nothing but another class of information hiding where buyer distinctive marks are inserted along with owners’ identification to prove ownership and be able to identify traitors. Fingerprinting (Li, Swarup, & Jajodia, 2005) is a technique used to deter illegal redistribution of valued digital information like documents, images, audio, video, databases etc. An important threat specific to fingerprinting is collusion attack along with other attacks like tuple insertion, deletion, mix and match, additive attack etc.

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 13: 6 Issues (2022): 1 Released, 5 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing