Data Randomization for Lightweight Secure Data Aggregation in Sensor Network

  • Authors:
  • Abedelaziz Mohaisen;Ik Rae Jeong;Dowon Hong;Nam-Su Jho;Daehun Nyang

  • Affiliations:
  • Electronics and Telecommunication Research Institute, , Daejeon, Korea 305-700;The Graduate School of Information Security, Korea University, Seoul, Korea;Electronics and Telecommunication Research Institute, , Daejeon, Korea 305-700;Electronics and Telecommunication Research Institute, , Daejeon, Korea 305-700;Information Security Research Laboratory, Inha University, Incheon, Korea

  • Venue:
  • UIC '08 Proceedings of the 5th international conference on Ubiquitous Intelligence and Computing
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Data aggregation is one of the main purposes for which sensor networks are developed. However, to secure the data aggregation schemes, several security-related issues have raised including the need for efficient implementations of cryptographic algorithms, secure key management schemes' design and many others. Several works has been introduced in this direction and succeeded to some extent in providing relatively efficient solutions. Yet, one of the questions to be answered is that, can we still aggregate the sensed data with less security-related computation while maintaining a marginal level of security and accuracy? In this paper, we consider data randomization as a possible approach for data aggregation. Since the individual single sensed record is not of a big concern when using data for aggregation, we show how data randomization can explicitly hide the exact single data records to securely exchange them between nodes. To improve the security and accuracy of this approach, we introduce a hybrid scheme that uses the cryptographic approach for a fraction of nodes. We study the efficiency of our schemes in terms of the estimate accuracy and the overhead.