Per node deployment based detection of controlled link establishment attack in distributed sensor networks

Publisher:
Inderscience Enterprises Limited
Publication Type:
Journal Article
Citation:
International Journal of Sensor Networks, 2011, 9 (3-4), pp. 192 - 208
Issue Date:
2011-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2009006981OK.pdfPublished Version423.23 kB
Adobe PDF
The great majority of processes in distributed sensor networks, such as data aggregation methods, routing protocols, distributed voting schemes, misbehaviour detection systems, and so on, can only be accomplished via collaborative efforts of sensor nodes. However, the success of these processes is undermined by an attack termed controlled link establishment attack whose goal is to gain significant portion or even full control of distributed sensor networks through the controlled link establishment. This attack can be exemplified by node replication attack and key-swapping collusion attack based on their final attack goal. Thus countermeasures against the two latter attacks can be utilised as remedies for the controlled link establishment attack. Despite the fact that a growing body of such countermeasures has been proposed over recent years, each of them exposes its own limitations such as high performance overheads, unsound assumptions, and security weaknesses. Therefore, we propose a series of evolutionary schemes comprising naïve, adaptive, and extended schemes in this paper to overcome those limitations. The first two schemes are designed to be light-weight in performance at the cost of slightly weaker security robustness while the extended scheme obtains much greater security by trading off small performance. Theoretical analyses, simulations, and extensive comparison with other schemes have been conducted to demonstrate the plausibility of our schemes with respect to security features and performance overheads.
Please use this identifier to cite or link to this item: