TITLE:
Investigating Approaches of Data Integrity Preservation for Secure Data Aggregation in Wireless Sensor Networks
AUTHORS:
Vivaksha Jariwala, Vishal Singh, Prafulla Kumar, Devesh C. Jinwala
KEYWORDS:
In-Network Processing; Integrity; Message Authentication Code; Secure Data Aggregation; Wireless Sensor Networks
JOURNAL NAME:
Journal of Information Security,
Vol.5 No.1,
January
10,
2014
ABSTRACT:
Wireless Sensor Networks (WSNs) typically use in-network processing to
reduce the communication overhead. Due to the fusion of data items sourced at
different nodes into a single one during in-network processing, the sanctity of
the aggregated data needs to be ensured. Especially, the data integrity of the
aggregated result is critical as any malicious update to it can jeopardize not
one, but many sensor readings. In this paper, we analyse three different
approaches to providing integrity support for SDA in WSNs. The first one is
traditional MAC, in which each leaf node and intermediate node share a key with
parent (symmetric key). The second is aggregate MAC (AMAC), in which a base
station shares a unique key with all the other sensor nodes. The third is homomorphic
MAC (Homo MAC) that is purely symmetric key-based approach. These approaches
exhibit diverse trade-off in resource consumption and security assumptions.
Adding together to that, we also propose a probabilistic and improved variant
of homomorphic MAC that improves the security strength for secure data aggregation
in WSNs. We carry out simulations in TinyOS environment to experimentally
evaluate the impact of each of these on the resource consumption in WSNs.