Abstract

Objective: To provide Secure Aggregation against Collusion attacks and Malicious or compromised aggregator.

Statistical Analysis: Wireless Sensor Network incorporate aggregation of data from multiple sensor nodes performed at the aggregator node or the cluster head due to limited computational power and energy. This aggregation technique is vulnerable to attacks of compromised nodes. Thus a secure Data Aggregation for Collusion Attacks (SACA) was proposed earlier that improved the previously stated IF algorithms by providing initial approximation of trustworthiness of the sensor nodes. This made the algorithm to be more accurate and faster.

Findings: The problem in this approach is that these improvisations were made with the assumption that the aggregator is not compromised. So in case if the aggregator node is compromised this method stands pointless in determining the security. Thus a new framework called Secure aggregation against Collusion Attacks and compromised aggregator (SACACA) is proposed to ensure that the Secure Date Aggregation scheme also fetch protection over the compromised aggregator node. The proposed work initializes with an aggregator node, that estimates the error and noise of the other sensor nodes in the cluster, and then calculates the reputation vector of each node and provides the information of trustworthiness of each node to the enhanced IF algorithm. The aggregator sends the aggregated information to the base station directly or to the aggregator of another cluster's aggregator and then reaches to the base station. This procedure is repeated by the replacement of the aggregator by a node in the cluster as the next aggregator node once in a time period. Thus the path of the aggregated information to the base station may vary according to the selection of the aggregator in each cluster. The Variance of each aggregator is calculated in the Base Station. The Calculated values are compared to a threshold value that determines whether the aggregator node is compromised or not. In order to find the variance value the enhanced IFs algorithm is utilized.

Improvements: Thus the proposed method improves the effectiveness of the enhanced IF algorithm over the compromised aggregator nodes. The change of aggregator in the network also saves a significant amount of computational power and energy.

Keywords

Full Text