Malleability Resilient Concealed Data Aggregation

Concealed data aggregation protects against passive attackers and ensures privacy of sensor readings at intermediate nodes. However, the use of inherently malleable privacy homomorphism makes it susceptible to active attackers. In addition, it is a well-known fact that encrypted data processing is vulnerable to pollution attacks where a single malicious node can flood the network by fake readings. Hence, there exists a need to authenticate the processed readings. Traditional authentication mechanisms are not viable due to the conflicting requirements like in-network processing and encrypted data processing. The need for en route aggregation of sensor readings, the need for encrypted data processing and the need for message authentication both at the base station and at aggregator nodes, make message authentication a formidable challenge. Homomorphic Message Authentication Codes (H-MACs) help to verify the integrity of processed sensor readings. However, the need to verify the integrity of sensor readings both at intermediate node(s) and at the base station cannot be realized simultaneously through the currently available techniques. In this paper, we combine the benefits of privacy homomorphism and H-MACs to provide malleability resilient concealed data aggregation in the presence of both insider and outsider adversaries. As per our knowledge, our solution is the first to achieve integrity protecting concealed data aggregation in the presence of both insider and outsider adversaries.