Concealed Data Aggregation in Heterogeneous Sensor Networks using Privacy Homomorphism

Data aggregation is implemented in wireless sensor networks to reduce data redundancy and to summarize relevant and necessary information without requiring all pieces of the data. The benefit of data aggregation can be maximized by implementing it at every data aggregator on the path to the base station. However, data confidentiality requires sensor nodes to encrypt their data prior to transmission. Moreover, once data is encrypted by a sensor node, it should be decrypted at the base station to maintain end-to-end security. This makes the implementation of data aggregation very difficult because data aggregation algorithms require encrypted data to be decrypted. Consequently, data aggregation and secure communication have conflicts in their implementation. To achieve data aggregation and secure communication together, this paper employs privacy homomorphism which offers end-to-end concealment of data and ability to operate on ciphertexts. In the proposed protocol, the computational overhead imposed by the privacy homomorphic encryption functions is tolerated by employing a set of powerful nodes, called AGGNODEs.