Energy Efficient Clustering and Secure Data Aggregation in Wireless Sensor Networks

Communication consumes the majority of a wireless sensor network's limited energy. There are several ways to reduce the communication cost. Two approaches used in this work are clustering and in-network aggregation. The choice of a cluster head within each cluster is important because cluster heads use additional energy for their responsibilities and that burden needs to be carefully distributed. We introduce the energy constrained minimum dominating set (ECDS) to model the problem of optimally choosing cluster heads in the presence of energy constraints. We show its applicability to sensor networks and give an approximation algorithm of O(log n) for solving the ECDS problem. We propose a distributed algorithm for the constrained dominating set which runs in O(log n log ?) rounds with high probability. We show experimentally that the distributed algorithm performs well in terms of energy us- age, node lifetime, and clustering time and thus is very suitable for wireless sensor networks. Using aggregation in wireless sensor networks is another way to reduce the overall communication cost. However, changes in security are necessary when in- network aggregation is applied. Traditional end-to-end security is not suitable for use with in-network aggregation. A corrupted sensor has access to the intermediate data and can falsify results. Additively homomorphic encryption allows for aggregation of encrypted values, with the result being the same as the result as if unencrypted data were aggregated. Using public key cryptography, digital signatures can be used to achieve integrity. We propose a new algorithm using homomorphic encryption and additive digital signatures to achieve confidentiality, integrity and availability for in- network aggregation in wireless sensor networks. We prove that our digital signature algorithm which is based on Elliptic Curve Digital Signature Algorithm (ECDSA) is at least as secure as ECDSA. Even without in-network aggregation, security is a challenge in wireless sensor networks. In wireless sensor networks, not all messages need to be secured with the same level of encryption. We propose a new algorithm which provides adequate levels of security while providing much higher availablility than other security protocols. Our approach uses similar amounts of energy as a network without security.

Researcher

Dr. Sanjay Madria

Julia Albath