Fuzzy Based Node‐to‐Node (N2N) and Intercluster Authentication Framework for External Attacks in UWSN

ABSTRACT In underwater wireless sensor networks (UWSNs), there is a need for providing security against internal and external attacks, apart from ensuring reliable data delivery. Due to the high computation cost and the low computing capabilities of UWSN, all present methods of device‐based identity authentication must be performed at the network layer and other layers. The main objective of this work is to protect the data against physical layer attacks using lightweight authentication techniques. In this paper, a fuzzy‐based node‐to‐node and intercluster authentication (Fuzzy‐N2N‐ICA) framework for UWSN has been proposed. Initially, the network is clustered using the minimum average routing path (MARP) algorithm. In N2N authentication technique, the cluster head (CH) collects the network layer and physical layer features from each node. It then determines the deviation index based on these cross‐layer metrics by applying a fuzzy‐logic decision model. Based on the determined deviation index, the corresponding node is authenticated as honest or malicious. In the intercluster authentication phase, each CH creates an authentication key using the secret key obtained from an offline trusted third party. When the aggregated data is received from a source CH, other CHs and the sink node perform authentication. Simulation outcomes have shown that the Fuzzy‐N2N‐ICA framework minimizes the miss detection ratio, authentication time, computational cost, energy consumption, and maximizes the correctness of packet delivery.