Summary
The design of energy‐efficient underwater wireless sensor networks (UWSNs) poses many challenges due to the intrinsic properties of propagation medium and limited battery power of sensor nodes. This paper proposes the concept of optimal clustering for three‐dimensional (3D) UWSNs leveraging compressive sensing (CS) and principal component analysis (PCA) technique of data compression. Optimal clustering reduces the energy consumption by selecting the optimal number of clusters whereas CS and PCA compression techniques reduce the energy consumption by considering a lesser number of samples and reduce the data redundancy at cluster heads (CHs) level, respectively. Moreover, three communication techniques like acoustic, electromagnetic (EM), and free‐space optical (FSO) wave are considered for communication in 3D UWSNs. We compared the energy efficiency for all three communication techniques by examining the three base station (BS) positions at the center, at the corner, and at the lateral midpoint of the 3D sensing area. Moreover, performance parameters (network lifetime, throughput, packet drop rate, and latency) are also evaluated for 3D UWSNs. It is observed that PCA outperforms the CS technique. The proposed technique is suitable for long‐term and densely deployed 3D UWSNs, in which saving energy is of crucial importance.