Energy efficient clustering with compressive sensing for underwater wireless sensor networks

Due to the challenges of the Internet of Things (IoT) enabled underwater communications, Underwater Wireless Sensor Networks (UWSNs) have been graced as a hot research topic. The energy efficiency, void communications, and packet collisions are vital challenges in using IoT-enabled UWSNs. To this end, we propose a novel cluster-based routing protocol called Energy Efficient UWSNs Clustering Protocol (EEUCP). The EEUCP is an integrated clustering with routing technique that aids in energy conservation for network lifetime enhancement in UWSNs. Initially, the underwater sensor nodes deployed in different layers of the ocean column are divided into clusters by a simple K means algorithm. The Fuzzy Logic (FL) approach is then implemented to select an optimal Cluster Head (CH) for each cluster in the network. The FL rules are designed using three input variables Residual Energy (RE), Distance to the Surface Sink (DSS), and Packet Delivery Ratio (PDR) of every sensor in the cluster. To address the problem of void communication, the reliable forwarding relay selection problem is formulated and has been solved by directly utilizing the periodic fuzzy trust values of the sensor nodes obtained during the CH selection phase. To further reduce the energy consumption and number of transmissions from the CH to the sink node, a hybrid data reduction, and Compressive sensing (CS) mechanism has been adopted. For data reduction, the lightweight high similarity data analysis mechanism is utilized. The hybrid CS method is implemented where the random CS metrics are applied to the periodically aggregated CH data before transmitting it to the surface sink node. The simulation results demonstrate that the EEUCP protocol significantly minimizes the energy consumption and improves the network Quality of Service (QoS) performances compared to the existing algorithms.