Underwater Digital Twin Sensor Network-Based Maritime Communication and Monitoring Using Exponential Hyperbolic Crisp Adaptive Network-Based Fuzzy Inference System

The underwater conditions of the coastal ecosystem require careful monitoring to anticipate potential environmental hazards. Moreover, the unique characteristics of the marine underwater environment have presented numerous challenges for the advancement of underwater sensor networks. Current studies have not extensively integrated Digital Twins with underwater sensor networks aimed at monitoring the marine ecosystem. Consequently, this study proposes a decision-making framework based on Underwater Digital Twins (UDTs) utilizing the Exponential Hyperbolic Crisp Adaptive Network-based Fuzzy Inference System (EHC-ANFIS). The process begins with the initialization and registration of an Underwater Autonomous Vehicle (UAV). Subsequently, data are collected from the sensor network and relayed to the UDT model. The optimal path is determined using Adaptive Pheromone Ant Colony Optimization (AP-ACO) to ensure efficient data transmission. Following this, data compression is achieved through the Sliding–Huffman Coding (SHC) algorithm. The Twisted Koblitz Curve Cryptography (TKCC) method is employed to enhance data security. Additionally, an Anomaly Detection System (ADS) is trained, which involves collecting and pre-processing sensor network data. A Radial Chart is then utilized for effective visualization. Anomalies are detected using the CosLU-Variational Shake-Long Short-Term Memory (CosLU-VS-LSTM) approach. For standard data, decision-making based on the UDT model is conducted using EHC-ANFIS, with a fuzzification duration of 21,045 milliseconds. Finally, alerts are dispatched to the Maritime Alert Command Centre (MACC). This approach enhances maritime communication and monitoring along coastal areas, with specific reference to the Coromandel Coast, thereby contributing to the protection of the coastal ecosystem.