Sensor and neural network-based optimization of fishermen’s data routing and boundary compliance

Abstract

Maritime safety remains a major concern, especially when fishermen inadvertently cross international boundaries due to unreliable communication and navigation systems. Existing boundary alert mechanisms often face challenges such as high latency, insecure data transmission, and limited environmental awareness. To address these issues, this study proposes an Intelligent Fishermen Boundary Alert System (FBAS) that integrates sensor-based environmental monitoring, optimized data routing, and secure communication. The system employs Pelican Optimization Algorithm (POA) to select efficient communication routes and Improved Recurrent Radial Basis Function Neural Network (IRBFNN) based security model for detecting anomalies and preventing cyber intrusions. Simulation results demonstrate significant performance improvements compared to existing models like Artificial Neural Networks (ANN), Radial Basis Function Neural Network (RBFNN), and Whale Optimization Algorithm-Radial Basis Function Neural Network (WOA-RBFNN). Overall, the proposed FBAS offers a reliable and secure solution that enhances boundary compliance and operational safety for fishermen in dynamic marine environments. The proposed work's footage is available at: https://youtu.be/XXic7CP9UBE.