Artificial Intelligence Techniques for an Optimized Causal Dilated Convolutional Neural Networks-Based Energy-Efficient and Delay-Sensitive Routing Paths Using Mobility Prediction in Mobile WSN: Trends and Challenges

Abstract

Mobile Wireless Sensor Networks (MWSNs) are increasingly utilized in applications such as smart cities, environmental monitoring, healthcare, and military surveillance. However, their dynamic topology, energy constraints, and delay-sensitive communication requirements present significant challenges in designing efficient routing protocols. Artificial Intelligence (AI) techniques, particularly deep learning models such as Convolutional Neural Networks (CNNs) and their advanced variants like causal dilated CNNs, have emerged as promising solutions for optimizing routing decisions. Recent studies highlight that AI-driven routing mechanisms can effectively analyse network conditions such as congestion, delay, and link quality to improve performance. Mobility prediction plays a crucial role in enhancing routing stability by forecasting node movement patterns, thereby reducing route failures and packet loss. Furthermore, hybrid deep learning approaches combining CNN, RNN, and attention mechanisms have shown superior capability in capturing both spatial and temporal dependencies in network data. This paper presents a comprehensive review of AI-based routing techniques focusing on causal dilated CNN models for energy-efficient and delay-sensitive routing in MWSNs. It analyses 30 studies published recently, highlighting trends, advantages, limitations, and future challenges. The findings indicate that integrating mobility prediction with deep learning significantly enhances network performance, reduces energy consumption, and improves Quality of Service (QoS). However, challenges such as computational complexity, scalability, and real-time deployment remain open research issues.