A Survey of Methods and Architectures for Optimized Causal Dilated Convolutional Neural Networks-Based Energy-Efficient and Delay-Sensitive Routing Paths Using Mobility Prediction in Mobile WSN

Abstract

Mobile Wireless Sensor Networks (MWSNs) are vital for modern applications such as environmental monitoring, healthcare, smart cities, and industrial automation. However, their characteristics—node mobility, limited energy, dynamic topology, and latency constraints—make designing efficient routing protocols challenging. Achieving both energy efficiency and low delay is crucial, as it directly affects network lifetime and performance. Traditional routing methods, including clustering, shortest-path, and metaheuristic approaches, improve certain aspects like energy consumption but often fail to adapt to dynamic conditions and typically address only one objective at a time. Recent advances in artificial intelligence and deep learning have introduced more effective solutions for intelligent routing. Causal dilated convolutional neural networks (CDCNNs) can capture temporal dependencies and predict mobility patterns, enabling proactive routing decisions that reduce delay and energy usage. Integrating mobility prediction further enhances performance by minimizing packet loss and improving resource utilization. AI-driven frameworks combining machine learning and optimization techniques have demonstrated superior efficiency compared to traditional methods. This survey reviews recent developments, highlighting advancements, comparing techniques, and identifying challenges such as computational complexity, scalability, and real-time implementation.