Wireless Networks Design in the Era of Deep Learning: Model-Based,\n AI-Based, or Both?

2019/02/05 by Alessio Zappone, Marco Di Renzo, Zappone, Alessio +3 · 9 citations
Computer Science · Engineering · #Distributed Sensor Networks and Detection Algorithms #FOS: Computer and information sciences #FOS: Electrical engineering #Information Theory (cs.IT) #Millimeter-Wave Propagation and Modeling #Signal Processing (eess.SP) #Speech and Audio Processing #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.1902.02647

openalex publication_date 2019/02/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This work deals with the use of emerging deep learning techniques in future\nwireless communication networks. It will be shown that data-driven approaches\nshould not replace, but rather complement traditional design techniques based\non mathematical models.\n Extensive motivation is given for why deep learning based on artificial\nneural networks will be an indispensable tool for the design and operation of\nfuture wireless communications networks, and our vision of how artificial\nneural networks should be integrated into the architecture of future wireless\ncommunication networks is presented.\n A thorough description of deep learning methodologies is provided, starting\nwith the general machine learning paradigm, followed by a more in-depth\ndiscussion about deep learning and artificial neural networks, covering the\nmost widely-used artificial neural network architectures and their training\nmethods. Deep learning will also be connected to other major learning\nframeworks such as reinforcement learning and transfer learning.\n A thorough survey of the literature on deep learning for wireless\ncommunication networks is provided, followed by a detailed description of\nseveral novel case-studies wherein the use of deep learning proves extremely\nuseful for network design. For each case-study, it will be shown how the use of\n(even approximate) mathematical models can significantly reduce the amount of\nlive data that needs to be acquired/measured to implement data-driven\napproaches. For each application, the merits of the proposed approaches will be\ndemonstrated by a numerical analysis in which the implementation and training\nof the artificial neural network used to solve the problem is discussed.\n Finally, concluding remarks describe those that in our opinion are the major\ndirections for future research in this field.\n

Cited by

• Deep Unfolding: Recent Developments, Theory, and Design Guidelines
• AURA: Adaptive Unified Reasoning and Automation with LLM-Guided MARL for NextG Cellular Networks
• Integrated Sensing and Communication: Towards Multifunctional Perceptive Network
• Deep Learning-based Techniques for Integrated Sensing and Communication Systems: State-of-the-Art, Challenges, and Opportunities
• A Deep Learning Framework for Joint Channel Acquisition and Communication Optimization in Movable Antenna Systems
• On the Compromise Between Performance and Efficiency in RIS-aided Communication Systems
• Lightweight Gradient Descent Optimization for Mitigating Hardware Imperfections in RIS Systems
• Edge General Intelligence Through World Models and Agentic AI: Fundamentals, Solutions, and Challenges
• Industrial Viewpoints on RAN Technologies for 6G

Related

• Wireless Networks Design in the Era of Deep Learning: Model-Based, AI-Based, or Both?
• Model-Aided Wireless Artificial Intelligence: Embedding Expert Knowledge in Deep Neural Networks Towards Wireless Systems Optimization
• Deep Learning for Wireless Communications: An Emerging Interdisciplinary Paradigm
• Application of End-to-End Deep Learning in Wireless Communications Systems
• Artificial Neural Networks-Based Machine Learning for Wireless Networks: A Tutorial