The modern networks are anticipated to connect billions of wireless devices due to the popularity of smartphones, wearable devices, and ubiquitous wireless sensors for various applications. In the last decade, the rapid deployment of massive devices and applications has caused a growing demand for wireless radio spectrum. Cognitive Radio (CR) was proposed to utilize the spectrum efficiently in an opportunistic way, which is evolved to be an intelligent radio that can change its transmitter parameters according to the interactions with the environment. CR network equips its users with cognitive capability to sense and gather information from the surrounding environment and with reconfigurability to rapidly adapt the operational parameters according to the sensed information. This can be formulated as a study problem of user: any user perceives its environment and takes actions that maximize its chance of successfully achieving its goals. To explore the study of intelligent agent, Artificial Intelligence (AI) defined as a machine mimics "cognitive" functions that humans associate with other human minds, such as "learning" and "problem solving" is brought to use in CR networks, named AI-enabled Radio and Networks. 

The ultimate design goal of future wireless networks is to meet diverse Quality of Service (QoS) requirements of end users. This calls for the network entities in nature to be cognitive of the network environment and autonomous in decision making. Different network entities in the network layer, control layer, and management and orchestration layer, such as wireless devices, base stations, and SDN controllers need to make local and autonomous decisions, including spectrum access, channel allocation, power control, etc. to achieve different goals of different networks, e.g., throughput maximization, delay and energy minimization. This decision-making problem can be modelled as an MDP, which then be solved by reinforcement learning algorithm that is a branch of AI algorithms. However, the reinforcement learning process, even though proved to converge, takes a lot of time to reach the optimal policy as it needs to explore and gain knowledge of the entire system. As the future wireless networks have been becoming large-scale and complicated, we face a more decentralized and diverse network environment. Hence, the network control problems become very challenging as the dimensionality and computational complexity rapidly increase beyond the manageable range of reinforcement learning, due to the dynamic and uncertain network status, as well as the coexistence and couplings among different wireless users with heterogeneities.

To address the curse of dimension problem in reinforcement learning, deep learning which is a part of a broader family of machine learning methods is proposed, and thus opens a new era for the development of reinforcement learning, namely Deep Reinforcement Learning (DRL). The DRL embraces the advantage of Deep Neural Networks (DNNs) as powerful function approximators, thereby improving the learning speed and the performance of reinforcement learning algorithms for high dimensional and continuous control problems. The integration of DRL into future wireless networks will revolutionize the conventional model-based network optimization to model-free approaches and to meet various application demands. By continuously interacting with the environment, DRL provides an autonomous decision-making mechanism for the network entities to learn and build knowledge about the network environment. Hence, the network entities can solve non-convex, complex and even model-free problems, e.g., spectrum access, handover, scheduling, caching, data offloading, and resource allocation, with minimum or without information exchange among each other.

The objective of this special issue is to explore recent advances in CR networks and AI algorithms employed in wireless communication networks. This special issue will bring together leading researchers and developers from to present their research on AI-enalbed Radio networks which include network framework and algorithms, network modeling and architecture, as well as AI algorithms inspired control problems in different layers, addressing various challenges related to the analysis and design for future wireless networks. High original research and review articles in this area are welcome. Potential topics include but are not limited to the following:

• Novel AI-enabled Radio and Network framework, algorithms, convergence, and performance analysis
• Testbed, experiments, and simulations of AI algorithms in communications and networking
• AI algorithms for physical layer issues, e.g., channel estimation, interference alignment, and coding
• AI algorithms inspired network architecture, MAC, and routing protocols
• AI algorithms in network access and transmit control, e.g., channel allocation, power and rate control
• AI algorithms for traffic engineering, scheduling, network slicing and virtualization
• AI algorithms for network coexistence, e.g., HetNet, cognitive radio, device-to-device networks
• AI algorithms in emerging networks, e.g., wireless powered networks, UAVs, ULLC, VANET, etc.
• AI algorithms in mobile edge computing, wireless caching, and mobile data offloading
• AI algorithms for network security and connectivity preservation
• AI algorithms for network forensic, fault detection, and auto-diagnosing
• AI algorithms for network economics, auction, multi-agent learning, and crowdsourcing
• Emerging technology on machine learning for communications

Submission Guidelines

Prospective authors should submit their manuscripts following the IEEE TCCN Guidelines. Authors should submit a PDF version of their complete manuscript to Manuscript Central according to the following schedule:

Important Dates

Submission Deadline: 1 June 2019 (extended)
First Reviews Complete: 1 September 2019
Revision Due: 15 October 2019
Final Review Decision: 15 November 2019
Final to Publisher: 1 December 2019
Publication: First Quarter 2020

Guest Editors

Yue Gao (Lead)
Queen Mary University of London, UK

Ekram Hossain
University of Manitoba, Canada

Geoffrey Ye Li
Georgia Institute of Technology, US

Kevin Sowerby
University of Auckland, New Zealand

Carlo Regazzoni
University of Genoa, Italy

Lin Zhang
University of Electronic and Science Technology of China, China