Call for Papers

The design of future wireless networks needs to meet diverse Quality of Service (QoS) requirements. This calls for the network entities in nature to be cognitive of network environment and autonomous in decision making. Different network entities in the network layer, control layer, and management and orchestration layer, such as mobile devices, base stations, and SDN controllers need to make local and autonomous decisions, including spectrum access, channel allocation, power control, etc. to achieve the goals of different networks, e.g., throughput maximization, delay and energy minimization. As the modern networks have been becoming large-scale and complicated, we face a more decentralized, ad-hoc, and diverse network environment. The network control problems are very challenging as the dimensionality and computational complexity rapidly increase, due to the dynamic and uncertain network status, as well as strong couplings among different wireless users with heterogeneities in, e.g., QoS provisioning, wireless resource, air interface, and mobility.

Deep reinforcement learning (DRL) has been developing as a promising solution to address high dimensional nd continuous control problems effectively, by the use of deep neural networks (DNNs) as powerful function approximators. The integration of DRL into future wireless networks will revolutionize the conventional model-based network optimization to model-free approaches and meet various application demands. By interacting with the environment, DRL provides an autonomous decision-making mechanism for the network entities to solve non-convex, complex model-free problems, e.g., spectrum access, handover, scheduling, caching, data offloading, and resource allocation. This not only reduces the communication overheads but also improves network security and robustness. Though DRL has shown great potential to address emerging issues in complex wireless networks, there are still domain-specific challenges that require further investigation. These may include the design of proper DNN architectures to capture the characteristics of 5G network optimization problems, the state explosion in dense networks, multi-agent learning in dynamic networks, limited training data and exploration space in practical networks, the inaccessibility and high cost of network information, as well as the balance between information quality and learning performance. 

The objective of this special issue is to explore recent advances in DRL and address practical challenges in wireless networks. This special issue will bring together leading researchers and developers to present their research on novel DRL framework, network modeling and architecture, as well as control problems in different layers, addressing various challenges related to DRL inspired 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 DRL framework, algorithms, convergence, and performance analysis
• Testbed, experiments, and simulations of DRL in communications and networking
• DRL for physical layer issues, e.g., channel estimation, interference alignment, and coding
• DRL inspired network architecture, MAC, and routing protocols
• DRL in network access and transmit control, e.g., channel allocation, power and rate control
• DRL for traffic engineering, scheduling, network slicing and virtualization
• DRL for network coexistence, e.g., HetNet, cognitive radio, device-to-device networks
• DRL in emerging networks, e.g., wireless powered networks, UAVs, URLLC, VANET, etc.
• DRL in mobile edge computing, wireless caching, and mobile data offloading
• DRL for network security and connectivity preservation
• DRL for network forensic, fault detection, and auto-diagnosing
• DRL 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: 31 March 2019
First Reviews Complete: 1 June 2019
Revision Due: 1 July 2019
Final Review Decision: 15 August 2019
Final to Publisher: 1 September 2019
Publication Date: Fourth Quarter 2019

Guest Editors