Call for Papers: Role of Deep Reinforcement Learning for Wireless Network Virtualization: Future Perspective
Network virtualization is the process of converging hardware and software resources to create a software-based virtual network. With the advancements in network and communication technologies, wireless network virtualization has become increasingly important in recent times. This is because there is an increasing demand for the network operators to efficiently manage their spectrum without compromising their cost and quality standards. Centralized radio access networks, cell virtualization, network function virtualization, and network slicing are considerable ways to achieve network virtualization. Towards a step forward, it is estimated that the future of wireless networks will be infrastructure-independent, aimed at offering server less computing services to the end-users. Thus, the future of networks will be the modules of software that can be customized based on the application requirements. To thrive in this dynamically changing environment, it is much needed for future networks to automate the resource provisioning processes and offer the agility that modern technologies provide.
Today, many business organizations are turning to deep learning as a source of advantage for network virtualization processes. The major concern here is the number of connected devices. Thus one has to find efficient ways of scaling the network to support a massive amount of the data. The need to assign resources in real-time based on the service needs in an automated manner is the key problem. Thanks to deep learning techniques, it constantly tests the network and its behavior over the resources assigned. Once the deep learning model has acquired knowledge on the exact service level a device needs, it automatically assigns the resources by relieving the unused ones and maintains a higher amount of resource utilization efficiency measures. However, building a deep reinforcement learning model requires thousands of process reputations with billions of data samples, which is highly a challenging process. On the other side, the benefits of deploying deep reinforcement learning for future wireless network virtualization are dramatic from a competitive standpoint. Hence, exploring more in this regard is crucially important, and it offers significant advantages to the future wireless communication networks.
In this special issue, we will take a closer look into advanced deep reinforcement learning paradigms for future network virtualization. The prime focus will be on increasing network performance, optimization, and efficiency measures.
Original research and review articles in this area are encouraged in the following topic areas including, but are not limited to:
- Novel deep reinforcement learning frameworks, algorithms, and tools for network virtualization
- Role of deep reinforcement learning in emerging networks virtualization
- Spectrum sharing in virtualized networks using deep reinforcement learning
- Wireless virtualization in future networks with deep learning
- Deep reinforcement learning to preserve security & privacy across network virtualization processes
- Effective ways of improving performance and network efficiency with deep reinforcement learning
- Deep reinforcement learning for traffic engineering, scheduling, network slicing and virtualization
- Deep reinforcement learning in mobile edge computing, wireless caching, and mobile data offloading
- Deep reinforcement learning assisted network paradigms and routing protocols for future networks
- Channel allocation algorithms considering QoS for Mobile Network Virtualization
Manuscript Submission Deadline: 15 August 2021
Authors Notification Date: 20 October 2021
Revised Papers Due Date: 25 January 2022
Final notification Date: 5 April 2022
Lead Guest Editor
Dr. Tu Nguyen
Department of Computer Science,
Purdue University Fort Wayne,
Fort Wayne, USA
Dr. Nam P. Nguyen
Department of Computer Science,
Purdue University Fort Wayne, USA
Dr. Claudio Savaglio
Department of Computer Science, Modeling,
Electronics and Systems Engineering (DIMES),
University of Calabria, Italy
Authors should follow the WPC Journal manuscript format described at the journal site. Manuscripts should be submitted online through https://www.editorialmanager.com/wire/default.aspx.