Project Description

5G mobile networks are expected to support a wide range of applications and services beyond traditional voice and data services. They will offer services with diverse QoS requirements, such as enhanced mobile broadband (e.g., Ultra-HD video streaming, Augmented/Virtual-reality communications), ultra-reliable low-latency communications (e.g., tele-driving, tele-surgery), and massive machine-type communications (e.g., smart cities). The capability to support these services will open new revenue streams for mobile network operators as well as enable new industry verticals. In this context, network slicing is an enabling technology for accommodating different QoS requirements on the same physical network. Network slicing allows the partitioning of the physical network into multiple network slices, each supporting QoS tailored to an application or industry vertical. For example, emergency services could operate a network slice independent from others with a specific QoS, increased security and reliability to accommodate their operational needs. Network slicing provides network operators with increased control over the management and orchestration of network services.

Providing network services using 5G network slices involves several network slice life-cycle management steps, including slice instantiation, configuration and activation, followed by a run-time phase, and finally slice decommissioning. The diverse, demanding, and stringent QoS guarantees offered by 5G networks cannot be satisfied with traditional reactive human-in-the-loop management and operation approaches. Artificial Intelligence (AI) and Machine Learning (ML) will play a major role in 5G network operations and management to achieve network automation. The overarching goal of this project is to realize automated and data-driven 5G network life-cycle management, powered by recent advances in AI, ML, and large-scale data processing. It will address several challenging problems spanning slice orchestration, slice monitoring, and fault and performance management.

Related Publications

• Md. S. Towhid, N. S. Khan, N.Shahriar, M. Tornatore, R. Boutaba, A. Saleh. A Token-Prioritization Strategy for Handling Data Imbalance in Network-Change Ticket Classification. IEEE/ACM/IFIP Conference on Network and Service Management (CNSM). Niagara Falls, Canada, October 30 - November 2, 2023.

• N. Saha, N. Shahriar, R. Boutaba and A. Saleh. MonArch: Network Slice Monitoring Architecture for Cloud Native 5G Deployments. IEEE/IFIP Network Operations and Management Symposium (NOMS). Miami, FL, USA, May 8-12, 2023.

• M. Sulaiman, M. Ahmadi, M. A. Salahuddin, R. Boutaba and A. Saleh. Generalizable Resource Scaling of 5G Slices using Constrained Reinforcement Learning. IEEE/IFIP Network Operations and Management Symposium (NOMS). Miami, FL, USA, May 8-12, 2023. Recipient of the Best Paper Award.

• A. Moyyedi, M. Ahmadi, M. A. Salahuddin, R. Boutaba and A. Saleh. Generalizable GNN-based 5G RAN/MEC Slicing and Admission Control in Metropolitan Networks. IEEE/IFIP Network Operations and Management Symposium (NOMS). Miami, FL, USA, May 8-12, 2023.

• N. Saha, M. Zangooei, M. Golkarifard and R. Boutaba. Deep Reinforcement Learning Approaches to Network Slice Scaling and Placement: A Survey. IEEE Communications Magazine. IEEE Press. Vol. 61(2), pp. 82-87, February 2023.

• S. S. Johari, S. Taeb, N. Shahriar, S. R. Chowdhury, M. Tornatore, R. Boutaba, J. Meetra and M. Hemmati. DRL-Assisted Reoptimization of Network Slice Embedding on EON-enabled Transport Networks. IEEE Transactions on Network and Service Management. Accepted November 2022.

• M. Sulaiman, A. Moyyedi, M. Ahmadi, M. A. Salahuddin, R. Boutaba and A. Saleh. Coordinated Slicing and Admission Control using Multi-Agent Deep Reinforcement Learning. IEEE Transactions on Network and Service Management. Accepted November 2022.

• N. Saha, A. James, N. Shahriar, R. Boutaba, A. Saleh. Demonstrating Network Slice KPI Monitoring in a 5G Testbed. IEEE/IFIP Network Operations and Management Symposium (NOMS). Budapest, Hungary, April 25-29, 2022.

• M. Sulaiman, A. Moayyedi, M. A. Salahuddin, R. Boutaba, A. Saleh. Multi-Agent Deep Reinforcement Learning for Slicing and Admission Control in 5G-CRAN. IEEE/IFIP Network Operations and Management Symposium (NOMS). Budapest, Hungary, April 25-29, 2022. Recipient of the Best Paper Award.

• S. S. Johari, N. Shahriar, M. Tornatore, R. Boutaba, A. Saleh. Anomaly Detection and Localization in NFV Systems: An Unsupervised Learning Approach. IEEE/IFIP Network Operations and Management Symposium (NOMS). Budapest, Hungary, April 25-29, 2022. Recipient of the Best Student Paper Award.

• R. Boutaba, N. Shahriar, M. A. Salahuddin, S. R. Chowdhury, N. Saha, A. James. AI-driven Closed-loop Automation in 5G and beyond Mobile Networks. In Proceedings of the 4th ACM SIGCOMM Workshop on Flexible Networks Artificial Intelligence Supported Network Flexibility and Agility (FlexNets). Virtual Conference, August 23-27, 2021.

Sponsors and Partners