Princeton University

School of Engineering & Applied Science

How and why to embrace artificial intelligence in wireless network operations?

Haris Gacanin, Bell Labs
B205 Engineering Quadrangle
Monday, June 11, 2018 - 10:00am

Today, network management is mainly reactive to enable corrective actions only when a problem has already occurred and detected by network counters or customer complaints. In the future, we envision self-organizing networks that exploit artificial intelligence (AI) to manage and optimize its performance in autonomous fashion. This talk explores challenges and future research directions related to knowledge-driven network automation with AI. The talk provokes new coming challenges and unveil necessary future directions across multi-disciplinary research areas. We start from challenges and requirements toward network self-organization (i.e., complex system-of-systems) and then discuss on practical examples of agent design with perception, reasoning and learning. Finally, we end with a case study using AI prototype implementation using commodity wireless access points.
Dr. Haris Gacanin is currently the Department Head of the Autonomous Networked Systems Group, at Bell labs. In this role, he is responsible for developing  and  leading  multi-disciplinary  research  related  to the application  of  artificial  intelligence and  machine  learning  to  autonomous  systems,  enabling  network  nodes  to  physically  move, communicate  and  reshape  with  user  activity  and  location.  This  multi-disciplinary  research targets  applications  such  as  networked  system  design,  customer  experience  management  and autonomous  network  operation  and  management. Prior to Nokia, Dr Gacanin was on the faculty at Tohoku University, Japan, and is currently a visiting professor at KU Leuven.