Workshop: INDIS Keynote: Applications of ML and AI in Next Generation Wired and Wireless Networks
Abstract: The emergence of new network technologies such as Intent based networks, coherent core optical networks, flow-based firewalls and 5G wireless has caused exponential increase in complexity and throughout of edge, access, metro and core networks. In such complex flow-based networks, manual and reactive adjustment of network parameters is not efficient and cannot guarantee reliable and efficient network deployment and operation. The monitoring and control of network parameters such as QoS policy assignment in access networks, modulation and symbol rate adjustments in optical networks, packet filtering and anomaly detection in security networks, beam-forming and scheduling in 5G wireless networks is extremely complex and challenging. Machine learning (ML) and Artificial Intelligence (AI) can be used to perform proactive prediction and adjustment from learned behavior in next generation networks to achieve high efficiency and reliability in networks. This session discusses the applications of machine learning in next generation networks and provides an overview of how machine learning models (supervised, unsupervised or recurrent) deployed on a programmable hardware can help to build automated, efficient and resilient next generation networks.