Focuses on predicting the network demand according to the spatial/temporal variation of the different regions of a city, and according to the crowd mobility patterns as well.
Published in Chapter:
Automating Network Management With Artificial Intelligence: In Software Networks and Beyond
Imen Grida Ben Yahia (Orange Labs, France), Jaafar Bendriss (Orange Labs, France), Teodora Sandra Buda (IBM, Ireland), and Haytham Assem (IBM, Ireland)
Copyright: © 2019
|Pages: 33
DOI: 10.4018/978-1-5225-7146-9.ch005
Abstract
Artificial intelligence (AI) and in particular machine learning are seen as cornerstones to automate and rethink network management operations in the context of network softwarization (i.e., SDN, NFV, and Cloud). In this regard, operators and service providers target the creation of service offerings, the customization of network solutions, and the fast adaptation to rapidly changing market demands. This translates into requirements for increased flexibility, modularity, and scalability in network management operations. This chapter presents a detailed specification of a cognitive (AI-based) network management framework applicable for existing and future (software-defined) networks. The framework is built upon the combined state-of-the-art on autonomic, policy-based management and big data. It is exemplified with two detailed use cases: the urban mobility awareness for today's mobile networks and SLA (service level agreement) enforcement in the context of NFV and cloud.