The new generation 5G wireless networks will have a huge impact on the society due to the high bandwidth and capacities they provide. The traffic volume is expected to grow significantly and new varieties of applications, e.g., Internet of Things and vehicular networking, are anticipated. As a result, effective management of the new networks will become much more complicated and challenging. Machine learning techniques have made unprecedented progress in recent years, as they are highly efficient for data-driven applications.
G networks have emerged as a promising solution for Mobile Network Operators (MNOs) to offer ultra-fast mobile broadband and ultra-low latency services with exceptional reliability for consumers. By leveraging softwarization, Software-Defined Networking (SDN) and Network Function Virtualization (NFV), MNOs can offset the high capital and operational expenditures incurred due the additional deployment of legacy equipment.
Optical fiber communications systems are used throughout the global communications network to transmit information over distances ranging from several kilometers to thousands of kilometers. This infrastructure is the backbone of the Internet that is used on a daily basis worldwide. Applications driving demand for increased capacity include (i) video streaming services, (ii) cloud based storage and services, and (iii) machine-to-machine applications.
Data centers (DCs) in network softwarization and 5G eras are significantly different from those operated nowadays by public cloud providers. They are massively distributed, closer to end-users, heterogeneous (e.g., multi-access edge, central office as a data center, etc.) and rely on much more complex technologies (e.g., Network Functions Virtualization [NFV] and Software-Defined Networking [SDN]). This makes their Operation and Management (O&M) much more challenging. Much more intelligence is required for automating the various tasks.
Smartphone based indoor navigation services are desperately needed in an indoor GPS-denied environment, such as in Combat-zone Surveillance, Health Monitoring, Fire Detection, etc. The Receive Signal Strength (RSS) based algorithms are commonly used in indoor localization, which rely on the WiFi fingerprint data built by the Mobile Crowdsensing approach.
The advent of 5G (fifth generation) telecommunication networks also brings new security challenges, in addition to many benefits to the community. Such is exemplified by its special nature of technology (as well as the new business model) and its deep involvement in people’s everyday life, hence more critical. We need proper security metrics to tell how secure these new networks are especially for decision makers. Our preliminary studies show that existing security metrics are insufficient for 5G networks (as they are not designed so).
One challenge to industry adoption of products and services based on Machine Learning and Artificial Intelligence is that their inner workings are often not discernable to human operators. When operators can’t reason about theit tools they tend to make poor decisions about how and when to rely on them. This ultimately limkitsthe effectiveness and efficiency of these technologies in industrial applications.
Artificial Intelligence techniques have been widely applied to solve real-world challenges, from autonomous driving cars, to detecting diseases. With the popularity of 5G wireless network, more and more AI systems are being developed to provide convenient services to everyone. It is important to ensure the reliability and quality of AI systems from every phase in software development cycle, i.e., development, integration, deployment and monitoring. In this collaboration with Ericsson GAIA, we will propose techniques to systematically improve the quality and reliability of AI systems.
There is an exponential increase in the network traffic worldwide due to the growth of social networks, multimedia sharing web services, streaming of video-on-demand (VoD) contents. However, the bandwidth isn’t growing at the same rate as the demand, resulting in a loss of Quality of Service (QoS) and Quality of Experience (QoE) for the users. Distributed edge caching provides an effective mechanism for mitigating the bandwidth requirements of the growing traffic demands by trading off bandwidth with storage.
The objective of this project is to develop techniques and tools that leverage artificial intelligence to automate the process of handling system crashes at Ericsson, one of the largest telecom and software companies in the world, and where the handling of crash reports (CRs) and continuous monitoring of key infrastructures tend to be particularly complex due to the large client base the company serves. In this project, we will explore the use of deep learning algorithms to classify CRs based on a variety of features including crash traces, CR descriptions, and a combination of both.