AI-Based Next Generation Edge Platform for Heterogeneous Services using 5G Technologies

  • Abdallah, Mohamed Mahmoud (Principal Investigator)
  • Abegaz, Mohammed Seid (Post Doctoral Fellow)
  • Hevesli, Muhammet (Graduate Student)
  • Student-1, Graduate (Graduate Student)
  • Saad, Miss Rahma (Consultant)
  • Assistant-1, Research (Research Assistant)
  • Assistant-3, Research (Research Assistant)
  • Mohamed, Dr.Amr (Principal Investigator)
  • Al-Jaber, Dr.Hessa (Principal Investigator)
  • Chiasserini, Prof.Carla Fabiana (Principal Investigator)
  • Al Fuqaha, Ala (Lead Principal Investigator)

Project: Applied Research

Project Details

Abstract

Promoting intelligent services for industrial automation and E-health is an integral part of the Qatar National Vision 2030. Providing smart, efficient, and secure services for improving Qatari citizens quality of life has been a top national interest in Qatar. Coinciding with this increasing interest, Tactile Internet has been envisioned to revolutionize several societal, economic, and industrial aspects of our future by enabling real-time machine-to-machine and human-to-machine interactions in the 5G era and beyond. The rapid evolution of Artificial Intelligence (AI), Internet of Things (IoT), and Big data is paving the path towards a plethora of Tactile Internet applications. Tactile Internet is considered as the next evolution of IoT that will enable the full potential of industrial 4.0 revolution. However, to enable such applications and build real-time interactive systems, we must provide a latency that matches the speed of our natural reactions (with a typical value of end-to-end latency being 1ms). Furthermore, the underlying network must support ultra-reliable services, since many critical tasks will be executed remotely. This calls for integrating multiple technologies at the network and application level. We envision that bringing the intelligence close to the users, using multi-access edge computing (MEC), along with transmitting the data over a heterogeneous 5G network is a key for supporting tactile applications and next generation systems’ requirements. Accordingly, this collaborative project aims at developing novel, intelligent platform, and algorithms for supporting reliable and effective services for diverse types of applications. The proposed platform and AI algorithms will adopt recent technologies of 5G to support a wide range of heterogeneous services in industrial automation and E-health. Leveraging ubiquitous sensing, heterogeneous network, intelligent processing and control schemes, the proposed platform can in real-time monitor people's daily life and provide intelligent services to both residents and travelers. In this project, we will consider several use cases from E-health applications (e.g., emergency detection and notification, remote health monitoring, epidemic prediction, and occupational safety), as representative applications for Tactile Internet. However, the proposed algorithms and methodologies can be easily applied to different types of applications. To support diverse-intelligent services with strict requirements, our project aims to adopt the recent 5G technologies. In particular, we propose the use of edge computing and network function virtualization (NFV) to fulfil diverse applications and services requirements. In this context, the proposed platform will integrate AI schemes in 5G infrastructures to enable a paradigm shift for improving efficiency of the healthcare systems in Qatar. In contrast to the previous literature on 5G and network virtualization, the adopted platform will consider context-aware MEC approaches to be implemented over 5G networks, in order to optimize the data delivery and support diverse services, along with their requirements. Specifically, this project aims to answer the following main questions: 1. How to integrate the wireless network infrastructure and application-layer characteristics to optimize the performance of the supported services? 2. Where and how to allocate the computational and network resources across the infrastructure (i.e., from the edge to the cloud) to fulfill diverse service requirements? 3. How to leverage the spectrum across multiple networks to fulfill diverse application Quality of Service (QoS) requirements? And, how to transfer the data across user devices and infrastructure with the ultra-high reliability and ultra-low latency levels needed for Tactile Internet? 4. How to reduce the incurred cost in supporting the targeted services, while ensuring the above fulfillment of the above requirements? We remark that the decisions made to answer the above questions are typically entangled in ways that lead to having complex and often counteractive phenomena. Thus, a key aspect of our work is to develop intelligent and flexible platform and algorithms that account for: (i) the different requirements of the supported services, (ii) the capabilities of the network and computing infrastructure, as well as (iii) the privacy of the acquired data. Such aspects will be thoroughly investigated and fully developed, also thanks to the collaboration with one of the top universities worldwide, Politecnico di Torino, Italy, which will further strengthen the scientific excellence in Qatar. Furthermore, we will depict the feasibility of the proposed platform and algorithms via detailed experimental use cases. Our demonstrations will be implemented using real-world data sets that will be acquired from our collaborators in Droobi Health, and Ooredoo Qatar.

Submitting Institute Name

Hamad Bin Khalifa University (HBKU)
Sponsor's Award NumberNPRP13S-0205-200265
Proposal IDEX-QNRF-NPRPS-38
StatusFinished
Effective start/end date11/04/2130/09/24

Collaborative partners

Primary Theme

  • Artificial Intelligence

Primary Subtheme

  • AI - Smart Society

Secondary Theme

  • Artificial Intelligence

Secondary Subtheme

  • AI - Smart Cities

Keywords

  • Machine learning
  • Network Function Virtualization
  • Edge computing; 5G technologies; Healthtech

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.