Periodic Reporting for period 2 - CogNet (Building an Intelligent System of Insights and Action for 5G Network Management)
Berichtszeitraum: 2016-07-01 bis 2017-12-31
The overall objective of the CogNet project was to make a major contribution towards autonomic network management through the use of network information such as state, topology, configuration and various key metric data, and relevant external data which we call situational context or out of band information, and use this to proactively manage the network, maintain sufficient capacity to meet demand levels and maintain Quality of Service at the levels required by the different types of network users. At the beginning of the project the key objectives were identified and cover intelligent data collection and filtering, intelligent policy management, machine learning applied to the problems of prediction and provisioning for energy and costs savings, machine learning for network resilience, anomaly detection and security threat detection and finally apply all of this knowledge to multiple demonstrations. The research that has been carried out within CogNet currently has and will continue to contribute towards more reliable mobile communications which in turn will facilitate the development and deployment of new applications such as richer communications experiences, smart mobile medical technology, autonomic driving and an ICT infrastructure that has the added benefit of being more energy efficient.
Work Package 1 managed the overall project from the management and technical perspectives. All agreements were put in place and the rhythm of the project was established. The deliverables covered the areas from project management reports to data management plans, IPR management plans and the data protection and privacy audit report.
Overall WP1 ensured to collaboration between the partners from all of the work packages in the form of weekly phone conferences, quarterly plenary and technical face to face meetings and various workshops that only involved relevant partners.
Work Package 2 concentrated on identifying and defining a set of challenges, use cases and scenarios and their requirements, related to the 5G network, specifically from a network management perspective.
The first deliverable D2.1 introduced six use cases of CogNet based on the challenges of the future 5G network management. D2.1 also introduced a total of eleven initial scenarios all pivoting around the use cases in order to facilitate more specific research questions of high impact value in a real life situation. These were reduced to seven scenarios in Deliverable 2.2.
Work Package 3 focuses on design and implementation of algorithms within the CogNet Smart Engine as well as integration and adaptation of off-the-shelf state-of-the-art machine learning modules.
The main outputs of WP3 have been presented in detail in four deliverables.
Work Package 4 aimed at researching and developing smart real-time analytics techniques for optimizing the performance of Virtual Network Functions (VNFs) in terms of energy efficiency, quality of service and resource elasticity by means of orchestration mechanisms.
Work Package 5 aimed at reaching an appropriate set of security mechanisms based on the machine learning addressing the dynamic network function environment of the software networks on top of the cloud. Each of the WP phases was reflected in the form of a deliverable.
Work Package 6 designed and implemented the CogNet Common Infrastructure. A set of scripts and Ansible playbooks which allows any network manager to deploy the full CogNet infrastructure ready to capture data, analyse specific key features and provide actuation suggestion according to the identified or predicted networking issues and the defined actuation providing a best-practice reference platform for developing cognitive management solutions with machine learning.
WP6 has also implemented a set of different demonstrators which meet the requirements and specific 5G challenges, explored in WP4 and WP5. The demonstrators, selected from real business plans from partners, generate or use representative datasets to meet target 5G challenges.
Work Package 7 was focused on guaranteeing a strong impact of the project achievements in the most relevant research and industrial communities, spanning across several categories of stakeholders in the cloud service provider, 5G and Machine learning areas by the use of the following mediums:
- Website and Social Media channels
- Communication and Dissemination
- Standardization
- Business exploitation
The dissemination plan of the CogNet partners included publishing high-quality papers in major international conferences and journals in the area of networking, security and autonomic systems. The consortium's final exploitation strategy has been periodically captured throughout the life time of the project and has the final strategy documented in the deliverable D7.9 where each partner has provided a plan to exploit the achievements of the CogNet project.
Overall WP1 ensured to collaboration between the partners from all of the work packages in the form of weekly phone conferences, quarterly plenary and technical face to face meetings and various workshops that only involved relevant partners.
Work Package 2 concentrated on identifying and defining a set of challenges, use cases and scenarios and their requirements, related to the 5G network, specifically from a network management perspective.
The first deliverable D2.1 introduced six use cases of CogNet based on the challenges of the future 5G network management. D2.1 also introduced a total of eleven initial scenarios all pivoting around the use cases in order to facilitate more specific research questions of high impact value in a real life situation. These were reduced to seven scenarios in Deliverable 2.2.
Work Package 3 focuses on design and implementation of algorithms within the CogNet Smart Engine as well as integration and adaptation of off-the-shelf state-of-the-art machine learning modules.
The main outputs of WP3 have been presented in detail in four deliverables.
Work Package 4 aimed at researching and developing smart real-time analytics techniques for optimizing the performance of Virtual Network Functions (VNFs) in terms of energy efficiency, quality of service and resource elasticity by means of orchestration mechanisms.
Work Package 5 aimed at reaching an appropriate set of security mechanisms based on the machine learning addressing the dynamic network function environment of the software networks on top of the cloud. Each of the WP phases was reflected in the form of a deliverable.
Work Package 6 designed and implemented the CogNet Common Infrastructure. A set of scripts and Ansible playbooks which allows any network manager to deploy the full CogNet infrastructure ready to capture data, analyse specific key features and provide actuation suggestion according to the identified or predicted networking issues and the defined actuation providing a best-practice reference platform for developing cognitive management solutions with machine learning.
WP6 has also implemented a set of different demonstrators which meet the requirements and specific 5G challenges, explored in WP4 and WP5. The demonstrators, selected from real business plans from partners, generate or use representative datasets to meet target 5G challenges.
Work Package 7 was focused on guaranteeing a strong impact of the project achievements in the most relevant research and industrial communities, spanning across several categories of stakeholders in the cloud service provider, 5G and Machine learning areas by the use of the following mediums:
- Website and Social Media channels
- Communication and Dissemination
- Standardization
- Business exploitation
The dissemination plan of the CogNet partners included publishing high-quality papers in major international conferences and journals in the area of networking, security and autonomic systems. The consortium's final exploitation strategy has been periodically captured throughout the life time of the project and has the final strategy documented in the deliverable D7.9 where each partner has provided a plan to exploit the achievements of the CogNet project.
3The initial period for the project afforded the work packages the ability to determine what the direction each was going to take with regards to objectives, innovations and initial finding on sate of the art. From the initial investigations each of the technical work packages (WP3, WP4, WP5 and to a certain extent WP6) were all able to being their analysis. Following on from the first year review each of the work packages were able to continue their investigation and to identify the gaps that exist in the current technologies and research. The architecture was finalised that included inputs from the major standardisation bodies such as ETSI and this architecture facilitated the application of the previously identified Machine Learning infrastructure. This was then coupled with a policy based network management framework that has the capabilities of enabling dynamic policies to be deployed based on the up to date learnings achieved by the machine learning models running within the infrastructure, all of which is running in a network infrastructure employing NFV and SDN technology. There have been many publications based on the work of CogNet, up to December 2017 there was eighty eight papers published and accepted by journals, conferences, etc. with many of them in top tier avenues. CogNet has also engaged with external network management experts, in particular those involved in the emerging IETF SUPA standard for policy based management expression and the architecture has incorporated this into its overall design. The potential impact and societal benefits from this research will include development of techniques to manage the larger and more dynamic network topologies necessary in 5G, facilitate 5G applications which maintain QoS and Security levels such as Network Slicing, and lower the capital and operational costs through improved efficiencies and the use of node, link and function virtualisation. These will eventually contribute towards more reliable mobile communications which in turn will allow new applications such as autonomic driving, smart mobile medical technology, richer communications which may help reduce the need for business travel, and a more energy efficient ICT infrastructure.