KCN: Knowledge Centric Networking

Lead Research Organisation: University College London
Department Name: Electronic and Electrical Engineering

Abstract

The recent advent of killer applications such as content distribution, cloud computing and Internet of things (IoT), all require for the underlying network to be able to understand specific service contexts. In this project we propose the Knowledge Centric Networking (KCN) paradigm, in which knowledge is positioned at the centre of the networking landscape. The objective is to enable in-network knowledge generation and distribution in order to develop necessary network control intelligence for handling complexity and uncertainty. In order to achieve this, specific algorithms and mechanisms/protocols will be developed for knowledge acquisition, processing, dissemination and organisation both within single and across homogeneous/heterogeneous administrative domains in the Internet.

The project will investigate three styles of knowledge exchange based on Software Defined Networking (SDN) principles: Knowledge as a Tool (KaaT), Knowledge as a Service (KaaS) and Knowledge as a Cloud (KaaC). KaaT will enable intelligent network operations in dynamic network environments driven by knowledge gathered at different vantage points. We advocate a hierarchical knowledge framework in which knowledge and control functions are distributed at the right places within the network for fulfilling specific control tasks. In addition, we will invetigate knowledge sharing between different players in the Internet marketplace. This can be achieved either through explicit knowledge transfer from a knowledge provider to a knowledge consumer (KaaS), or based on open knowledge clouds where knowledge prosumers may publish or subscribe to information through an open but controlled knowledge ecosystem (KaaS).

The proposed KCN architecture will be validated through two complementary use cases. KCN-driven content traffic offloading between heterogeneous radio access technologies for the future mobile Internet aims to achieve adaptive resource control by taking into account a wide variety of knowledge associated with content, users and network conditions. In addition, KCN-driven energy management targets cross-layer energy saving techniques at both the IP and the physical optical layer according to the derived knowledge and dynamically changing context information.

The project provides direct contributions to the TI3 sub-challenges 1, 2, 3 and 4. First of all, the KCN-based knowledge ecosystem will equip the next generation Internet with necessary intelligence for handling complex requirements under dynamic conditions. Such an ecosystem, seamlessly coupled with the SDN architecture, will be able to gracefully support the ever increasing complexity and heterogeneity of future networked services and multitude of users. The two complementary use cases demonstrate how the proposed KCN framework will be instantiated in two different application domains, content traffic offloading in mobile/wireless access networks and energy efficiency in IP/optical transport networks. Use case 1 contributes to the 3rd sub-challenge, with knowledge-based content caching and traffic offloading techniques for the future content-oriented mobile Internet. Use case 2, on the other hand, contributes to the 2nd sub-challenge with intelligent energy saving mechanisms at both the IP and optical layer. Finally, with in-network knowledge inference and learning based on raw context information, the project also addresses the 4th sub-challenge of extracting understanding from data. In summary, context information captured during network/service operation will be used to derive systematic in-network knowledge and intelligence in order to deal adaptively with both complexity and uncertainty and enable near-optimal network operation.

Planned Impact

The KCN project is highly timely, targeting both short to medium-term (5-10 years) and long-term (10-20 years) impact. It aims to push knowledge and intelligence into the network infrastructure, enabling real-time in-network decision-making and paving the way to "intelligent" future networks which will support new applications and services and open up new business opportunities. Knowledge as a Tool (KaaT) will accelerate the development of intelligent support for the future fixed and mobile Internet, the latter ranked first in terms of economic potential according to the Government's Information Economy Strategy, June 2013. In addition, Knowledge as a Service (KaaS) and as a Cloud (KaaC) will provide an open environment to different players in the Internet marketplace for exchanging knowledge, resulting in globally optimal network performance which will enable new applications and in new business opportunities. For example, contextual data could be used for large-scale user and content profiling activities while network/service operational data could be used to enable "service nesting" by infrastructure-less service providers.

Overall, the KCN project is expected to have the following impact in different sectors/areas:

* Major ICT stakeholders: All major communication stakeholders will benefit from the KCN results. Infrastructure providers (e.g. ISPs) will be able to deploy network platforms with intelligent control functions through KaaT, reducing substantially the complexity and cost of network operations and management. Knowledge exchange through KaaS/KaaC will enable contextual information and knowledge to be disseminated to infrastructure-less service providers (SPs), enabling them deploy and provision their own services. The beneficiaries in this case will not only include large SPs (e.g. Skype) but also small and medium enterprises (SMEs) which would like to deploy their own networked applications and services with minimum CAPEX/OPEX. Last but not least, end users will also benefit in terms of services with enhanced quality of experience and reduced cost thanks to more efficient network resource control by ISPs (KaaT) and to "harmonised" interworking between ISPs and SPs (KaaS/KaaC). In fact, world-class content services with affordable cost and consumer safety are key targets in "Britain's digital platform for growth", July 2013.

* The UK society: First of all, the KCN project will directly address energy efficiency in the ICT sector, which will contribute to the reduction of power consumption and CO2 footprint in the telecom industry. In addition, KaaS/KaaC will stimulate the generation of future business opportunities in the Internet marketplace, no least for SMEs to create revenues either through the deployment of own networked applications or through retailing user/content profiles (e.g. user preferences/interests, content popularity etc.). The resulting business opportunities will contribute to job creation and economic prosperity. Finally, interactive real-time applications in the fixed and in mobile worlds will not be a taboo discussion among Internet researchers and engineers but will become a reality.

* The academic and research community: Future Internet architecture has been a key topic in the networking research community in recent years. A key observation from relevant research activities is that the border between core networking and network management/control functionality tends to become increasingly blurred. We believe that KCN can play a key role towards the realisation of future network architectures with intrinsic knowledge and intelligence support. We expect that our approach will stimulate fundamental rethinking on how knowledge and control functions should be placed and used across different network entities. Furthermore, the resulting changed philosophy of network design will stimulate new related fields of research.
 
Description In its three years, the Knowledge Centric Networking (KCN) project has progressed significantly its research agenda and produced the following key contributions:

1) A novel framework in which distributed management entities adapt the network configuration based on their knowledge of the network state and the development of algorithms for the placement of management and control functionality (IEEE TNSM'2015). In addition, a novel knowledge gathering approach for software-defined networks that can provide management applications with frequent and consistent network state updates of fine granularity (IEEE/IFIP IM'2017 and CNSM'2017 papers). The solution relies on a decentralized architecture that can support a wide range of measurement tasks and requirements. Evaluation has shown that lossless monitoring can be achieved with minimal overhead, which can improve reconfiguration reactivity by reducing control-loop delays.

2) A new signalling framework and protocol that facilitates the knowledge exchange between decentralised decision-making entities (IEEE/IFIP NOMS'2016 and IEEE TNSM'2017 papers). The protocol enables the communication between management applications that execute at different network locations and allows them to form the necessary view of the resource state and subsequently achieve near optimal configurations. The flexibility and extensibility features offered by the protocol can enable different types of applications to share knowledge and control heterogeneous network resources.

3) In the context of D2D-enabled mobile social media delivery, we identified application scenarios on D2D-based connection relay service in not-spot areas which can benefit from the same knowledge maintained at the mobile network edge. Customised D2D connection policies have been derived that cater for both delay-sensitive and disruption-sensitive connection requirements. Based on different human mobility models we have shown that the proposed scheme is able to provide connectivity coverage through D2D for up to 96% of user requests (IEEE WiMob'2016 conference paper and IEEE Access 2017 paper).

4) Also in the context of media delivery, we developed a scalable approach to control the placement of content in distributed caching points (IEEE JSAC'2016 journal paper). This is based on parallel decision-making processes that enable fast reconfigurations. Experiments with exogenous knowledge on future request patterns, in a Knowledge-as-a-Service (KaaS) scenario, have shown that cache hit ratios can be significantly improved. Another outcome of this work is a tool that visualises cache management strategies and knowledge across the content delivery chain (IEEE/IFIP IM'2017 Demo paper). The tool has been made available online: https://github.com/cacheMAST/cacheMAST.

5) We developed a user Quality of Experience (QoE) estimation tool which is embedded at a mobile edge node and is able to continuously infer DASH video quality and buffer status on the client side. This tool has been used for DASH video caching and prefetching towards substantially improved user QoE performances (ACM ICN IC5G'2016 workshop paper). In a KaaS scenario, we have developed a mechanism that automatically reports radio network conditions to a mobile edge node, such that the corresponding content caching/prefetching/adaptation operations can promptly react to the captured dynamicity.

6) We developed an energy management scheme for fixed ISP networks aiming at adaptive and incremental reconfiguration of a small number of network links between sleep and active modes based on the knowledge of up-to-date traffic conditions (IEEE Systems Journal 2015 paper). This enables traffic diversion for run-time energy saving without causing traffic disruptions.
Exploitation Route The results achieved in the project have received positive feedback from the research community and papers have been accepted in high quality publication venues (e.g. IEEE JSAC, IEEE TNSM, IEEE Systems, IEEE CNSM, IEEE IM, IEEE NOMS and IEEE WiMob among others). In the ICT sector, the research carried out by KCN is primarily of interest to network operators (both fixed and mobile) and service providers. The former can use the solutions for automating (re-)configuration processes to a large extent, performing those at a faster pace than today and improving the overall user Quality of Experience. The KCN knowledge ecosystem is also expected to drive new types of applications and thus create new business opportunities for service providers. An additional area of applicability of our caching approaches is in railway transport where content is pre-fetched at stations and consumed while the train is on the move.
Sectors Digital/Communication/Information Technologies (including Software),Transport

URL https://www.ee.ucl.ac.uk/kcn-project/
 
Description We expect that the KCN knowledge ecosystem will equip the future Internet with the necessary intelligence for performing dynamic adaptation of resources automatically. At this stage of the project we have made significant steps towards that direction. The developed user QoE estimation tool and the DASH video prefetching technique (see Key Findings v) were demonstrated to the public during the 5GIC annual workshop in October 2016 and received very positive feedback from the industrial partners. In addition, the team has had specific in-depth technical discussions with a number of partners who have potentially further interests in using such techniques in their infrastructures (Telefonica/O2, BT and BBC). In January 2017 the team also conducted a demonstration to the delegates from ETSI Next Generation Protocol (NGP) Industry Specification Group. In 2018 the enhanced platform was further demonstrated in the annual Mobile World Congress (MWC) event which attracted attentions from a wide range of audience in the telecom sector. We envisage that our developed KaaT/KaaS/KaaC techniques will create significant impact on the realisation of cutting-edge 5G networking technologies circa 2020. The signalling framework and protocol developed (see Key Findings ii) can serve as the east-west interface in a distributed SDN setting and can be easily extended to facilitate the communication requirements of a wide range of applications. This can simplify the deployment of new services that require knowledge from different vantage points and thus create additional revenue streams in the Internet marketplace. The dynamic adaptation schemes that have been developed in the context of the project can allow network operators to make more optimal use of their resources and thus reduce the capital expenditure in the form of infrastructure investments. The availability of knowledge at different granularity and frequency levels contributes to the automation of decision-making processes, which is a key driver towards the reduction of operational expenditure. The cache management visualisation tool that was demonstrated during IM'2017 received positive feedback from researchers in both academia and industry. The tool simplifies the analysis of caching choices and allows the design of strategies for optimised content delivery.
First Year Of Impact 2016
Sector Digital/Communication/Information Technologies (including Software),Transport
Impact Types Societal,Economic

 
Description Demonstration of project's prototype in the Mobile World Congress, Barcelona, Spain, February 2018 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact Researchers from the KCN project gave a demonstration of the project's produced prototype in the Mobile World Congress held 26 February to 1 March 2018 in Barcelona, Spain. This demonstration made the congress participants, mainly people from industry, aware of KCN's concrete achievements and generated interest for further information on the project's approach and findings.
Year(s) Of Engagement Activity 2018
 
Description Demonstration of project's prototype in the TI3 Workshop - September 2017 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Researchers from the KCN project gave a demonstration of the project's produced prototype in the TI3 Workshop held on the 8 September 2017 at the Royal Society. This demonstration made the workshop participants, mainly academics researchers from other TI3 projects, aware of KCN's concrete achievements and generated interest for further information on the project's approach and findings.
Year(s) Of Engagement Activity 2011,2017