The UK5G Innovation Network, a UK government-led programme to bring companies together for the development and deployment of next-generation cellular connectivity in the nation, has released a report into network architecture and security – and calls for new technologies to prevent the need for security compromising 5G’s promise of high performance.
A collaborative effort drawing in results from the AutoAir, 5G Rural First, Worcestershire 5G Testbed, and the University of Surrey’s 5G Innovation Centre, as part of the Department for Digital, Culture, Media, and Sport (DCMS) Phase 1 Testbeds and Trials Programme, the 5G Network Architecture and Security technical report describes an environment where a “combination of increased network-to-network complexity, end-to-end cross layer system security and critical applications will mean conventional security methods will not be feasible.
“New technology will be required to meet these challenges to prevent conventional security approaches compromising the required 5G performance,” the report concludes. “Context-aware networks and artificial intelligence (AI), which can process context transfer patterns and correlate them with user, device, application, and security context meta-data to make predictive decisions, present the UK with a significant innovation opportunity. This will assist the network to make sure the system set up is one step ahead of the dynamics of the UE [User Equipment] behaviour and context, therefore predicting and pre-validating the required end-to-end security and connection in advance of the UE requesting the service.”
The paper describes the four UK 5G testbeds and evaluates the scope of 5G’s security surface. In doing so, it makes four recommendations: “[The development of] cross-layer standards and framework enabling end-to-end security for prioritised critical or vertical segment use cases; a new approach to predict and pre-validate cross-layer user equipment (UE) connections, utilising Artificial Intelligence (AI) and context-aware networking, to ensure 5G performance is not compromised; [the formation of] an organisation that is tasked to help monitor and encourage good security-by-design practice, and set out and document an approach to designing secure 5G networks, applications and services; [and] standardisation and security tests and trials.”
The concept of context-aware networks with integrated artificially intelligence capabilities is not a new one, with organisations around the globe currently working on projects in that precise field for everything from improving security to making better collaborative use of increasingly-crowded radio frequency spectrum allocations. For this, the marriage of software defined radio (SDR) and general-purpose processing (GPP), to bring computational capabilities to the edge of the radio access network (RAN), will be key. Initiatives like the Telecom Infra Project’s CrowdCell, on which Lime Microsystems has partnered with Vodafone, and the LimeNET family – recently expanded with the LimeNET Micro, an ultra-low-cost all-in-one open source SDR and GPP device bringing the technology to within easy reach for smaller enterprises, education, and even hobbyist users – are perfect examples of the innovation that will be required to implement the recommendations made in the UK5G report.
The full 70-page paper is available now from the UK5G Innovation Network website.