The session focussed on these themes:
Product Development and Engineering
Product Demonstration & Performance
The following provides a summary of issues discussed in the themed break-outs.
Product Development & Engineering
Consideration should be given to future projects funded by DCMS spreading the net wider to draw in technologies from adjacent industries (like automotive), but the duration of the projects needs to be longer to achieve technology transfer. Setting of Standards has always been a strong aspect of the UK tech sector. New applications of AI (an area of UK strength) can come into smaller equipment suppliers doing R&D in the UK. RAN Intelligent Controller (RIC) is an emerging area - the UK has a chance to contribute here, specialised niche software will be needed.
Awareness of enabling technologies is important, but the engineering skills in product platforms and a smoothing of the transitions from small cell to macro (re-use strategy) are critical. Increasing ecosystem visibility and consensus on technologies and architectures that work is critical. Larger OEMs manage production processes and supply chains that have 10s of thousands of component/suppliers - this is a highly complex logistics value proposition that is synergised with global R&D.
The limited attendance of UK MNOs in the discussions is disappointing; it would be beneficial to have more evidence of engagement at these industry dialogues. The Government must be realistic as there is huge complexity in the existing public networks, 1000s of engineering years of capital investment and R&D in the current set up.
Cloud super scalers have successfully entered the technology platforms of the MNOs, especially in the Core, they are interfacing direct to the RAN and so an important part of the opening of the interfaces in the RAN. New architectures will emerge here due to different paradigms of containerisation; this architecture thinking is penetrating into the RAN.
MNOs will always rely on their in-house acceptance testing, can we envisage this changing? Don’t rely too heavily on case studies from MNOs with limited legacy to show the way forward. However, new alternative Operators are entering as a result of the different ways in which spectrum can be accessed. Increased general awareness of the technology platform choices, and interworking with existing MNOs, can show a route to migration towards an Open RAN tipping point.
Energy Efficiency, Security, System on Chip (SoC), RIC (different variants of), adaptive fibre vs non-fibre xHaul solutions are worthy of special focus. There are niche and smaller players in the market in the UK; in both chip IP and product, enhancing their ability to scale and deliver some of the broader “tough stuff” is important. The shortages in the global semiconductor supply chain are a threat to expanding the supply side, larger established buyers are first served and even they are struggling with supply.
Private Networks (from which sector(s) needs evidence) can provide the diversity of deployment scenarios and use cases to prove out the Open RAN approach. Enterprise mature usage of cloud, their indoor coverage needs, offers routes to JOTS and neutral host which is an area of growth in the UK, can this be leveraged?
Can the UK Government more effectively signal International cooperation and collaboration mechanisms that set the appropriate policy direction?
NFVI and Core are becoming essential to the expansion of the ecosystem due to E2E (End to End) delivery requirements placed on System Integrators (SI)s. Mobile Network Operators (MNO)s prefer to deal directly with the Original Equipment/Design Manufacturers (OE/DM)s and thus SIs are brought in by OE/DMs, SIs as standalone scale entities is a slowly developing market. SIs will be better placed as new vendors enter the market. Commoditisation of the Open RAN components will help SIs, too much differentiation of vendors creates substantial costs. Commodity platforms and not just openness are important for the SI sector.
Neutral Hosts (NH) are challenged by the disaggregated model with software update handling. NH takes responsibility for E2E and has a large ecosystem testing, mirror labs are a high CAPEX (substantial barrier to entry). SIs may move to standardised components with commercial partners, with some customisations. SIs establish “standard” pre-integrated components themselves - this is their productivity innovation for competitive advantage. Continuous integration (CI) and continuous delivery (CD) strategies can help here.
SIs can have different operating models from MNOs, carrier grade and 5’9s reliability is not a universal need - enterprise / office environments can relax some of the requirements. Incumbent vendors have a loud voice on the strengths and weaknesses of Open RAN - SIs need to find ways to have their own voice.
The UK can and must develop technologies, not just the use cases. Private networks are a useful proving ground for E2E solutions, re-use and transfer to leverage solutions from MNO to Private is an area that SIs work on optimising. Approaches to accreditation at individual and company level should be explored.
Innovation is happening between new entrants, the edge is showing particular promise in this regard, still control of incumbents can limit scaling of solutions, especially when standards need
to be enhanced. The established business and operational models regarding sales and maintenance of test equipment can sometimes play against the UK due to challenges with skills retention.
MNOs driving for cell site gateways and interoperability of components is a noticeable trend, innovation in the UK could potentially go global. Radio Unit (RU) is a challenging component with lots of global OEMs looking at these, other components are worth looking at. Large OEMs are managing their in-house solutions and 3rd party solutions, how to encourage them to bring in more 3rd party solutions?
It is essential to note the importance of the service to the customer, it is everything to the MNOs, and any solution that they use in-place of the current model will need at least the same level of performance, show good integration, and have to demonstrate it at scale to ease concerns and speed-up deployment.
Highly integrated macro sites are not an easy win for Open RAN. In the early stages keep focus on simpler solutions in the rural and indoor. Different business models and solutions based on Open RAN that target venues could be useful. Neutral hosting in large buildings, campus like networks. JOTS is worth keeping in the Open RAN ecosystem. Whilst keeping pressure on Macro openness, the UK can pursue a dual strategy of building a roadmap from small cells to indoor to macro. Look at the performance of the smaller deployments to drive credible scaling of expectations of performance in the macro in the future.
UK may be able to have a good control of outcomes by maintaining focus on the performance of systems that leverage deep understanding of RF, data, AI, photonics. Open source helps with the supply and maturing of solutions, but skilling up is required. Commercial-of-the-Shelf (COTS) hardware is very restricted in suppliers. Is there an Open RAN equivalent for spectrum?
Substantial coverage networks are needed to provide proof points of readiness, Internet of Things (IoT) oriented use cases must also be supported by the network to scale device connections. Establishing roadmaps of proof points towards maturity of products is essential to build confidence. New open ecosystem strategies are emerging around some Universities leveraging “place” and locally required use cases, building on pools of excellence in skills.