Standardizing a new paradigm in base station architecture
Around the turn of the last decade, a wave of excitement was sweeping through the corridors of Ericsson Research labs. New antenna-integrated base station architectures were emerging and looking forward, an exciting breakthrough in the feasibility of using millimetre wave technologies was on the horizon. This move up the spectrum would ultimately change how we approach base station architecture. Yet, no concept or standard had ever existed for operating or regulating these new kinds of base station architecture and such ultra-high frequencies. Below, Thomas Chapman, RAN4 standardization expert, reveals Ericsson’s role in bringing the Over-The-Air standard to 3GPP and laying the groundwork for today’s great technological revolution.
The base station sits at the heart of the network platform.
Traditional 4G LTE base stations contain one, two or possibly even four transmitters and usually operate on core band frequencies of up to 2.5 GHz, sometimes even 3.5 GHz and 5 GHz. This enables high-capacity and low-latency communication, enough for video streaming and optimal user experience of online services, among other use cases. With 5G, we have moved even higher up the spectrum, up to 7GHz and ultra-high millimetre wave (mmWave) frequencies between 24-53 GHz. This enables ultra-high capacity, high bandwidth and super-fast connectivity – signalling a watershed moment on our journey to future technologies.
The breakthrough in beamforming technology came around the turn of the last decade with the emergence of antenna-integrated base stations. At Ericsson, we realised before most that, as transmission technologies would soon change quite rapidly and drastically, this would also mark a paradigm shift in how we engineer base station architecture. To enable this shift, the industry would need to move from just a couple of transmitters, to literally hundreds of transmitters in a single base station.
Laying the groundwork for the active antenna system
Up until this point, the methods used to specify, measure and regulate traditional base station transmission and reception had been fully conducted (i.e. taken at the antenna connector). This meant that activity of testing would in theory require up to 256 cables to be manually attached to the base station. Furthermore, such testing would not capture antenna characteristics of the BS, which is an integral part of AAS performance.
The move into super-high frequencies meant that the traditional conducted measurements – used over many years for characterizing the behaviour of previous system generations – would no longer be feasible owing to the lack of a connector and the fact that radiated requirements include antenna performance (especially EIRP/EIS) and so simply cannot be measured conducted. The challenge was thus in engineering a new common approach to specifying, measuring and regulating transmission and reception using over the air measurement.
Today, of course, we know this as the AAS Over-The-Air (OTA) specification. Not only is OTA testing now an essential part of the mmWave specification and regulation, but it is also highly beneficial for low-band base stations with beamforming arrays. Yet, back in 2011, the very concept of OTA was not even part of the conversation for mobile network equipment.
Agreeing on a common standard for OTA testing would lay the groundwork for the evolution and deployment of active antenna systems (AAS). This new form of base station architecture is engineered on the concept of Massive MIMO (multiple input, multiple output) and beamforming, which enable mmWave transmission and, as such, enable the evolution of commercial 5G and beyond. By utilising hundreds of transmitters in an array and potentially operating on ultra-high frequency, AAS brings a more powerful user experience, as well as significantly increased capacity, coverage and performance in both the uplink and downlink.
As early advocates of this shift, we led the industry on a journey into unknown technologies and paradigms.
Moving from research to 3GPP
Moving up the spectrum presents a multitude of challenges. It is these kinds of challenges which we address at Ericsson System and Technology. Here, we occupy the space between Ericsson Research and Product Development. Typically, we build test beds, demonstrators and assess which technologies are likely to eventually be deployed on the market. We do this by collaborating internally, across our research and product development teams, and also externally across 3GPP and technology vendors.
Following the initial breakthrough by our standards teams in developing a specification for how the air interface may work and perform, we then assess which technologies are on the market, both in the short-medium and long term. As part of that, we also assess how the performance of silicon, filters, power amplifiers etc. are likely to develop. From that we need to write practical requirements so that we can regulate, for example, the production of unwanted emissions. Thus, we take the concept from the theoretical level (i.e. “how do we use this spectrum”) to the practical level (i.e. “what will the available technology actually achieve”) and on to the 3GPP agenda (i.e. “which standardized requirements are needed for successful deployment and operation of the technology”).
Finding a common vocabulary for progress
With the AAS and NR specifications, we lay the key groundwork for new OTA testing methodologies and specifications (for both frequency ranges, i.e., bands below 7GHz as well as mmWave), as well as the core requirements for the mmWave spectrum. Back when we started in 2011, this was a profoundly forward-looking process which, with hindsight, was to become extremely relevant in the 5G era. However, owing to the technical challenges of 3GPP standardization, the journey up the spectrum was extremely complex, delicate and multi-faceted.
At the time, our proposed approach to testing and requirements were new to the industry, which naturally led to some resistance. I’m sure that parts of the industry thought we were racing ahead and doing something rather exotic. However, I believe these concerns masked a poor understanding as to the scale of the changes needed to enable the new antenna array and mmWave technologies.
To overcome this early resistance, it was clear that we needed to establish a common vocabulary and understanding of the issues. We achieved this by leveraging and linking our expertise in research, product development and practical testing. For me, these first years of building a common consensus was how we eventually succeeded in bringing 3GPP on our journey. This, in retrospect, was critical in allowing us to progress the conversation. When standardisation of the New Radio began in 2014, and due to the urgency of 5G, there was then a strong need to quicken the pace of the discussion.
Developing an Over-The-Air standard
It is actually, technically speaking, not obvious how to make OTA measurements on base stations. Yet, the entire deployment of 5G hinged on this very challenge. This is where our research, system and technology and product development teams aligned to bring real technical leadership to the discussions and drive that agenda in 3GPP.
On this journey, we spent extensive time to understand the measurement uncertainties. If they would be too low, then extremely specialized test chambers would be needed. If they would be too high, then the products would need to allow additional margin for measurement uncertainty. Therefore, we set out to understand what was feasible to measure reasonably in order to set an expectation as to which tolerances and margins would be needed for testing.
In 2015, with LTE Release 13, we succeeded in including partial OTA testing on the specification. This profoundly forward-looking move, to include OTA under the LTE umbrella, allowed us to put a stake in the ground and establish the basic principles of OTA testing in 3GPP. We had started early, arguably much earlier than the industry. In doing so, we could simply port our LTE findings to the NR specifications which would come later.
From here on, we were also able to begin a more concrete discussion on testing and test methods, which we could then begin to present to the external world. As such, the whole industry, from the test industry to customers and regulators became familiar with the notion of a new specification for air interface testing. This paved the way for the specification of full OTA testing as part of LTE and NR in Release 15. At this stage, for the first time, it had become clear to the entire industry why this work was essential as part of NR.
Throughout the process, we ensured close alignment with research, product development, test bed platforms, technology and test equipment vendors, and measurement chamber manufacturers, to understand what was needed and what was possible. We invested heavily in practical testing, by testing various hypotheses in test chambers. As such, we were able to develop a number of novel concepts for assessing AAS performance, which we then drove in 3GPP. This included comprehensive technical analysis, simulation results, and field measurement results, as well as co-existence and power accuracy simulations, and the concepts such as co-location, directional requirements specification and an advanced method for determining emissions total radiated power (TRP). It is this approach which differentiates us. Our drive to focus on the real technology, which is underpinned by test bed and product performance, makes us a leader in the 3GPP arena.
At the beginning, we were proposing OTA concepts which other major competitors saw as unfeasible and unnecessary. Today, they are considered integral to the deployment of Massive MIMO base stations.
Looking ahead to future base station architecture
Today, we can look ahead to Release 17 and a move to new parts of the spectrum between 7-24 GHz and above 53 GHz. Again, this will bring with it a number of challenges. However, just as our technology leadership was critical in developing a new standard for the mmWave, we will continue to drive efforts to ensure the co-existence, interoperability and minimum performance benchmark for all radio frequency operations moving forward.
We’re invested in this for the long term. When we began on this journey, our motivation was based on a conviction that this would be needed in the future. Indeed, this is an example of our philosophy and role when it comes to technology and 3GPP.
Find out more about Ericsson’s work in 3GPP on our standardization page.
Learn about the advances in base station architecture in this advanced antenna systems white paper.
Read more about the work of Ericsson Research on our Future Technologies page.