Open RAN manager Arnaud Baron cultivates ecosystems
Open RAN has a yawning gap to make up on traditional RAN and engineers are often purists whose idealism is not a protocol suited to the environment of the networking business, which operates on a model of ‘make it timely, make it proprietary’. This is the unacceptable truth that the industry will deny until it’s too late to change. Arnaud Baron, (pictured) is the OPEN Ran and Product Line Manager for Amphenol Antenna Solutions, AAS, one of the more influential cultivators of the Open RAN ecosystem. Baron said he’s seen this blindness to reality even from one of the head gardeners of the networking system that trained him. In the interview below, Baron outlines his hopes and fears for the Open RAN movement.
Interview: Ever since Ethernet beat Token Ring, elegantly engineered systems seem to have lost out. Doesn’t the complexity of OpenRAN handicap progress? It gives engineers far more variables to think about while the dirty-hack brigade of proprietary tech merchants always seem to win out against the more inclusive and thoughtful alternatives, would you agree?
Open RAN is a big thing so it cannot happen overnight. The number one challenge with Open RAN is to catch up with traditional RAN technology. Open RAN is evolving quickly, but the gap with traditional RAN is huge. I like your statement about elegantly engineered systems. When in engineering high school a long time ago, we challenged our teacher because he was saying Ethernet was a rubbish system, but we could see that it was already successful. The teacher eventually admitted we were right and it was successful because of the cost benefits. I could tell you similar stories about people challenging Unix versus MS-DOS or ATM versus IP. You only have to see who’s won!
Fast forward to Open RAN’s tribulations. It’s disruptive and difficult to integrate. The so-called vendor lock-in that everybody complains about also comes with the benefit of single point of contact and end-to-end performance insurance, which is quite comfortable. Having worked for Nortel I can appreciate the pros and cons of Open RAN and traditional RAN – it isn’t black or white.
What policy recommendations did you make when worked for regulator Arcep? Was it an easy life compared with working for a vendor?
I was in charge of checking the reliability of the coverage maps issued by the operators. I also contributed to the template of open data files provided by the operators and describing their radio access networks, such as the location of the sites, antenna reference, azimuth, antenna height and type of backhaul. I also contributed to the content of the 5G licences at 3.5 GHz. In these licences, we specify the number of sites to roll out and their timeline. People do work a lot at Arcep. It is a pressurised job supporting the ministry of industry and dealing with media queries. The time around the 5G licence awarding was an especially frenetic period.
You were then recaptured by Amphenol, so they obviously need you! What challenges do they want you to address now you’re back home?
Our business is undergoing transformation. It is basically about antennas becoming active, meaning they are radio hosted in the antenna radome. [A radome is a portmanteau of radar and dome, a structural, weatherproof enclosure that protects a radar antenna.]
As an antenna manufacturer making passive antennas, we face the challenge of integrating active antennas into our products. The market has pivoted towards combined passive/active solutions. This dynamic was hurried by the arrival of 5G and the massive MIMO antennas.
My challenge is to drive the portfolio in line with this and anticipate what the antenna market will look like in the second half of this decade. So we do run an in-house active antenna development programme and we also work on integrated active-passive solutions.
Another key challenge is Open RAN. This is an opportunity for Amphenol, as it enables new players in the ecosystem, typically new radio vendors and new partnerships. Unlike rival antenna makers Huawei and Ericsson, we do not make radio units, so there is the requirement to partner with a radio unit specialist.
What are the major constraints that your customers/clients face when trying to build infrastructure? Is there a lack of skilled engineers? Lack of support from town planners? Is the government supporting you?
The main challenge with site acquisition is to find a new site at the right location in the search area. Or to upgrade existing sites with new technologies, like 5G for instance. Anything that can ease site acquisition is welcome.
The staffing challenge is in sub-contracting, often several layers of people, resulting in low skilled workers and a huge pressure on costs. Riggers are not paid much and this creates a vicious circle when combined with low expertise. In fact our customers, the mobile network operators, do not pay their installers much because of the permanent Capex and Opex savings they have to make.
On the other side, the average revenue per European user is very low in comparison to their US counterpart. A top executive of one European Tier 1 operator told us some years ago that he didn’t expect 5G to generate more revenue, so there’s be no rise in ARPU. And this is just the start of the story.
Forgive me if this is a stupid question, why do antennae have to be so ugly? No wonder the public reject them! Couldn’t they be shaped into more aesthetically pleasing guises? (I swear I saw one disguised as a tree in Las Vegas and I don’t think I was hallucinating). The French are brilliant at creating iconic public works of art. Could an antenna be shaped into some arty sculpture, without impeding its function?
We can shape antennas or create camouflage solutions but again the main driver is cost. Camouflage solutions are very expensive and so they are avoided.
Antenna are built in line with the law of physics and performance expectations, so they can’t be drastically changed. The appearance can be changed but, in my opinion, the mast, mounting hardware and cabling to the antenna are the biggest contributors to the entire aspect of an antenna system. We can’t control or dictate all of that. Again, a standard installation is much cheaper than any specific antenna integration. It invariably comes down to money.
What are the worst misconceptions about Antenna?
It’s a one-word fits all, like platform. People usually use the word antenna for: a complete cell site with radios, cabling, mast and one or several antennas. But an antenna is purely passive. Its technical definition is an interface between a conducted mode and radiated mode of signal transport. So it has no power, no electronics, no active component. Even in a so-called active antenna, there is a radio unit and a radiating part, the radiating part being the antenna itself.
Another aspect is the complexity of the design of a multi-band antenna. If you visit our factory you’ll discover the inside of a passive antenna and how complex and high-end technology it is. It’s not just a box with some technical bits and pieces. Most people ignore this complexity and technology.