Didn’t I say that before 5G has changed my life, there are already people gathering together to fight over 6G?
The thing is, although 6G is not yet known, the news a while ago has given people a smell of gunpowder.
Advertisement
First,A group of 10 countries including the United States and the United Kingdom issued a joint statement supporting the so-called 6G principles – “open, free, global, interoperable, reliable, resilient and secure connections”, but did not bring China into play.
After that,European and American technology giants such as NVIDIA, Microsoft, and Nokia announced the formation of the AI-RAN Alliance, intending to jointly promote the application of AI in future 6G networks, so that members can all have a bite of the “cake”, and there are no Chinese companies in it.
Advertisement
Good guy, 6G is not there yet, but the group has already picked it up.
In fact, as netizens said, foreigners have been working together in communication networks for more than a day or two. In addition to the “10-country 6G principles” and AI-RAN mentioned earlier, they have also engaged in various projects such as O-RAN and xRAN. Although it has been a group for several years, it has not achieved any great success.
Speaking of this, I think of Open RAN, which has been particularly popular in the communications field in recent years. This is also because it includes major operators and communications giants Cisco, Ericsson, Nokia, etc., but does not involve Huawei, and it has made a lot of big news. .
“Um. So what is this bunch of RAN and RAN stuff? ”
I know there must be a lot of question marks on everyone's head, but don't worry, if we want to understand the success of Open RAN back then, we have to start with what RAN is.
RAN is actually the English abbreviation of Radio Access Network, that is, radio access network, an important component of which is what we often call base stations.
So you can simply and crudely equate RAN with base stations.
Before the 5G era, the technology for building base stations was firmly in the hands of communication equipment manufacturers.
The role of an operator is just as it sounds, just responsible for operating the network.
Since the mobile communications equipment industry is an industry that requires heavy investment and accumulation, there are only a few major players until now – the global market is basically divided by the four giants: Huawei, ZTE, Ericsson and Nokia.
Due to various considerations such as industry status, technical barriers, profits, and product stability, these players have made the base station a “black box” model.
Although we can roughly know that this “black box” contains components such as antennas, radio frequencies, and basebands, and operates through dedicated software, the specific internal module divisions, component interfaces, and what the software looks like cannot be seen from the outside. Arrived.
Only the equipment manufacturer knows how the base station works, how to repair it if it breaks down, etc.
Everyone can see here that operators don't actually have much say in front of those communication equipment giants.
In order to break this situation, operators began to work together to seek changes.
Especially on the eve of the arrival of 5G, operators have just completed the construction of 4G networks. Before they can catch their breath, they are facing the pressure of 5G network construction.
“Who can withstand this?”
And compared with 4G, the construction of 5G network is more expensive, because although 5G is fast and has high bandwidth, its penetration is far less than that of 4G network.
According to industry insiders, areas that used to be covered by one 4G base station now require 4 to 5 5G base stations to complete network coverage.
You know, it has been difficult for operators to do communication business in recent years, and they cannot find new growth points, so they must find ways to reduce costs.
For operators, this idea became extremely strong before the arrival of 5G, which eventually led them to form a group in 2018 to form the O-RAN Alliance (the protagonist finally appeared), which includes China Mobile, China Telecom, AT&T, Vodafone has it, basically all major operators are involved, and there are also some equipment manufacturers.
However, the main reason is that operators are more enthusiastic in participating.
Because O-RAN advocates a “white box” open standard that is completely opposite to the traditional base station “black box” model, that is, Open RAN (open radio access network/open base station).
They have completely opened up the closed mode of traditional base stations and made the internal principles completely public. By formulating standardized interfaces and protocols, all manufacturers can build equipment according to the organization's standardized technical documents.
in other words,Operators want to turn base stations into “open assembly computers” where the accessories can be selected from different suppliers.
For example, choose Intel for the chip, Flextronics for the radio frequency unit, Dell for the processing unit, Cisco for the system integration solution, and so on.
The specifications and standards for connecting equipment from different manufacturers are something that O-RAN needs to bring together to discuss.
The green solid line is the interface standard developed by O-RAN ▼
In the eyes of operators, opening interfaces, promoting open source software, and introducing more small and medium-sized enterprises to participate in equipment manufacturing will help break the monopoly of large communication equipment manufacturers and reduce costs.
The reality is indeed that in such a large communications market in the world, only four or five giants play in the end, and the monopoly profits are very ridiculous. The operators can't wait to raise more fish to disrupt the market.
But everyone is a black box, and it is difficult for new suppliers to join, so operators have come up with the idea of forming the Open RAN Alliance.
emmm, what should I say? This idea of operators looks good, but it is very difficult to implement it.
The immaturity of technology is one aspect, and the other aspect is due to various frivolous operations by O-RAN members.
Let’s start with technical factors.
One of the core of the technical standards introduced by O-RAN is to use general-purpose chips such as X86 CPU to replace the ASIC-specific chips previously used in traditional base stations.
But anyone with a little bit of hardware knowledge should know that,Although general-purpose chips can do any job, they are not specialized. Once a specific task is involved, the energy efficiency is not as good as that of specialized chips.
For example: when Bitcoin mining first started, everyone used graphics cards for mining. However, after only a year or so, graphics cards fell out of favor, because specialized Bitcoin-specific computing cards only cost a few hundred yuan, but mining The efficiency is hundreds of times that of a graphics card.
Although the dedicated chip can only do one or two specific tasks, because most of its algorithms are hard-coded into the chip, it achieves true integration of software and hardware, so it is very efficient when working in its own professional field. High, the power consumption is much lower than that of the X86 chip.
After all, there are specialties in the arts ~
Someone has tested it before:The power consumption of 4G base stations using “white box” equipment equipped with Intel X86 CPU will be 10 times higher.
If it is a 5G base station, this gap may be more than 20 times because 5G base stations have more antennas and higher power requirements.
This is a bit uncomfortable.
Even more distressing:After RAN is decoupled, the number of interfaces and the number of software and components also increase. After mixing software and hardware from different vendors, debugging will inevitably become more complex, and consistency and reliability will also decrease.
Although O-RAN later introduced conformance testing standards, for operators and equipment suppliers who have no previous relevant experience, if they want to ensure everything is foolproof, they must either spend money to find a more professional third-party testing agency, or set up their own Support a team that can fully understand all aspects of O-RAN knowledge and skills.
Vendors providing O-RAN solutions ▼
However, either way, operators will spend the costs saved elsewhere in another form.
From an operation and maintenance perspective, Open RAN has more software and hardware providers, so accountability and maintenance will become more difficult, and it may not be possible to find faulty products as soon as possible.
There are also security risks. With the participation of so many suppliers, how to protect user privacy under the communication network is a huge challenge for operators.
After talking about the technical issues, let’s talk about the “human” factor.
First of all, the members of O-RAN all have their own thoughts.
Let’s take the three communication equipment giants who have joined the group as an example:
ZTE has added organizations, does not work but fishes, and has a relatively negative attitude.
Ericsson supports the intelligence and automation pursued by O-RAN, but is not too concerned about openness.
Nokia is the most active among them. It not only participates in the technology development of the alliance, but also cooperates with operators to promote product implementation.
Nokia and Vofunda’s Open RAN pilot in Italy ▼
Of course, this is not because Nokia wants to accelerate its own revolution, but because they have other plans.
Previously, Nokia failed in the domestic 5G centralized procurement and was at the bottom of the market. As a result, they could only rely on the European, American, Japanese and Korean markets to generate revenue. Open RAN technology is strongly promoted by these markets, so Nokia wants to seize the lead in other communications giants. Before entering Open RAN, you should first take advantage of the pitfalls.
Nokia’s share of the winning bid was even less than 2% ▼
Secondly, although operators actively embrace O-RAN, Open RAN equipment has rarely been used in 5G construction in the past.
why?
Still the problem mentioned earlier, the immaturity of Open RAN technology will slow down the construction of 5G networks.
For example, in the early stages of 5G network construction, US operator T&Mobile chose to use traditional 5G equipment in order to complete equipment upgrade and deployment before AT&T and Verizon.
In addition, although O-RAN pursues openness, because the alliance itself is not binding enough, it is still difficult to avoid closed operations by various companies.
For example, previous cooperation between Open RAN technology company Altiostar and Nokia and NEC (Nihon Electric) did not comply with O-RAN interface standards, but used private interfaces based on CPRI and eCPRI respectively.
In addition, due to the high cost of mixing and matching products from different suppliers, some operators will still try to choose a single supplier or dual suppliers when purchasing Open RAN equipment.
For example, German operator 1&1 chose Rakuten Symphony as a software and system integration provider.
Open RAN Technology Company, a subsidiary of Japan's Rakuten ▼
“If there is only one system integrator, then the risk is that you choose this integrator and take advantage of all the expertise they have. You are adopting Open RAN technology but you are dependent on them.” – Director of Network Architecture at WoFunda Yago Tenorio
In essence, if you only choose Open RAN products from a single supplier, it is no different from choosing Huawei or Ericsson in the past.
Eh. In my opinion, this Open RAN may just be lonely in the end.
As an old netizen who has spent many projects in the open source community for free, I adhere to the spirit of the Internet and am willing to vote for O-RAN, which pursues openness and open source.
But from a practical point of view, I still have the same view as before, and I am not very optimistic at the moment.
There is no other reason, because according to the past technological development trend, advanced technology must replace backward technology, low-cost technology replaces high-cost technology, and simple technology replaces complex technology.
But compared with traditional RAN, Open RAN is neither advanced, cheap, nor simple. Even the advantage of openness is almost being used by suppliers.
“If we want to overturn the existing industry structure in this state, I can only say that there is a long way to go. ”
But having said that, Open RAN is not completely without a future.
Some people say that if O-RAN members are really serious about developing Open RAN, they can bring relevant technical standards into 3GPP (standardization organization) and sit together with global companies, experts and scholars to discuss a more standardized , more open and more complete global technical standards, there is still hope.
However, there are so many vested interests in non-Open RAN, and it is difficult for newcomers to overcome the huge technical and patent barriers to replace them. No matter how you look at it, the vision of Open RAN is difficult to realize.
Unless…unless?
“Huawei is also coming in?”
Editor in charge: Shang Shangwen Q