More than just 1G more than 4G, what are the obstacles for 5G commercial use?

Although there is still a long way to go before 5G is officially commercialized, the challenges facing 5G commercialization are an issue that the industry has to think about now. At the 5G industry salon held by Intel recently, Wu Geng, Intel Fellow and Chief Technical Expert of Wireless Standards of the Communications and Devices Division, pointed out that 5G will not only face problems with the air interface, but also the industry needs to consider everything from terminals to networks to the cloud. In other words, 5G provides both new wireless functions and new network functions.

As we all know, 5G has advantages that 2G to 4G do not have, such as high bandwidth, low latency, and low power consumption. Therefore, 5G is also considered to be the basis for large-scale applications in fields such as virtual reality and driverless cars. However, it is not easy to realize the application of 5G. Wu Geng gave an example from the field of Internet of Things. He believes that the realization of Internet of Things needs to go through three stages:

• The first is to connect things together;
• The second stage is intelligence;
• The final step is to make the device autonomous.

Connectivity is the foundation of every generation of mobile communications, but what does it mean to be intelligent and autonomous?

[[191466]]
Wu Geng, Intel Fellow and Chief Technology Expert of Wireless Standards of the Communications and Devices Division

Wu Geng told me, "The requirements for computing and operation in this process are constantly and gradually upgraded, which requires the integration of communication and computing, especially the addition of artificial intelligence."

Obviously, when all devices have autonomy, the system will become more complex. Wu Geng pointed out that in the next few years, the complexity of system software and hardware will be at least 20 times that of the current level, and the risk of error will also be higher. Therefore, the autonomous system must be smart enough to detect and control errors.

For a machine, if it only stays at a very low-level form of connection and has no wisdom or intelligence, when we make the entire industry in the information society fully intelligent on a large scale, we will actually need artificial intelligence, more intelligence, integrated communication and computing designs to create a trustworthy autonomous system.

Therefore, compared with the past few generations of mobile communications which took the design of communication networks as the starting point, starting from the 5G era, the focus of the entire industry has been evolving towards the integration of communication and computing, which actually means that the requirements for computing power will become higher and higher.

Since last year, the upstream and downstream of the industry chain have proposed the concept of edge computing. The purpose is to migrate large-scale computing and high-performance computing that originally only existed in data centers to the edge of the network, thereby releasing the value of the network. At this time, the network is obviously not a pipeline in the traditional sense.

Therefore, 5G is not just an upgrade on the air interface (5G≠4G+1G), it affects the entire industry chain from network to terminal.

The following are some of the topics that received much attention during the event. Leifeng.com has made some deletions without changing the content:

About the Standard

What is the progress of 5G standard setting?

Luo Zhendong, an expert from the IMT-2020 Promotion Group of the Ministry of Industry and Information Technology: 5G is defined by two standardization organizations. One is the ITU (International Telecommunication Union), but the ITU does not specifically make technical standards. It discusses and evaluates the 5G standard proposals submitted by all parties and finally determines which one is 5G. The real specific discussion of details is in another organization called 3GPP. There are some differences between these two standardization organizations because 3GPP sets standard plans based on the needs of the ITU.

ITU will start the evaluation of 5G standards in the second half of this year, which means that all parties can submit proposals for 5G standards. They will go through a series of evaluations and discussions and finally be finalized in 2020. The industry is more concerned about 3GPP, which is discussing specific technical details. The first version of 5G standards will be completed in the middle of next year, which will target some 5G scenarios. The second version of 5G standards will be completed in the second half of 2019. This version includes the requirements of all 5G scenarios, and will form the ITU 5G standards based on this in the future.

Mainstream companies at home and abroad have a common voice: they hope to make a global unified 5G standard in the 5G stage. This has never been achieved in the past. To achieve this goal, both the domestic government and the IMT-2020 5G Promotion Group are very open to 5G work. We are very actively cooperating, including many foreign companies participating in our 5G R&D activities. Intel has also actually participated in some of our domestic 5G technical trials and has also participated in the application of many national projects. In this regard, I mainly want to emphasize the issue of unified standards.

Where exactly do domestic organizations have a say in 3GPP?

Luo Zhendong: We just mentioned unified standards. Although we hope to enhance our own strength and voice through technological innovation and incorporate more into international standards, we still need a broad environment of international cooperation. Of course, we also need to reach a consensus. On the basis of consensus, all parties should incorporate new technologies into the 5G standards through their own technological innovation.

So far, China's industry has done very well, including establishing good cooperative relations with foreign companies. For example, we use industry organizations as a platform, and China has signed some strategic cooperation agreements with Europe, the United States, Japan and South Korea. I don't want to emphasize who is stronger here, but I hope we can jointly define the standard, because no matter who owns the technology, the most important thing is to serve the entire industry and every user, and all parties need to reach a consensus. The emphasis is on considering innovation on this basis.

What role do operators play in standard setting?

Liu Guangyi, Director of Wireless and Terminals Research Institute of China Mobile Research Institute: From the perspective of standards, China Mobile has truly participated in the formulation of standards. Starting from TD-LTE, we have established our own professional team to participate in the formulation of standards. At that time, together with our partners, we finally achieved the goal of synchronous development of TDD and FDD, helped the entire TD-LTE industry to develop globally, and won the support of many operators around the world, achieving the situation that we have one-third of the world so far.

China Mobile started to lay out 5G research and standardization five or six years ago. In the 5G era, we have gained everyone's recognition and trust, allowing China Mobile to be responsible for the future needs and vision of 5G development and to organize everyone in 3GPP to formulate the vision and standards for the future development of 5G. At the same time, we are also recognized to lead the entire 5G future network architecture, because operators have a greater say in the development of future services.

Will 5G standards be unified?

Liu Guangyi: From the perspective of 5G standards, we and the world's mainstream companies advocate for a globally unified 5G standard. We believe that only a globally unified standard can maximize the scale of global development, and enable the entire upstream and downstream of the industry chain to benefit from this scale, reduce the cost of research and development, and ultimately reduce the cost of equipment, so that ordinary people can afford the network. This should also be the common goal pursued by the global mobile communications industry.

At this year's MWC, we took the initiative to co-host a summit with other operators and enterprises, including operators such as AT&T and Vodafone, as well as communication industry chain enterprises such as Intel, Ericsson, Nokia, ZTE, and Huawei. We advocate global unified standards, and we hope that the future will be a harmonious and unified ecosystem. Only in this way can we lay a solid foundation for the future development of 5G.

Dr. Luo also mentioned earlier that an early version of the 5G standard will be released in December this year, and the first version will be officially released in June next year, including two deployment plans. The second version will be released around the end of 2019, and these two may be what everyone is looking forward to.

About frequency band

How do you view the unification of 5G frequency bands?

Luo Zhendong: Spectrum is the basic resource for wireless communications, a bit like the concept of land in real estate. Now from 2G to 4G, how many frequency bands does a mobile phone need to support?

China Mobile may have experienced this well. Dozens of channels are concentrated in one mobile phone, and most of the space in the phone is occupied by these things, such as antennas and radio frequency. If 5G cannot form a unified frequency band, in the future, one mobile phone platform will have to support both 4G and 5G, and it will have to switch constantly in different countries. This is a very large project. Even chip companies like Intel may not be able to completely and perfectly solve these problems.

Therefore, the unified frequency will not only greatly reduce the R&D costs of communication companies and chip companies, but also greatly facilitate the use and purchase of mobile phones by ordinary consumers, including when you travel abroad. This is a win-win process.

In fact, the ITU held a conference on radio spectrum allocation in 2015. At that time, the ITU was not very successful, and the industry has been working hard to achieve that goal. For example, China was the first country in the world to announce that the 3.4-3.6, or 3.5G, frequency band would be used for 5G trials. We have long promoted the unification of the frequency of the 5G low-frequency band. The US government and the European Union have some plans for the high-frequency band. Now the frequency range is also converging, between 20 GHz and 40 GHz, and several unified frequency bands may be formed.

Once Europe, China and the United States can reach a consensus, I believe that other countries will follow the trend. Although the official coordination is not very successful, from the perspective of the industry, it is better than 4G and 3G, as if there are not so many frequency bands. This is also very good news for the development of our industry.

How to coordinate different frequency bands to meet the needs of different 5G applications?

Liu Guangyi: From the perspective of operators, we hope that the global frequency bands are unified, but this is a good ideal. Since different countries and different industries have their own development demands for the allocation of frequency bands, such as satellites and the military, the division of frequency bands in each country is different. For example, in the United States, the focus of 5G development is the 28G millimeter wave frequency band; China is currently considering 3.4-3.6, 3.3-3.4, or 4.4-4.5 and other frequencies. Of course, we are also considering high frequency bands.

From the current situation, different operators may have different conditions in the future, which may have restrictions or lead to different considerations. From a technical point of view, we hope to make full use of these spectrums and solve the needs of different scenarios mentioned above through collaboration between them.

For example, from our perspective, we are considering using lower frequencies: the lower the frequency band, the better it can meet the needs of IoT applications, because its coverage requirements are very high, so we are currently deploying low frequency bands; for human communication, after the development of 4G, we need larger bandwidth, higher speed, and lower latency. We need continuous bandwidth, first of all, we deploy technology, such as 3.4-3.6, and in the future we need 100M bandwidth as business support capabilities.

If we want to meet the needs of AR and VR in the future, the speed must reach 1Gbps, which requires a higher bandwidth. For example, the frequency band for future millimeter waves can provide a speed of several hundred megabits or even several Gbps.

From this perspective, no matter which operator, in the future development, they will adopt a deployment method that combines high, medium and low frequency bands. It’s just that everyone has different timings or the time points of the spectrum. But the final state should be like this, using different frequency bands to meet the needs of different deployment scenarios.