7 major uses of 5G, how many do you know?-first part

7 major uses of 5G, how many do you know?

The 7 major uses of 5G are described in detail as follows:

I believe everyone may be familiar with Moore's Law, but Shannon's theorem in the field of communications is not so popular.

I remember that this formula was strictly derived in a graduate course "Information Theory". The formula of this theorem indicated the elements related to the communication rate and where the limit value is.

When transmitting data signals on a channel with random thermal noise, the relationship between the channel capacity Rmax and the channel bandwidth W, and the signal-to-noise ratio S/N is: Rmax=W*log2(1+S/N).

Note that log2 here is the base 2 logarithm.

The above formula is simply that if you want to increase the channel capacity, you can increase the bandwidth or increase the signal-to-noise ratio.

Increasing the bandwidth is easier to understand, but the spectrum resource itself is limited and it is impossible to allocate unlimitedly.

Even if it can be allocated indefinitely, there is a key factor that limits the signal-to-noise ratio.

There are many ways to improve the signal-to-noise ratio, which can be done by increasing the transmission power. However, the country has strict limits on the transmission power of base stations and cannot be increased without limit. Even if it can, there are high requirements for devices, etc. High-frequency amplification is not a simple matter, and it can be improved by improving source coding and channel coding.

7 major uses of 5G, how many do you know?

Regarding some related technologies of 5G, I have compiled a brain map, hoping to let everyone understand 5G technology more systematically.

You can look at it by comparison. If this article is not very clear, you can search it online based on keywords.

There are many key technologies of 5G, so combining these technologies with the three major scenarios, each technology is developed to solve some problems in actual scenarios, which may be easier to understand.

7 major uses of 5G, how many do you know?

1. Standardization of 5G

1.1 KPI

★ Peak rate reaches 20Gbps

There is no stipulation on how much bandwidth to use. It can be met by 32 carrier aggregation. This rate is the peak rate of the base station, not the rate of a single user. This rate is shared by users within the coverage of a base station.

★ User experience data rate (urban area) reaches 100Mbps

There is also a description of the requirements for the user experience rate in more subdivided areas in the standard. For example, the rate requirements in the concert area where we are more concerned about the high density of people. The description of Broadband access in a crowd in 5G is that the overall user density is 50. When the activity factor is 30%, it needs to meet the user experience rate of 25Mbps downlink, 50Mbps uplink, regional capacity downlink [3,75]Tbps/km2, and uplink rate [7,5]Tbps/km2.

★ The spectrum efficiency is 3 to 5 times higher than that of IMT-A

1) IMT-A is a 4G mobile communication standard specification formulated by the International Telecommunication Union (ITU). The spectrum efficiency of 4G can be viewed in this way. The vision of 4G is to achieve a rate of 100Mbps at 20MHz, so the spectrum efficiency of 4G can generally be considered as 5bps/Hz.

2) According to KPI, the improvement of spectrum efficiency in 5G is still very obvious, and the improvement of spectrum efficiency will guide the improvement of technology more directly, because according to the Shannon formula, the increase rate can be used to increase the bandwidth, which is easy to do in many cases , But the problem is that spectrum is a scarce resource, and the supply can only be increased to a certain extent. According to this spectrum efficiency of 3 to 5 times that of 4G, 5G should theoretically provide a rate of 1.5Gbps to 2.5Gbps at 100MHz. On May 9th last year, Guizhou Unicom's first 5G base station was opened. In the field environment, the peak downlink rate of the 5G network tested by a single terminal under 100MHz broadband reached 1.8Gbps.

★ Mobility up to 500km/h

In the 4G era, this problem was actually not solved well, and the network on the high-speed rail was intermittent. The speed of 500km/h will produce a serious Doppler effect, and there will be some challenges for frame format processing. In addition, the extremely fast speed may produce frequent handoffs, which also poses some challenges to the stability of the data link.

★ The delay reaches 1 millisecond

TDD may not be satisfied.

★ The connection density reaches 1 million per square kilometer

The Internet of Things is blooming everywhere. This indicator does seem to be very powerful, but in fact, the advancement of the Internet of Things standard mMTC in 5G is relatively slow. The reason may be that explosive products have not yet appeared, and the current NB-IoT has not yet reached saturation. Although everyone recognizes the importance of the Internet of Things, it may take a while to wait patiently.

★Energy efficiency is 100 times higher than IMT-A

The more base stations are built, the more operators spend on electricity, which reduces energy consumption and is environmentally friendly.

★ The traffic density reaches 1Mbps per square meter.

Need more bandwidth and other new technologies to meet together.

7 major uses of 5G, how many do you know?

1.2 Standardization

The 3gpp organization is responsible for standardization. You can go to the official website to learn about the standardization process. 5G NR physical layer protocol, if you want to know the detailed technology of the protocol, you can download the protocol document to view.

The main content of the NR physical layer protocol is outlined in 3GPP TS 38.201V15.0.0 (2017-12). The physical layer contains an overview document TS 38.201, six protocol documents: TS 38.202, TS38.211, TS38.212, TS38.213, TS38.214 and TS38.215.

7 major uses of 5G, how many do you know?

2. A rational view of 5G speed increase

All kinds of speeds are blowing up, one will be 20Gbps, one will be 4.6Gbps, and the other will be 6.5Gbps. Why is there such a big difference? Are those 4.6Gbps definitely worse than 6.5Gbps?

1) 5G peak download rate is 200MHz 4.6Gbps below 6GHz.

This 6GHz refers to the carrier frequency below 6GHz, and 200MHz refers to the bandwidth. Students who are confused about the concept of carrier frequency and bandwidth can Baidu by themselves.

4.6Gbps is the peak rate. According to the 5G KPI, the spectrum utilization efficiency needs to be 3 to 5 times that of 4G. What is 4G?

The 4G spectrum efficiency is 5 (that is, the 20MHz bandwidth achieves a peak rate of 100Mbps), so according to the 5G KPI, let's calculate how much the 200MHz bandwidth should reach the standard. A simple formula calculates that the standard rate should be 3Gbps ~ 5Gbps.

2) Millimeter wave 800MHz 6.5Gbps (10 times the experience rate of 4G LTE).

Millimeter wave refers to the frequency band, the international mainstream is 28GHz, this refers to the carrier frequency.

800MHz refers to the bandwidth, and the high frequency band is good, the resources are quite abundant, and the bandwidth is 800MHz at every turn.

Actually, the spectral efficiency of this calculation is only 1.625 times that of 4G. This may be mainly due to the wider bandwidth, so the OFDM sub-carrier bandwidth used is also wider, and the spectral efficiency decreases when the sub-carrier spacing increases. But this rate is still very impressive.

It is worth noting that these rates are peak rates, which are shared resources under a base station, so your actual experience rate will not be so fast. There will be a scheduling algorithm on the base station side to ensure fairness, but it is possible in 5G. There will be no absolute fairness, and paying enterprise users may get more resource scheduling, which is no longer a one-shot.

In addition, it should be noted that the communication rate is bit, not byte, there is a gap of 8 times, including when you install broadband in your home, it is also bps, not Bps.

In summary, everyone remembers the 5G spectrum efficiency KPI, and then adds the bandwidth to know the peak rate, and this peak rate can only indicate your total capacity, which is different from the personal perception rate, but It will be clear where the bottleneck is.

Regardless of how much bandwidth is achieved, how much speed is achieved is rogue.

PS: Carrier frequency and bandwidth are two different things. The peak rate is related to bandwidth and spectrum efficiency, and has nothing to do with carrier frequency.

Just like a train, it is the number of cars, not the speed, that determines the size of the load.

3. Business integration point

3.1 Development of VR/AR Technology

With the continuous expansion of the AR and VR market, video streaming is bound to show significant growth, and the next-generation content format similar to 6DoF will place higher requirements on the network, and the upper limit of personal data rate demand will also jump from 200Mbps Up to 1Gbps, these will require more bandwidth to support.

Many companies doing AR and VR have already started gearing up and are ready to seize this opportunity. Everyone is enthusiastic about 5G, and they all want to get the ticket as soon as possible, create explosive models, and occupy the market.

5G is a communication technology that solves transmission problems. Many experience problems, content source problems, and resource problems that VR and AR need to solve still need to be solved by the industry. Of course, whoever solves the best solution is to cooperate with 5G. If consumers are willing to pay, they will take the lead in the market.

Let's calculate the bandwidth following the previous one. Take the 100MHz bandwidth currently allocated by China Unicom/Telecom Operators as an example. According to 5G requirements, it can provide a maximum bandwidth of 2500Mbps. If a smooth VR requires 50Mbps, the maximum is It can withstand the use of 50 people at the same time. Of course, this is a theoretical value. With some signaling-related overhead, this value may not be reached. This is still a long way from the concert site where everyone can watch the performance in real time from multiple angles. Of course, This is only the initial stage of the network, and more millimeter-wave frequency band resources will be allocated in the later stage. At the same time, non-interactive broadcasting can be used to ease the situation.

7 major uses of 5G, how many do you know?

In fact, AR and VR are not exactly the same, and the problems faced by the scenes are different. There is no distinction here.

And the future form of this product, whether it is the google glass or the projection, or a more advanced interactive method, remains to be seen, and I look forward to seeing more amazing solutions.

When I participated in a 5G exchange meeting, everyone talked about the application of holographic projection in 5G. It is indeed a cool thing for a late singer to stand on the stage again and continue to sing for everyone. In fact, at present It has been implemented on a small range of specific IP, but the promotion and development of this is currently limited by the content. The content needs time and energy to be carefully created, and the creation process often takes more time.

The actual equipment can be obtained through various leases, which is easier to handle, but today, looking at the three-dimensional effect created by the way of stage curtain construction, it is still a big challenge to apply it in reality on a large scale. It can be expected that there will be better ways of interaction in the future, just like in science fiction films, interacting directly in the air, on a full-length mirror, etc.

Today’s holographic projection needs to solve more of its own projection technology, content creation and other issues. Transmission is only part of it. In other words, today you can use optical fiber without 5G. If the problem is solved, only that optical fiber will affect the experience. , Then just go directly to 5G to create a perfect experience, but what needs to be solved today is more about the field of the scene itself.

Of course, the improvement of the transmission link will also bring many benefits. For example, large bandwidth and low latency transmission can be achieved. Then a lot of calculations and rendering can be completed in the cloud. Cloud machines can be expanded quickly and conveniently, even at any cost. Expansion of the cost.

Creating a more perfect user experience through this cloud rendering method is also a research direction. Another advantage of cloud rendering is that multiple isolated scenes can be combined to form a more interesting virtual world, just like we are playing online. Like games, VR not only requires bandwidth, a good experience is also more sensitive to latency.

Therefore, network slicing for VR may be a compromise network slicing between large bandwidth and low latency.

3.2 Application of network slicing technology

As a very important technology in 5G, network slicing is widely loved by operators.

Because through this technology, data packets can be classified, service levels can be established, and differentiated services can be realized.

The vernacular is that the rich can provide good services and guarantee the bandwidth guarantee delay, and those who don't have the money can put it later.

In the previous 4G standard, differentiated service QoS was also defined, but the original differentiated service is only for the access network, that is to say, differentiated service is only for the segment from mobile phone to base station. The original differentiated service is also Derived from demand, when the base station allocates Internet user resources and voice phone user resources, they are definitely different, and priority will be given to protecting voice users.

In the 4G era, operators use qci to classify services, but the granularity is relatively coarse. 3gpp specifies a total of 9 levels, 4 GBR (Guranteed Bit Rate), and 5 Non-GBR.

Network slicing can be divided into network slicing in the core network and network slicing in the access network. The network slicing in the core network is closely related to virtualization technology.

NFV (network function virtualization) and SDN (software defined networking), as the main technical support for network slicing in the core network, have received extensive attention and research.

The realization of network slicing in the access network is more challenging.

In addition to being used in different business models, the network slicing and access network need to provide performance indicators such as low latency, large connection, and high reliability at the same time for different business index requirements, and to ensure the isolation between network slices.

Alibaba Group is exploring applications related to network slicing. In the 4G era, there is actually no complete network slicing solution. The 3GPP protocol defines QCI (QoS Class Identifier). Different QCIs promise different packet delays and packet error rates. Operators approved QCI provides differentiated services for users and businesses.

The typical scenario of users with QCI=3 defined by 3GPP is Real Time Gaming (real-time gaming). Generally, data services of ordinary users are carried on QCI=6.

As shown in the figure below, in a certain stress test, the guaranteed user with QCI=3 is significantly better than ordinary users with QCI=6 in terms of average packet delay and jitter. When bandwidth resources are tight, users with QCI=6 cannot preempt. Enough bandwidth resources to complete the business, and users with QCI=3 can maintain a stable rate of 800kbps.