The next decade: changes and prospects of mobile communications

What are we talking about when we talk about the future? At the opening forum of the China International Information and Communication Exhibition at the end of September, I suddenly remembered this sentence, because the theme of the opening forum was "Discussing the Next Decade". The invited guests (software developers, equipment manufacturers, consulting companies) expressed their views on this topic. At first, I felt that the prospects were extremely broad, but after careful consideration, I found that the value that communication operators can share is getting lower and lower, and the trend of mobile communications from high profits to low profits or even micro profits is becoming more and more obvious.

Recently, I heard a term called "digital deficit", which means that the expected future earnings of enterprises will decrease due to the substitution effect of digitalization and mobile Internet, just like the substitution of e-commerce for physical shopping malls and mobile games for traditional games. Unfortunately, operators are in the hardest hit area of ​​digital deficit. OTT applications have led to a rapid decline in voice and SMS revenues, and the trend is expected to accelerate further in the next two years. But relatively fortunately, the demand for mobile broadband (data traffic demand) has grown rapidly, and as a result, mobile operators can still survive, but their lives are more difficult than before.

As a member of a telecommunications operator, I have witnessed the earth-shaking changes in mobile communications over the past two decades. The current communication technology revolution has driven the great development of mobile Internet and affected all aspects of social life. In this wave of technological innovation, I have been thinking about this question: Will we be better or worse in the next decade?

Because the changes are so fast, it is difficult to make an accurate prediction. But it does not prevent us from connecting the past and the present into a line, and then making various guesses through the extension line. Even if we cannot guess, we can summarize the various opinions and conclusions at this stage and let everyone think and judge. The following will try to answer this question from the aspects of network technology evolution, Internet of Things, cloud and big data impact.

1. LTE will exist for a long time, and ONELTE will become mainstream

2014 is generally considered the first year of 4G in China, the first year of official commercial use, and 2010 should be the first year of LTE globally. Basically, the industry generally recognizes that China Mobile's heavy investment in 4G has promoted the development of the entire industry chain. There are several data that can cheer up China Mobile, which has chosen TD-LTE technology, and finally return to the mainstream. There are already 41 TD-LTE commercial networks in the world (18 of which are TD/FDD joint networks), and more than half of the world's 150 leading operators will use TD-LTE networks. From January to August, the national TD-LTE mobile phone shipments reached 67 million units, and China Mobile's TD-LTE users exceeded 40 million. The number of TD-LTE base stations worldwide reached 540,000, accounting for 60% of LTE base stations, of which China Mobile contributed more than 400,000.

If the emergence of the first in-car mobile phone (based on analog technology) in 1983 solved the problem of mobile communication, then as in-car mobility shifted to personal mobility, capacity became a bottleneck. The second-generation mobile communication technology GSM, which was determined in 1994, achieved the transition from analog to digital, solving the problem of capacity, but as voice demand shifted to data demand, insufficient bandwidth became a bottleneck. In 2001, the third-generation mobile communication technology CDMA (including W and TD) increased data bandwidth, but high-speed and large-volume data communication still had limitations, and high costs became a bottleneck. In 2010, the fourth-generation mobile communication technology LTE better solved the problem of low-cost, high-speed, and large-volume data communication. So where is the bottleneck now? Although many equipment manufacturers and research and development institutions have begun to discuss the next generation (5G) of mobile communication technology, from the perspective of demand, personal mobile terminals have basically been saturated. For personal customer needs, LTE and its evolving versions will be able to meet personal customer needs for a considerable period of time, that is, the current bottleneck has not yet appeared.

From the perspective of the evolution of technical standards, when GSM was determined as the second-generation mobile communication standard in 1994, Qualcomm had already proposed the CDMA standard (actually the third-generation mobile communication standard). When the 3G mobile communication standard was determined in 2001, Intel had already begun to propose the 4G mobile communication standard (WiMAX) to challenge the existing system. However, when the LTE standard was determined in 2010, the 5G standard was still a long way off, and it was not until 2014 that a preliminary "5G Vision and Requirements White Paper" was released. This shows that the newer generation of mobile communication technology (5G) is more complex, the content is more scattered, and the technology and standards themselves are still a long way from maturity.

Although there are currently two LTE standards (TD and FDD), almost all equipment vendors and operators unanimously call for "ONELTE", that is, TD/FDD converged networking. An expert report from Ericsson said that TD and FDD are 95% completely consistent in technology, and within Ericsson, LTE products only use one set of standards. The author understands the converged networking needs of China Telecom and China Unicom, because the country has allocated TD frequency resources and there is no need to waste them, but I don't understand why China Mobile also needs converged networking. Later, when I talked with an expert from Huawei, I found that China Mobile also needs it, because GSM will be decommissioned in the future, and the spectrum after GSM is decommissioned has been divided into FDD mode. What should I do if I don't converge this part of the spectrum resources? The current carrier aggregation technology is relatively mature, and making full use of spectrum resources will bring a service experience with greater capacity and higher peak rate. Moreover, the 5-mode 10-frequency terminal (it is said that the iPhone 6 already has 6-mode 20-frequency) has long been no obstacle to converged networking, so it can be determined that ONELTE (TD/FDD converged networking) will become the mainstream in the future.

2. Mobile communication technology will develop to the fifth generation, with a qualitative breakthrough in communication density and latency

(I) 5G mainly solves the problems of communication equipment density and traffic density

It can be seen from the "5G Vision and Requirements White Paper" released by the IMT-2020 (5G) Promotion Group in May 2014 that the development of mobile Internet and Internet of Things are the two main driving forces for the development of mobile communications in the future, which will provide broad prospects for 5G. On the one hand, mobile Internet will promote the further upgrading of the information interaction mode of human society, and provide users with more immersive and ultimate business experience such as augmented reality, virtual reality, ultra-high-definition (3D) video, mobile cloud, etc. This will lead to a thousand-fold development of mobile traffic. On the other hand, the Internet of Things has expanded the service scope of mobile communications. Looking to 2020 and the future, mobile medical care, Internet of Vehicles, smart home, industrial control, environmental monitoring, etc. will promote the explosive growth of Internet of Things applications, and tens of billions/hundreds of billions of devices will be connected to the network to realize the true "Internet of Everything".

The white paper gives several data. It is expected that by 2020, the number of mobile terminals (excluding the Internet of Things) will exceed 10 billion worldwide, of which China will exceed 2 billion (the author estimates that China currently has at least 1.2 billion), and the number of Internet of Things users will exceed 50 billion worldwide, of which China will exceed 10 billion (currently estimated to be 50 million). Based on the 2010 traffic volume, global data traffic will increase 200 times in 2020, of which China will increase 300 times, and hot cities such as Beijing and Shanghai will increase 600 to 1,000 times. Of course, traffic growth is exponential. By 2030, global traffic is expected to increase 20,000 times that of 2010.

According to the above vision of future mobile communication demand, the next generation of mobile communication mainly solves two density problems. The first is the device access density. More devices can be connected in the same community. It cannot be like a certain prefecture-level city in central China in 2002. During the Spring Festival, there were too many people going home from work, so that the system capacity collapsed and the machine could not be used after booting. The second is the traffic density problem. Not only can it be used, but more devices can use traffic at a faster access speed. Of course, the premise for realizing more access devices is the widespread application of the Internet of Things, the access delay of the SIM card in the device is lower, and it is truly integrated into industrial applications. This involves the third problem, the delay problem (service quality and experience).

2. 5G technology innovation comes first from antenna technology, followed by modulation technology

If we decompose the communication technology (wireless part), it can basically be decomposed into four parts: signal coding, channel coding, modulation and demodulation, and antenna. From the first generation to the second generation, the main change is in signal coding (the control channel separation has also changed significantly). GPRS in the second generation should be an important evolution of channel coding technology. The third generation mobile communication technology Code Division Multiple Access (CDMA) is a breakthrough in modulation and demodulation technology. The important technical feature of the fourth generation LTE is multi-carrier orthogonal frequency division multiplexing (OFDM), which is essentially an innovation in modulation and demodulation technology.

According to an expert report, the technologies in the first three aspects have reached their limits, and there is still a lot of room for improvement in antenna technology. Therefore, the technological innovation of 5G will have major breakthroughs in antenna technology.

I know that there is still room for improvement in modulation and demodulation technology, and it can be improved by 4-6 times in the next decade. However, the spectrum efficiency improvement brought by technological progress cannot meet the more than 1,000-fold capacity improvement requirements described in the 5G demand vision (rate, density, and latency each increased by more than 10 times), so experts generally focus on idle high-frequency spectrum resources (such as the 3.5GHZ segment, and even the 6GHZ segment). Therefore, more high-frequency resources and higher spectrum utilization efficiency together constitute the technological innovation of the next generation of mobile communications.

"Multi-stream convergence" will also be the focus of innovation in future mobile communication technology. Huawei held a mobile exhibition in a hotel next to this communication exhibition. It vividly demonstrated the efficient use of multiple network resources brought by "multi-stream convergence" in the future. It was mentioned above that TDD and FDD can achieve hybrid networking through multi-carrier aggregation, so is it possible to integrate 4G, 3G, 2G and even WIFI?

"Multi-stream convergence" is the solution to the integration problem of different standards and networks. To put it in a figurative way, after the application of this technology, when customers pass through the 4G coverage area, they can enjoy 100MB of mobile bandwidth. When entering the overlapping area of ​​4G and 3G, due to the superposition of dual network resources, they can enjoy 110MB of mobile bandwidth (assuming that the 3G bandwidth is 10MB). When entering the 4G, 3G, and WIFI coverage at the same time, they can enjoy 210MB of bandwidth (assuming that the WIFI bandwidth is 100MB). When leaving the overlapping area, it is automatically released back to the initial situation. This is definitely an ideal technology that is beneficial to operators. It should be noted that the coexistence of a large number of networks will cause a huge waste of investment and management resources. For example, in order to cope with the unfavorable competition of 3G in the past few years, China Mobile carried out large-scale WIFI construction. A simple estimate will show that the investment efficiency of WIFI will be as low as a jaw-dropping level.

3. Advances in communication technology do not necessarily bring benefits to operators, because they cannot break the "ARPU no growth" curse

If operators fail to open up new areas and stick to communication operations, there are only two situations that can bring benefits. The first is that more and more people use SIM cards (more and more users), and the second is that user ARPU continues to increase. But the reality is that the first situation is easy to achieve, and the second situation is difficult to achieve, basically impossible to achieve. The ARPU of existing users who have been used is basically stable, and the ARPU of unused potential incremental users is generally low. Ultimately, the average ARPU level is constantly declining.

Will you incur more communication expenses after using a 4G mobile phone? Obviously the answer is no. It should be said that communication expenses, like water, electricity and gas, have a basically stable proportion in everyone's disposable income. To some extent, as disposable income grows, the proportion is still declining. This is a general consumption rule.

There are reports that after upgrading to 4G, the traffic of 3G users in the United States increased from 0.5G to 1.4G, and after upgrading to 4G, the traffic of 3G users in South Korea increased to 1.6G. It is said that the situation in China is similar, and the average traffic of customers increased by 3-5 times after upgrading from 3G to 4G. The fundamental reason for the increase in traffic is not due to technological progress, but the adjustment of operator tariffs. Part of it is the adjustment of tariff structure, with the voice part reduced and the traffic part increased; more is the direct decline in traffic tariffs. There are reports that in the 3G era, 1 yuan consumed an average of 5MB of traffic, and 1 yuan consumed 15MB of traffic in the 4G package, and the actual 4G traffic tariff is 3 times cheaper. So far, the author has not been able to obtain representative cases to verify the consumption changes of Chinese operator users after upgrading to 4G.

It can be seen that a 10-fold increase in 4G bandwidth does not increase user traffic by 10 times. What determines user traffic is the tariff, that is, the cost of the operator. If the cost of traffic is broken down, it can basically be divided into the main equipment cost, wired backhaul, tower (property and electricity), management and maintenance, and business sales. The technological progress of the wireless part can only bring about a decrease in the first part of the main equipment cost, while the costs of the other four parts continue to rise as the cost of resources and labor increases.

For communication operators, the profit rate of SMS business exceeds that of voice business, and that of voice business exceeds that of traffic business. When the business changes from voice business to traffic business, the overall profitability changes from high interest rate to low interest rate, which is obvious. In traffic business, if the pressure of competition forces the further reduction of tariffs, but the cost cannot be reduced at the same time (such as the inability to reduce electricity bills, tower costs, and wages), then technological progress may lead to the transformation of low interest rates into small profits or no profits. The real benefit of technological progress lies in equipment manufacturers and users, but not necessarily in the interests of operators.

There are two possibilities to change the above trend. The first is the explosion of the Internet of Things, which will infinitely expand the saturated market space. The second is to break the "ARPU no growth" curse, that is, to enter new fields, replace other industries, or replace other parts of user spending.

4. The Internet of Things is a real opportunity for operators, but the portion of the Internet of Things pie that belongs to operators is very small

The rapid development of wearable devices directly promotes the prosperity of the Internet of Things. The remote ECG monitoring exhibited by China Mobile at this communication exhibition is very representative. A 2 cm sensor (built-in SIM card) is attached to your chest, and you can open the mobile phone APP to read your ECG status and analyze abnormal points. You should know that heart disease cannot be detected when it is not attacking. Is it more convenient to go to the hospital for ECG monitoring or to bring this sensor with you to detect the real cause as long as you feel uncomfortable? Obviously the latter. At the same time, various "children's bracelets" and "smart patches" have also promoted the development of the Internet of Things. Mobile medical care, Internet of Vehicles, smart home, industrial control, environmental monitoring, etc. will promote the explosive growth of Internet of Things applications.

So how big is the incremental space created by the Internet of Things for operators? Take China Mobile as an example. Currently, the lowest monthly package for China Mobile's machine card is 3 yuan (a certain amount of SMS or data package), and the most commonly used one is 5 yuan (data service package). If the current average ARPU of mobile phone users is 50 yuan, 10 IoT cards are equivalent to one mobile phone user. According to the "5G Vision and Demand White Paper", China's Internet of Things users will reach 10 billion in 2020, equivalent to 1 billion mobile phone users at the current level (currently China's three operators have more than 1.2 billion mobile phone users, of which China Mobile has 800 million), which is at least equivalent to creating another China Mobile.

Although the Internet of Things brings opportunities to operators, the value of the communication part that operators get only accounts for a small share of the Internet of Things pie. Take a child safety bracelet as an example. The price of a bracelet is only 200 yuan, and the communication fee for one year is between 36 yuan and 60 yuan. Assuming 50 yuan, the ratio of communication to non-communication is 1:4. Take "budiu" children's shoes as an example. The package with a total value of 720 includes 4 pairs of shoes (used within 2 years) and 2 years of communication fees. The actual expenditure of 2 years of communication fees is only 120 yuan at most, accounting for 1/6 of the total value. It can be seen that in terms of pure communication channels, operators only account for less than 20% of the big pie of the Internet of Things. For higher-value industrial control, Internet of Vehicles, etc., the value share of the communication part may be even lower.

Unwilling to be just a stupid pipe, operators are eager to enter the field of Internet of Things applications, such as the Internet of Vehicles. China Unicom has established an Internet of Vehicles company and developed an Internet of Vehicles application platform. China Mobile is also about to establish an Internet of Vehicles company. The concept of Internet of Vehicles became popular six years ago, but more than six years have passed and there is still a lack of necessary killer applications. However, facing the current 140 million motor vehicles, no one will ignore such a huge potential market. What exactly is connected to the Internet of Vehicles? What applications are there? Before answering this question, we need to clarify what cars can be connected to? It is this question that has restricted the real prosperity of Internet of Vehicles applications for many years.

Tesla has given a direction to some extent, and currently high-end cars are basically developing in this direction, that is, large screens in the car (including front and rear seats). Media playback, broadcasting, air conditioning systems, and navigation systems can actually be integrated into a large screen (the car computer is tablet-like and large-screen). Part of the vehicle control and management can be achieved through the touch screen. If the communication module is loaded at the same time, it is true in-vehicle interconnection. Promoting and occupying large screens will be an important means of future car networking competition.

In addition to the large screen on the car, the OBD (On-Board Diagnostics) interface is another entry point that can be connected to the Internet. Insurance companies prefer to obtain real-time information such as the speed, route, and area of ​​the driving car through the OBD entry point, so as to achieve more accurate personalized insurance assessment. At present, many companies have launched OBD boxes, such as Tencent's Lubao box, etc. Customers can read various performance data of the car in time through their mobile phones, but the current application value is not great, because the information that customers need most, such as whether the tire pressure is normal, whether the windows are closed, whether the handbrake is pulled, whether the engine is turned off, etc., cannot be read out under the existing OBD. If it can be done one day, then the entire interface standard system, safety and legal risks will need to be rebuilt.

In addition, there are also "after-installed" devices that directly load equipment to create an Internet of Vehicles entrance, such as various navigation devices, vehicle service equipment, location anti-theft and other "after-installed" equipment, which is also the main content of Internet of Vehicles applications in the market in recent years (because car manufacturers are too closed, pre-installed applications are very troublesome, and the market is difficult to open up). If we want to make a prediction about the future of Internet of Vehicles, the market space and capacity of the first two are the largest, especially the first one, which is the key to the success or failure of Internet of Vehicles competition. For operators, no matter which direction the application of Internet of Vehicles develops, the demand for machine cards is certain, because mobile networks are the most suitable communication method for vehicle-mounted modules.

China Mobile's construction of a dedicated IoT network and dedicated number that started many years ago was a very far-sighted strategic investment. Once I asked the relevant product manager why we need to build a dedicated IoT network (only the core network, not the wireless network), and whether the existing network and card management system can meet the needs? The product manager gave me a scenario, and I immediately realized it. Take children's safety bracelets as an example. Since the tariff/charging model of the Internet of Things is completely different from that of ordinary mobile phones, when China Mobile sold a batch of cards to 360, China Mobile did not know which specific customer 360 sold the bracelet to. This is a typical B2B2C model. Here is a question. What if the customer removes the SIM card after getting the bracelet and uses it on the mobile phone? Can we charge the customer directly? The answer is obviously not. The dedicated network has the advantage here. The data has been prepared before the card is issued. The SIM card can only communicate with a certain server (IP or domain name) in the background of 360 for traffic. The dedicated network can realize this type of control at a low cost and conveniently.

The development of the Internet of Things will lead to "card dealers" having more say for the first time. Because IoT machine cards have different forms and different requirements, such as some require high temperature and high pressure resistance (or low temperature and low pressure), some require multiple plug-in and pull-out, etc., and the demand for machine cards is greater than that of ordinary mobile phone cards, "card dealers" have become more important and can bypass operators and directly face customers.

5. "Cloud" will become an important way of production and life, and software and applications are the future

What is "cloud"? Even experts who have been engaged in cloud computing for many years may find it difficult to explain it in one sentence, because "cloud" involves too many fields. As a result, many "cloud" products have appeared on the market, some of which have little to do with "cloud". Apple's "iCloud" is for individual users of Apple phones, and so is China Mobile's "Color Cloud", which mainly solves the problem of "cloud storage". "Alibaba Cloud", "Baidu Cloud", "Tianyi Cloud" and "Mobile Cloud" are mainly for corporate users, especially small and medium-sized enterprises, and mainly solve the problem of low-cost and rapid deployment of networks and IT for small and medium-sized enterprises, involving IT and network resources such as "virtual machines" and "elastic storage", as well as various cloud-based SaaS applications. The ultimate destination of "cloud" should be "big data", because the final result of "cloud" will inevitably lead to the concentration of data, which will be concentrated in one or several "resource pools". "Big data" is actually a value and application proposition, and "cloud computing" is a way to implement big data applications. Only the "cloud" method can realize the implementation of big data applications economically and efficiently.

How will it develop in the future? For software companies, the "private cloud" currently used by Huawei's R&D personnel may be a model. It is said that Huawei has 60,000 R&D personnel (accounting for 40%), and the office computer they use is actually just a monitor and a data cable. Every code operation is stored in a centralized resource pool in real time. After get off work, you can apply for background computing to test the program code of the day's work as needed, and you can know the test results of yesterday's code the next morning. The advantage of doing this is to make full use of computing resources and minimize waste. Of course, there is another benefit, that is, all actions are in the internal resource pool, which fully guarantees the overall information security of the enterprise.

For individuals, the "virtual mobile phone" exhibited by NEC is also representative. That is, after the mobile phone is turned on, you can only click on a "desktop application", and all operations can only be performed after entering the application. All videos, images, and file data are synchronized to the "cloud platform" through the network in real time. The mobile phone is actually equivalent to a monitor + calculator, and does not require any storage resources. The advantage of this is that on the one hand, it saves the cost of the mobile phone, and you don't have to worry about 8G or 16G. On the other hand, you don't have to worry about losing it because there is no local storage. Of course, this is an extreme case, but from the author's personal experience of using China Mobile's "Color Cloud", the mobile phone opens the "Color Cloud" client, and there is almost no difference between reading pictures and videos in the cloud and local in the WIFI environment. When LTE is popularized to a certain extent, like the ubiquitous WIFI, the "cloudification" of mobile phones and various personal terminals will inevitably become a common lifestyle.

Are there any applications that cannot be "cloudified"? Of course there are. When the application is closely connected to proprietary equipment, it cannot be "cloudified", such as base station equipment. When the application does not involve proprietary equipment, everything can be "cloudified" into an X86 server and an application board. According to this logic, the core network of mobile communications can actually be "cloudified" or "virtualized". The "virtual EPC" mentioned by NEC at this exhibition is actually exploring this direction. The ultimate goal is of course to reduce deployment and maintenance costs and more flexible configuration instructions and resources.

For communication operators, the direct result of "cloudification" is that deployment and operation and maintenance are more convenient and cost-effective, which means that the number of network operation and maintenance personnel will be greatly reduced. Of course, with the popularization of electronic channels, the number of front-line customer service personnel will also be greatly reduced. Another result of "cloudification" is that applications are completely separated from capability platforms and control platforms, and the design and development of various "SaaS" applications based on basic capabilities is what communication companies really need to do. Some experts have proposed that the first-class communication companies in the future will all be software companies. After listening to this, the author was deeply moved. If time could go back, should I learn software technology in college?

The future is uncertain and full of variables. What we are talking about when we talk about the future of mobile communications depends on the perspective and assumptions of our analysis and prediction. If we limit our focus to communication operators, China Mobile, which is one step ahead due to 4G, will be optimistic. When we look further, technological progress does not necessarily bring benefits to operators, and 4G does not necessarily make our lives better. For companies and individuals alike, the most important factor in whether they will have a good future is their own efforts and ability to adapt to the environment. Various predictions about the future only allow us to understand ourselves and the surrounding environment more objectively. (This article was first published on the WeChat public account "Technology Talk")

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.