Home > News content

Depth analysis: Lenovo is wronged!

via:博客园     time:2018/5/16 17:53:53     readed:120

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146400-305358158.jpg

There is a little misgivings about this topic.

Most of the comments made by our team are based on research, and many old readers should be aware of them. But with regard to Lenovo's comment, there is a little bit more or less

But to be honest, the recent attack on Lenovo has left a lot of people in disarray and, to be exact, very angry. Angry at individual opinion leaders abusing their right to speak, pretending to be angry at individual media, chasing onlookers who do not know the truth

1. In the context of the Sino-US trade war, the relationship between Huawei and Lenovo has sinister intentions: the incident has actually been restored in two words, the first being whether Lenovo supports Huawei or not, and the second, whether Lenovo deserves respect. The first point of this article is attached to a lot of authentic materials you can read and distinguish, Huawei officials have refuted rumors, the second point we can actually change a point of view, what is patriotic? Lenovo beat Dell and Hewlett-Packard to win the first place in China's PC market in 1997, and in more than 30 years it was estimated that taxes were also in the tens of billions. Is it a bit strange for many well-paid but rarely taxed media to smear Lenovo's unpatriotic status? Many of these ideas can't stand scrutiny, obviously with malicious attacks on businesses and entrepreneurs. Collectively called

2, Lenovo's business is not as bad as the outside world says: it may be that the public is too high on the Lenovo national brand, so it is very sensitive to the cyclical fluctuations in its business. But our team has deeply studied PC, mobile phone, enterprise data center service, cloud service and many other sectors, and believe that other aspects are still healthy and hopeful in addition to the mobile phone's obvious decline. Moreover, Lenovo is undergoing a profound strategic transformation. Once the transformation is done, the transformation will be a rebirth of the epic level. It's too early to evaluate Lenovo as a loser. We will have a depth analysis in the back of this point.

3. Public opinion has long suppressed Lenovo.

4, Lenovo's current transformation is the forefront of China's high-tech enterprises and globalization Enterprises: Lenovo Group's current transformation is simple to say there are three aspects, the first is the PC based B2C business in the new retail business strategic adjustment, which is in line with the trend; the second is enterprise level cloud services and data services. Business expansion and application development is similar to Microsoft's cloud business input; the third point is strategic layout and investment for AI. As a whole, as a giant of global operation and business globalization, as well as a year of about 300000000000 sales, this transformation is very profound, and it takes time. The challenges facing the global science and technology giants are very common, and the criticism of this is often due to their lack of global vision;

5. Lenovo does not speculate: the outside world often simply thinks that not sulking on R & D is speculation, but if you think the other way around,

In the end, the outside world has been biased towards Association, and of course the bias has two objective factors. First, the overall business and stock price of Lenovo is falling; the second is that there is no immediate effect, or on the way.

In fact, the above views can be expanded, we will expand in the later in-depth study of Lenovo, but here please allow me to make a perceptual assessment.

Obviously, the current incident has caused irreparable damage to Lenovo's reputation, which reflects many deep-seated social problems: first, China's self-media environment has long included Lenovo, Huawei, Enterprises such as Xiaomi and Baidu have obvious malicious criticism, which has something to do with the professional ability of individual media, and also with values. Second, some members of the public who do not know the truth have always failed to understand a logic

Why wrong association? He fought for 34 years for China's high-tech industry, and he was the first to challenge a Chinese company.

Lenovo needs time and Lenovo needs fairness. They also need more love and care for the country and people they are striving for.

Attached 1: the 5G Laboratory of the comprehensive Lenovo Research Institute and several authoritative media contents:

1, background: channel coding is one of the most important components of the 5G communication standard. It has a great impact on the implementation of the chip and the cost of the patent. All the companies have paid great attention to it. In the process of 3GPP making 5G standards, three coding schemes are introduced, namely, Turbo code (turbo code), LDPC code (low density parity check code), and Polar code (polarization code). The three codes are as follows:

In communication systems, channel coding is required for data channels and control channels. Because these two requirements (including block size, bit rate, reliability, encoding and decoding speed, retransmission mechanism) are different, they are usually carried out in different coding modes and are implemented by separate circuit modules in the chip. In 3G and 4G systems, Turbocode is used in the data channel, and the octave convolutional code (TBCC) is used in the control channel. Convolutional code has a long history, simple coding and decoding, enough performance for control channel performance, and no effective patent. It is necessary to point out that, for each channel, under the premise that a coding method can meet the requirements, a variety of coding methods will not be necessary to increase the complexity of the system, improve the cost of hardware, and also use the manufacturer and the household to face a higher patent fee. It is a bad way to avoid it.

In the process of 3GPP 5G standardization, the decision on channel coding was made in two sessions. The first meeting was held in Lisbon in October 2016 with RAN 1 #86 bis. Only data channel coding in eMBB scenarios is fully discussed in this meeting. At this meeting, three data channel coding schemes were proposed, namely, Qualcomm, Samsung, LDPCs from companies such as Alcatel-Lucent and Shanghai Bell, with Huawei's Polar codebook, and Ericsson and several Japanese companies recommend a combination of LDPC Turbo code. At the start of the meeting, each of the three camps put forward their own proposal that way forward).Way forward was a formal working procedure for 3GPP that could be passed without objection from other companies. Since this is the first time a positive statement has been made (only companies that support individual programs are recorded), all three are the original ones and none of them have made it through. For the consideration of LDPC technology and patent, Lenovo chose to support LDPC. It is important to point out that this time China Mobile did not appear in any of the camps. The three way forward come from three different camps. Lenovo supports LDPC's way forward R1-1610767.

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146435-923138076.jpg

Since no consensus was reached on the first vote, the chairman asked the companies what they wanted. At this point, many companies, including Huawei terminals, have switched to LDPC Polar hybrid solutions, while only Huawei still insists on using Polar code as the only solution for data channel coding (see chart below). It was only a test of the meeting's meeting by the chairman, not a formal vote. In the test intention, Lenovo expressed support for LDPC as the only coding scheme.

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146439-881033036.jpg

The first two intentional voting did not make final decision on data channel coding. The three camps made their respective technical statements. At this point, LDPC's technological superiority has been recognized by more companies. The meeting then conducted a reverse vote (several options were recorded for those opposed to the company). HUAWEI sees the Polar code as the only coding scheme for the data channel without the support of other companies. It also proposes to vote on the size and size of the two code blocks, resulting in the emergence of a data channel long code and a short code. In the voting option, data channel only using Polar code has been excluded. The original Polar code camp, which has agreed that LDPC is used on long codes, is only trying to use Polar code on short code (Alt2). The record of this vote is as follows:

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146429-1412909812.jpg

Lenovo and Motorola voted against the LDPC Polar hybrid scheme, Alt2, for concerns about the use of both coding schemes in data channels. In this vote, all companies have no objection to LDPC's use of long codes, while three views on short codes remain at loggerheads. On the basis of this vote, the meeting finally reached a decision on the long code and left the short code to a later meeting. The meeting also confirmed that the encoding methods for other scenarios URLLC, mMTC) and for eMBB control channel are all reserved for subsequent meetings. The figure below shows the final decision of the RAN1#86bis conference on the coding scheme:

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146424-1378843621.jpg

From the above facts, it can be clearly seen that in the Lisbon conference, Lenovo was chosen entirely from the interests of technology and company. The failure of HUAWEI's Polarcode on long codes is because the technology itself is not mature enough, and has not been approved by the market and most companies. HUAWEI finally voluntarily gave up the Polarcode short code, which has nothing to do with Lenovo.

After this meeting, in the United States, out of strategic considerations, Lenovo changed its view on Polar code, so that it gave Huawei Polar code full support at the next RAN1#87 meeting held in Reno in the United States, and gave short codes to Huawei's data channel. And control channel coding scheme, both voted in favour. Regarding short codes for data channels, Huawei's two proposals at the RAN1#87 meeting were recorded as follows:

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146425-1317488944.jpg

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146536-1091564622.jpg

It can be seen that Lenovo gave support to HUAWEI's two proposal on data channel short code. Because of opposition from Qualcomm, Samsung, Ericsson and other companies, HUAWEI's proposal has not been approved. The final meeting decided that the short code of the data channel is also LDPC, which is consistent with the long code, which is conducive to the simplification of the terminal implementation.

HUAWEI's proposal for the control channel coding scheme has also been strongly supported by Lenovo and finally passed.

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146550-586421673.jpg

The final resolution of the conference on data channel short codes and control channel coding schemes is as follows: The working assumptions in these conferences were confirmed in subsequent conferences. The data channel and control channel coding scheme of 5G in the eMBB scenario has so far been Finally ended. It is clear that at the Reno meeting, Lenovo gave its full support to Huawei. Huawei's failure in data channel short code can be more regarded as LDPC wins because of its technical advantages. Huawei's Polar code finally won the control channel. Lenovo's support was indispensable.

Remarks: the drawings in this paper are recorded by the chairman of the conference (Chairman

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146558-930528383.jpg

Content of the 3G 5G dispute: according to the 3GPP plan, 5G will be achieved by 2020 and the first version of the 5G international standard will be completed in June 2018. 5G for the entire industrial chain

From the perspective of development speed, the current 5G standard has defined three major scenarios, namely, eMBB, mMTC and URRLLC. Among them, eMBB corresponds to a large number of mobile broadband mobile services such as 3D/ ultra high definition video, and mMTC corresponds to the large-scale Internet of things services, while URLLC corresponds to such services as low delay and high reliability, such as pilotless and industrial automation. Among them, video scenes and human driving will become the most competitive areas.

For consumers, the highlights of 5G services will also be concentrated on two aspects: Ultra HD video application and AR/VR application. For enterprises, 5G will create value for automobile transportation, energy / utility monitoring, security, finance, medical health, industry and agriculture, especially for the large-scale and key application areas that require low delay.

4. Other data: on January 30, 2018, the State Council Information Office held a news conference. Zhang Feng, chief engineer of the Ministry of Industry and Information Technology, pointed out at the conference that Y5G was in the critical stage of setting standards. The International Standards Organization (3GPP) will complete the first version of 5G International Standard in June this year. 2018 is the key year for 5G standard determination and commercial product development. It will rely on the third stage of 5G technology research and development experiment and focus on

Attached 2: know how technical experts comment on how to look at Lenovo's vote on 5G standard? "

Author: Guo Yuechao

Links:Https://www.zhihu.com/question/275922735/answer/392102424

Source: knowing

First of all, I graduated from 14 in July to August in 16. I was lucky to work in the standard team of Beijing in about two years. It is now about 16 years' 10. In November, it is almost a year and a half. Later, after leaving the association, it has gone out of the communication circle to do some isomeric calculations and learn something about it deeply, but something that happens in this circle is probably a bit of a understanding.

At the time of the work in the association, we did some research on waveform, millimeter wave, access technology, and developed simulation software to evaluate HUAWEI's SCMA access technology. From my own research experience, there are still some understanding of the channel coding technology routes and technical solutions.

Looking at a lot of the replies above, there are a lot of emotional things, and more importantly, everyone is identical, or intentionally or unintentionally ignoring the mode of operation of the 3GPP organization itself, and what is the realistic driving force of the organization's work and a large amount of money to take part in the meeting.

If you want to understand these issues and why Lenovo adopted such a voting strategy, you need to start from scratch to understand the 3GPP organization.

First, 3GPP is a purely business - oriented technical meeting for business interests. It is simply that many companies around the world spend a large amount of money to register for this meeting. Each meeting is held in various cities around the world, and the travel costs of the company's employees are a lot of money. The company is willing to pay such a cost, and there is only one purpose to achieve the maximization of the company's own interests. If it is divorced from the main line, it is impossible to understand the behavior of each company.

The way to achieve corporate benefits in 3GPP is to make as many reserve patents as possible, as well as derived technical solutions in the final standard manuscript. Here is a very professional concept called read on (I don't know how Chinese should be translated). If the text description of a patent and the text description of the standard text can match, the patent is a standard patent, the amount of gold is very high, if there are a lot of standard patents can form a standard patent pool, that is, can attack, such as high pass, everywhere by its own patent pool charge, make a big deal. The family is very angry and can be used to defend. Although the patent pool is not big enough, it can still be used to defend, and you don't want to argue with me. And the purpose of the patent is, get free action of business, so that the company in a certain field can do business freely, in white, is free to make more money, do not have to worry about access issues.

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146577-1793978023.jpg

The next thing is to introduce a very important concept called a proposal, equivalent to a technical solution. In 3GPP, the company proposes its own proposal (proposal) for a particular technology. As in the picture above, the blue is the proposal number. In fact, there are a lot of proposals on each issue at each meeting, if each company has a proposal. It is not enough for the proposal to discuss the time of the meeting. Therefore, before the meeting begins, the chairman will choose some typical proposals for discussion. So the chairman of the conference has a certain discretion, which can be used to draw the direction of the discussion to the favorable direction of the company that the chairman belongs to. To ensure the fairness and justice of the whole organization.

Similarly, the company consumes the resources to write the proposal, also in order to make the technical scheme of the patent of the home, including the entry, so that the technical plan of the own proposal is passed, the editor of the related conference will be based on the proposal and write the final standard manuscript, then the reserve patent of the own company There is a greater probability that read on standard manuscripts will become standard patents.

Here I am using a bigger probability, not a certain proposal that you dominate, that you can make your own technical scheme a standard, and usually the company will compromise and modify a lot of content to allow other companies to support the proposal. On behalf of people around a circle of other companies, the purpose is to add or cut down the proposal by modifying the proposal to include the possible reserve patents of other home companies in exchange for support. And it's just a compromise on the meeting site. More compromise may happen outside the meeting, for example, in order to get support, you can reduce a bit of patent costs. The supply chain company gives you a priority, and the opposite is that business is a compromise and can be talked about. A bit like the modern parliamentary political model, political issues are advancing like compromise and trade-offs.

The biggest vote on the Lenovo dispute occurred at the RAN186bis conference, on the eMBB data channel (data channel), and in order to better understand the vote, a discussion of a meeting was needed.

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146570-785319372.jpg

It was basically the three camp leading LDPC camp, the Polar led by HUAWEI, and the Turbo camp led by LG, and the respective camps were supported by their respective camps. Each camp had both domestic and foreign companies. But this time there was no voting, because the field technical discussion was very intense, so the chairman of the conference voted to put the vote. At the next meeting.

Then there are the most controversial vote screenshots:

orgsrc=https://images2018.cnblogs.com/news/66372/201805/66372-20180516172146579-949567209.jpg

The voting is to determine the direction of technology. There are four main directions:

1, high pass lead pure LDPC scheme

2, the HUAWEI led pure Polar scheme

A Twin Turbo Polar dual Code Scheme

A dual Code Scheme for Polar

Among them 2, 3 basically no one to support, should be directly sentenced to death, the rest of the dispute, most Chinese companies support the program 4, and Lenovo supported the high Qualcomm program 1, so there are many emotional descriptions, only to vote with the high Qualcomm proposal to conform the association, in fact, it is very good Something the matter。 First of all, the vote of the vote association does not work. In the final Agreement of the last figure, Polar is still a short code candidate. In fact, if you carefully look over the back of the record, Lenovo invested HUAWEI when deciding whether to use the Polar code as a short code coding scheme for the data channel. Because the weight of voting is not enough, the short code part of channel coding is still LDPC. But soon, the China led Polar scheme, which is the coding scheme of the control channel, won a victory. That vote Lenovo also voted for HUAWEI.

Back to this controversial vote, this vote is about the direction of technology. As far as I know, Motorola has always had a 4G layout in the direction of LDPC, and MotM has some patent reserves for studying the layout of LDPC. The vote is also in line with one's own interests, that is, if the technical route moves towards pure LDPC, then motM's patents will be useful. In addition, don't we wonder whether Xiaomi and MediaTek supported Qualcomm at the last meeting (see above) and even did not use Polar codes as candidate codes for channel coding? why did they support LDPC Polar at this meeting? Nothing more than compromise with each other, Lenovo as a rational voter feel that under their own patent, option 1 is more revenue, I think it is not to blame.

A step back, even if Lenovo polls Qualcomm's lead program 1 must have hurt the interests of the country? Here, I first clarify that the national interest is an abstract concept, rather than an instantiation of a company, a person, but a general concept. In a variety of 3GPP research, a large number of companies are betting, HUAWEI ZTE has Polar code patents, LDPC patents, high Qualcomm and other companies, and other companies that are not very powerful are often studying coverage far less comprehensive, so for a new technology direction Polar code, large companies can invest a lot of human and material research in a short time, the result is that Polar code technology patents may be more centralized, and LDPC is a relatively early technology, relatively more people, relatively scattered patents. This is why the vote in the upper drawing, the reason that millet cast high Qualcomm, although LDPC is a high - pass lead, but because of the dispersion of patents, high pass can not be so strong to the argue patent fee, this is a small company for their own interests to consider a voting strategy.

Therefore, the choice of a simple LDPC scheme may be due to the problem of patent decentralization, and the comprehensive earnings of all Chinese and small and medium-sized companies may be better than the emergence of several patent giants (HUAWEI, ZTE, Qualcomm, and HUAWEI's patent fees for Chinese companies). Of course, this is just a possible path of game. In reality, everyone does not know the actual situation of the patent on the other hand. It is not the game of the other side, the game is not visible, it is complicated, the voting strategy is very complex, and the result of voting is varied.

But I think we have to recognize that the multilateral conference framework is an effective way to solve the complex interests of different countries and industries. The premise is to admit that the company must be a rational voters based on their own interests and fight for their own interests, but the final result is to compromise through a number of complex games. All participants can benefit from it and technology can also develop. Just like in the free market, everyone is most interested in his own interests, and the end result is the welfare of the whole society.

The worst way is to instantiate the national interest , to make exclusive narrative , and this exclusive narrative is easy to enter

So, let go of the company's own interests in the 3GPP framework to play a full game, ultimately all the companies have benefits, the corresponding national interests can be reflected, and what really needs to do is to monitor the operation of these meetings, to ensure that these meetings can be open, transparent, fair organization operation.

China IT News APP

Download China IT News APP

Please rate this news

The average score will be displayed after you score.

Post comment

Do not see clearly? Click for a new code.

User comments