If you're going to spend $3tn, what's another billion? Congress urged to inject taxpayer dollars into open anti-Huawei 5G radio tech A concerted push for the US government to fund research into open 5G technologies has gained additional momentum – with a bipartisan letter from 38 lawmakers to the leaders of the House of Representatives urging them to support the Open RAN initiative. Open RAN is an attempt by mainly American corporations to open up 5G …
How many of SCROTUS' stupidities will be easy to reverse or cancel when they get adult in as president?
I imagine that he wins (aaargh!) in November or finds some semi-legal way to block an election, it will be even harder to undo the damage when he does go as even the most blatantly stupid ideas become more acceptable with time. (See, for example, how many counties in the USA are still "dry"and how they do not allow those under 21 to even drink beer 86 years after the, semi, repeal of the 18th amendment.)
If they get rid of him, they can undo so many of his bad ideas that the USA might even lead the world into recovery after the pandemic. We can always hope!
This post has been deleted by its author
"Currently that technology is dominated by proprietary standards..."
Ironic that is stated to be a problem as the USA leads the world in making it so. Whenever somebody thinks of an improvement in anything, somebody (not necessarily the originator) will grab a patent on it - even if somebody else was already applying the idea, it's the first to patent that stops it being used without stumping up cash. It's well and good rewarding companies for spending on research, but few patents are awarded with that ever in mind.
I'm just waiting for a patent to be issued for the coronavirus DNA, followed by demands for licenses in order to produce vaccines.
Ironic that is stated to be a problem as the USA leads the world in making it so. Whenever somebody thinks of an improvement in anything, somebody (not necessarily the originator) will grab a patent on it..
This isn't really the US's fault, it's just the way the game is played in standards setting. The 3 main players in 5G aren't US companies after all, and part of the push towards 5G has probably been to escape patents, eg Qualcomm. But it's something that favors the big operators, so patent pooling can act to shut out smaller players, who may have to negotiate expensive licences to keep their tin in the game. But the tin may still end up being made in China, so back to the same risks. Apple's apparently moving some production to India, but still with Foxconn doing the manufacturing.
So open sourcing design might be one step, but to fully de-risk, it also needs production to be shifted.. Which also feeds into the end cost due to labor rates. And I suspect it'd be difficult to create an open system given the amount of IP already involved in components and elements of typical RAN designs.
This isn't really the US's fault, it's just the way the game is played in standards setting. The 3 main players in 5G aren't US companies after all,
You present the two statements in the wrong order. Because the US did not participate in GSM standard development, it shut itself out of future development. So, yes, it very much is the US' own fault. When Bell Labs and Nortel R&D, et al. were closed to save money the US effectively froze itself out of most of the development. At the same time, China saw the importance of being onboard and consequently through money at R&D.
You present the two statements in the wrong order. Because the US did not participate in GSM standard development,
But it did, or at least US companies did. Sort of. So GSM was an ETSI innovation, and the US kinda went along with it anyway. Then it morphed into ITU-land for 3G(PP)/LTE/4G & 5G. Then of course there was NTT who pioneered a lot of mobile stuff. US companies are still very much involved in the process(es) though. Huawei's main advantage has been a combination of R&D, and being able to churn out millions of bits of tin at price points attractive to network operators.. Which has been the challenge the US and RoW has been facing, ie China morphing from a simple low-cost manufacturing centre into an innovator and competitor.
Compare the amount of resources devoted to defining, protecting and enforcing "intellectual property" today with the situation 100 years ago. Heck, even 50 years ago.
"Intellectual property" has been pushed by the USA, among all countries, as a way to pretend that the inexhaustible realm of "ideas" is an eternally-expanding frontier that anyone can exploit. Because what's the alternative? - letting the new generation of go-getters have actual land and real property would mean - (gasp) - the rich of today letting go of some of the stuff.
Which they have no intention of doing voluntarily, and no American government has anything like an appetite to force them to.
So instead they've built up this edifice of "intellectual property", where people can stake out their claims and exploit them without dispossessing the already-rich. The only people disadvantaged are newcomers, and the poor, and who ever gave a toss about them?
So yeah, it's very much "America's fault". Fifty years ago we wouldn't have had to worry about this kind of nonsense. But this is the world America built, and now we all have to live in it.
"Currently that technology is dominated by proprietary standards..."
There's a bit of journalistic smokescreening going on there, the sort of stuff you'd expect from politicians but not engineers. Put simply, international standards are agreements that include a patent pool as a key component -- all the proprietary intellectual property needed to implement the standard goes into the pool which then becomes equally avaialbe to all members for an agreed single fee. Literally everything importat goes through this process.
The only reason why Huawei has got the lion's share of IP for the 5G standards is that they put in the leg work. US manufacturerxs were missing in action and what implementations they had were years behind Huawei's. Its a nuisance but also a fact of life -- and it proves that its obviously easier and cheaper to hire lobbyists and PR people than engineers in the US. This might be the undoing of Open-RAN -- I normally welcome open standards but you just know that as things stand at the moment if the US government has anything to do with it then it will not be open, they'll claim jursdiction over anyone and anything that's even peripherally involved in the project. (Fortuantely, as I noted its currently a whole lot easier to hire PR types than engineers in the US and the traditional way of making up the shortfall by importing engineers from overseas is currently not that attractive -- the visa system's a shambles, there's neither job nor immigration security, if you do become a permanent resident (and then a citizen) the IRS will 'own' you for life and the FBI will be snooping on you to make sure you're not talking to the 'wrong' people.)
Ah but these are proprietary standards not created by the US because the companies involved in those sectors believed they owned the market and did not need to innovate.
This is yet another poke at Huawei and a belated attempt to get the US outfits to try an come up with competitive technologies. The outcome being that the US prevents any or its "friends" from using Huawei so they have to use second rate alternatives that cost more. This is not going to be something that happens overnight unless (god forbid) they attempt to reverse-engineer Huawei.
That pressure was largely ignored, however, with Germany and the UK, for instance, saying they will not ban Huawei equipment from their national networks. A main reason for that ispure cost: it would cost each country billions of dollars more to go with more expensive alternatives.that the Americans have consistently failed to any evidence to what are therefore by very definition unfounded allegations, and it is generally believed that the Americans are trying to exclude a competitor with better functionality at a lower price for unproven "security reasons" because they can't otherwise compete.
That would be a more accurate paragraph from a European perspective.
The National Cyber Security Centre run by GCHQ gets given copies of the code, so i'd be surprised if the NSA hasn't got copies already.
The National Cyber Security Centre has reportedly found that the code quality is "sloppy" so it's probably not really worth stealing. They haven't found any backdoors though, and one imagines that with the source code and ten years they'd have found any backdoors if they were any.
So, many of those who complain about the Chinese government subsidising Huawei et al, now want the US to subsidise them.
Aside from the hypocrisy, I see a danger of 'Boeing Syndrome' occurring, where the government has invested so much in attempts to get things developed that they can't allow them to fail. So these companies become part of a never ending gravy train where they don't really try that hard because they are making money whatever they do.
Regarding the injection of $3Tn to support the post C19 economy, what do they think had brought about the coming economic collapsr thst was imminent before C19 reared it's ugly head?
A downturn for the near future, will not be due to C19, though that will definitely have exacerbated the situation, the downturn was coming because of three times as much bad debt in the US and elsewhere as that which kicked off the last recent financial crisis.
The Fed and other central banks just keep printing money and lending to anyone without a care in the world because they won't be the ones who need a wheelbarrow to collect yhe day's wages and they won't be standing in line for soup kitchens.
I have so many mixed feelings about this. On the one hand, yes, a consortium pushing for open standards is better for consumers and will bring down cost and minimise frustration and waste. But...the motive seems to largely be a xenophobic "Anybody but China" approach which isn't very helpful.
Considering it was just announced it's a bit early to declaring everything covered by prior art. It's not inconceivable they develop patentable technology. Also it seems unlikely Huawei would want to spend millions or even billions to invalidate patents just so it can implement a standard royalty free in a market it's effectively barred from unless the new standard has substantial advantages over it's own that can used else where.
First, they are a chinese company. There are laws on the books that require all Chinese people and businesses to assist in spying on the west. So there's that.
Second, they are IP thieves. I worked for a large semiconductor some time back and we had a number of our designs not just copied but outright stolen by Huawei.
In the case of the second issue, it's easy to see why it's so cheap: low R&D costs. In the case of the first, well there's always a cost for something, and clearly spying is a great opportunity to sponsor.
"They spend more on R&D than several of the top US tech companies combined."
For fiscal 2019, Huawei spent 15.3bn in R&D. Apple alone spent over $17bn for fiscal 2019.
Source: Their balance sheets are public. Also, I have stock in both (among other tech companies) and I attend earnings calls.
"The Chinese also educate way more people in STEM than anyone else."
They have 1/8th of the world's population. The next is India, with a little under that, then it cuts sharply to less than a quarter of that number for the US. So... yes between China and India having 1/4 of the world's population I'm pretty confident they have more of ANYTHING than anyone else. What a dumb statement either way since it doesn't impact the fact that both countries are rife with IP violations.
The US is still playing catch-up in the 5G world. They will do anything to stifle China's better and more advanced technology, AND they still need the ECHELON backdoors in their own gear, and the other 5-eyes countries too!
VERY convenient how they claim to have mobile phone data from Wuhan that shows research lab shutdown.....
So what Open RAN seems to be is a US 5G patent pool, which in order for it to work either needs to steal patents from Nokia, Ericsson and Huawei (and other non-US contributors) or bully them to agree to license their Patents to Open RAN on terms favourable to the US.
Naturally, the companies that will have goods stopped by US trade officials will be consumer devices from companies such as Samsung - all above board and in accordance with the US's take on WTO's rules.
On a technical basis ANY MODERN CPU CHIP running between 1.5 GHZ to 5.2 GHZ already has the ability to do completely SDR (Software Defined Radio) where I can implement EVERY WiFi/Edge/2G/3G/4G and 5G protocol completely in software and SHAPE and INTERPRET or EMIT ANY WAVEFORM and hardware-level modulation schema such as QAM-1024, OFDM, PCM, etc simply by using the high speed PCI-3/4 bus already on the chips!
For the mere cost of 25 US dollars up to a few hundred dollars I can simply put any modern ARM cpu or AMD Ryzen-3/5/7/9 chip on a simple mobo with a 5 to 120 watt power amplifier and transformer system and antennae on-board connected directly to the PCI lanes of the chip and do ALL the radio functions in cache! This is TRIVIAL for me to program! I've already done so and SDR on a Ryzen-7/9 works GREAT up to 2.5 GHZ! You use Nyquist sampling as a natural antialiasing DSP so that half the frequency of the CPU is your maximum data-rate (for a 4.3 GHz Ryzen-7, its 2.15 GHz as my maximum frequency which is usable for Wifi, 3G and many 4G bands)
Since I also use cheap older AMD FX-series chips (now only $85 US) which I have been able to overclock to as a high as 6 GHZ which means I can do up to 3 GHz SDR communications with a single chip.
If you need higher speeds, just interleave an ODD number of CPU chips (i.e. use 3 or 5 cpu's) and DOUBLE or even TRIPLE your sample rate by doing nano-second synchronized offsets (the CLOCK clocks on many AMD CPUs can be synched down to as fast as 100 nanosecond offsets) which means I can do interpolation of signals to obtain up to 8 or even 12 GHz radio comms bi-directional (i.e. full duplex!). You just use a fast switch (i.e. 7 to 10 nanosecond IGBT) to switch between and mix the emitted digital signals from the PCI buss of each CPU which then forms an equivalent 8 to 12 GHz waveform containing the modulation scheme required for transmission comms.
For reception, have multiple chips sample 100 nanosecond slices on an interleaved basis, and re-construct a combined waveform representing the incoming Wi-Fi, QAM-1024, OFDM, PCM or other data communications schema and interpret the re-digitized waveform as bits, bytes and then as data packets of various higher level communications protocols such as IPV4/6!
THIS IS TRIVIAL TO DO IN SOFTWARE and using simple multi-cpu motherboards I can print myself!
With 9 of the Ryzen-series 5 GHz-clocked CPU chips I can create the software equivalent of a $1.5 million carrier class router in something that costs me less than $2500 to build and populate with CPUs and RAM! I could handle THOUSANDS of simultaneous users on multiple frequencies up to 12 GHZ by doing interleaved time-slice sampling. The more CPUs I use, the more ACCURATE the reconstruction of my received waveforms and more accurate the emitted digitized waveforms which means my Signal-to-Noise ratio will be better AND I could do advanced in-memory signal processing to get rid of or correct for incoming signal multi-path errors and even signal drop-outs to a high level of accuracy!
Maybe I should just release the plans and software I already have ....
Why are we depending upon Huawei again when the above SDR techniques are a LOT cheaper and easier to do?
V
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
