Comment by owenversteeg | Hacker News Reader

Comment by owenversteeg | Hacker News Reader

Parent Full thread

As the HN guidelines say, "Assume good faith."

I understand the MIMO designation, and I am not trying to misrepresent it in my comment. I know that 128x128 is not sixteen thousand (!!!) antennas on one tower, but rather 256, half for rx and half for tx.

That said, two hundred and fifty six antennae on one tower is far, far more (orders of magnitude, plural!) than the typical 2x2 or 4x4 4G towers deployed today. And keep in mind that this is certainly not "just" 256 antennae for one mall, because 5G is blocked by anything thicker than a post-it (the article mentions how future plans include deploying massive MIMO to each floor of a building, as drywall and glass both block many 5G frequencies!)

The reality is that no matter how you want to slice it, 5G means moving from widely spaced 4G towers transmitting at frequencies we have used for around a century to hyper-dense massive deployments of large numbers of antennae transmitting at new and relatively un-studied frequencies. 4G towers are (mostly) kept away from humans getting too close; 5G towers will be on lampposts and on the same floor of your office where someone can touch them.

I'm not saying 5G is inherently bad, or good, I have no idea. I haven't run a longitudinal study, but _neither has anyone else_, and that's what scares me a little. Look, the #1 post on HN right now is how highly trained, expert engineers accidentally made a bridge (one of the oldest human inventions, around for millennia) create an extremely loud eerie noise currently blanketing San Francisco.

And sure, transmit power is low. But it's not just about the power. Today I went out and enjoyed a warming 1000 W/m2 of hundreds of-nm EM radiation all over my body. Then I accidentally looked at the source, which hurt my eyes and disoriented me. Later I played with a device of only 0.01W emitting highly temporally coherent EM radiation. I had to put on safety equipment, because it had the possibility to harm me, including permanently damaging my sight.

Lots of work has already been carried out.

I think you misunderstand the concept of directional beamforming. The antennas in question in a phased array are 'elements' that cooperate to steer a more focussed beam in a given direction.

The phone in your pocket when transmitting (whether 3G/4G/5G) is the biggest contribution to your personal health. Not a base station.

I'd be concerned much more about low and mid band that can penetrate much further than high band. I am concerned about high band too (skin surface effects) but know nothing about this and assume work has been done to quantify and assess.

My biggest problem both with 5G hype and those that seek to cast doubt on it is that both sides misrepresent the 'facts' to suit their agenda. Sadly that's true in all walks of life.

I've messed around with radio a bit, so I have some idea of beamforming. Basically, you use interference in the wavefront to your advantage. I am not sure what relevance it has here - at the end of the day, you are still dumping lots of power into new and unknown frequencies in every direction, very close to humans.

I am going to disagree with you that the phone is always the bigger risk. Apparently Huawei 5G base stations are above 11 kilowatts, and these are being rolled out closer to humans than ever before.

Sure, some research has been done, but the vast majority of it has been simply pointing some mm-waves at people for a few hours and checking if they're OK afterwards. I haven't seen anything for babies (baby skin is thin enough that mm-waves can go through the dermis, unlike adults) and not even a hint of any quality longitudinal study.

And yes, I agree, the 5G hype/hate is so rabid that it makes it much harder to find any actual science on the internet about the subject. Personally, I don't have any agenda here. I was generally positive on 5G until I read the details in this great and informative article, and thought "wow, that's a lot of antennas, and a lot of new frequencies, and they're closer to us than ever before"

If you know of any decent studies on this, I'm all ears - please drop any links.

Do you have a source for 11 kW base stations?

Assuming we are talking about RF power output here, that is far higher than anything I've ever dealt with, and I'm not sure that it's at all right. A typical macro cell has 20 Watts of RF power output (for example).

The relevant international standards are those from ICNIRP, and they've recently updated and expanded on these in great detail around safety margins of various frequencies.

The UK telecoms regulator, Ofcom, recently did a safety study of actual deployed equipment, looking at measuring the RF output from base stations.

Therefore I'm not sure where figures like 11 kW are coming from. Maybe there are some very hot (and inefficient) early base stations somewhere using 11 kW of mains power to produce 20 or 40 W of RF energy? But considering Huawei is heavily advertising 5G as being "greener", that wouldn't seem likely either. But even then, this is far higher than even the wall socket consumption of the equipment I've dealt with.

11kW is the figure for power consumption of the entire cell site, not RF output power. The same source gives about 7kW for a current 4G site.

https://www.fiercewireless.com/tech/5g-base-stations-use-a-l...

Do keep in mind that those ~20 W typically apply to a single MIMO band (so, 10 W +45 deg linear polarization, 10 W -45 deg linear polarization, for a, say, 120 deg wide and 5-20 deg high sector). One eNodeB can have multiple frequency bands, that are aggregated when communicating with a recent high-end smartphone. Of course this restricts deployments where humans can easily walk up to the antenna, but these power levels are only worth it for long-range scenarios.

The 11 kW seem realistic when you take a multi-sector mast that fills 200 MHz TDD with 1W/Mhz in 3 separate 120 deg sectors (600W), at least once you include the compute to handle aggregate data at a realistic maximum spectral efficiency of ~10 (bit/s)/Hz (6 Gbit/s). These power amplifiers are typically between 10 and 40% efficiency.

11kW in the 1GHz-1THz range, in a location where a civilian bystander could accidentally move himself into the beam, isn't going to happen. Not even 1kW.

If you look at a microwave magnetron (don't), you go blind in seconds from your eyeballs getting cooked and turning opaque. Surgery could theoretically replace them (and leave the retina), but I'm not aware of that ever happening.

It's quite feasible for a sizable eNodeB to consume 11kW, but keep in mind that they have significant amounts of compute hardware.

It'd help if you'd make a bit more of an effort to check/validate/cite the concrete numbers you name, if you don't want to be seen as a "confused anti-5G person". I don't think you are malicious, considering your history, but no everyone will.

> Later I played with a device of only 0.01W emitting highly temporally coherent EM radiation. I had to put on safety equipment, because it had the possibility to harm me, including permanently damaging my sight.

Are you talking about a laser? The mechanism of injury is the same as any RF injury -- localized heating. You stick a soldering iron on your retina, you're going to have permanent damage to your vision. The reason you only need a low power to do damage is because the power is spread out over a very small space, because your cells absorb the energy instead of letting it pass through, and because the retinal cells are extremely sensitive to temperature (a 10C increase is apparently all you need to permanently damage them).

> Today I went out and enjoyed a warming 1000 W/m2 of hundreds of-nm EM radiation all over my body.

Skin cancer from exposure to sunlight kills people. UV light has enough energy to rip electrons off of atoms in your body, causing much damage. ("Ionizing radiation.") From physics, we know where the boundary between ionizing light and non-ionizing light lies (in the middle of the UV spectrum). We know ionizing radiation is dangerous -- we understand the mechanism and the effects. (Notice how much safety equipment is involved in getting a dental x-ray. Minimum dose required to get the necessary image. The technician stands in a separate room lined with lead. You wear a lead vest to avoid exposure anywhere that's not needed for the image.)

We also know when ionization stops happening. The energy carried by a photon is directly related to the wavelength of the light, see https://en.wikipedia.org/wiki/Photon_energy. Below the middle of UV, photons don't have enough energy to ionize anything. Thus begins "non-ionizing radiation", which as far as we know, just burns you. You can, of course, die from burns, so it's not harmless. That is how microwaves cook foods -- water absorbs energy from 2.4GHz photons quite well ("dielectric heating"). The RF energy is converted to heat, and makes you a hot meal. But you sure wouldn't want to have your hand in there when it's on.

Anyway, these effects are well understood. Any "5G is harmful" argument needs to propose a new mechanism for causing damage, or show how a known mechanism works at 5G frequencies. You could show that it turns out we were wrong about ionization (experimentally testable). You could show that certain tissues convert all the energy to heat (again experimentally testable).

Sure, there could be a new mechanism for causing damage that only happens at some random location in the EM spectrum not related to any other effects. Someone has to be the first to discover it. Portal 2 has a good quote: "Marie Curie invented the theory of radioactivity, the treatment of radioactivity, and dying of radioactivity."

It seems unlikely, though. You can do an infinite amount of research without discovering something that doesn't exist. People want fast YouTube downloads more than they want a 0% chance of risk. The risk is already so close to zero that it doesn't matter.

> Look, the #1 post on HN right now is how highly trained, expert engineers accidentally made a bridge (one of the oldest human inventions, around for millennia) create an extremely loud eerie noise currently blanketing San Francisco.

I believe they knew it would make a noise. They just didn't think anyone would care.

A deeper problem is that humans are capable of creating much more complicated systems than nature. Ever debug some code and wonder "how did that ever work?" It's just a lot more complex than basic physics, and so there are a lot more unexpected results.

>> Sure, there could be a new mechanism for causing damage that only happens at some random location in the EM spectrum not related to any other effects. It seems unlikely, though.

And there lies the body of your argument.

Now I will disprove it:

- Photochemically-induced retinal injury (NOT tissue heating) by 430nm EM radiation [1]

- Exposure to ~600 THz EM radiation disrupting the circadian rhythm [2]

- Living within 50m of a <100 kHz EM radiation source more than doubles the chance of Alzheimer's disease [3]

- Spinal cord stimulation at 10 kHz significantly helped chronic pain [4]

[1]: https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1751-1097....

[2]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4734149/

[3]: https://academic.oup.com/aje/article/169/2/167/95445

[4]: https://academic.oup.com/painmedicine/article/20/10/1971/541...

The EM spectrum is vast and complex. Deploying unprecedented numbers of towers, each with more antennas, at higher powers than ever before, at new and untested wavelengths [5], strikes me as risky. You may, of course, draw your own conclusion. As with many technological improvements, we may not know the true cost for many years to come.

[5]: from the above article - up to 2 orders of magnitude more antennas per tower, orders of magnitude more towers, power significantly higher than before, and 48 GHz.

> Exposure to ~600 THz EM radiation disrupting the circadian rhythm [2]

You see, if you hadn't switched the units around, this would have been a good example. But I'm failing at searching a way for how this could have been deliberate, yet non-malicious.

Also, 10kHz is something you get with modern power electronics, not RF. Take a look at the way they couple that energy into the subject, and what the power levels are.

Can you please give a concrete citation for the power being "significantly higher than before"?

There are some interesting radio transmissions in the ~10kHz region: https://en.wikipedia.org/wiki/Very_low_frequency

I am not sure how these signals come from power lines, which is what the study looks at. I would assume those are in the 50-60Hz range. (But we can use that band for communication too! https://en.wikipedia.org/wiki/Extremely_low_frequency)

Well, keep in mind that the Swiss 50Hz do fall into the <100kHz designation.

I'm aware of it's usage for pinging currently-diving submarines, and in the slightly higher region, providing wireless bidirectional low-bandwidth comms into deep mines. The power lines do have quite some magnetic stray fields, though. Not that it's too massive; it's not more than a few (3~4) kA at most.

Wow, you are really struggling with the whole "Assume good faith" thing.

https://news.ycombinator.com/newsguidelines.html

Here's a helpful link. HN can be a great place full of interesting people, but you really do have to assume good faith.

Look, I came into this discussion trying to have an intelligent discussion about the possibile risks of 5G. When I opened the article I was somewhat positive about 5G, now I am somewhat negative. I am not some crackpot saying it will damn us to Hell eternally, I'm saying hey, we've messed around with novel parts of the EM spectrum before, why not at least evaluate the risks first? Never before in human history has anyone lived with a mmwave source, and now we are planning to roll them out at unprecedented speed to all the world's cities without a single longitudinal study on the risks.

I've been playing with different ways to represent things partly for fun, and partly as a filter for this discussion. Anyone who can't trivially figure out that 1 Hz ~= a wavelength of 299792 km does not have the knowledge necessary for this discussion.

Tens and hundreds of kHz RF has been created by humans. The first of studies have been coming out in the last decade about its effects. So far, it looks like living within 50m more than doubles your risk of Alzheimer's.

>> Can you please give a concrete citation for the power being "significantly higher than before"?

Sure. Both MTN and Huawei say that just with current systems we're looking at 200-300% of existing levels. And it's their job to make it sound -good-.

https://www.fiercewireless.com/tech/5g-base-stations-use-a-l...

That link is talking about total power used by the entire site, not RF power emitted. A lot of that is going into things like additional switching capacity, advanced FPGA and DSP hardware, etc. Also, it is additive on top of older cell technology (look at the graph) - it makes sense that power consumption will go up at a site supporting 2G/3G/4G and 5G, instead of just 2/3G or 2/3/4G.

I'm sorry for not seeing

> and partly as a filter for this discussion

coming. As I said earlier, I am not assuming bad faith from you in this discussion, but didn't see a good reason for mixing these up. It made me consider whether you might have soaked those numbers from popular media or (worse) echo chambers.

I guess it's just another example for sarcasm/humor causing problems on the internet, but this time making me worry whether you got captured by some echo chamber, instead of making the sarcastic author look like a troll.

And, just to note, I personally try hard to not behave badly on HN, so feel free to assume a misunderstanding instead of me assuming/exhibiting bad faith.

The first two links are about blue light affecting your eyes. Indeed there are many light inputs that affect your eyes. You have to understand that the cells in your eye are uniquely tuned to produce a chemical response to light. That's what seeing is. One also cannot discount the effects of your brain's interpretation of various light sources -- there are lots of horrific images that you can see, but the brain damage doesn't come from a reaction to the energy in the radiation, it's from your interpretation of what you're watching.

The next link suggests an increase in Alzheimer's disease from living near power lines. I am not sure how meaningful the results are; the entire document is caveats on how they selected which data points to use, the field strenghts are vastly different from wireless signals, and the frequency ranges are off by 5 orders of magnitude.

The last link is about a device designed to receive radio signals and apply electrical current to your spinal cord. If the radio waves worked by themselves, why would they implant a receiver?

You latch on to the "large number of antennas" thing, but it's not really a big deal. All the antenna array does is create interference patterns that control the direction of the signal. The antennas are excited with the desired signal with different phase delays, and then the interference results in some spots receiving 0 power (in an ideal case) and another spot receiving all the power. A friend of mine wrote a small simulation you can play with in your browser: https://apenwarr.ca/beamlab/. (The reality is that channel capacity depends on SNR, and while these do increase the signal strength at the receiver, they more importantly reduce the amount of noise that reaches the receiver. It is easier to null out received noise in a certain direction than to transmit power somewhere.)

In the past, we did this with carefully placed antenna elements that receive and re-radiate signals from another antenna element: https://en.wikipedia.org/wiki/Yagi%E2%80%93Uda_antenna (among many other designs). Now that DSP has become more powerful, it is possible to do it in software instead of hardware, and get better results. (Phased array radar have existed since the 70s. WiFi and 4G networks also use MIMO.)

I guess my summary continues to be: some sort of mechanism of causing damage will have to be proposed and tested. Maybe things on the very low end are bad (as your study [3] suggests). We know things on the high end are bad (don't get too many x-rays). But those effects roll off, and 5G is in an area where those effects probably don't exist.

Anywhere current flows through a conductor, radio waves follow. We have been using radio for 100 years, and we kind of know how this stuff works. I don't think there is any reason to worry (and many experts in medicine and RF agree with me).

Your microwave uses 2.4 GHZ radiation, does your router burn you?

Even though that is indeed a simple, quite trivial example, it does not really apply to the case of 30~50 GHz in combination with significant beamforming, because the average power density deposited into surface tissue can be much higher than with 2.4 GHz. You just can't focus 2.4 GHz onto a square inch (~6.5 cm^2) with a far-field-optimized antenna. According to [0], humans can sense an energy dose of 4J (0.2 K threshold, 5 cm^2 area, 1 cm penetration depth) in their palm due to the temperature rise. Assuming, say, 1 W focused onto the spot, that'd be just 4 seconds until it's perceptible. Measurable, harmful effects at thresholds slightly lower than what's perceptible don't see that far-fetched, once you consider the shallow sub-surface heating.

I don't think there'd be any frequency-dependent effects that don't stem from the interaction of penetration depth, triggered circulation (transporting heat away), and deposited power. Neglecting triggered circulation, it'd seem viable to just test thermally-mediated health effects from focused, 0.01~10 W @ 20~50 GHz (spot size limited due to a reasonable (0.1~0.8) numerical aperture of the beam focusing system) radiation.

[0]: https://books.google.de/books?id=4dL6CAAAQBAJ Fig 3.a page 437

> Below the middle of UV, photons don't have enough energy to ionize anything. Thus begins "non-ionizing radiation", which as far as we know, just burns you

Photodissociation is not simply heating. Red light is for example known to affect the mitochondrial electron transport chain

Reply on news.ycombinator.com