Hacker News new | past | comments | ask | show | jobs | submit login
Manhole Covers Serve as Antennas Expanding Wireless Network Coverage (ieee.org)
125 points by Varcht on Dec 19, 2018 | hide | past | favorite | 53 comments

The more I hear about 5G, the more I think it's an industry trying to sell a product that nobody is asking for. I have no complaints about the speed I get on my iPhone from T-Mobile here in the US. I wish I had better LTE coverage, but I'm happy with the performance. Most of the time it's snappier than my Wi-Fi (which is weird). I get a solid 8 to 20 mbits under fair conditions- I can stream movies and music just fine.

All I hear about 5G is that it requires a large amount of antennas on both your phone (4 because a hand covering the antenna can cause a drop in signal) and around you. I don't want to be near a high power transmitter that's low frequency, and I really don't want to be near a high frequency one.

What's the sell of 5G? I guess the speed, but what? 1 Gbit to my phone? Why. Considering that I only get 15GB of "Unlimited" data now, it isn't going to replace Wi-F. Because of the density required of the towers, it's not going to replace 4G/LTE as some areas it's just unrealistic to have the required density.

I think there are a few selling points.

First, 5G should increase available bandwidth. Industry analysts are predicting that data usage will go up 7x between now and 2024. More speed/bandwidth can help with that. Without increasing capacity, would we just see our usage stagnate?

Second, 5G should have much better ping times. Not only will this make browsing noticeably better, but it also means that latency applications can work well over wireless.

Third, there's a decent possibility that 5G will bring home broadband competition. If you're fed up with Comcast or Charter, 5G could mean choice in your internet service.

Fourth, I believe that 5G has a lot more flexibility than LTE. Part of that might mean low-power narrow connections that will allow cheap and battery-efficient sensor connections. Part of that might mean ultra-wide-band connections allowing for high-bandwidth devices that can't be served by the more limited carrier aggregation of LTE.

Fifth, sometimes applications happen when the technology is there to accommodate it. The broad availability of broadband let Netflix pivot from shipping DVDs to streaming. Broadband has meant online gaming, richer online applications, digital distribution, etc. As mobile bandwidth increased, we went from loading music from our computer, to streaming music, to streaming video. I think the broadband speed you have to support in your applications tends to be 10Mbps+, but if that floor raises over the next 5 years, it could change which applications and experiences people invent.

I mean, before the iPhone was announced, a lot of people wanted a phone dial pad on one side and an iPod click wheel on the other. People wanted Blackberries. The iPhone laid the technological foundation for all sorts of new experiences that people hadn't really thought about. Successive iPhones have been comparatively minor upgrades, but even then the faster processors, better screens, better wireless connectivity, etc. have allowed for better experiences to be built.


It's also important to remember that there are different deployments of 5G. While Verizon has pushed out its ultra-high-band, short-range 5G, other companies are going to be rolling out 5G on more traditional spectrum offering much the same range as LTE. 5G NR is a new radio interface and it promises better spectral efficiency than LTE.

I think it's also likely that many areas will see high-density deployments. Companies like Comcast run a wire to every house they serve and require coordination with property owners to get it into the building. If you could run a wire to one point every 18 acres, that's a lot easier (that's assuming a 500 ft transmission range). Higher density than that is certainly achievable in dense cities.

It can certainly be hard to predict what a new technology might make happen. It's also possible that 5G won't make a huge impact beyond what LTE has already done.

5g is everything, and 5g is nothing. 5g is. To find the true meaning of 5g, one must look to the government subsidies.

This is an excellent koan, but I thought the true flow of money was in the other direction with the spectrum auctions.

I have no idea why you're being downvoted. You have successfully summarized 5g.

What 5g will give webdevelopers will take.

I start to observe, that previously I was pretty much covered with HSPA. But now more and more without LTE it's hard. I guess when most technologists are within LTE coverage one is at disadvantage outside of it.

Sadly it seems that at least where I live mobile transfer limits are stagnant.

> Fifth, sometimes applications happen when the technology is there to accommodate it.

I like to call it the "infrastructure pattern", via [0]. If something serves (or could serve) as an enabler for other activities, it's important to make it more capable than the immediate needs would dictate, as this allows for new applications to appear. Internet speed & bandwidth is a great example of this, and so is CPU speed, RAM, and storage size. I'd go as far as saying that the history of civilization is one big series of "I know we don't need this" followed by new applications spawned by the surplus.


[0] - https://wiki.hackerspaces.org/The_Infrastructure_Pattern

'Moar speed' is not really the point of 5g, and as usual the main advantages are not widely talked about because a lot of the media doesn't understand them and how to communicate more advanced concepts to readers. In my mind the biggest advantages and shifts in thinking are virtualization and network slicing [1], and enabling IOT [2].

[1] https://www.sdxcentral.com/5g/definitions/5g-virtualization/

[2] https://www.ericsson.com/en/about-us/india/authored-articles...

Bandwidth is tremendously important in wireless communications because the more data you can transmit in less time, the less time you need to be transmitting.

In general, only one radio can transmit on one frequency at a time. So a city wide network quickly becomes congested. Reducing the time everyone spends transmitting greatly increases the quality of service and the amount of users that can be serviced.

The increased antenna count is also a benefit in crowded situations. If each antenna covers a smaller range, then more people can use the network in a given area- two antennas that would interfere with each other in a longer range system will operate simultaneously.

They've solved that issue by deploying more towers in high density areas and carriers have swapped allocations in areas to get more continuous spectrum. Upping the frequency just gives you wider bands.

I think a better argument could be made about spending R&D on making 4G/LTE require a smaller band. I think LTE (assuming 850MHz) can be 1.4, 3, 5, 10MHz per band- which is non-optional. If you could cut the required Tx/Rx band width in half; I think that would be better use of effort.

The challenge is not so much to deliver 20mbit to one person, as to deliver anything at all to the several thousand people in the area of a tower, all of whom are waiting for their next push notification.

With 4G I routinely see situations in busy areas where the phone claims "full bars" but data performance is sub-dialup.

5G will not solve this problem. It cannot load much more bits/Hz of spectrum than LTE (iirc it is 10% or so more efficient, in ideal circumstances). More spectrum will solve this problem, and to be honest no reason LTE couldn't take advantage of that too, it is very good with channel aggregation.

You're still thinking in 2018 terms. What about 2025?

640K ought to be enough for anybody.

When I was in college, I remember taking a class in silicon design. At one point, a student remarked, "man, silicon is like... black magic."

"No," the teacher sharply rebuked. "It's governed by well understood and reliable mathematical equations. Though difficult, it is far from magic."

There was a brief pause, the teacher was normally easy going and congenial. His harsh words caught us off guard.

"...antennas on the other hand, that is black magic."

Indistinguishable From Magic: Manufacturing Modern Computer Chips -- https://www.youtube.com/watch?v=NGFhc8R_uO4

Same talk (slightly updated perhaps?) in higher quality: https://www.youtube.com/watch?v=KL-I3-C-KBk

Didn't know about that, thanks!

Are these proper manhole covers? I mean the big heavy metal type that can support vehicle traffic? The one in the picture seems the type for covering access to a "manhole" but not one under a street. I do have to wonder whether the local cell will drop out every time a huge metal object, a bus, rolls over the cover.

>>It eliminates traffic disruptions from street construction, and there are no antennas awkwardly placed on buildings, marring the appearance of a neighborhood.

Said by someone who has never attempted to install anything inside a manhole. They will need to close the streets. You cannot just creep through the sewers installing stuff without disrupting the traffic above. Workers are not moles. They will need to run new lines. And in modern cities, the type without the huge walk-in storm sewers of NY/London/Paris/SF, there will still be digging.

With many innovations these days it is hard to determine feasibility from initial articles/claims. Lots of inventions seem to exist only to get grant money, with no hope of practicality (solar roads, solar windows). If these are lightweight manhole covers that cannot support traffic (ie just an antenna mounted atop an aluminum cover) then I'll chalk it up alongside those sidewalk tiles that generate electricity from pedestrians: might work, but will never be practical.

https://en.wikipedia.org/wiki/Solar_Roadways https://www.6sqft.com/pavegen-opens-worlds-first-smart-stree...

I don't think London has huge walk-in storm sewers. And there are certainly people digging up everything they can see, all the time, at least on the roads i use to get to work.

It does. The old cities have combined systems, storm and sewage together. To accommodate rainwater and maintenance, they are essentially brick corridors. Some are huge.


More modern cities have separate systems for sewage and stormwater, and these are contained in small pipes that people cannot walk around in. Upgrading from a combined system is a massive project, well beyond something like 5g imho.


I'm curious how this affects the NIMBY fights about antennas. Some of those fights are about aesthetics of the antennas, but some of the concern is around the radiation, which is not a finished debate. Mobile phone signals have been designated as possibly a cause of cancer by the WHO if I recall correctly. It hasn't been possible to rule out that mobile phones do not cause harm, which is something I was surprised to learn. I thought it was a settled debate that mobile phones were safe.


> phone signals have been designated as possibly a cause of cancer by the WHO

No, it has not. "To date, research does not suggest any consistent evidence of adverse health effects from exposure to radiofrequency fields at levels below those that cause tissue heating" [1].

[1] https://www.who.int/news-room/fact-sheets/detail/electromagn...

I see as another datapoint the fact that we've had broadcast radio for over a century, some using several orders of magnitude higher power levels[1] than mobile phones today, yet no corresponding increases in cancer were seen.

On the chance that there is an effect, it might not even be detectable, which means for all practical purposes it doesn't.

[1] https://en.wikipedia.org/wiki/WLW

When you extract temporal and frontal lobe glioblastoma from the rest of brain tumor numbers you get a graph like this:


Not saying it's a direct correlation, this is a very difficult topic to analyze, but if you are a heavy phone user you take your chances. Cancer has a long latency period, so you may die of other causes before the damage is realized.

Personally I think other more immediate non-thermal effects of pulsed RF are of more concern than cancer at the moment. This may change, but that's what the current research seems to point towards.

What does that have to do with radio waves? And how often do people actually put cell phones close to their head compared to 10 years ago? For all we know it could be related to increasing obesity rates.

Mine's right in front of my head (frontal lobes), as I squint to read your comment.

The radiation pattern isn't the strongest there, however. Antennae on mobiles will be strongest emitting from the back and sides, with less out the front. Any RF that gets absorbed by the body is wasted energy, so they optimise the antennae to emit away from the body in the typical orientation.

Absorbance characteristics of sub 100 MHz are substantially different from those of multiple GHz.

And yet the second bullet in that article:

> The electromagnetic fields produced by mobile phones are classified by the International Agency for Research on Cancer as possibly carcinogenic to humans.

So it's not settled science, even though there is some evidence pointing toward it not causing cancer.

There is a lot of evidence, long term studies, showing they do not do cancer. So the question is why are they listed as a possible cause? It invalidates the entire list. My car keys may well "possibly" cause cancer if the bar is so low.

The specific item that is on the IARC Group 2B ("possibly carcinogenic") is "radiofrequency electromagnetic fields", which includes exposures due to cell phone use-- cell phone radiation specifically is not classified.

So, if your car has keyless entry, your car keys would technically also be on the list.

From wikipedia, it says "limited evidence of carcinogenicity in humans and less than sufficient evidence of carcinogenicity in experimental animals. It may also be used when there is inadequate evidence of carcinogenicity in humans but there is sufficient evidence of carcinogenicity in experimental animals. In some instances, an agent, mixture or exposure circumstance for which there is inadequate evidence of carcinogenicity in humans but limited evidence of carcinogenicity in experimental animals together with supporting evidence from other relevant data may be placed in this group."

Anecdotally, I see alongside some unpleasant-sounding items, a lot of common things on that same list like: aloe vera, Ginkgo biloba, engine exhaust, pickled vegetables, and cocamide DEA. It seems to me that mere presence on that list doesn't mean much without further context -- why it was put on the list, the relative danger of it compared to other things on that list, etc.

What has IARC excluded from causing cancer?

I've been following this debate for a long time, and it's definitely not settled science.

In fact The Lancet just published this article:

"Planetary electromagnetic pollution: it is time to assess its impact" https://www.thelancet.com/journals/lanplh/article/PIIS2542-5...

There is plenty of evidence of non-thermal and non-ionizing biological effects [1]. The problem is at the moment most people are focused on cancer, when in fact there are a lot of other published effects that need to be looked into.

ICNIRP and FCC are both focused solely on thermal effects, with human modeling as a blob of jelly, when the latest science shows there also exists non-thermal effects on a cellular and bio-molecular level.

[1] An example of 500 published papers on EMR (electromagnetic radiation) since August 2016: https://www.saferemr.com/2018/04/recent-research.html

I cannot speak about the biological effects of RF but I'm pretty sure that the denser the coverage, the less power will end up being used by the phone (or base stations) to complete the link (since on average phones will be closer to base stations) -- leading to less RF exposure.

I wonder if there's any Monsanto-esque funny business going on in this debate about EM radiation and cancer. If there was, how could we tell?

The answer is objective and trusted third party analysis. But since we don't have the time or resources as individuals to know and follow the work of the myriad researchers in the world, we need at least third party certification agencies to certify their ongoing work and not just their initial education.

Thus a gap could be filled by a foundation which had the resources to provide a brand to peer review and follow up study for the most impactful publications/findings of the year. If this foundation had a significant amount of funding for research well beyond typical peer review, it would inspire some level of confidence in results that otherwise we cannot have from the peer review performed by people working for the publishing companies that profit from publication.

We don't let farmer joe and farmer bob review each other's operations, yet that is often what we do with the food for our minds.

Personally I think the better reference is Wakefield - trying to fabricate a danger for fiscal gain. They keep on trying to find a smoking gun so they can have 'the next tobacco lawsuit' but it just isn't there.

The beauty of science is that we never know anything with absolute certainty. However, things like burnt toast and cured meat have been shown to be carcinogenic. On the other hand, non-ionizing radiation has not been shown to be cancer causing despite a lot of studies. In that context, I would say it is absolutely absurd to worry about mobile phones without some clear new evidence.

A side issue with 5G is that there are all sorts of fun / unusual effects being discovered with GSM/wifi suffusing our environments - i remember Stanford being able to detect people moving behind a wall using the wifi signals. Simply Monitoring signal strength has had many commercial opportunities and now we are going to put millimetre wave radar throughout our cities - expect novel uses to crop up

Wouldn't traffic poles serve as better publicly-owned antenna mount points? They are powered and have better line of sight to everything (broader range). I wonder why manholes are considered better.

As I understand it, they're owned by different entities; the street lights/traffic poles are often owned by local authorities (in the UK) while manhole covers are owner by the utilities.

I suspect that one aspect of this is to give them leverage in getting access to antenna sites.

The article doesn't state it was better than traffic lights. I think small cells in traffic lights are already reality, even if not widespread, the small cells exist.

The article presents a novel, non-obvious approach in some ealier-stage development.

I'm far from an expert, but aren't manholes (and streetlights and traffic lights, by extension) effectively grounded, hence incurring into insane loss?

Many Antennas are grounded (eg for lightning protection), but that has no relevance for their effectiveness as an antenna.

Because the size is much less than a wavelength, the body can be at ground, while the active parts can be at RF potential.

Example: Your phone is grounded, but it still works.

They are grounded with a precise length of coax (multiples of 1/4 or 1/2 wave, IIRC) to keep the VSWR ratio low for the intended frequency. In a pole you can get to do that by probing the signal at certain height, but the active part of a patch antenna (which I assume a manhole works as) is at the extremes of it (as such, probes are always off center for circular polarization, and at the extreme for linear), and those are grounded. Patch antennas are normally grounded at the center, as the voltage/current there should be close enough to 0 anyway unless there's an electrostatic discharge.

Came here to wonder the same. Do they wind a coil around it?

That's pretty clever, and appears to be already in use even for 4G.

Yes, I know Swisscom does this already.

For a little background [0]:

"Swisscom and Ericsson have proved that city manholes can be used worldwide to improve capacity with small cells – even below street level – using the Ericsson Vault Remote Radio Unit and Kathrein’s Street Connect, an in-ground microcell antenna system. The use of existing street manholes where fiber and power already exists lowers total cost of ownership by 50 percent.

This, the world’s first vault site for LTE and small cells has been approved by the Swiss authorities, and 250 new rollouts are due during 2016 in the country’s cities. The solution effectively addresses cities’ needs by enabling the reuse of existing assets and underground space."

Really since 2016. Good framing by Voda to present this as "Building on its heritage of innovation".

[0] https://www.ericsson.com/en/news/2016/3/swisscom-and-ericsso...

[1] https://ieeexplore.ieee.org/document/7481351

This means we may have to run power supplies to existing fiber manholes...

Wow, that must be why I start itching whenever I pass a manhole cover.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact