Hacker News new | comments | ask | show | jobs | submit login
One Man's Quest to Prove Vermont Has Terrible Cell Service (npr.org)
88 points by evo_9 17 days ago | hide | past | web | favorite | 46 comments

I believe this is the case everywhere in the U.S.

That is, major cell carriers claim widespread coverage in rural and exurban areas that is simply not the case.

People have had an incentive to expose this because of the federal funding for wireless expansion, but really it is a "truth in advertising" issue.

I'm sure it is everywhere. I'm in NW Montana and was recently checking out coverage maps of various carriers. I had never noticed the fine print telling me something like "coverage" was defined as slow and would only work outdoors. If I'm in my house, I'm lucky to get a signal on one side of it.

I really do count it as lucky. My neighbor doesn't get anything at all - inside or out, but the coverage map says he's good. A few years ago, before buying a phone, he borrowed a Verizon phone (from them) to make sure it worked at home. It did, he bought it and a few months later it quit working. The phone is fine, I guess they are no longer using the tower that served him. I don't think they updated their coverage map.

I've had a good experience with Verizon in rural areas in New England and Virginia/West VA. I used to use T-Mobile, and lose coverage frequently, but now I can go days without losing signal, even in the middle of nowhere.

I have both a T-Mobile phone and a Verizon phone, and Verizon often provides me with the experience of my phone saying that I have 4 or 5 bars, but being unable to get any data to flow, or sometimes to even make a call.

I don't know what causes that, but I have learned to ignore the signal strength display on Verizon.

Maybe Verizon does something similar to Sprint? https://forums.androidcentral.com/sprint/210510-bars-lie-lte...

> Did you know that with Sprint LTE Android devices, the signal strength indicator at the top does not show your LTE signal strength? Even if 4G is displayed next to it?

> This signal displayed here is always your 1x (voice signal), and it is not your 3G EVDO signal strength, nor your LTE signal strength. Regardless of whether it says 3G or 4G next to it. This is the cause of a lot of confusion.

edit: looks like they might https://forum.xda-developers.com/showthread.php?t=1142503

created an account just for this:

i spend a lot of time in airports, one of my conversation starters is 'tell me something about your work i can't find on google'.

i said this to a person who inspects/installs cell towers. this person told me that a few years ago instead of installing new or upgrading towers for 4G LTE, on a lot of the towers the company had them install a chip that makes it look like users have 4g LTE signal but it's only 4g or 3g and not LTE.

i'm sure someone will come along in the comments and tell me this person lied and this is impossible, and that may be true, just throwing my two pennies out there.

>the company had them install a chip

Yeah no.

As the other posts said below, AT&T has had a history of pushing minor changes in early technology as a new generation of that technology. The most recent is their 5Ge nonsense.

Back in the early 2010's they pushed an update to iOS to change HSPA+ 3G to display as 4G. It's not, but the behavior carried over to other carriers. T-Mobile was closest to being able to argue it as 4G as they deployed DC-HSPA+ which could provide 42mbit in ideal conditions.

It's possible that there was something lost in translation between the installer and you that there was a radio component upgrade on the antenna that allowed HSPA+ to be used over the older UMTS and HSPA. Now most towers have the radios and antennas integrated as one component, but on weak towers and older installs the radios and the antennas were separate.

I know a bit more about T-Mobile USA, and one of the things they did a few years back to bring up 2G-only regions was get better backhaul in then plug LTE radios into the 2G antennas, making the sites broadcast 2G and LTE on band 2, 1900mhz. Later on they would add band 12 LTE by hooking in an antenna that could be tuned for 700mhz.

AT&T will put a fake 5G logo on its 4G LTE phones


If AT&T can rebrand LTE as 5GE then I’m sure this could be true.

During the transition to LTE, T-mobile did something similar -- they updated phones to report HDSPA+ as "4G".

AT&T (with HSUPA) spearheaded that, T-Mobile followed but was better able to actually argue it as their HSPA+ implementation was faster due to use of DC-HSPA+ (42mbits).

ITU relaxed their definition of 4G to allow for "significant upgrades to 3G technology", so blame them for everyone jumping on the bandwagon: https://fcw.com/blogs/mobileplatform/2011/03/att-hspa-4g-pro...

And I’ve used most every major carrier, and they all do that fairly frequently. The strength display is especially inaccurate when you’re moving or when you are in a particularly unpopulated place.

That said, the iPhone I used for a while did better than my old Nokia and my current Samsung, so it certainly seems to pertain to the hardware, too. I’m guessing many people change carriers at the same time as they change phones, which might make it hard to figure out which one is to blame.

Anecdotally, I was able to make calls from the summit of Mt. Whitney as well as at Forester Pass. I'm impressed that my cell phone could send a signal for miles.

Well the northeast mountains have a ton of iron in the rocks is my theory. It makes it almost line of sight for signal. I use to live in Connecticut and it was HORRIBLE. Living in PA it is so much better.

When I travel through to my In-laws in WV we get no internet signal for about 2.5 hours. That has to be due to the low population in Western Maryland.

Many of the hilly areas in the Eastern US are plains that have been dissected by water and ice. Where I live in upstate NY, the hilltops are all around 2000 feet and are flanked by valleys such that at best a hilltop has a line of site for the adjacent valleys.

Last summer I found I could get about 3Mbps down from an AT&T tower that was two valleys away that was using 700 Mhz spectrum -- with a directional antenna.

In some of the more rural areas, the hilltops are roadless so any attempt to put in wireless infrastructure isn't just a matter of running wires for power and data, but actually building roads.

It's probably the different type of terrain more than the composition. It isn't like cellular signals blast through most earth surface materials (or even just vegetation).

Iron ore is neither conducive nor magnetic.

Well, magnetite (Fe3O4) is, but you're right that its a silly theory to use to explain poor signal. For one, signal would have to be able to penetrate a decent amount underground in the first place, which it can't.

Yes and since it isn't conductive it causes for weak radio signal. Source - My Dad is an old HAM Radio operator.

Ham is not capitalized. It’s not an initialism or an acronym.

For some OFs (old farts), HAM is written as all uppercase because of the (apocryphal) story of the a station being operated under the "HAM" call sign by Albert Hyman, Bob Almy, and Peggie Murray that came to stand for all of "ham radio" because they made an impassioned plea against a bill in Congress that would have revoked the frequency allocations for all of amateur radio and given them over to the military.

Further, since call signs are always written in uppercase (WRR-FM, KIRO-AM, KBTC-TV), some hams are apt to write the name of the hobby as HAM out of habit/solidarity.

So while you're technically right that "ham" is more of "ham-fisted" or "what a ham" variety, you're not following the internal culture.

I know that the story about Hyman and company is false. That’s why I put the word “apocryphal” in there, which means “erroneous or fictitious.” But people still write HAM in all uppercase because of the legend, false or not.

Look at their advertised claims closely.

They say they cover 98% (ish) of Americans. You can cover that amount of Americans while only covering 60% of the land mass. This means rural areas are typically lower priority ... But they always soon to get interstates.

This map matches 1-1 in the areas I drive regularly. it orrectly shows no coverage at home and 2-3 places on the way to work 20 miles away where there is zero signal

> At stake in the state's project [of having an Vermont Department of Public Service employee drive around the state for six weeks taking download speed tests] is potentially millions of dollars in federal money. The FCC has $4.53 billion available to use around the country.

> "What we were looking at was being almost completely foreclosed from any of it because our state was considered nearly completely covered with 4G LTE service," [Clay] Purvis, [the state's telecommunications director] said. "And now we've opened up a significant amount of that space."

Just wrote a friend who is a telecom professor and suggested they run a summer experiment here in Michigan like this guy did with students collecting the data.

I don't know what he will think of the idea but I'd sure like to see a coverage map with real data, not the carriers version of the truth.

Cell signal strength would be an interesting addition as a layer to Open Street Maps.

OpenStreetMap doesn't have layers.

It's also not really a great fit for data that you'd want to gather systematically. For signal strength, you'd get lots of data in places that people frequent (and that tend to have good signal) and little data in more remote places, where the measurements would be of more value.

It's not perfect, but cellmapper.net has signal strength and the bands pulled in all on a crowdsourced map.

I think this is great but would having the phones all lined up in a box like that cause any significant detriment / interference in their networking / radio communications?

I hope the state gets the funding it needs.

would having the phones all lined up in a box like that cause any significant detriment

When I saw that picture I thought the same thing.

But since their goal was to document how bad the reception is, any extra interference could only help make their case. :)

I put a smiley there, but this did seem very haphazard; not a well thought out setup.

I've spent a lot of time in Vermont.. It has terrible cell service and no quest is needed to prove that. As a vacation spot, I consider it a positive. I know the residents feel differently.

This is a bit off topic but the article made me think about it:

I am considering getting a landline for my next home, and just realized I have no idea what a fair price is. I have not had a landline my entire life (always used a cell).

Is a hardline cheap in these areas or do they charge about as much as a cell since they know there is demand for a reliable connection?

Reasonable is between $20-30/month.

What is the advantage of landline over VOIP?

> What is the advantage of landline over VOIP?

A few things.

First, there is a small bit of power going over traditional copper lines, and you are able to dial 911 even if the power goes out or the internet goes down.

Second it's a sociological mechanism. People in modern times are incredibly entitled, so saying "I didn't want to answer your call because I was busy" makes them irrationally angry.

But if you give out your landline you can just say you weren't at home.

Much more likely to work in an emergency-voip won't work in a power outage.

I wonder how the cell coverage maps are generated. I think it is common to use terrain data and tower positions to determine line of site to a tower. Is that flawed in some way, or are the results being manipulated to improve them?

God bless the guy, and too bad that the state is too cheap to buy better equipment for the purpose.

All carriers have this data for all of the carriers in each region. If you’re a big customer, they will even show it to you under NDA.

>and too bad that the state is too cheap to buy better equipment for the purpose.

What would be "better equipment" in this case? Professional/enterprise/labatory radio equipment, rather than cell phones? CNC milled/3d printed racks rather than a cardboard box? I don't think there's much you can improve here.

We should tweak the law then to require carriers to report this data to the FCC for public consumption as a condition of radio spectrum licensing and usage.

I agree.

They would fight hard to keep that data proprietary, as certain property owners might be more aggressive about lease pricing or royalties.

All carriers have this data for all of the carriers in each region.

Many many years ago the FM radio station at college wanted to get FCC approval to increase broadcast power to 10 KW. In order to justify that, some students drove around taking signal strength measurements all over the area. They then picked the worst measurements to submit to the FCC.

This wasn't a violation of any rules. Let's say (for example) that the FCC submission needs 50 measurements spaced apart a certain distance. If you have actually taken 150 measurements then you can easily select a coherent set of 50 that show just how bad things are and justify why you need that 10 KW non-commercial license.

It's possible for the cell carriers to do something like this, but in reverse. It may make business sense for them to only retain/report measurement data that makes them look good?

Cell towers aren't cheap to put up. If carriers can obtain "decent" coverage with fewer towers, then it's smart for them to do that?

Of course, the carriers already have very very very granular data. They should be able to gather, and perhaps actually do gather, data from every single cellphone transmission as it establishes both voice and data connections. The carriers could/should use that information to erect more towers and to tweak antenna orientations on existing towers.

In my experience, however, this doesn't seem to happen. The same areas around me have dead spots that persist for years. This is in relatively flat suburban terrain, without too many large hills or tall buildings to interfere.

So in my anecdotal experience, Verizon doesn't give a shit about poor signal strength. And they're perhaps? the best at caring, so what does that say about the others?

Verison is the product of Bell Atlantic which also acquired GTE: https://en.wikipedia.org/wiki/GTE

Applications are open for YC Summer 2019

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