SpaceX and T-Mobile Plan To Connect Mobile Phones To Satellites, Boost Cell Coverage (reuters.com) 96
U.S wireless carrier T-Mobile will use Elon Musk-owned SpaceX's Starlink satellites to provide mobile users with network access in parts of the United States, the companies announced on Thursday, outlining plans to connect users' mobile phones directly to satellites in orbit. From a report: The new plans, which would exist alongside T-mobile's existing cellular services, would cut out the need for cell towers and offer service for sending texts and images where cell coverage does not currently exist, key for emergency situations in remote areas, Musk said at a flashy event on Thursday at his company's south Texas rocket facility. Starlink's satellites will use T-Mobile's mid-band spectrum to create a new network. Most phones used by the company's customers will be compatible with the new service, which will start with texting services in a beta phase beginning by the end of next year.
Air interface (Score:3)
Re:Air interface (Score:5, Insightful)
Similar frequencies so the same antenna can be used on the satellite end. With only a standard phone antenna on the ground the data rate is going to be very low, and it won't work indoors. It will be a battery killer too.
Still, if it's enough to make an emergency phone call from some remote location, it might be useful.
Re: Air interface (Score:2)
Re: Air interface (Score:5, Interesting)
Starlink satellites are ~340 miles away... but it's a straight line, with nothing besides precipitation in the way.
I'll be shocked if it's usable for actual voice calls or "normal" internet usage, vs an occasional few hundred bytes for sms (at least, in the phone to satellite direction). Even purely terrestrial CDMA2000 could pull off 100+ miles at 850MHz if your phone allowed it. The main constraint was thermal. A fraction of a second every now and then? No problem. A minute+ at a time? Problem.
GSM's ~30 mile distance limit was imposed by timing and the speed of light. LTE probably was defined with an alternate mode for countries like Australia & Canada (where CDMA2000 voice lived alongside 3G GSM precisely because of the rural-distance problem).
Re: Air interface (Score:5, Informative)
Re: (Score:3)
TFS is really sloppily written and skipped a lot of details:
would cut out the need for cell towers and offer service for sending texts and images where cell coverage does not currently exist
The service is no substitute for cell towers, rather it's more like an "in a pinch" situation. The bandwidth is very limited, basically 2.4mbps per cell downstream, upstream is much more limited. Voice services will work, and if nobody else is using a lot of bandwidth within the cell then you might be able to get downstream video as well, but don't count on it. It uses a standard 5g protocol on a band that most existing phones already use. The reas
Re: (Score:3, Informative)
And in those "in a pinch" situations it would be invaluable. Think of the people carrying emergency beacons in the back country now, with expensive sat plans, in case of hiking/skiing/hunting accidents. The ability to carry just your regular mobile device with an "I need help, here are my coordinates" app/button/text/etc., would be great. Less to carry, and less to pay so more people will have access to it.
You could even set up an app to monitor the accelerometer, and send an emergency message if a certa
Re: Air interface (Score:2)
I'm not downplaying it btw, this is a really cool thing, just it feels like TFS is not creating a very accurate representation of what it actually is, what it's good for, and what it isn't.
Re: Air interface (Score:5, Informative)
On a related, but interesting, note... future lunar residents will probably find themselves unable to use any chunk of RF spectrum that isn't blocked by walls or set aside for their exclusive use and silent on Earth itself. Why? Most RF, even microwave, eventually ends up shooting off into space... and the collective manmade RF background noise of Earth is *staggering* on the moon.
HF frequencies (below ~50MHz) are likely to be completely unusable by Lunar inhabitants... not even for ham radio, because someone on the moon will get hammered by ALL the band's planet-wide activity at once, whereas 2 hams on Earth can only hear activity within skip distance, so a given frequency might have a half-dozen simultaneous users who are oblivious to the others, but all stacked on top of each other from the perspective of a ham radio operator on the moon.
More trivia: given sufficiently-sophisticated software, Earth's GPS constellation can technically work for much of the moon, too... it just makes the math a lot harder, and requires additional downloaded ephemera data that's absent from the "pure" satellite transmissions. We'll unquestionably put more satnav satellites in orbit around the moon when the need arises, but the ability to use terrestrial ones for additional resolution is a nice bonus.
Re: Air interface (Score:5, Informative)
Not actually true. Most importantly, you have free space loss, which are on the order of 300+dB (it's 260ish dB to geostationary orbit). The earth effectively radiates in all direction, and covers a tiny fraction of the sky from the PoV of the moon. The result is that only an infinitesimally small amount of the RF energy radiated by earth ever reaches the moon.
If what you said was true, it would be impossible to use rf communications on earth, because the Sun is the biggest of all RF emitters in the solar system. This is actually detectable in geostationary satellite communications. Twice a year, the geometry works out that the sun will pass directly behind the satellite your dish is pointed at. When this happens, the RF energy from the sun drowns out the satellite and you lose the link for about 6-8 minutes. For the rest of the year, it's not a problem.
Yes, if you were trying to communicate with earth using directional antennas on frequencies that are used on earth, there could be a problem. Even then, though, with enough gain on the transmitter, it's not impossible to overcome the noise. A couple of kW into a large directional antenna, with known coding, is going to be pretty easy to pick out.
Re: (Score:2)
https://www.amateurradio.com/new-versions-k1jt-weak-signal-digital-modes/
Don’t forget that we’re already using moonbounce as hams. We just work around the problem with well-written protocols.
http://jcar.us/the-story-of-ham-radio-moon-bounce/
https://moonbouncers.org
https://en.wikipedia.org/wiki/Earth–Moon–Earth_communication
http://www.hamclass.net/ranv/moonbounce.pdf
If I may quote from the last one, some relevant factors here:
Re: Air interface (Score:5, Informative)
Inverse square law is a thing. A local radio transmitter within the lunar horizon will absolutely overpower any transmission from Earth (by a factor of tens of billions for the same output power). The combined effect of all terrestrial transmissions would raise the noise floor, but it's not a problem unless you're trying to listen to some faint source on the far edge of the visible universe.
Re: (Score:2)
The article says SMS and data transfer (slowly). So as long as you can get a connection, this will work. Voice, not so much.
Re: (Score:2)
No special high power phone needed. Just standard cell phones.
Re:Air interface (Score:5, Informative)
It's exactly this: LTE directly via satellite.
They DO run into problems with the Doppler shift, but they offset it by adjusting frequencies to compensate for it.
The DO indeed run into problems with distance. They kept stressing how massive the antennas on the new generation of Starlink satellites will be (although we weren't given any dimensions). Apparently they're huge.
The other problem they run into is that each cell is massive, so if there's any density then you're sharing it with a lot of other people. There's no way to compensate for this except for "If there's more than a handful of people, everyone's bandwidth is really low". So you can message, SMS, get and send text emails, etc, but not, say, stream HD videos off Youtube or anything. Contrary to the Slashdot header, they made it sound like even sending images might be too much early on.
But still, it'll basically let you stay in touch even when you're out in the middle of some God-forsaken wilderness or deep out at sea.
Unspoken, but I wonder: will it work on airplanes? The altitude difference isn't that great. Will they be programmed to handle the extra Doppler shift?
Re: Air interface (Score:2)
What gets me is why dont they just deploy solar/wind powered cell towers into those remote areas using starlink as the back haul. Same net effect, but you can use current phones.
Re: Air interface (Score:4, Informative)
Uh, have you ever looked down from your plane over flyover country? There's hills, trees, F'd up terrain. It's basically impossible unless you build tens of thousands of towers hundreds of feet high that would cost more per than a satellite. And even then you'd have numerous dead zones. Take a look at a topographic map of the Pacific Northwest, you seriously think you can put cell towers there and get coverage? Even drawing fiber across the mountainous region would cost millions and you won't reach most of the valleys. That terrain and topography isn't easy. We've given the telcos tens of billions of dollars over decades and they haven't wired a single rural household with broadband.
Satellite is the ONLY viable solution for a large percent of rural areas.
Re: Air interface (Score:5, Insightful)
We've given the telcos tens of billions of dollars over decades and they haven't wired a single rural household with broadband.
That's not because they can't, though. It's because they won't. The formula for doing it is pretty simple. Put fiber on pole, run it to a cabinet, put DSLAM in the cabinet. They have lightweight DSLAMs now. But part of that formula is paying for the equipment, and since we didn't actually put any consequences for failure into the handouts, they just gave the money to executives and shareholders, and The People can go fuck themselves.
We need to stop handing money to corporations without consequences for failure. At minimum, we need to get paid back when they don't deliver. If they can't afford to pay us back in a reasonable period, they can be nationalized.
Re: (Score:2)
>"That's not because they can't, though. It's because they won't."
Correct, and they shouldn't. Because there is no way it can ever make them any money. Instead, they will lose tons. With very low density, customers won't pay thousands of dollars a month.
>"We need to stop handing money to corporations without consequences for failure."
I couldn't agree more.
>"At minimum, we need to get paid back when they don't deliver."
Or perhaps we shouldn't offer money for something that anyone with any reasona
Re: (Score:1)
People also have to realize there are consequences to living out in the boonies. Some things just aren't feasible.
Okay, but that doesn't include providing functional internet access to everyone who has a phone line. That's totally feasible, and we already paid for it.
we seem to always get the worst, tiny, caveat exceptions. For example, why are there no reasonable 600 to 700 Mhz available for WiFi?
There's an LTE band there...
Re: (Score:2)
>"Okay, but that doesn't include providing functional internet access to everyone who has a phone line. That's totally feasible, and we already paid for it."
All depends on definitions and assumptions. It is certainly technologically possible. The question is cost- how much is reasonable and who has to pay for it. For example, it already is possible through DSL to provide at least 6Mbs to just about anywhere but often options are more like 50 to 100Mbs. Right now, the average is 45Mbs for rural areas.
Re: (Score:2)
For example, it already is possible through DSL to provide at least 6Mbs to just about anywhere but often options are more like 50 to 100Mbs.
You are simply sadly mistaken. There are still huge areas where there is no DSL at all, or DSL is limited to 5-6 Mbps — that's the case in Humboldt, for example. Even if you're right in Eureka, that's all you can get out of DSL. If you're out of town, you typically can't get DSL at all. Around here luckily there is cable internet, and while satisfaction is very low I have so far found it to be extremely good. I'm paying for 400 Mbps, and getting it too, and even at peak times I can easily get over 200
Re: (Score:2)
>You are simply sadly mistaken. There are still huge areas where there is no DSL at all, or DSL is limited to 5-6 Mbps
Well, areas that have no DSL would be a major problem and should be addressed. I wasn't aware there were that many areas where DSL wasn't offered. I know there are issues with using DSL on super-long/old copper where the provider has no sub-stations. But I also thought that was the whole point of the projects the public interest was pushing. I have no interest in tax money being used
Re: (Score:2)
> This is why I'm still renting a modem, while DOCSIS is in theory standardized, in practice what brand of device you have still matters a lot.
Yeah, that's actually complete bullshit. In my experience Comcast sabotages non-Comcast modems by disabling flags that allow higher power output back up the line. I had to spend a year telling Comcast to get a truck out here to fix their shitty fucking lines and by the time they did, they ended up calling other trucks in to help them track down the fault which w
Re: (Score:2)
For example, it already is possible through DSL to provide at least 6Mbs to just about anywhere but often options are more like 50 to 100Mbs
You are simply sadly mistaken. There are still huge areas where there is no DSL at all, or DSL is limited to 5-6 Mbps.
Part of the reason for this is that DSL operates over copper wire. That wire needs to be in good condition and there are significant limits on how long that wire can be - for any usable bandwidth, the limit is like a mile or two. Here is an interesting article [ccexpert.us] discussing some of the limits. For DSL to work you have to be within a mile or so of either your central phone office, or a "remote office" - one of those cabinets on the side of the road that has the DSL termination equipment (DSLAM). The remote cabi
Re: (Score:2)
Re: (Score:2)
I have Suddenlink and not Comcast, which I mention only for completeness. I just came to say that it doesn't matter why one modem will work and another won't. The provider is in control of the configuration, not me. If I have to account for an extra $10/mo, then I just roll that into the math. For me it's a very simple situation though because I currently have really high quality internet access, as residential connections go. I have just absolutely no reason to change to anything else, because nothing els
Re: (Score:2)
Part of the reason for this is that DSL operates over copper wire.
Look, I actually live out here and have been paying attention to this since it was the Pac Bell days, OK? And they were notorious for overspliced copper, and indeed they originally sold DSL to something like 17,000 feet and eventually cut back to around 14,500 (although sometimes you could get them to do a longer installation if you knew the right people) because their copper was so shit. And then they promised to extend DSL to every POTS customer by 2000 as their "lightspeed" project, which was itself made
Re: (Score:2)
I have no interest in tax money being used to push hundreds of Mb/s to "everyone",
Do I write you down as "selfish" or as "wants to live in the past"?
but would certainly concede that some amount of broadband, even if just 5 to 10Mbs, should be, at this point.
We literally paid the telcos hundreds of billions of tax dollars to accomplish that, and they haven't.
Re: (Score:2)
You not only don't know what you are talking about, but you are whoring for the telcos.
I see we are a bit touchy. On the technical side, I was actually supporting and adding context to your statement for other readers. On the political side, stating that it's not cost effective is hardly "whoring for the telcos", it's a business fact. Sure they have been given lots of money, but if they spend that money it's less profit for them. Businesses are going to do what maximizes their profit (usually) within the legal and regulatory system they operate in. If you want to be angry, you should be angry
Re: (Score:2)
On the technical side, I was actually supporting and adding context to your statement for other readers.
Your comment was full of factual errors that offered excuses to the telcos. If that's the kind of help you have to offer, don't.
Re: (Score:2)
Your comment was full of factual errors that offered excuses to the telcos. If that's the kind of help you have to offer, don't.
Other than using layman's terms (which I clarified for context) of the name of the street side remote cabinet which terminates the DSL wiring, what factual errors do you refer to?
Re: (Score:1)
Oh yeah and
perhaps we shouldn't offer money for something that anyone with any reasonable business background knows cannot pan out in the first place.
The telco is a utility, and it has special obligations in exchange for a monopoly. If it cannot meet those obligations, then it should not exist. Someone more competent should.
"If they can't afford to pay us back in a reasonable period, they can be nationalized."
No, that is not the way we should be doing things. The government is already completely inept,
No, this is dumbfuckery of the highest order. Public utilities consistently offer higher levels of service at a lower cost. All utility infrastructure should be nationalized, period.
Re: (Score:2)
Problem is that we keep expanding the term utility to cover more and more.
Re: (Score:2)
What do you think we're now erroneously referring to as a utility?
Re: (Score:2)
"It", meaning rural access, does not necessarily have to make money itself. The profits made in the high-density areas subsidize the access in the low-density areas. That's part of the trade-off for the limiting competition.
Re: (Score:2)
would cost more per than a satellite
Wrong. A single Falcon 9 launch currently costs $67M. A broadcast antenna is going to top out at a few hundred thousand to maybe $1M at most. Thats a lot of antennas you can put up for the cost of a single launch. Being generous for arguments sake, if each Starlink launch puts 60 satellites in orbit, thats 200 launches to complete the original Starlink plan for 12,000 active satellites, costing $13Billion to complete. Thats 13,000 $1M long range antennas, more than the number of satellites you could have fo
Re: (Score:3)
Now do the rest of the math. Each tower has a range of 35 km. What sort of area does that cover? What state would that be equivilent to?
For range estimate, I used the 22 mile figure here:
https://smallbusiness.chron.co... [chron.com]
The 12k satellites cover the entire world with internet service BTW.
Re: (Score:2)
Re: (Score:2)
Look at the topography of many parts of the US! That range is line of sight, it can't penetrate hills/mountains .. let alone foliage.
Re: (Score:2)
Pick your number, do the math. Those 13,000 towers won't cover much of anything, the sats do it better for cheaper in the end.
I chose the number that the site stated was the most common max range due to time delay, you can pick whatever range you like, and it doesn't change the math any. Towers won't cover anywhere near what the Starlink sats will cover, so comparing the two makes no sense.
Re: (Score:2)
Re: (Score:2)
My parents, who live on a farm, in very rural Iowa, with fiber to the home, will be shocked to hear they don't have broadband.
Re: (Score:2)
Aerostats are another option. Project Loon may be a failure, but it was used in Puerto Rico after a hurricane, and in Peru after an earthquake. Loon was shitcanned because it’s not commercially competitive, but that’s in normal market conditions. By integrating Loon with Starlink for backhaul, flying cell towers and electronic surveillance systems could be mobilized swiftly into a disaster area by cargo aircraft, simply tossed out the back door and allowed to inflate already at altitude and del
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
The biggest advantage for starlink operating over longer ranges is likely that it can use a highly directional antenna. All the customers are within a fairly small angle of straight down.
Re: (Score:2)
Maybe 30 degrees in any direction. Satellite goes horizon to horizon in maybe 4-10 minutes and might be somewhat north or south.
Re: (Score:2)
What gets me is why dont they just deploy solar/wind powered cell towers into those remote areas using starlink as the back haul. Same net effect, but you can use current phones.
To the consumer, needing a new phone is a bug. To the providers, it's a feature!
Re: (Score:2)
Not worth the investment. Not many people there. Plus, wilderness areas. Hard to build and maintain a cell tower in the middle of nowhere.
Re: (Score:2)
Two reasons:
a) It's insanely expensive
b) In many areas it's illegal. I work with a remote site surrounded by Federal Wilderness. While we're responsible for maintaining the trail network in the entire watershed, we are only permitted to use hand tools in the wilderness areas (ie no chainsaws). If we have a hazard tree that can't be safely hand cut, there is a whole process to get permission to either bring in a chainsaw, or blast the tree.
There are two communications sites in the wilderness area (one is a r
Re: (Score:2)
If you're using Starlink as the backhaul for towers you can't serve any higher density of users than direct Starlink.
Re: (Score:2)
My question is power usage. It takes a lot of power to send to the satellite. People using this in remote areas won't have their phones last that longif they use this for any sustained period of time.
Re: (Score:3)
It'll be like having a weak LTE signal, and the corresponding battery drain of that.
Re: Air interface (Score:2)
Re: (Score:2)
The other problem they run into is that each cell is massive, so if there's any density then you're sharing it with a lot of other people.
Cell size would ultimately be limited by the need for beam separation, rather than satellite spacing. If the satellite antennas are just 10x the ground station size, then you'd have a beam footprint of a few kms.
Unspoken, but I wonder: will it work on airplanes? The altitude difference isn't that great. Will they be programmed to handle the extra Doppler shift?
I think compared to the satellite velocity, the aircraft velocity would be a minor effect.
Re: (Score:2)
Unspoken, but I wonder: will it work on airplanes? The altitude difference isn't that great. Will they be programmed to handle the extra Doppler shift?
I think compared to the satellite velocity, the aircraft velocity would be a minor effect.
Also, don't airplane bodies kind of act like faraday cages?
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
No dimensions were given but Elon estimated 25 square meters for the antenna.
Re: (Score:2)
"if there's any density then you're sharing it with a lot of other people"
While plausible, that turns out to not be a practical problem since you are unlikely to get that density of people in a location where you could not get traditional tower+backhaul infastructure to go along with them..
Re: Air interface (Score:2)
Re: Air interface (Score:5, Interesting)
Re: (Score:2)
It takes much less power than you think.
As a ham radio operator, I regularly reach the international space station with 100 milliwatts of power.
(Former ham here).
Just curious: What kind of antenna are you using?
Re: (Score:2)
I had the same question on a recent post about AirTags transmitting to space. Turns out that's not unreasonable, and people are already doing it [slashdot.org] with a battery life of 10 years without charging. It's not streaming video, but TFS says this is going to start with just text messages.
Re: (Score:2)
Uses standard cell phones. Not special high power phones so no difference in what you have now.
Re: (Score:2)
Re: (Score:2)
Yes, they program the satellite to act like a cell base station.
Yes, lots of technical issues but apparently they have solved them.
"Official space helmet on, Captain Video!" (Score:4, Informative)
I'm old enough that this would be like having a "space phone", and incredibly cool. I'm also old enough that I sometimes remember to appreciate the awesome power of the personal computers at my disposal.
https://www.imdb.com/title/tt0... [imdb.com]
Re: "Official space helmet on, Captain Video!" (Score:2)
Re: (Score:2)
We carry a phone that has the computing power of what? 100x what got us to the moon? 1000x?
>1000x
Wiki says the 15 bit Apollo Guidance Computer clocked at 2MHz, which was pretty hot stuff for it's day! And portable at only 70lbs! The contemporary PDP-10 only clocked at 1MHz.
Re: (Score:2)
Re: (Score:2)
"We carry a phone that has the computing power of what? 100x what got us to the moon? 1000x?"
Processor, several thousand at least. Memory, try several *million*. The Apollo Guidance Computer had a 2 MHz processor, 2k words of RAM and 32K words of ROM with a 15-bit word. A modern smartphone will have a processor with around 3000 MHz and a design that gets more done in a clock cycle, at least 6 gigabytes of RAM and 32 gigs of storage or more.
Re: (Score:2)
You've been able to buy a sat phone from Iridium for some time now. They have been very expensive, however. This will be a hell of a lot cheaper since the satellites will be doing other stuff, and the launches are cheap.
Re: (Score:2)
You can buy satellite phone devices at any electronics store these days for around $100 or so. They usually use GlobalStar, though there some using Iridium. They're things like the Garmin/DeLorme InReach and many others which use a satellite data link and GPS.
They're often used by hikers and
Re: (Score:2)
You can get text relatively cheaply. Voice is expensive. I'm guessing that this new service is going to murder the existing satellite text/locator services, though.
Re: (Score:2)
You know, this post is refreshing.
Sometimes I forget to step back and marvel at how far we have come. I need to work a lot on my cynicism; it really prevents me from enjoying the wonders of our world.
RIP InReach and Spot (Score:2)
Re: (Score:2)
Re: (Score:3)
My cellphone, however, is not rugged at all. Good luck in an are
Re: (Score:2)
Re: (Score:2)
I have to take issue with your appraisal of cellphone ruggedness... they're pretty rugged. I used to get a 'ruggedized' phone like a Galaxy "Active" but now most any regular midlevel phone like my Galaxy A57 is IP67 rated. I have tortured a number of phones at length on the handlebars of my dirbikes and the only problems I've seen are from vibrations - damaging the camera's ability to focus, an
Re: (Score:3)
Yes, will be amazing (Score:1)
I also have an emergency communication device (I forget the brand) that I have to activate a subscription for - so I would activate it for a few months while hiking, or before a really big trip. It uses the existing satellite phone network.
Being able to do this with the phone I already carry would be amazing and I wouldn't have to remember to turn on service before I go anywhere, or remember to bring the device.
Seems like this spells doom for the existing satellite phone companies...
Re: (Score:1)
Agreed! This is exciting. I go to rural areas to hike, explore, and get away from tech for a bit. Being able to send a text message is a fantastic safety net. Even just one time. Especially when pared with things such as smart watch fall detection.
Cost per bit (Score:2)
My first thought was that satellite-phone connections would be very expensive, but then I realized that's already the case. My phone and broadband plans cost about the same, and I probably move 30x the data across the broadband connection. So, Elon might well compete with my phone provider!
I miss that have to get this due to 3G's death (Score:4, Interesting)
Considering the death of 3G means I can't get service in a lot of places, I might have to pony up for this and drop Verizon
If it works (Score:2)
It'll be a solar plexus punch for a certain specialized (and expensive) real estate market
Radio Towers
Re: (Score:2)
On the one hand, radio towers will be vastly more capable than cell-to-sat communications. On the other hand, anything that would let me afford to colocate a GMRS repeater near my home would be kickass. There’s plenty of ham radio nearby, but the only test you need for a GMRS license is one question long — “Will your check clear?” — which makes it nice for people who are intimidated by the amateur radio exam or just want simple, channelized comms with minimal thinking, effort,
Re: (Score:2)
ham radio doesn't use commercial tower sites, not even repeaters. Nor does GMRS or FMRS
I am discussing commercial radio towers, and those things are expensive to get onto... IF they have space.
I am a ham and I've done engineering for commercial radio
Might take (Score:2)