Posted on Sep 18, 2026
Short answer: when a phone with satellite emergency messaging cannot find a cell tower, it
sends a compressed emergency text through a satellite instead. That session reaches your
center one of two ways. If your PSAP accepts text, it arrives through your existing Text Control
Center like any other text-to-911 conversation.
If your PSAP is voice only, an emergency relay center texts the caller, then calls you on a regular 911 trunk with the caller's number and location in the usual ANI and ALI fields. That second path is the one that surprises people. A satellite emergency can land on your console as a plain voice call from a third party who is reading a caller's texts aloud to you.
Knowing which of the two paths your center is on, and what each one changes about your call handling, is the entire point of this blog from NGA.
For most of the history of 911, the rule was simple: no bars, no call. Coverage maps have
improved for decades, but large parts of the country still have terrain, distance, or weather
conditions that leave a caller with a phone that cannot reach anything. Roughly speaking, that gap has always been where search and rescue begins rather than where 911 begins.
Satellite emergency messaging closes part of that gap. A modern handset that cannot see a tower can point itself at a satellite and send a short burst of text. It is not a phone call.
Bandwidth is tiny, latency is measured in tens of seconds rather than milliseconds, and the caller usually has to stand still, outdoors, holding the phone toward a specific patch of sky.
There are two distinct flavors of this, and conflating them causes real confusion in the
operations room:
All three end up at a public safety answering point eventually. They just take very different
roads to get there, and only one of them looks like a normal 911 call when it arrives.
This is the part worth reading twice, because it determines what your telecommunicators will
actually see.
Path one: your center accepts text. The satellite session is delivered to the PSAP responsible for the caller's location through your existing Text Control Center, and location arrives through the TCC interface. Functionally it behaves like the text-to-911 traffic you already handle, with slower message timing. If your center already runs text-to-911 services, you are most of the way there.
Path two: your center is voice only. An emergency relay center handles the text side of the
conversation with the caller and then contacts your PSAP by voice. Per Apple's PSAP
documentation, those relay calls arrive on your normal 911 trunks and display the caller's phone
number and location in the standard ANI and ALI fields. The relay telecommunicators are
trained to national protocol standards and will keep working the caller until the emergency
resolves, and if you release the call after dispatch they will hand you an incident number and a
callback line.
Neither path requires you to buy a satellite anything. What both paths require is that your
service area boundaries are correct with the text control centers and call routing providers
serving your jurisdiction, because that boundary data is what decides whether the session lands
on your console or a neighbor's.
If you take one operational item away from this article, make it that one: satellite routing is
only as good as the boundary and GIS data you have already given your routing providers.
The regulatory picture is more settled than most people assume. In March 2024 the Commission adopted interim 911 requirements for what it calls supplemental coverage from space, the arrangement where a terrestrial carrier uses satellites to extend its own coverage footprint. The FCC's summary page lays out the framework, and the rule text sits at 47 CFR 9.10(t).
The core obligation is worth quoting in plain language. A carrier delivering 911 voice calls or 911 texts over satellite must do one of two things:
Those two options map exactly onto the two paths described above. The rules describe an
interim framework rather than a finished one, which is a fair description of where the
technology sits as well.
Location expectations are the other regulatory thread to keep an eye on. Satellite sessions
generally deliver a device-based location estimate along with an accuracy radius, which is a
different animal from the indoor location accuracy regime the Commission has built for
terrestrial wireless calls under its indoor location accuracy benchmarks. If your center's
protocols assume a dispatchable location, satellite traffic will not fit that assumption
Telecommunicators who have worked one describe satellite sessions as unfamiliar rather than
difficult. Four differences stand out.
short questionnaire before the session opens, so the first data you receive typically already contains the nature of the emergency, whether anyone is injured, and how many people are involved. That is a gift, and it exists precisely because round trips are expensive.
None of this is exotic once a center has trained for it, which is exactly why training for it before the first one arrives is the difference between a smooth incident and an improvised one. Our field notes on what dispatchers say they need at cutover apply here too.
The install base is now large enough that this is no longer a niche concern for wilderness
counties. Satellite emergency messaging ships on recent iPhones and on recent Pixel and other
Android handsets, and carrier satellite service is being marketed directly to consumers as a
coverage feature rather than a safety edge case.
The practical consequence for a PSAP is not about brands. It is that a growing share of the
phones in your jurisdiction can now originate an emergency contact from places your coverage maps say are unreachable, including places outside your normal response footprint entirely.
Visitors, hikers, boaters, hunters, and stranded motorists on rural highways are the populations
most likely to generate this traffic.
It also means the volume is lumpy. Many centers will go months without one and then see
several during a single storm or holiday weekend. Low frequency plus high stakes is the classic
argument for written policy rather than institutional memory.
Everything else depends on this. Stale service area boundaries send satellite sessions to the wrong agency, and the caller has no easy way to correct that. Our overview of the role of GIS in NG911 covers the underlying data work.
Satellite messaging is a reminder of something the NG911 transition has been arguing for years:
emergencies no longer arrive exclusively as phone calls, and a system built only to switch voice
will keep needing bolt-on workarounds for every new format.
A center operating on an ESInet with an i3 architecture treats an inbound emergency as a
session carrying data rather than as a circuit carrying audio. Text, location, additional data, and
eventually media all travel the same path and land in the same workflow. That is the difference
between adding satellite as a new exception and simply receiving it as another session type. Our explainer on what i3 compliance means unpacks the standards side, and NENA's NG911 guide for 911 authorities is the definitive reference document.
For centers still on legacy infrastructure, the relay center path exists precisely because that gap
is real. It works, and it saves lives, but it inserts a human translation layer between the caller and the agency that will respond. Removing that layer is a solid, concrete argument for
modernization, alongside the more familiar ones about text-to-911, caller location, and
Can someone text 911 with no cell service at all?
Yes, if their phone supports satellite emergency messaging and they are outdoors with a clear
view of the sky. The message travels by satellite rather than through a cell tower, and it reaches
either your PSAP directly or an emergency relay center that contacts you by phone.
Does the caller need a subscription?
Generally no for the emergency function. Handset manufacturers have included emergency
satellite messaging without a separate fee, and carrier programs have made emergency texting
available broadly, including to people who are not their customers. Non-emergency satellite
messaging is usually the part that costs money.
Will my PSAP see it as a text or as a call?
It depends on your center. Text-capable PSAPs receive it through their existing text control
center. Voice-only PSAPs receive a voice call from an emergency relay center, with the caller's
number and location in the standard ANI and ALI fields.
How accurate is the location?
Expect a device-based estimate with an accuracy radius rather than a precise address. It is
usually good enough to direct a search, and it should be communicated to responders as an area rather than a point.
How long do satellite messages take?
Significantly longer than SMS. Delays of tens of seconds per message are normal, and terrain or tree cover can extend them further. Sessions should not be treated as abandoned simply
because the caller has gone quiet for a minute.
Do we need new equipment to receive these?
No. Both delivery paths are designed to work with what centers already have. What you may
need to update is your boundary data, your text timeout policy, and your written procedure for
relay center calls.
Satellite emergency messaging is the rare public safety development that arrived quietly and
worked. It did not require PSAPs to buy anything, it did not demand a new standard before it
could launch, and it is already producing rescues in places where a 911 call was previously
impossible.
The risk is precisely that quietness. A capability that shows up without a procurement cycle also
shows up without training, without policy, and without anyone checking whether the boundary
data will route it correctly. The centers that handle their first satellite session well will be the
ones that treated it as a known scenario rather than a surprise.
That preparation is cheap. Verify your boundaries, write the relay procedure, fix the timeout
rule, and run one tabletop. Everything after that is the same job your telecommunicators
already do extremely well, conducted through a slower and stranger channel. For more on
building a center that absorbs new call types without drama, see our 2026 NG911 resource
guide or explore NGA's full solution set.