US Mobile gives you access to the three major nationwide cellular networks in the US: Warp, Dark Star, and Light Speed. We have also connected satellite infrastructure, including Starlink.
Your phone number, plan, account, and billing stay with US Mobile. The network providing the connection can change.
That is the basic model.
Fig. 1 · One service, several networks
Your US Mobile service stays in place. The network underneath it can change.
You are using the underlying cellular networks
When your line is on Warp, your phone connects to Warp's underlying cell sites. On Dark Star, it connects to Dark Star's. The same is true for Light Speed.
The radio network is the network's own infrastructure: its towers, spectrum, antennas, and coverage footprint.
There can still be differences in how service is configured. Network policy, roaming arrangements, device support, and data priority can all affect the experience. But there is not a separate, smaller US Mobile radio network sitting underneath the service.
Your phone connects to the underlying network directly.
With US Mobile, that network is not a permanent choice.
Choosing your primary network
For most people, one of the three networks will be the best place to start.
Enter your address and we will recommend a primary network based on the coverage we expect around that location.
Check coverage at your address
The recommendation is specific to your location. It is not a ranking of the networks nationwide.
Cellular coverage is local. Tower placement, spectrum, terrain, buildings, congestion, and the device itself all affect performance.
One network may work best at your house. Another may work better at your office. A third may have stronger coverage along a route you drive regularly.
The address check gives you a good starting point. Your phone and your actual use tell you the rest.
Fig. 2 · Coverage is local
HOMEOFFICEGROCERYHIGHWAY
Coverage is local. The best starting network depends on where you use your phone.
Changing networks
If another network works better, you can move your line to it.
A switch between Warp, Dark Star, and Light Speed keeps the US Mobile service in place. Your phone number, plan, account, and billing remain the same. The network subscription underneath the line changes.
So if you move, change jobs, travel somewhere regularly, or find that another network performs better, you do not need to move your service to another carrier.
You can change the network instead.
Fig. 3 · What stays and what changes
The service stays. The network underneath it changes.
Using more than one network
Some locations do not have one clear winner.
With Multi-Network, a compatible device can have subscriptions on more than one US Mobile network while sharing the same plan.
The networks remain independent. Their towers are in different places. They use different spectrum. Their capacity and congestion vary by market.
That gives the phone more than one available path.
If one network has weak indoor coverage, another may work better. If one is congested, another may have more capacity. If one has a coverage gap, another may have a site nearby.
Multi-Network does not combine the networks into one radio system. It gives the device access to more than one of them.
Fig. 4 · Two independent paths
Warp
Dark Star
More than one network can be available. Cellular data still uses a selected network path.
Coverage outside the US
The same model extends internationally.
US Mobile provides connectivity in more than 180 destinations, with access to multiple networks in many places.
Today, some of that access is provided through roaming. We are also building deeper integrations with international networks so that more of the connection can happen directly through local network infrastructure over time.
From the user's side, the goal is simple. Your phone should have useful networks available when you travel, without requiring you to manage separate carrier relationships in every country.
The platform handles the network-specific work underneath.
Satellite is another access network
We have also connected satellite infrastructure, including Starlink, to the platform.
Satellite and terrestrial cellular solve different problems.
Terrestrial networks are very efficient where many people need capacity. Lower-frequency cellular spectrum can cover large areas and work well indoors. Satellite can extend connectivity into places where building terrestrial infrastructure is difficult or uneconomic.
These systems do not need to replace one another.
They can both be part of the same service.
Fig. 5 · Different infrastructure for different places
Different networks are good at different jobs.
The platform underneath it
Making several networks work under one service requires a software layer above them.
US Mobile handles subscriber identity, SIM and eSIM provisioning, plans, billing, usage, device support, network configuration, and the systems that connect those pieces to each network.
Each network works differently underneath. The platform keeps the customer-facing service consistent while handling those differences behind the scenes.
Warp is connected.
Dark Star is connected.
Light Speed is connected.
Starlink is connected.
International networks are connected, with more integrations to come.
That gives us another way to improve coverage. We can connect more useful networks to the platform.
US Mobile does not depend on one physical network being best everywhere.
It gives you access to the networks that work where you are.
Fig. 6 · The complete model
The networks are separate underneath. US Mobile connects them at the service layer above.
US Mobile sits above several independent access networks.
Today, that includes Warp, Dark Star, and Light Speed in the US, satellite infrastructure including Starlink, and international network access across more than 180 destinations.
Your US Mobile service exists separately from the network carrying it.
Your number, plan, account, billing, and service configuration stay at US Mobile. Underneath that, a line can be provisioned onto a particular network, moved to another one, or paired with additional network subscriptions through Multi-Network.
That separation is the basis of how our coverage works.
Fig. 1 · The service layer and the access networks
US Mobile owns the service relationship. Each underlying network operates its own infrastructure.
The phone connects to the underlying RAN
When a line is on Warp, the phone attaches to Warp's underlying radio network. Dark Star and Light Speed work the same way.
The RF environment comes from that network's infrastructure: site locations, spectrum, antennas, sector configuration, capacity, backhaul, and network topology.
US Mobile is not inserting another radio layer between the phone and the serving cell.
There can still be differences above the RF layer. Network policy, QoS treatment, roaming, device support, and provisioning all affect the service. But if the question is whether a tower exists near your house, the relevant towers are the underlying network's towers.
That matters because the three networks have different physical layouts.
Fig. 2 · Same location, three RF geometries
NET—
SITE—
BAND—
RSRP—
SINR—
ILLUSTRATIVE VALUES
The phone did not move. The network geometry did.
Coverage is more than signal strength
A coverage map mostly answers one question: is a usable radio connection likely to exist here?
Actual performance depends on more than that.
A UE can have strong RSRP and poor SINR. It can have an excellent radio link to a loaded sector. It can be attached on low-band spectrum that reaches a long way but has less available bandwidth than a nearby mid-band deployment.
The serving cell has finite radio resources, and every active device on that sector is competing for some share of them.
Can the phone reach the network?
Can the network move useful traffic once it does?
They are related, but they are not the same measurement.
Fig. 3 · Strong signal, limited capacity
ACTIVE UEs ON SECTOR1
The radio link can stay the same while the available capacity changes.
Choosing a primary network
For most users, one of the three networks makes the most sense as a primary.
Our coverage tool evaluates the address and recommends where to start.
Enter an address
That recommendation is a prediction based on the network conditions we expect around that location. It is not a permanent assignment.
Propagation is local enough that no remote model can know everything about the user's actual environment. Building materials, floor, device antennas, local congestion, and even which side of a structure the phone is on can change the result.
The address recommendation gets the line onto a sensible first network.
The phone provides the real-world test.
If another network works better, the line can be moved.
What changes during a network switch
Consider a Warp to Dark Star transfer.
From the customer's point of view, the service stays the same. The phone number, plan, account, and billing remain at US Mobile. What changes is the network subscription underneath the line.
A simplified sequence looks like this:
US Mobile line
▼
validate destination network + device
▼
provision destination subscription
▼
prepare SIM / eSIM profile
▼
associate the line and its features
▼
UE installs or enables the profile
▼
register on destination network
▼
verify service
Each stage belongs to a different part of the system.
US Mobile maintains the persistent service state. The destination network maintains its own subscriber state. The SIM or eSIM provides the credentials the UE uses to authenticate. The device then registers on the new network and establishes voice, messaging, and data service using that network's configuration.
This is not an inter-network radio handover. The UE is leaving one operator domain and registering in another.
The control plane coordinates those steps so they appear as one operation to the user.
Fig. 4 · What stays, what changes
The service stays in place. The access subscription changes.
How Multi-Network works
A network switch replaces the primary access network.
Multi-Network does something different. It gives the device separate subscriptions on separate networks and ties them together at the US Mobile service layer.
Each subscription is provisioned independently. It has its own SIM or eSIM profile, authenticates against its own operator domain, and registers with that network normally.
The sharing happens above the networks.
US Mobile knows that the subscriptions belong to the same plan, so usage and eligible plan entitlements can be managed together even though the underlying networks have no direct relationship with one another.
On a Dual SIM Dual Standby device, multiple subscriptions can remain registered at the same time. Cellular data generally uses one selected subscription as its active path.
The handset decides which line carries that data.
This gives the device access to more than one RF environment without pretending those RF environments are one network.
Fig. 5 · One plan, two independent paths
Separate subscriptions. Shared plan. One selected cellular data path.
Why multiple networks improve coverage
The value comes from the networks being independent.
Their sites are in different locations. Their spectrum portfolios differ. Capacity is deployed differently by market. An outage or overloaded sector on one network does not necessarily affect the others.
The failures are not completely independent. Operators can share towers, fiber providers, utility infrastructure, and other dependencies.
But the overlap is incomplete enough to be useful.
If one network has weak indoor coverage in a particular building, another may have a better site nearby. If one sector is loaded, another network may still have capacity. If one network has a geographic hole, the others may not.
Multi-Network gives the handset another available path when those differences matter.
International coverage uses the same basic model
Outside the US, the set of available networks gets larger.
US Mobile already provides connectivity in more than 180 destinations. The domestic networks have different roaming footprints and partner relationships, so the networks available in a country can vary depending on the underlying subscription.
Roaming separates radio access from the subscriber's home service.
The visited operator supplies the local RAN. Subscriber authorization, policy, charging, and traffic routing are coordinated across the operator boundary.
US Mobile also has the option to integrate international networks more directly over time.
That lets the access implementation become more local while the US Mobile service remains the persistent layer above it.
Fig. 6 · Coverage becomes a network graph
The service can stay the same while the available access networks change by country.
Satellite fits at the access layer too
Satellite and terrestrial cellular have very different physical characteristics.
Terrestrial networks can place large amounts of capacity close to dense demand. Low-band macro networks can cover large areas. Satellite can reach places where terrestrial infrastructure is sparse, difficult to build, or uneconomic.
US Mobile already has Starlink connected to the service platform.
That does not require satellite to behave like a terrestrial RAN.
At the platform layer, many of the same objects still exist: customer, service, device or hardware state, plan, usage, billing, activation, diagnostics, and support.
The physical access implementation can be completely different underneath.
This is why terrestrial and satellite connectivity do not need to be treated as competing end states.
They solve different parts of the coverage problem.
Fig. 7 · Different access systems under one service
The common point is the layer above them.
The software layer
Connecting another network involves more than making its coverage available.
Each network has its own subscriber systems, provisioning interfaces, SIM requirements, device rules, feature support, roaming behavior, usage feeds, and operational constraints.
The platform has to translate those differences into a common service model.
That includes systems for:
subscriber identity
SIM and eSIM lifecycle
network provisioning
device compatibility
plan state
entitlements
usage accounting
billing
roaming
network-specific configuration
diagnostics
support tooling
The networks do not need to become identical.
They need to be integrated well enough that the customer can use them as parts of one US Mobile service.
That is the architecture already running underneath Warp, Dark Star, Light Speed, satellite connectivity, and international access.
When another useful network becomes available, the platform gives us somewhere to connect it.
What this changes about coverage
A network operator improves coverage by improving its physical network: adding spectrum, building sites, increasing capacity, improving backhaul, or expanding roaming.
US Mobile benefits when those networks improve.
We also have another option.
We can connect another network.
That does not make every dead zone disappear, and it does not remove the limits of RF, devices, spectrum, or network capacity.
It gives the subscriber more available infrastructure to work with.
That is what we mean when we say US Mobile has all the networks.
The networks remain separate.
The service above them does not.
US Mobile sits above several independent access networks.
Today, that includes Warp, Dark Star, and Light Speed in the US, international network access, and satellite infrastructure including Starlink.
Each of those networks is its own system. Warp, Dark Star, and Light Speed have separate RANs, spectrum portfolios, cores, subscriber systems, roaming relationships, QoS policies, provisioning interfaces, and operational behavior. Satellite and international networks introduce different architectures again.
US Mobile provides the layer above them.
At that layer, we maintain the customer-facing service: the account, lines, plan, entitlements, usage, billing, device state, and provisioning state. We also maintain the logic that maps those objects to the network-specific systems underneath.
That separation is important because the US Mobile line and the network subscription carrying it are not the same thing.
A line can move from Warp to Dark Star while the number, plan, account, and billing remain unchanged. A Multi-Network setup can attach more than one network subscription to the same plan. International access can use networks outside the US without creating a separate customer relationship for each one.
The underlying networks remain independent. The platform handles the work required to make them part of one service.
Fig. 1 · The platform boundary
One service layer above several independent networks.
The radio link still belongs to the underlying network
When a US Mobile line is on Warp, the UE attaches to Warp's radio access network. The same is true for Dark Star and Light Speed.
Cell search, synchronization, RRC establishment, scheduling, mobility, carrier aggregation, power control, and handover are handled inside that operator's network. US Mobile is not inserting another radio layer between the device and the serving cell.
So the first part of coverage is still a conventional RF problem.
At a particular location, the quality of the link depends on things like:
serving-cell geometry
neighboring-cell interference
frequency
channel bandwidth
terrain and building loss
antenna configuration
UE modem and antenna performance
uplink margin
current cell load
Those variables can produce very different results even when two locations look similar on a coverage map.
The uplink is especially easy to overlook. A macro site has much more transmit power and antenna gain available than a handset. Near the edge of coverage, the UE may still receive the downlink while its own transmission back to the site is becoming marginal.
That is why a useful cellular connection has to be evaluated in both directions.
Fig. 2 · One UE, one cell
DISTANCE—PATH LOSS—DL RX—DL MARGIN—UL MARGIN—
The connection has to close in both directions.
Spectrum changes the geometry
Low-band and mid-band spectrum solve different parts of the coverage problem.
Lower frequencies generally propagate farther and penetrate structures better. Mid-band gives the network access to much more bandwidth, which can support substantially more capacity where the site geometry is favorable.
Those tradeoffs shape how an operator builds the network.
A rural site may rely heavily on low-band spectrum because the priority is reach. In a dense market, mid-band capacity and spatial reuse become more important.
The band itself is only part of the result. Site placement, antenna configuration, channel bandwidth, UE support, terrain, and surrounding structures all matter with it.
That is why two networks using different spectrum from different sites can behave very differently at the same address.
Fig. 3 · Same site, different bands
BAND—CH BANDWIDTH—RX AT UE—CAPACITY—ILLUSTRATIVE RF FOOTPRINT
Spectrum changes both how far the signal travels and how much capacity the network can deliver.
Strong signal does not imply spare capacity
A UE can have excellent RSRP and still perform badly on a loaded sector.
The serving cell has a finite amount of radio resource to schedule. That resource is shared across active devices according to radio conditions, traffic demand, QoS treatment, scheduler behavior, MIMO state, and the spectrum available on the cell.
The RF link can therefore remain almost unchanged while throughput falls.
Backhaul can impose another limit. A lightly loaded radio sector is not especially useful if the transport behind it is constrained.
So there are at least two separate questions when we talk about coverage:
Can the UE establish a usable radio link?
And once it does, how much service can the network actually deliver?
Fig. 4 · Strong signal, loaded scheduler
ACTIVE UEs ON SECTOR1
The radio link can stay healthy while the available capacity changes.
The three US networks are physically different systems
Warp, Dark Star, and Light Speed have different physical networks underneath them.
Their sites are in different places. Their spectrum portfolios differ. Their sector geometry, capacity deployment, roaming relationships, and operating behavior differ by market.
That is why the same address can produce very different results across the three networks.
A house may sit close to a favorable Warp sector but behind terrain relative to Dark Star. An office may be served by a nearby Dark Star mid-band site while Warp reaches it from farther away. A highway corridor may line up better with Light Speed's site spacing and low-band coverage.
None of those examples require one network to be better overall.
They only require the local RF geometry to be different.
This is also why national coverage maps can only tell part of the story. What matters to the handset is the network that exists around the places where it is actually used.
Fig. 5 · Coverage is local
NET—
SITE—
BAND—
RSRP—
SINR—
OBSTR—
ILLUSTRATIVE VALUES
Same location. Different network geometry.
What the address recommendation knows
The address checker is making a prediction about which network is most likely to be a good primary network around that location.
It can use information such as expected coverage, network characteristics available to the model, device capability, and observed performance data where available.
What it cannot see is the exact RF environment around the handset.
It does not know whether the UE is in a basement, behind low-E glass, inside a concrete stairwell, or on the side of the building facing away from the serving site. It also cannot know exactly what sector load will look like at the moment the customer uses the phone.
Those local details can change the result.
So the recommendation is best understood as a starting point.
It puts the line on the network we expect to work well around that address. The handset then provides the measurement that matters most: how the network actually performs where the customer uses it.
If another network works better, the line can be moved.
Enter an address
Fig. 6 · Prediction and observed RF
The model chooses where to start. The handset measures what is actually there.
A line maps to several different identities
A customer thinks of a line as one thing. The network does not.
The phone number is one identifier.
The device has an IMEI.
The eUICC has an EID.
An installed SIM or eSIM profile has an ICCID and contains the credentials used to authenticate the subscription.
The operator maintains its own subscriber identity and authentication state behind that profile.
US Mobile has to keep those objects associated with the correct line, plan, features, and billing state.
That distinction becomes important when the access network changes. The customer can keep the same US Mobile line while the network-side subscription and SIM profile underneath it are replaced.
Fig. 7 · Identities inside a line
One customer-facing line maps to several technical identities. Click a layer to highlight it.
eSIM provisioning has two sides
Installing an eSIM is only one part of an activation.
The destination network also needs a valid subscriber record with the correct service state.
On the network side, the subscription has to be created and configured with the features the line is supposed to have.
On the device side, the eSIM profile has to be prepared, delivered, installed into the eUICC, and enabled.
The Local Profile Assistant handles the device-side profile flow. The profile gives the UE the credentials it needs to authenticate to the destination network.
Those two paths meet when the UE is able to register successfully.
US Mobile coordinates the service state around both.
Fig. 8 · Subscription provisioning and profile delivery
The network has to know the subscriber, and the device has to hold the right credentials.
Registration turns RF access into service
Once the UE has a usable radio link and valid subscriber credentials, it still has to register with the operator.
In a 5G Standalone network, the UE establishes radio access through the gNB and sends registration signaling toward the AMF. Authentication reaches the operator's subscriber and authentication functions, including AUSF and UDM. After registration, session management establishes the packet-data path through functions such as the SMF and UPF.
The sequence is different in LTE/EPC. There, the corresponding path runs through the eNodeB, MME, SGW, and PGW. A 5G NSA device can use NR radio while remaining anchored to LTE/EPC for parts of the control and core-network path.
From the US Mobile side, the important part is the operator boundary. The SIM credentials, subscriber record, and provisioned services all have to agree before the UE can register successfully.
Once registration is complete, the network can establish the services associated with that subscription.
Fig. 9 · From radio access to packet data
Signal bars only describe part of the stack
A phone can show usable signal without having a working data connection.
The UE may have found and camped on a suitable cell while something farther up the stack is incomplete. Registration can fail. The packet session may not establish. APN or DNN configuration may be wrong. Policy or entitlement state may prevent service. The user plane may not be passing traffic.
Those states look similar from the top of the phone because the radio layer is already working.
This is why signal strength is useful diagnostic information, but it is not a complete service test.
A functioning cellular data connection requires the sequence above it to complete as well.
Fig. 10 · The service stack
Voice adds another registration path
Modern cellular voice is delivered through IMS.
A UE can have an active packet-data session while voice remains unavailable. IMS registration, voice provisioning, device support, and the relevant network entitlements all have their own state.
VoLTE and VoNR both depend on this layer. SMS over IMS, Wi-Fi Calling, voicemail, supplementary services, and emergency calling introduce additional configuration and network-specific requirements.
Those capabilities are not provisioned identically on every operator.
US Mobile therefore has to maintain the feature state for the line and map it correctly into the voice capabilities supported by the selected network.
Fig. 11 · Packet data and IMS
What happens during a network switch
Consider a Warp to Dark Star transfer.
The customer keeps the same US Mobile line, number, plan, account, billing, and entitlements. What changes underneath is the access subscription.
The destination network first has to be valid for the device and the requested service. A subscriber record is provisioned there, the appropriate SIM or eSIM profile is prepared, and the line's features are associated with the destination state.
The UE then installs or enables the destination profile and registers with the new operator.
A simplified sequence looks like this:
US Mobile line
▼
validate destination network + device
▼
provision destination subscription
▼
prepare SIM / eSIM profile
▼
associate line and features
▼
UE installs or enables profile
▼
register on destination network
▼
verify voice, messaging, and data
The systems involved do not all update at the same moment.
The destination network may finish provisioning before the UE installs the profile. The device may have the new profile before registration completes. US Mobile keeps track of those states and moves the line toward the requested final configuration.
This is different from a radio handover. The UE is leaving one operator domain and registering in another.
Fig. 12 · Handover versus network switch
A handover moves the UE inside one network. A network switch changes the operator domain underneath the line.
Multi-Network keeps separate subscriptions active
Multi-Network uses a different model.
Instead of replacing one access subscription with another, the device has separate subscriptions on separate networks.
Each subscription has its own SIM or eSIM profile. Each authenticates against its own operator domain and registers with that network normally.
US Mobile maintains the relationship between those subscriptions at the service layer.
That is where the shared plan, usage, and eligible entitlements are applied.
The underlying networks do not need to coordinate with one another. Warp operates the Warp subscription. Dark Star operates the Dark Star subscription. Light Speed operates the Light Speed subscription.
On a Dual SIM Dual Standby device, more than one subscription can remain registered at the same time. Cellular data generally uses one selected subscription as the active path.
Fig. 13 · Multi-Network architecture
Changing the data subscription changes the packet path
When the handset moves cellular data from one subscription to another, traffic starts traversing a different operator network.
The serving RAN changes first. The packet core underneath it changes as well. Internet egress may be different, along with source addressing, NAT state, latency, and other path characteristics.
An application using the old path now has to deal with that change.
Some transports and applications recover cleanly. Others reconnect. An existing session can also break and be established again over the new path.
None of this requires the two mobile networks to behave as one core network.
The handset has access to two independent paths and selects which one carries cellular data.
Fig. 14 · Packet path before and after
Shared usage is accounted above the operator subscriptions
Each underlying network meters its own subscription.
US Mobile associates that network usage with the plan above it.
In a Multi-Network configuration, usage from several subscriptions can therefore contribute to the same plan balance even though the networks are measuring those subscriptions independently.
The same structure applies to other plan-level state where supported, such as eligible entitlements or roaming allowances.
The accounting relationship belongs to US Mobile. The operator networks continue to maintain their own subscriber and usage records underneath.
Fig. 15 · Shared usage
QoS matters when the cell gets busy
Under light load, two subscribers on the same sector may see similar performance even if the network applies different scheduling treatment to them.
The difference becomes easier to see once radio resources are constrained.
LTE uses mechanisms including QCI. In 5G, QoS flows and 5QI provide the corresponding framework. Operators map subscriber products and traffic classes into those systems according to their own network design.
The scheduler then has to decide how limited radio resources are assigned among active UEs.
Two phones can have nearly identical RSRP and SINR while receiving different amounts of scheduled capacity under congestion.
That is the condition where priority treatment becomes most relevant.
Fig. 16 · Same RF, different scheduling
SECTOR LOAD2
Roaming separates local radio access from the home subscription
When a subscriber arrives in another country, the UE searches for an allowed visited PLMN and registers through the roaming relationship associated with its subscription.
The visited operator supplies the local radio access. Authentication and subscriber policy still depend on the home side of the roaming relationship.
Packet routing can follow more than one architecture. In a home-routed configuration, user traffic returns through the home network before reaching the Internet. With local breakout, the user-plane path can exit closer to the visited network.
Voice and messaging add their own IMS and roaming state on top of that packet path.
From the handset, this still looks like an ordinary network registration. The operator boundary is visible in the signaling and routing underneath.
Fig. 17 · Visited network and home network
International access can use a more direct local integration
Roaming is one way to reach local infrastructure. It is not the only one.
A deeper integration can provision the customer into a local operator environment more directly. Depending on the partner and architecture, that can involve a local profile, a different subscriber identity, local packet breakout, or another operator-specific arrangement.
The implementation does not have to be identical from country to country.
US Mobile still maintains the line, plan, billing, entitlements, and customer relationship above it, then maps that state into the local network integration underneath.
This gives the platform room to move international access closer to the local network where the integration supports it, rather than depending on one roaming model everywhere.
Fig. 18 · Roaming and direct local integration
Satellite uses a different access stack
Starlink terminal-based broadband does not resemble a cellular RAN underneath.
The terminal, radio link, satellite constellation, routing, and capacity model are different from the stack used by a handset on Warp, Dark Star, or Light Speed.
The integration point is higher up.
At the US Mobile service layer, the familiar objects still exist: customer, hardware, activation state, plan, usage, billing, diagnostics, and support.
That allows satellite connectivity to participate in the same service platform without forcing its physical network to look like a cellular operator.
Fig. 19 · Cellular and satellite access
Direct-to-device satellite has different radio geometry again
Direct-to-device satellite is a different technical path from terminal-based satellite broadband.
A cellular payload in low Earth orbit serves a handset over a much longer path than a terrestrial macro site. The satellite is moving relative to the UE, so slant range, radial velocity, Doppler, timing, and beam geometry change continuously during a pass.
The handset still has handset-class transmit power and antenna constraints.
3GPP NTN mechanisms address those conditions with support for satellite position, timing, moving coverage, and NTN-specific mobility behavior.
For the UE, the unusual part is that the serving cell can move even when the handset does not.
In NTN, the serving cell can move while the UE stands still.
Terrestrial and satellite networks scale capacity differently
A satellite beam can cover a large geographic area without putting terrestrial sites throughout it.
The available capacity in that beam is shared across the users and traffic inside the footprint.
A dense terrestrial network works differently. Operators divide the geography into much smaller cells and reuse spectrum across those cells, placing capacity close to concentrated demand.
In a remote region, the large satellite footprint can be a practical way to provide reach. In a dense city, terrestrial site density and spatial reuse support a very different traffic load.
The two systems are solving different capacity geometries.
Fig. 21 · Sparse geography and dense demand
The control boundary matters
US Mobile controls the service and provisioning state that determines how a line reaches an underlying network.
That includes network selection, subscription lifecycle, SIM and eSIM state, plan and entitlement mapping, usage, billing, roaming eligibility, Multi-Network relationships, and customer-facing diagnostics.
The underlying operator controls its own network. It decides where to build sites, which spectrum to deploy, how to configure sectors, how to schedule UEs, how to run mobility, and how to engineer its core and backhaul.
The device adds another boundary. Modem capability, supported bands, antenna design, simultaneous-SIM behavior, and UE transmit power all affect what is possible on the handset side.
Those boundaries are important when diagnosing a problem or deciding which part of the system can actually change it.
Fig. 22 · Ownership boundary
Diagnostics follow the same layers
A service problem becomes easier to narrow once the working layer is known.
No Service keeps the investigation near RF availability, SIM state, PLMN selection, registration, device support, or a network outage.
Bars with no data moves the search upward. Registration, packet-session state, APN or DNN configuration, policy, entitlement, and the user plane become more relevant.
If packet data works but voice does not, IMS is the next place to look. A roaming-only problem brings visited-network availability, roaming entitlement, steering, and partner state into scope.
US Mobile does not have every internal counter inside every underlying network. Some state is directly visible to us, some arrives through partner interfaces, some comes from the device, and some requires the operator to inspect its own systems.
Fig. 23 · Diagnostic stack
Integrating another network
A new network has to be mapped into the service model US Mobile already uses.
The integration usually includes subscriber provisioning, SIM and identity lifecycle, device eligibility, feature provisioning, plan and policy mapping, usage, charging, roaming, operational telemetry, and support tooling.
The exact interface is network-specific.
One operator may represent a feature as a subscriber attribute. Another may require a separate provisioning operation. Another may not support the feature at all.
The platform therefore needs both a common service model and an explicit capability model for the differences underneath.
That lets the rest of the system ask practical questions without assuming that every network exposes the same controls.
Can this device activate here?
Does this subscription support roaming?
Is hotspot available?
What usage has this line generated?
Which states mean activation is complete?
Fig. 24 · Adding a network
The resulting coverage model
US Mobile benefits when any underlying network improves its own infrastructure.
A new site, additional spectrum, more backhaul, a denser mid-band deployment, or a broader roaming footprint can all change the service available to a subscriber on that network.
The platform adds another path: connect another useful network and make it available through the same service layer.
Today, that set includes the three nationwide US cellular networks, international access, and satellite services.
They remain separate networks with different RANs, cores, spectrum, topology, and operating models. The service layer keeps the customer-facing objects and network relationships together above them.
That is the architecture behind network switching, Multi-Network, international access, and satellite integration.