How can a phone built for nearby cell towers work with a satellite hundreds of kilometres away? The difficult part is keeping the handset compatible while providing enough link margin, timing control and capacity for a useful service.

This page follows the problem from the phone antenna through the radio procedures to the delivered service. Read D2C Overview first for the direct-access configuration. The examples here are engineering calculations, not advertised performance figures.

• Why Is an Ordinary Phone the Main Constraint ?
• How Does a Small Phone Antenna Reach Space ?
• Worked Example: The Uplink Before Receiver Noise
• A Coverage Footprint Does Not Guarantee Indoor Service
• How Can the Network Fit Terrestrial Timing and Frequency Behavior ?
• Why Does One Successful Call Not Prove Large-Scale Capacity ?
• Worked Example: Shared Resources in One Beam
• Coverage, Mobility and Battery Life
• Spectrum, Interference and Deployment Cost
• What Should a D2C Trial Actually Prove ?
• References

Why Is an Ordinary Phone the Main Constraint ?

Conventional D2C tries to connect a phone using a cellular interface that it already supports. The satellite system has to fit that interface. It cannot assume that the phone accepts arbitrary timing changes or implements a new satellite-specific compensation procedure.

This is the scope of this page: direct satellite access using existing cellular handsets. Standardized NR NTN and IoT NTN add satellite-specific capabilities and are discussed separately in NTN Challenges. The terms are compared in D2C Overview.

Figure 1 separates three conditions for service. A phone may support the radio band but fail to acquire the cell. It may register but lack the required service support. These are different failures and need different evidence.

Figure 1. Handset compatibility, successful access and usable service must all be verified.

An unmodified handset demonstration establishes feasibility for its tested device and configuration. It does not establish support for every model, firmware build, SIM or country. Start a device matrix with band support and modem behavior, then add operator provisioning and the intended services. See Device Support.

A phone's GNSS receiver does not automatically give its cellular modem a satellite compensation feature. The modem must implement the relevant behavior, and the network must use procedures that the device actually supports.

How Does a Small Phone Antenna Reach Space ?

The phone has limited transmit power and no large directional satellite antenna. Its orientation also changes in the user's hand. The satellite receiver must recover that weak uplink while the downlink must remain detectable by the same small handset antenna.

Worked Example: The Uplink Before Receiver Noise

Assume a 2 GHz carrier, 600 km slant range, 23 dBm conducted phone power, 0 dBi effective transmit antenna gain and 30 dBi satellite receive gain. Add 3 dB of other losses. These are illustrative inputs, not the specification of a commercial constellation.

Free-space loss = 92.45 + 20 log10(2) + 20 log10(600) = 154.0 dB. Received power = 23 + 0 + 30 - 154 - 3 = -104 dBm. At 1,200 km, the same calculation gives about -110 dBm because doubling range adds 6 dB of loss.

Is -104 dBm enough? We cannot answer without receiver noise, occupied bandwidth, interference and the required decoding threshold. For example, thermal noise plus a 3 dB noise figure is approximately -111 dBm in 1 MHz at room temperature. That gives roughly 7 dB SNR for the assumed received power, before interference. This is not a throughput prediction.

Large satellite arrays, sensitive receivers and beamforming can improve the link. However, additional gain has costs in deployed aperture, payload power, calibration and beam control. A stronger downlink alone cannot repair an uplink deficit. See Radio Link for the separate uplink and downlink questions.

A Coverage Footprint Does Not Guarantee Indoor Service

A building, vehicle roof, tree canopy or the user's body can consume the available margin. A phone can therefore work at one outdoor orientation and fail a short distance away. Tests need representative obstructions and orientations rather than only an open-sky setup.

This also explains why directional satellite broadband terminal experience does not transfer directly to D2C. The antenna gain, installation and power budget are different. State the tested environment whenever quoting an access success rate or service availability result.

How Can the Network Fit Terrestrial Timing and Frequency Behavior ?

A conventional cellular receiver expects particular acquisition ranges, timing relationships and response behavior. The satellite network must accommodate long paths and rapid motion within those supported behaviors. A vendor can adapt the network extensively, but software cannot remove signal travel time.

At 600 km, one service-link traversal takes about 2 ms. If the relevant protocol endpoint is on the ground behind a transparent payload, the feeder path also contributes. If it is onboard, the timing budget is different. Always identify the endpoint before quoting a round-trip delay.

At a 2 GHz carrier and 7 km/s radial velocity, the Doppler magnitude is about 46.7 kHz. A network correction can reduce the common component, but users at different locations have different radial velocities and ranges. Beam size and placement therefore affect the residual error that each handset experiences.

Initial access is particularly demanding because the network has not yet completed the normal per-UE timing adjustment. After access, it still needs to track motion and maintain the timing of uplink data and feedback. A successful downlink decode does not prove that the response will arrive in the correct receive window.

The January 2024 implementation example in References describes standard LTE compatibility, large arrays and choices involving altitude, beam geometry and elevation. It does not publish a complete compensation algorithm. Avoid presenting one assumed timing workaround as the method used by every provider. Technology and Procedures discuss these boundaries.

Why Does One Successful Call Not Prove Large-Scale Capacity ?

A working link answers whether the receiver can decode a transmission. Capacity answers how many transmissions can share the available resources with acceptable delay. A large satellite footprint can include many users, particularly during a terrestrial outage when demand rises at the same time.

Worked Example: Shared Resources in One Beam

Assume a beam has 5 MHz of usable scheduled bandwidth and achieves an aggregate net spectral efficiency of 0.5 bit/s/Hz after the overhead included in this example. Its resulting shared rate is 2.5 Mbit/s. These assumed values illustrate resource sharing; they are not a provider measurement.

Real scheduling is not equal sharing. Some users need more airtime because their link is weaker, and control traffic, retransmissions and service priorities change the result. Short messages can tolerate a queue that would make an interactive voice session unusable. The number of registered users is also different from the number actively transmitting.

Narrower beams and frequency reuse can increase total capacity, but sidelobes and adjacent beams create interference constraints. Satellite power must be distributed across the active beams. More visible satellites add usable capacity only when spectrum, antenna resources and network coordination allow them to serve the demand.

Measure loaded throughput and delay at the service boundary. A peak rate from one favorable test does not establish a busy-hour rate or continuous availability. See Performance for the distinction between link rate, shared capacity and user experience.

Coverage, Mobility and Battery Life

The satellite moves even when the phone does not. The network must maintain service as geometry changes, beams change and satellites leave view. The phone also has to return to terrestrial coverage when appropriate without becoming trapped in repeated access or selection attempts.

A beam change can be handled internally or can require a cell change, depending on the design. Where handover is used, its interruption and failure behavior matter to the application. An ongoing voice call has a different tolerance from a message waiting for delivery. Do not infer seamless continuity from a successful initial connection.

A sparse constellation may provide service opportunities separated by gaps. In that case, message delivery delay includes waiting for coverage, not just propagation. Search behavior and retry policy also affect battery consumption. Repeated unsuccessful access at the edge of coverage can use more energy than a short successful transaction.

Service integration adds another set of conditions. The network needs subscriber authentication, correct routing and support for the intended messaging, data or voice service. Emergency services require their own end-to-end verification, including location and routing where applicable. A radio attach result does not establish that every service is available.

A useful trial records time to discover service, time to register, transaction completion probability, interruption duration and battery use. Repeat it through an entire pass and through a return to terrestrial coverage. Architecture helps identify which satellite, gateway and operator components participate in each step.

Spectrum, Interference and Deployment Cost

Using a cellular band helps the phone remain compatible, but it creates coexistence questions for the network. The satellite illuminates a wide area, and the same or adjacent frequencies may already support terrestrial users. A border or a terrestrial coverage gap does not eliminate interference paths.

Beam shaping, sidelobe control, power limits, scheduling and geographic restrictions can reduce interference. These measures also limit capacity or coverage. The engineering assessment needs both directions: satellite emissions affecting ground networks and ground transmissions affecting the satellite receiver. Aggregate interference from many transmitters may matter even when each individual contribution is small.

Authorization is jurisdiction-specific. The US FCC's 2024 supplemental coverage from space framework is one example of coordinating satellite access with terrestrial licensees; it is not a worldwide permission to use cellular spectrum from orbit. Check the applicable authorization and operating conditions for the actual country and service. See Spectrum.

Commercial scaling adds satellite manufacture, launch, replacement, gateways, network integration and customer support. Larger antennas may improve the radio link while increasing spacecraft complexity. More satellites may reduce coverage gaps while increasing continuing fleet costs. The useful business measure is cost per delivered service under load, not only the cost of placing one satellite in orbit.

Deployment claims should therefore state supported devices, geographic authorization, service type, expected coverage gaps and performance assumptions. The same constellation can offer different services in different markets because radio capability, operator integration and authorization do not progress together.

What Should a D2C Trial Actually Prove ?

Start with the proposed service rather than a generic satellite connection. Define which phone, which band, which subscription and which environment are in scope. Then test the complete transaction and record why each failed attempt stopped.

Include weak and strong conditions. Test the cell center and edge, open sky and representative blockage, quiet and loaded beams. Keep failed trials in the statistics. Reporting only successful sessions makes access time and availability look better than a user will experience.

Finally, distinguish a demonstration, a supported device list and an operating service. Each answers a different question. A demonstration proves a configuration can work; a service assessment needs repeatability, support, authorization and enough resources for its users. Those distinctions make comparisons much more useful than comparing satellite antenna size or a single peak speed.

References

The implementation example is historical and is used for its engineering constraints, not as a current service-availability announcement. The regulatory source describes the US framework adopted in 2024.

• January 2024 Direct to Cell technical update : SpaceX, conventional LTE handset access, link constraints and satellite implementation.
• DOC-401208A1 : FCC, announcement of the supplemental coverage from space framework, March 2024.