Satellite Ground Segment: Gateways, Teleports, and Handoffs
This page maps the Earth-side boundary of a satellite service: the earth-station RF path, baseband or waveform processing, traffic gateway, network control and management, terrestrial service handoff, and the facility that hosts those functions. It is an architecture and acceptance reference, not a universal bill of materials.
In regulatory language, an earth station is a station on or near Earth intended to communicate with one or more space stations. In project language, gateway, teleport, hub, and point of presence (PoP) describe different roles but are often used inconsistently. The working definitions below keep the physical site, satellite access function, control function, and terrestrial handoff separate; the signed interface-control document and licence remain authoritative for a deployment.
Do not assume that every ground segment is a transparent star network carrying IP. Transparent, mesh, regenerative, and non-geostationary systems can place traffic, control, and switching functions at different locations. The design must therefore show the actual user, control, management, and—when in scope—spacecraft tracking, telemetry, and telecommand paths.
Scope note: Organization fact check updated 26 August 2026 against the primary references below. This is not a named expert approval. Gateway terminology, licences, payload and waveform design, terrestrial handoffs, security controls, and acceptance limits must be confirmed for the actual network.
Ground-Segment Functions and Boundaries
- Earth-station aperture and RF interface — antenna, feed, polarization, pointing or tracking, transmit chain, receive chain, filtering, and the declared reference planes used for power, gain, noise, and loss measurements.
- Baseband and access system — modulation, coding, framing, multiplexing, encapsulation, synchronization, and access control required by the selected air interface. A modem or hub platform may implement some or all of these functions.
- Traffic or service gateway — the user-plane function that connects satellite access to a terrestrial service. Its handoff may be routed IP, Ethernet, a private network interface, or another service-specific boundary.
- Network Control Centre (NCC) — access-control functions such as terminal admission, signalling, synchronization, and resource assignment where the chosen architecture requires them.
- Network Management Centre (NMC) or operations platform — configuration, inventory, alarms, performance data, service policy, and reporting. Control and management can share equipment but should remain separate logical roles.
- Terrestrial and facility infrastructure — backhaul, power, timing, cooling, grounding, physical security, and site access. The required implementation follows the availability target, local hazards, and ownership model.
- Spacecraft operations boundary — tracking, telemetry, and telecommand are space-operation functions. They may share a site with a service gateway, but they are not automatically part of the customer traffic path and should be shown separately when included.
Gateway, Teleport, Hub, and PoP Are Not the Same Boundary
Use gateway for the functional bridge between satellite access and a terrestrial network; teleport for the physical facility that can host one or more earth stations and customer or operator equipment; hub for the shared access, baseband, control, or traffic platform used by a network; and PoP for a terrestrial interconnection point. A provider may combine all four at one address or distribute them across several sites.
A service diagram should identify the radio reference plane, baseband boundary, customer or carrier handoff, control endpoint, management endpoint, owner, and acceptance test for each interface. A street address or equipment rack is not by itself a service demarcation.
Forward and return describe direction relative to a terminal; uplink and downlink describe the direction of a radio hop. Because a path can use different gateways, beams, satellites, or terrestrial routes in each direction, document both directions instead of drawing one reversible arrow.
- Transparent star example — terminal traffic traverses the satellite to a hub or traffic gateway. NCC and NMC functions may be colocated with the gateway, but their logical roles and failure effects should still be identified.
- Mesh or regenerative example — user traffic may not pass through the control station, and on-board processing can change the uplink and downlink air-interface relationship. Do not infer the user-plane route from the location of the NMC or NCC.
- NGSO or multi-gateway example — satellite visibility, feeder-link availability, handover, and terrestrial routing can change the active path. State whether continuity is provided by the access system, the service network, the application, or a combination of them.
Antenna and RF Design: Specify Reference Planes
Antenna diameter and amplifier power are outputs of the design, not universal gateway specifications. Size the earth station from the assigned frequency and polarization, required transmit EIRP or EIRP density, receive G/T, satellite contours, carrier plan, propagation model, interference constraints, emission limits, pointing loss, and availability objective.
The pointing system follows the apparent motion and accuracy requirement of the selected spacecraft and service. A nominally geostationary link can still require tracking or repointing, while the mechanism for a non-geostationary link depends on scan range, slew rate, handover strategy, and regulatory authorization. Record the assumptions rather than selecting an antenna solely from the orbit label.
- Receive chain — identify the antenna/feed reference plane, polarization network, filters, LNA or LNB where used, frequency conversion, distribution losses, receiver input range, and the point at which G/T or system noise temperature is specified and verified.
- Transmit chain — identify the modulator output, conversion stages, BUC or separate HPA where used, output back-off, filters, switching, feeder loss, antenna gain, and the plane at which carrier power and unwanted emissions are measured.
- Shared RF hardware — diplexers, orthomode transducers, combiners, waveguide, cables, and switches add loss, isolation limits, power limits, and possible single points of failure. Include their actual configuration in both the link budget and failure analysis.
- Pointing and polarization — define acquisition, tracking, pointing-loss allowance, polarization alignment, exclusion or keep-out constraints, and the acceptance measurement. The selected satellite operator and administration may impose additional procedures.
- Protection scheme — state the protected service, failure-detection criterion, switching authority, spare capacity, configuration-state handling, expected impairment, and return-to-service procedure. Labels such as 1+1 or N+1 do not prove hitless failover.
Baseband, Control, Management, and Service Edge
Baseband converts the selected service representation into the framing, modulation, coding, and access procedures used across the satellite link, and performs the inverse receive functions. DVB-S2, DVB-S2X, DVB-RCS2, MF-TDMA, SCPC, ACM, and IP are possible design choices—not properties of every gateway.
Keep the user plane, control plane, and management plane visible even when one platform implements all three. In DVB-RCS2, for example, the Hub/NCC, NMC, and traffic-gateway roles are distinct: access signalling and resource control are not the same as service traffic or operational management.
The terrestrial service edge requires its own interface definition. Record physical and link layers, encapsulation, addressing and routing responsibility, maximum transmission unit, security boundary, QoS policy, multicast behaviour where relevant, timing dependency, and the exact point at which service performance is measured.
- Carrier and waveform plan — forward and return carriers, symbol rates, coding and modulation modes, access method, synchronization source, and permitted adaptation behaviour.
- Control contract — terminal identity, admission, authorization to transmit, resource assignment, handover, and behaviour when the control path is unavailable.
- Management contract — configuration source of truth, software and parameter versions, inventory, telemetry, alarms, audit records, and time synchronization.
- Service handoff — owner and test point for routing, security, QoS, availability, latency, loss, throughput, and any acceleration or proxy function.
Operations, Observability, and Failure Proof
Reliability is demonstrated at the service boundary, not inferred from the presence of standby equipment. The operations design must correlate RF, baseband, control, management, terrestrial, power, and environmental state with a common time basis and a documented ownership model.
- Measurement map — collect RF power or quality, modem state, access-control state, queue or traffic state, terrestrial interface status, and facility alarms at named reference points. Define sampling, retention, clock accuracy, and data ownership.
- Alarm and incident model — distinguish symptom from cause, map dependencies, state escalation and communication responsibilities, and preserve configuration and event evidence for post-incident analysis.
- Spectrum and emission checks — define how authorized carriers, spectral masks, polarization, frequency, and unexpected emissions are verified. Monitoring scope and response procedures must match the licence and operator agreement.
- Change and failover tests — exercise equipment, link, site, terrestrial, power, and control-path failures that the design claims to survive. Measure detection time, switching impairment, capacity in the degraded state, session impact, and recovery.
- Site operations — derive spares, maintenance access, physical security, grounding, cooling, weather protection, and backup-power arrangements from the local risk assessment and restoration objective rather than a generic teleport checklist.
Ground-Segment Design and Acceptance Checklist
1. Freeze scope and terminology: list every site, earth station, gateway, hub, NCC, NMC, PoP, service edge, TT&C function, owner, and contractual demarcation. Mark functions that are out of scope.
2. Qualify the site: confirm spectrum authorization, satellite visibility and pointing envelope, interference environment, propagation inputs, terrestrial routes, power, grounding, environmental loads, access, security, and restoration logistics.
3. Baseline interfaces and budgets: publish the RF reference planes, link budget, frequency and carrier plan, interface-control document, timing plan, addressing and routing, security boundary, and capacity model for both directions.
4. Define normal and degraded service: specify availability and performance measurement points, protected failure cases, surviving capacity, handover or reroute authority, maintenance constraints, and restoration objectives.
5. Prove acceptance: record calibrated RF and spectrum measurements, end-to-end service tests, control and management tests, failover results, alarm delivery, configuration versions, open exceptions, and the party accepting each result.
6. Hand over operations: provide as-built diagrams, licences and operator approvals, baselines, backups, spares policy, escalation contacts, maintenance procedures, evidence retention, and a schedule for revalidation after change.
What a Complete Ground-Segment Design Must Prove
A complete design shows where the satellite access system ends, where the terrestrial service begins, how user traffic differs from control and management, and whether spacecraft operations are inside or outside the boundary. Every interface has an owner, reference plane, configuration baseline, measurement, and acceptance criterion.
Gateway diversity, redundant equipment, and alternate backhaul are useful only when their failure domains, capacity, switching logic, control dependencies, and measured service impact are known. The final evidence should let an operator reproduce the test and determine which component or organization owns a failure.
This reference intentionally avoids universal dish sizes, amplifier powers, waveforms, topology counts, and availability promises. Those values must come from the actual satellite network, link and propagation budgets, licences, equipment specifications, site conditions, and service contract.