Satellite Gateways, Teleports, and Points of Presence | Design, Redundancy, and Procurement Guide
Technical guide to satellite gateways, teleports, hubs, and PoPs. Covers terminology, reference architecture, site design, redundancy patterns, operations, and procurement checklist.
Satellite Gateways, Teleports, and Points of Presence
If you operate, procure, or design satellite networks, you will encounter the terms gateway, teleport, hub, and Point of Presence (PoP) in almost every vendor proposal, RFP response, and architecture diagram. These terms are frequently used interchangeably — and incorrectly — which creates confusion during design reviews, service-level negotiations, and procurement decisions. This guide disambiguates the terminology, walks through a reference architecture, examines the design trade-offs that shape real deployments, and provides an actionable procurement checklist. It is written for satellite network engineers, IT procurement managers, and systems integrators who need to evaluate ground infrastructure options and ask the right questions before committing to a provider.
Key terms used in this article — For complete definitions, see the Glossary G–L.
- Gateway: The ground station that terminates the satellite link and connects traffic to the terrestrial network.
- Teleport: A facility hosting multiple antennas, RF chains, and baseband equipment — often serving several satellites and customers.
- Hub: The baseband processing platform (hardware and software) that manages bandwidth allocation, modulation, and traffic engineering for a VSAT network.
- Point of Presence (PoP): A colocation site where satellite traffic is handed off to IP transit, peering exchanges, or enterprise networks.
- Backhaul: The terrestrial link (fiber, microwave) connecting a gateway or teleport to the nearest PoP or internet exchange.
- NOC: Network Operations Center — the 24/7 monitoring facility responsible for fault detection, performance management, and incident escalation.
- Landing rights: Regulatory authorization to transmit and receive satellite signals from a specific country or territory.
What Is a Satellite Gateway?
A satellite gateway is the ground-side anchor point of a satellite link. It receives traffic from user terminals via the satellite (forward/return channels), processes that traffic at baseband, and routes it onward to the terrestrial internet or a private enterprise network. In the reverse direction, it accepts traffic from the terrestrial side, encapsulates it into the satellite protocol, and transmits it up to the spacecraft for delivery to remote terminals.
The gateway's role differs depending on the satellite payload architecture. With a bent-pipe (transparent) payload, the satellite simply frequency-translates and amplifies the signal — all routing, encryption, traffic shaping, and protocol processing happen at the gateway. With a regenerative (processing) payload, the satellite demodulates, decodes, and may re-route traffic onboard, reducing the gateway's processing burden but increasing the satellite's complexity and cost.
In practice, the vast majority of commercial VSAT networks today use bent-pipe transponders, which means the gateway is where the intelligence lives. Encryption (often AES-256), IP routing, bandwidth allocation, Quality of Service enforcement, and traffic engineering all execute at gateway-side equipment.
Signal Reception
The gateway antenna (typically 7–13 m for GEO Ku/Ka-band) receives the downlink signal from the satellite. A low-noise amplifier (LNA) boosts the weak received signal, and a down-converter translates it from RF to an intermediate frequency for baseband processing.
Baseband Processing
The hub platform demodulates and decodes the signal, recovering the original IP packets from the satellite waveform. Error correction (typically LDPC or turbo codes), de-encapsulation, and decryption happen at this stage.
Routing and Hand-off
Recovered IP traffic is routed through the gateway's network infrastructure — firewalls, traffic shapers, and edge routers — before being handed off via fiber or microwave backhaul to the nearest PoP or internet exchange point.
Return Path (Outbound)
Traffic destined for remote terminals traverses the same chain in reverse: IP packets are encrypted, encapsulated in the satellite protocol, modulated, up-converted, amplified by a high-power amplifier (HPA), and transmitted via the antenna to the satellite.
Teleport vs Gateway vs Hub vs PoP
The satellite industry uses these four terms loosely, and many professionals treat them as synonyms. They are not. Understanding the distinctions matters when evaluating vendor proposals, negotiating SLAs, and designing for resilience.
| Term | Primary Function | Typical Scale | Key Differentiator | Who Owns It |
|---|---|---|---|---|
| Gateway | Terminate one satellite link (or beam) and connect it to terrestrial infrastructure | 1–4 antennas, single satellite/beam | The RF + baseband chain for a specific link | Satellite operator or service provider |
| Teleport | Host multiple gateways and services in a single facility with shared infrastructure | 10–50+ antennas, multiple satellites and orbits | Facility-level infrastructure: power, cooling, security, fiber diversity | Teleport operator (may lease capacity to multiple providers) |
| Hub | Manage bandwidth allocation, protocol processing, and traffic engineering | Software + hardware platform, not a physical site | The brains of a VSAT network — vendor-specific (e.g., iDirect, Hughes, Newtec) | Service provider or enterprise |
| PoP | Hand off satellite traffic to IP transit, peering, or enterprise WANs | Rack space in a carrier-neutral data center | No RF equipment — purely a network interconnection point | Service provider or content delivery network |
A single teleport may house dozens of gateways serving different satellites, each running one or more hub platforms, with traffic backhauled to one or more PoPs in nearby metro areas. When a vendor says "we have a gateway in Frankfurt," they might mean a teleport facility, a single antenna, or just a PoP with no RF equipment at all. Always ask for specifics.
Why does precision matter? Because each component has different failure modes, redundancy requirements, and SLA implications. A PoP failure (router outage) can often be resolved in minutes with remote intervention. A gateway failure (antenna or HPA fault) may require on-site technicians and several hours of downtime. A teleport-level event (power outage, fiber cut) can affect dozens of customers simultaneously.
Reference Architecture
A typical end-to-end satellite connectivity path follows this chain:
User Terminal → Satellite → Gateway (at Teleport) → Backhaul → PoP → IP Backbone → Internet / Enterprise Network
Each node performs a distinct function. The user terminal (VSAT, flat-panel antenna, or maritime stabilized dish) establishes the RF link to the satellite. The satellite relays the signal to the gateway, where baseband processing converts it to IP traffic. The gateway sits within a teleport facility that provides shared power, cooling, physical security, and fiber connectivity. A backhaul link — typically redundant dark fiber, though microwave is used in some regions — carries the IP traffic to a PoP located in a carrier-neutral data center or internet exchange. At the PoP, traffic is handed off to IP transit providers, peering partners, or enterprise MPLS networks for onward delivery.
The backhaul segment between teleport and PoP is often overlooked in architecture reviews but represents a critical single point of failure. Best practice is to maintain two physically diverse fiber paths from the teleport to at least two PoPs, ideally served by different fiber providers and entering the data center through separate cable vaults.
For deeper coverage of the full three-segment satellite model, see our End-to-End Architecture guide. For a practical introduction to how the signal path works, see How Satellite Internet Works.
Key Design Considerations
Path selection is the single largest driver of user-experienced latency. A GEO gateway adds approximately 600 ms round-trip time from the physics of the 35,786 km orbital altitude alone. MEO constellations (O3b mPOWER) reduce this to approximately 150 ms, while LEO systems (Starlink, OneWeb) target 20–40 ms. The gateway's geographic proximity to the PoP and the PoP's proximity to content servers further influence total latency — every additional 1,000 km of terrestrial backhaul adds roughly 10 ms of round-trip delay.
Throughput planning requires understanding oversubscription ratios. Most satellite service providers oversubscribe gateway capacity by 4:1 to 20:1 depending on the service tier. Ask for committed information rates (CIR), peak information rates (PIR), and the contention ratio during busy hour. A gateway rated at 10 Gbps aggregate throughput with 20:1 oversubscription supports only 500 Mbps of sustained committed traffic.
For a detailed comparison of latency across orbital types, see Satellite Latency Comparison.
Rain fade is the dominant weather-related impairment for satellite links, particularly at Ka-band (26.5–40 GHz) where atmospheric attenuation during heavy rainfall can exceed 10 dB. Ku-band (12–18 GHz) is more resilient but still affected in tropical regions with intense precipitation.
Geographic diversity is the primary mitigation strategy. By placing two gateways at least 300 km apart (ideally in different climate zones), the probability of simultaneous rain fade at both sites drops to near zero. This is standard practice for Ka-band HTS systems where individual spot beams cover areas of 200–500 km diameter.
Site separation must account for the satellite's beam coverage footprint. Both diverse gateways must be within the same beam or within beams that can be interconnected via onboard switching, which adds architectural complexity.
For more on frequency band trade-offs, see Ku-Band vs Ka-Band Satellite. For the mathematics behind link margin and fade mitigation, see Satellite Link Budget Calculation.
ITU coordination is required before any gateway can begin transmitting. The process involves registering the earth station's frequency assignments with the International Telecommunication Union, coordinating with adjacent satellite operators to avoid harmful interference, and obtaining national spectrum licenses from the local regulator. This process routinely takes 6–18 months and is often the longest lead-time item in a new gateway deployment.
Landing rights — the authorization to operate a satellite earth station in a specific country — vary dramatically by jurisdiction. Some countries require that a gateway be owned or majority-owned by a local entity. Others restrict which satellite operators can provide service within their territory. These regulatory constraints directly influence the geographic placement of gateways and PoPs.
Interference monitoring is an ongoing operational requirement. Gateways must continuously monitor for adjacent satellite interference (ASI), terrestrial interference from 5G or radar systems (particularly in C-band and portions of Ku-band), and coordinate with the satellite operator's spectrum management team when anomalies are detected.
Reliability and Redundancy Patterns
Satellite ground infrastructure redundancy operates at two levels: equipment redundancy within a single site and geographic diversity across multiple sites.
Equipment redundancy (N+1) means that for every N active units of a critical component, one additional standby unit is available. For a gateway with two active HPAs, an N+1 configuration provides a third HPA on standby with automatic switchover via a waveguide switch. This pattern applies to HPAs, LNAs, up/down-converters, modulators, and demodulators. For the details of individual 1+1 RF chains and switchover mechanisms, see Ground Segment & Hubs.
Geographic diversity protects against site-level failures: power outages, fiber cuts, natural disasters, and widespread weather events. Two (or more) geographically separated gateway sites provide coverage for the same satellite beam, with traffic automatically re-routed if one site becomes unavailable.
Automatic failover mechanisms include:
- Make-before-break: The standby gateway is continuously receiving and processing the satellite signal in parallel. When the primary gateway degrades, traffic is seamlessly shifted to the standby with no packet loss. This is the gold standard for high-availability services.
- Warm standby: The standby gateway is powered and configured but not actively processing traffic. Failover requires the standby to acquire the satellite signal and synchronize, resulting in 30–120 seconds of service interruption.
- Cold standby: Equipment is installed but not powered. Failover requires manual intervention and may take 15–60 minutes. This approach is only acceptable for best-effort services.
| KPI | Target (Tier 1) | Target (Tier 2) | Measurement Method |
|---|---|---|---|
| Service availability | 99.95% | 99.5% | Monthly uptime excluding scheduled maintenance |
| Failover time (equipment) | < 50 ms | < 5 s | Measured from fault detection to traffic restoration |
| Failover time (site) | < 2 s | < 120 s | Make-before-break vs warm standby |
| MTTR (Mean Time to Repair) | < 4 hours | < 8 hours | Time from fault detection to full N+1 restoration |
| Packet loss during failover | < 0.01% | < 1% | Measured at PoP egress point |
| Jitter during failover | < 10 ms | < 50 ms | Measured at PoP egress point |
Operations and Monitoring
Running a satellite gateway or teleport requires continuous oversight across RF performance, network health, and physical infrastructure.
Key operational KPIs extend beyond the RF metrics (Es/No, BER, packet error rate) covered in the ground segment equipment guides. Gateway operations teams track aggregate throughput utilization (percentage of total capacity in use), session counts (active terminals), backhaul utilization, and PoP interconnection health. These metrics are monitored via SNMP, syslog, and proprietary hub management platforms.
Alarm management follows a tiered escalation model. Tier 1 alarms (link loss, HPA failure, complete site outage) trigger immediate automated failover and require human acknowledgment within 5 minutes. Tier 2 alarms (Es/No degradation, elevated error rates, partial capacity loss) may indicate developing weather fade or equipment aging and require investigation within 30 minutes. Tier 3 alarms (non-critical warnings, capacity thresholds) are logged for trending and reviewed during regular operations meetings.
Change management is critical in satellite operations because the shared nature of the satellite medium means that misconfigured carrier parameters can interfere with adjacent services. All changes to modulation, symbol rate, frequency plan, or power levels require a formal change request, peer review, and execution during an agreed maintenance window. Unplanned capacity augmentations carry particular risk — increasing transmit power without coordinating with the satellite operator can trigger adjacent satellite interference and regulatory penalties.
Staffing models vary by facility scale and service tier. Large teleports typically maintain a 24/7 on-site NOC with shift engineers. Smaller gateways may rely on remote monitoring from a centralized NOC, with on-call field technicians dispatched for hardware faults. The trade-off is response time: a remotely monitored gateway with a 2-hour technician response time cannot meet a 4-hour MTTR SLA for equipment failures, so the staffing model must align with the contracted availability targets.
Procurement Checklist
When evaluating gateway, teleport, or managed satellite services, use the following questions to assess provider capabilities and identify gaps before signing a contract.
- Service Level Agreement: What is the committed availability percentage? Does it include or exclude scheduled maintenance windows? What are the financial penalties (service credits) for SLA breaches?
- Geographic diversity: How many gateway sites serve your beam/region? What is the physical separation between sites? Is failover automatic or manual?
- PoP locations: Where are the provider's PoPs? Are they in carrier-neutral data centers with multiple transit and peering options? Can you cross-connect directly to your cloud provider (AWS Direct Connect, Azure ExpressRoute, Google Cloud Interconnect)?
- Backhaul architecture: How many fiber paths connect the teleport to the PoP? Are they physically diverse (different cable routes, different fiber providers)?
- Oversubscription: What is the contention ratio on the gateway? What CIR and PIR do you receive? How does the provider manage congestion during peak hours?
- Regulatory compliance: Does the provider hold all necessary landing rights and spectrum licenses for your operating region? Who is responsible for ITU coordination if you expand to new markets?
- Monitoring and reporting: What visibility do you get into gateway and PoP performance? Do you receive real-time dashboards, or only monthly reports? Can you integrate their monitoring data into your own NOC tools?
- Scalability: How much additional capacity is available on the current gateway? What is the lead time to provision additional bandwidth? Are there capacity caps per customer?
Common pitfalls to watch for:
- Providers quoting availability based on satellite uptime only, excluding gateway, backhaul, and PoP components from the SLA calculation.
- "Diverse gateways" that share a common fiber backhaul path or terminate at the same PoP — eliminating the diversity benefit.
- PoPs located in a single data center with no alternative interconnection if that facility has an outage.
- Oversubscription ratios that look reasonable on paper but are measured at the satellite level rather than the gateway level, hiding local congestion.
For a broader comparison of satellite service providers and their infrastructure, see Satellite Service Providers.
Frequently Asked Questions
What is the difference between a satellite gateway and a teleport?
A gateway is the RF and baseband equipment chain that terminates a single satellite link or beam. A teleport is the physical facility that houses multiple gateways, shared infrastructure (power, cooling, fiber), and often serves multiple satellite operators and customers. A teleport always contains at least one gateway; a gateway does not require a full teleport facility.
What does a Point of Presence (PoP) do in a satellite network?
A PoP is the network interconnection point where satellite traffic is handed off to the terrestrial internet, cloud providers, or enterprise networks. It contains routers and switches but no RF equipment. Its location determines the last-mile terrestrial latency and the available peering and transit options for satellite traffic.
What availability SLA should I expect from a managed satellite service?
Tier 1 services with geographic gateway diversity and make-before-break failover typically commit to 99.95% availability (approximately 4.4 hours of downtime per year). Standard services with single-site gateways and warm standby generally offer 99.5% (approximately 43.8 hours per year). Always confirm whether the SLA covers end-to-end service or only individual components.
What are the main cost drivers for gateway infrastructure?
The largest cost components are antenna systems and RF equipment (30–40% of CAPEX), the teleport facility lease or construction (20–30%), fiber backhaul (10–15%), and hub platform licensing (10–20%). OPEX is dominated by staffing, power, satellite bandwidth, and IP transit costs. Geographic diversity roughly doubles the infrastructure cost.
Can satellite gateways connect directly to cloud providers?
Yes, many teleport operators and satellite service providers offer direct cloud interconnection at their PoPs via AWS Direct Connect, Azure ExpressRoute, or Google Cloud Interconnect. This bypasses the public internet, reducing latency and improving security for enterprise traffic. Confirm that the provider's PoP is in a data center where your cloud provider has a presence.
How do LEO gateways differ from GEO gateways?
LEO constellations require many more gateways than GEO systems because each satellite is only visible from a given ground location for a few minutes. A LEO constellation like Starlink or OneWeb operates dozens to hundreds of gateways globally to maintain continuous coverage. Each LEO gateway must also handle rapid beam handoffs as satellites pass overhead, requiring more sophisticated tracking antennas (or electronically steered arrays) and faster baseband processing.
How many gateways does a typical satellite network need?
For a GEO network serving a single region, two geographically diverse gateways are standard (one primary, one backup). A multi-region GEO operator may operate 4–10 gateways globally. LEO constellations require 40–100+ gateways worldwide to provide continuous coverage, with the exact number depending on orbit altitude, constellation size, and inter-satellite link capability.
Summary
Key Takeaways:
- A gateway terminates the satellite link, a teleport is the facility hosting gateways, a hub is the processing platform, and a PoP is the terrestrial hand-off point — they are distinct components with different failure modes and SLA implications.
- The backhaul between teleport and PoP is a frequently overlooked single point of failure — demand physically diverse fiber from different providers.
- Geographic diversity (sites 300+ km apart) is essential for Ka-band services to mitigate rain fade and provides site-level disaster resilience.
- Make-before-break failover is the gold standard for high-availability services, achieving sub-second site switchover with near-zero packet loss.
- Regulatory lead times (landing rights, ITU coordination) of 6–18 months often gate new gateway deployments — start early.
- Use the procurement checklist to verify that vendor SLAs cover end-to-end service, not just individual components, and that stated diversity is genuine.
Related Articles
- Satellite Communication Basics — Foundational concepts and terminology
- How Satellite Internet Works — End-to-end signal path explanation
- End-to-End Architecture — Three-segment system overview
- Ground Segment & Hubs — Equipment specifications and RF chain details
- VSAT Network Architecture — Network topology and design patterns
- Ku-Band vs Ka-Band Satellite — Frequency band comparison
- Satellite Latency Comparison — Latency across orbital types
- Satellite Link Budget Calculation — Link margin and fade analysis
- Satellite Service Providers — Provider comparison and evaluation
Author
Categories
More Posts
BER, FER, and Packet Loss Explained: How Satellite Link Errors Affect Real Network Performance
Engineering guide to bit error rate, frame error rate, and packet loss in satellite communication covering RF-to-IP error propagation, FEC recovery, ACM interaction, and practical troubleshooting.
Satellite Jitter Explained: Why Delay Variation Matters in SATCOM Networks
Engineering guide to jitter in satellite networks covering causes, GEO vs LEO behavior, VoIP impact, measurement methods, QoS mitigation, and troubleshooting.
BUC vs LNB vs LNA in Satellite Systems Explained
Engineering guide comparing BUC, LNB, and LNA satellite RF components covering signal flow, selection criteria, failure modes, and practical troubleshooting.
Newsletter
Join the community
Subscribe to our newsletter for the latest news and updates