Chapter 40: Satellites: Communications, Positioning, Observation
Era span: 1957 Sputnik → mega-constellations · Difficulty: extreme
Requires: Ch 34, Ch 35, Ch 39
Unlocks: Ch 47
Data snapshot: volatile figures in this chapter (prices, capacities, deployment counts, regulation, and capability claims) reflect published sources through 2024 unless dated otherwise; check current data before planning.
A satellite is a robot in vacuum doing one job for fifteen years without repair. This chapter covers the bus (everything shared), then the three payload families that dominate a planet: communications, positioning, observation.
40.1 Orbital Mechanics Minimum
- LEO (~400–1,200 km): ~7.5 km/s, ~90–110 min periods; latency low, coverage narrow, drag real.
- MEO (~20,000 km): navigation's home (GPS constellation).
- GEO (35,786 km altitude): matches Earth's rotation, giving a nominally fixed position. Propagation alone for a user→satellite→gateway route is on the order of a quarter-second; end-to-end latency depends on distance to the gateway, routing, processing, protocol, and any additional terrestrial path. A fixed satellite does not “cover one third of the planet” in a service sense without antenna gain, interference, capacity, and visibility constraints.
- Station-keeping burns fuel against perturbations; when propellant ends, the satellite dies — orbit lifetime IS mission lifetime. Sun-synchronous orbits precess to keep constant solar time over ground (remote sensing favorite).
| Regime | Period | Footprint | Latency | Enemies |
|---|---|---|---|---|
| LEO 500 km | ~95 min | ~2,000 km wide | ~ms | Drag, debris, handoff complexity |
| MEO 20,200 km | ~12 h | Continental | ~100 ms | Radiation belts, clock stability |
| GEO 35,786 km | 24 h (fixed) | Constrained by antenna gain and gateway reach | ~0.25 s propagation on one route; route-dependent total | Launch cost, slot crowding |
| Sun-sync ~700 km | ~100 min | Narrow strip | ~ms | Downlink volume, dawn/dusk power |
40.2 The Bus
Every satellite shares:
- Power: solar arrays (III-V cells >30 % efficiency on premium birds) + batteries through eclipse seasons; power budget rules everything else.
- Thermal: vacuum has no convection — radiators face deep space, louvers/heat pipes move heat, MLI blankets buffer solar cycling.
- Attitude control: reaction wheels slew precisely using stored momentum (magnetorquers unload it against Earth's field); star trackers give arcsecond knowledge; thrusters for large maneuvers.
- Command/data: TT&C links, radiation-hardened or redundant computers (Ch 35 parts screened against single-event upsets).
Power-budget law: payload watts + bus watts + margin (≥20 %) ≤ array output at end-of-life (degraded by radiation) through worst eclipse on batteries. Size arrays for year 15, not year 1 — a satellite that starves at year 8 dies of optimism, not age. Thermal twin rule: every watt generated must be radiated; size radiators to the hot case, heaters to the cold.
40.3 Communications Relays
Lineage: Echo balloon (passive bounce) → Telstar (1962, active transponder) → Syncom (1963, GEO) → Intelsat global telephony. Transponders receive uplink, shift frequency, amplify, downlink — bandwidth sold by MHz. Antenna engineering (spot beams reuse frequencies geographically) and encryption layers serve broadcasters, carriers, and remote regions fiber can't reach (Ch 41 carries the bulk; satellites own the last mile everywhere else).
UPLINK (6/14/30 GHz) → transponder: filter → shift frequency → amplify (TWTA/SSPA)
→ DOWNLINK (4/12/20 GHz) → spot beam footprint → dishes ([§40.7](#407-ground-segment))
Link arithmetic (one line): received power = transmit × gains ÷ path loss (∝ range² × frequency²). GEO's 36,000 km costs ~200 dB — hence big dishes, premium amplifiers, and frequency-reuse spot beams. Rain fades Ka-band; size margin for the local monsoon, not the datasheet's desert.
40.4 Navigation: GPS as Civilizational Utility
Architecture: ~24+ satellites in MEO, each carrying atomic clocks, broadcasting precise time + orbit ephemerides continuously.
- Trilateration: receiver solves position from signal travel times to ≥4 satellites (x,y,z + clock offset). Your phone's clock is garbage; the fourth satellite corrects it.
- Relativity is not optional: satellite clocks tick ~38 μs/day fast net of special+general relativistic effects — uncorrected, positions drift ~10 km/DAY. One of the most elegant proofs-by-necessity in engineering.
- Accuracy tiers: open service meters-level; augmentation (WAAS/RTK) reaches centimeters for survey/agriculture/machine control.
- Dependence audit (do this seriously): aviation approaches, shipping, banking timestamps, power-grid sync, logistics all quietly assume GNSS. Jam/spoof resilience = critical-infrastructure planning (Ch 47).
| Tier | Accuracy | Needs | Serves |
|---|---|---|---|
| Open standalone | ~3–5 m | Any receiver | Hiking, cars, phones |
| SBAS (WAAS/EGNOS) | ~1 m | Geostationary corrections | Aviation approaches |
| RTK / PPP | ~cm | Base station / precise orbits | Survey, autosteer, machine control |
| Holdover (no GNSS) | Drifts | Chip-scale atomic + inertial | Jam/spoof survival |
40.5 Observation
- Weather satellites: TIROS (1960) began continuous atmospheric watching; hurricane warnings transformed mortality statistics — a quiet but major life-saving program. Geostationary imagers loop storms live; polar orbiters feed forecast models globally.
- Land/ocean remote sensing: Landsat continuity since 1972 gives decades-comparable surface records; spectral bands diagnose crops (NDVI vegetation index), water quality, mineral signatures; radar (SAR) images through cloud/night measuring ground deformation millimeter-scale (earthquake/volcano/subsidence science).
- Climate accounting (ice mass, sea level, carbon flux) exists ONLY via these platforms — policy-grade data infrastructure.
Spectral-band logic: visible (what eyes see) → near-IR (plant vigor: NDVI = (NIR−Red)/(NIR+Red)) → shortwave-IR (minerals, moisture) → thermal (fires, heat islands) → microwave/SAR (all-weather shape change). Calibrate against ground truth plots seasonally; an uncalibrated index is wallpaper.
40.6 Constellations and Debris
Operational hazard: orbits clog permanently when operators treat them as infinite — dead hulks, separation junk, and intentional debris cascade into Kessler collisions no cleanup can afford. Fly only maneuverable plus trackable birds with deorbit and passivation built in, answer conjunction warnings, and design for demise; never create intentional debris.
Mega-constellations (thousands of LEO birds) blur satellite-to-ground latency toward terrestrial feel. The externality they manage: orbital debris — Kessler syndrome (cascade collisions rendering orbits unusable) is the tragedy-of-commons failure mode. Mandatory doctrine: deorbit capability at end-of-life, collision avoidance coordination, trackable-and-maneuverable designs. A clogged LEO is a self-inflicted wound generations can't undo cheaply.
| Rule | Implementation | Verifies by |
|---|---|---|
| Deorbit ≤25 y (tighter now) | Propulsion or drag sail, passivation | Tracked reentry, no dead hulks |
| Maneuverable + trackable | GPS + thrusters + reflectors | Conjunction warnings answered |
| No intentional debris | No ASAT, no separation junk | Norms + Ch 47 licensing |
| Design for demise | Burnable materials | Casualty risk < 1:10,000 |
40.7 Ground Segment
Satellites are half the system: dishes/gateways, network operations centers, telemetry analysis, anomaly response procedures ("save the bird" playbooks). Ground stations networked globally hand off passes; scheduling software treats contacts like appointments. Budget reality: ground segment ≈ half of lifecycle cost — plan it with the launch, not after.
Anomaly playbook (write before launch): safe-mode triggers, power-positive attitudes, stored command loads, ground-simulator twin for rehearsing recoveries, 24 h staffing through early orbit. The bird that survives its first eclipse season usually survives its design life — staff accordingly.
Capability gate: when positioning, weather warning, and intercontinental connectivity operate as maintained public or organisational services—with backups, governance, and failure recovery—the society has a resilient planetary information layer. Space infrastructure supports global capability; it does not by itself define political power or “dominance” (Appendix D).
40.8 The Satellite Papers
- Telstar launched July 10, 1962 — the day after the Starfish Prime high-altitude nuclear test injected energetic particles into the magnetosphere; radiation damage contributed to Telstar's failure months later. Space-weather risk entered engineering through weapons testing; every subsequent rad-hardening spec descends from that lesson.
- Syncom 3 (1964) carried the Tokyo Olympics across the Pacific — geostationary broadcasting demonstrated on the world stage; Early Bird/Intelsat I (1965) made GEO commercially routine.
- Reconnaissance existed before civilians knew: CORONA film-return satellites flew 1960–1972 and were declassified in 1995 — their imagery ended the "missile gap" debates and seeded civilian remote-sensing practice.
- Debris became measurable: Kessler's 1978 cascade paper named the syndrome; the 2007 Chinese ASAT test on Fengyun-1C added thousands of cataloged fragments; the 2009 Iridium 33–Cosmos 2251 collision proved accidental cascades occur. Tracking networks (US Space Surveillance, ESA conjunction warnings) now operate as traffic control — governance infrastructure built post hoc, as usual (Ch 47).
- GPS lineage: Timation and 621B experimental programs merged into the 1978 Block Is; the Gulf War's 1991 use forced Selective Availability debates (civilian accuracy degraded deliberately until May 2000); GLONASS, Galileo, and BeiDou followed as sovereignty investments — navigation joined currency and language among state-function monopolies, then pluralized.
- Landsat's continuity argument: five decades of comparable multispectral imagery created climate and land-use science that no single mission could have; program survival fights recur every decade — long-duration measurement needs institutional champions, not just hardware budgets (Ch 20's metrology culture scaled to orbit).