Study Guide — Geostationary Earth Orbit (GEO) Satellite Systems

Unit Overview

At exactly 35 786 km above the equator, a satellite's orbital period equals one sidereal day — and the spacecraft appears to hang motionless in the sky. This coincidence, unique in orbital mechanics, has made GEO the workhorse of satellite communications for half a century: a single satellite stares at a third of the Earth, ground antennas never move, and a three-satellite ring covers all but the poles. GEO's weaknesses are equally distinctive: a quarter-second of delay, a 200 dB path, rain that swallows Ku-band signals, and a sun that briefly blinds receivers every equinox. This guide develops the physics, the architecture, the link budgets, and the operational realities of the fixed point in the sky.

Learning Outcomes

After studying this unit and completing the tutorials you should be able to:

Derive the GEO radius from the sidereal-day condition and state the three GEO conditions.
Compute look angles (elevation, azimuth) and slant range for any Earth station.
Explain station-keeping boxes, eclipse seasons, and graveyard-orbit disposal.
Describe the GEO space segment: power systems, transponders (TWTA vs SSPA), antennas.
Build GEO link budgets including backoff and intermodulation constraints.
Explain C/Ku/Ka band trade-offs, ITU coordination, and orbital-slot allocation.
Quantify rain fade, sun outages, gaseous absorption, and their mitigations.
Defend GEO's niche against LEO/MEO challengers with quantitative arguments.

Indicative Study Time

Activity	Time
Guided reading (Sections 1–9)	5 hours
Figures studied / examples reworked	2 hours
Tutorial problems (Section 10)	3 hours
Revision of key equations & glossary	1 hour

Prerequisite knowledge: Kepler's third law; coverage geometry (arccos formula with elevation mask); link budgets (EIRP, G/T, C/N₀); basic antenna theory (gain, beamwidth).

Section 1 — GEO Fundamentals

1.1 The GEO Condition

T_sidereal = 86 164 s ⇒ a = (μT²/4π²)^1/3 = 42 164 km ⇒ h_GEO = 35 786 km

Three simultaneous conditions make a satellite stationary (not merely synchronous):

Circular orbit (e ≈ 0) — constant altitude;
Period = one sidereal day — a geosynchronous satellite with e > 0 or i > 0 traces a daily figure-of-eight (analemma) around its mean longitude; only e = 0, i = 0 removes all apparent motion;
Zero inclination (i = 0) — the orbit lies in the equatorial plane.

1.2 Perturbations and Station-Keeping

GEO satellites are not truly fixed — they drift and must be actively parked:

Perturbation	Effect	Mitigation
Earth triaxiality (uneven mass)	East–west drift toward two stable longitudes (~75°E, 105°W)	E–W station-keeping thrusters
Lunar–solar gravity	Inclination growth ~0.9°/year → N–S drift, daily figure-of-eight	N–S station-keeping (the propellant hog)
Solar radiation pressure	Eccentricity oscillation	Absorbed in the E–W box

Typical station-keeping box: ±0.05° – ±0.1° in longitude and latitude; fuel budget sets ~15-year lifetime

1.3 Eclipse Seasons and Power

Near the equinoxes, the satellite passes through Earth's shadow for up to ~72 minutes per day around local midnight — batteries must carry the entire load.
Solar array sized for beginning-of-life (BOL) with degradation budget (~2–3%/yr); eclipse seasons drive the battery cycle count.

1.4 End of Life: The Graveyard Orbit

Disposal: raise by ~300 km into the graveyard orbit — a dead GEO satellite must never drift through the operational arc

With propellant exhausted, a GEO satellite becomes an uncontrolled 2-tonne object crossing every operational slot at its inclination — the graveyard requirement is the GEO industry's self-policing debris rule.

Check Your Understanding 1

Why must the GEO period reference the sidereal day (86 164 s) and not the solar day (86 400 s)?
A satellite is geosynchronous with i = 5°. Describe its apparent motion as seen from the ground.
Which station-keeping axis consumes most propellant, and why?

Answers

After one orbit the satellite must realign with the same Earth-fixed meridian; Earth rotates 360° relative to the stars in 86 164 s but 360°+0.986° relative to the sun in 86 400 s — synchronism is with the stars.
A daily figure-of-eight (analemma): ±5° in declination combined with east–west libration — fixed dishes would need 5° of tracking margin.
North–south: inclination grows secularly (~0.9°/yr) from lunar-solar torques; E–W drift is a mild oscillation about stable points.

Section 2 — Coverage Geometry and Look Angles

2.1 Elevation, Azimuth, Slant Range

cos γ = cos φ · cos Δλ (γ = central angle to sub-satellite point)

sin(el) = (cos γ − R_E/(R_E+h)) / √(1 + k² − 2k cos γ), k = R_E/(R_E+h) = 0.1511

Slant range: d = √(R_E² sin²el + h² + 2R_Eh) − R_E sin(el) → at el = 10°: d ≈ 40 600 km

Fig. A — Elevation map from one GEO satellite. Coverage dies poleward of 81°; beyond ~60° latitude elevation falls below ~15° with its rain/noise/path penalties. Three satellites at 120° spacing (blue markers on the equator) overlap generously at the equator and mid-latitudes.

2.2 Three-Satellite Global Coverage

Footprint half-angle at 10° mask: ρ = arccos(k cos10°) − 10° ≈ 71.4° → sub-points at 120° spacing overlap by 2×71.4° − 120° ≈ 23°

Three satellites cover all latitudes to about ±70° with ≥10° elevation — everything except Antarctica and the Arctic Ocean. This is the famous GEO marketing figure, and it is geometrically exact.

Example 2.1 — Nairobi looks at a 33°E satellite. φ = −1.29°, Δλ = 36.82 − 33 ≈ 3.8°. cos γ = cos(−1.29°)cos(3.8°) = 0.9976 → γ ≈ 4.0°. sin(el) = (0.9976 − 0.1511)/√(1.0228 − 0.3015) = 0.8465/0.8493 ≈ 0.9967 → el ≈ 85.3° — nearly overhead, the friendliest GEO geometry on Earth. Slant ≈ 35 820 km. Compare London (51.5°N, Δλ = 0): el ≈ 31°.

Check Your Understanding 2

Compute elevation and slant range for a station at 40°N looking at a satellite 20° away in longitude.
Why do three GEO satellites leave the poles uncovered — and can anything fix this?

Answers

cos γ = cos40°cos20° = 0.7198 → sin el = (0.7198−0.1511)/√(1.0228−0.4340) = 0.5687/0.7672 = 0.7413 → el ≈ 47.8°; d ≈ √(6371²sin²47.8 + 35786² + 2·6371·35786) − 6371·sin47.8 ≈ 37 900 km.
The GEO plane is the equator: central angle to a pole is 90° > 81.3° maximum. Only inclined/HEO or LEO/polar constellations see the poles (Iridium, Molniya) — GEO cannot, by geometry, not by engineering.

Section 3 — The GEO Space Segment

3.1 The GEO Bus

Power: 10–20 kW solar arrays (GEO comms sats); dual-axis sun tracking with eclipse-season battery (NiH₂/Li-ion) ~ 100% depth-of-discharge per eclipse.
Thermal: radiators dump 10+ kW; traveling-wave tube amplifiers (TWTAs) at the focus of the thermal design.
Propulsion: chemical apogee engines historically; modern sats launch directly to GEO or use electric orbit-raising (months-long spiral, mass savings ~40%).
Lifetime: 15 years limited by propellant (N–S keeping) — the satellite is retired with transponders still working.

3.2 The Communications Payload

Element	Function	Engineering notes
Antennas	Spot beams (Ku/Ka), shaped/contoured beams (C), global beams (L/S)	Spot beams ≈ 1°–2° → reuse ×20+; feed arrays with on-ground beam-forming (digital payloads)
Low-noise receivers	Downconvert uplink with minimal noise	~1.5–3 dB noise figure sets G/T ceiling
Channelization	36 MHz C-band / 27–72 MHz Ku transponder filters	Guard bands; channel plan is the spectrum business
High-power amplifiers	TWTA (50–60% efficiency, high power, needs linearization) vs SSPA (30–40%, linear, lower power)	Multi-carrier operation forces output back-off (IBO/OBO) — the link budget's hidden loss
Beacons	Tracking & propagation reference	19.7/29.5 GHz beacons enable rain-fade monitoring at gateways

3.3 Bent-Pipe Transponder Chain

Uplink (6/14/30 GHz)
      │
      ▼
[Rx antenna]─►[LNA]─►[mixer]─►[channel filters]─►[HPA: TWTA/SSPA]─►[Tx antenna]
      │                                                    │
   G/T ~ −5…+10 dB/K                                OBO 3–10 dB for multicarrier
      ▼                                                    ▼
Downlink (4/12/20 GHz)  —  no demodulation: frequency translation only ("bent pipe")

Digital payloads (the modern trend): on-board demodulation/regeneration, digital channelizers, and beam-forming convert the bent pipe into a software-defined router — processing traffic in orbit, enabling dynamic capacity allocation. GEO is absorbing the LEO lesson of upgradability.

Check Your Understanding 3

Why do multi-carrier transponders operate backed off, and what does this cost the link budget?
Why do GEO satellites carry beacon transmitters separate from the comms payload?

Answers

Multiple carriers through a nonlinear TWTA generate intermodulation products; back-off (OBO) trades ~3–10 dB of available EIRP for linearity — hence multicarrier EIRP budgets use the backed-off value, not the saturated rating.
Beacons provide a continuous, stable, unmodulated carrier for gateway tracking and — critically — for measuring rain attenuation in real time to trigger uplink power control.

Section 4 — The Ground Segment and Operations

4.1 Elements

Element	Role
TT&C stations	Telemetry read-out, command uplink, ranging — every satellite monitored continuously (typically two diverse sites)
Teleports / gateways	RF interface to the terrestrial network: antennas 4–13 m, redundant chains, located in coordinated interference "hot spots"
User terminals	DTH dishes 0.45–1.2 m; VSATs 0.6–2.4 m; flat-panel electronic-steered arrays (new)
SOC / NOC	Satellite operations (station-keeping plans, eclipse prep) and network operations (carrier plans, rain-fade response)

4.2 The VSAT Hub-and-Spoke

The GEO economic engine: one hub with a big antenna serves thousands of small remote terminals. Time-division multiplexing on the outbound (hub→terminal) and Alova/TDMA inbound (terminal→hub) with a 250 ms double-hop penalty when two VSATs talk — the reason peer-to-peer VSAT applications (VoIP between branches) feel laggy.

4.3 Operational Calendar

Equinox seasons: battery-managed eclipses + sun-outage schedules published to customers.
Rain season: uplink power control and ACM (adaptive coding & modulation) profiles.
Conjunctions: tracked objects passing within km of the slot → temporary box excursions.

Check Your Understanding 4

Why are TT&C stations duplicated at diverse sites?
A VSAT-to-VSAT VoIP call over one GEO hop suffers what delay, and what design mitigates it?

Answers

A single site has a satellite setting below its horizon daily (and local rain/utility failures); two sites ~120° of longitude apart give continuous command authority.
2× ~119 ms space path + processing ≈ 250–300 ms one-way perception; mitigation: terrestrial breakout at the hub (single satellite hop per direction is impossible for mesh — hence regenerative payloads or LEO for true mesh).

Section 5 — GEO Link Budgets

5.1 The Standard Chain

C/N₀ = EIRP_sat + (G/T)_ES − L_FS − L_atm − L_misc + 228.6 (dBHz)

L_FS = 92.45 + 20 log f(GHz) + 20 log d(km); GEO: d ≈ 35 800–40 600 km

Example 5.1 — DTH at Ku (12 GHz). EIRP 52 dBW (per spot beam), user dish 0.9 m → G/T ≈ 12 dB/K, d = 37 500 km (mid-lat): L_FS = 92.45 + 21.6 + 91.5 = 205.6 dB; misc 2 dB.
C/N₀ = 52 + 12 − 205.6 − 2 + 228.6 = 85.0 dBHz.
DVB-S2 8PSK at 27.5 Msym/s needs C/N ≈ 10 dB (code 3/4): available C/N = 85.0 − 10log(27.5×10⁶) = 85.0 − 74.4 = 10.6 dB → 0.6 dB clear-weather margin — tight and typical; rain is handled by ACM stepping down to QPSK (−5 dB threshold) rather than by margin.

5.2 The Uplink Is the Hard Direction

Downlink EIRP is cheap (solar power + TWTA); the uplink from a 1.2 m terminal at 14 GHz has EIRP ≈ 44–47 dBW and fights rain twice (up and down). VSAT systems budget 6–8 dB uplink margin or site-diverse gateways.

5.3 The Latency Line

One hop ≈ 2×d/c ≈ 240–270 ms RTT for hub-spoke; interactive limits: voice tolerable, TCP needs acceleration

Performance-enhancing proxies (PEPs) split TCP at the gateway — acknowledging segments over the satellite leg — the standard fix without which GEO throughput collapses on long fat pipes.

Check Your Understanding 5

Why is ACM (adaptive coding and modulation) better than building in rain margin?
Compute C/N₀ for a Ka-band downlink (20 GHz, d = 37 000 km, EIRP 60 dBW, terminal G/T 15 dB/K, misc 2 dB).

Answers

Rain margin is paid 100% of the time for a rare event; ACM spends capacity only when fading (lower code rate + lower order modulation on demand) — average throughput rises ~20–30%.
L_FS = 92.45 + 26 + 91.4 = 209.9 dB; C/N₀ = 60 + 15 − 209.9 − 2 + 228.6 = 91.7 dBHz.

Section 6 — Frequency Bands, Spectrum, and Orbital Slots

6.1 The Band Trade-off

Band	Uplink	Downlink	Pros	Cons
C (4/6 GHz)	5.85–6.725 GHz	3.4–4.2 GHz	Rain-immune, mature, wide beams	Congested, interfered, big dishes, wide beams (little reuse)
Ku (12/14 GHz)	13.75–14.5 GHz	10.7–12.75 GHz	Smaller dishes, spot beams, DTH scale	Rain fade moderate; pointing tighter
Ka (20/30 GHz)	27.5–31 GHz	17.7–21.2 GHz	Huge spectrum, very narrow beams → massive reuse	Severe rain fade; 0.1° pointing; gateway density

6.2 Spectrum Management

ITU coordination: filings at WRC conferences; the arc is divided into planned (C-band) and unplanned (Ku/Ka) bands; "bring into use" deadlines force real satellites.
Orbital slots: 2° nominal spacing at Ku (ITU Radio Regulations; national regulators may grant tighter coordination) — the arc from 45°W to 165°E is the valuable real estate.
Interference discipline: off-axis EIRP masks, coordination arcs between adjacent satellites, and EPFD (equivalent power flux-density) limits protect terrestrial services sharing Ka.

Check Your Understanding 6

Why did DTH migrate from C-band to Ku even though rain fade worsens?
What enforces the 2° spacing, and what happens when two operators want the same slot?

Answers

Reuse: Ku spot beams give an order-of-magnitude more capacity per satellite, and 0.6 m consumer dishes replaced 2–3 m C-band ones — capacity economics beat rain statistics (with ACM as the insurance).
National regulators + ITU coordination; disputes proceed through bilateral coordination, then ITU procedures (priority by filing date) — and are ultimately settled by money or beam-shaping.

Section 7 — Propagation Impairments and Mitigations

7.1 Rain Attenuation

A_p = A_0.01 · 0.12 · p^{−(0.546 + 0.043 log p)} (ITU-R form, 0.001% ≤ p ≤ 1%)

Fig. C — Rain fade exceedance at Ku and Ka (tropical schematic). Ka carries >2× Ku's fade; the 0.01%-of-time figure (≈53 min/yr) is the standard design point.

Mitigations: uplink power control, ACM, site diversity (two gateways 30–100 km apart — rain cells are local), larger margins at C-band.

7.2 Sun Outage

When the sun — an ~11 000 K noise source — passes within the antenna beamwidth of the satellite, the receiver noise floor rises tens of dB and the link drops. Predictable: near the equinoxes, near local noon, for a few minutes over several days.

Fig. B — Simulated sun-outage calendar for Nairobi, 1.2 m Ku dish, satellite at 35.9°E: outages last up to ~2 min/day and cluster around the equinoxes. Larger dishes (narrower beams) have shorter, deeper events.

Mitigations: none for physics — but operators schedule (bank settlements, telemedicine sessions avoid the window), and diversity sites on different longitudes escape it entirely.

7.3 Other Impairments

Impairment	Magnitude	Mitigation
Gaseous absorption	0.03–0.1 dB at Ku; 0.1–1 dB at Ka	Included in L_atm; elevation-dependent
Clouds/fog	0.5–3 dB at Ka	Part of ACM margin
Tropospheric scintillation	±0.5–1 dB (Ku), more at low elevation	Small-signal design margin
Depolarization	XPD degrades 5–15 dB in rain	Circular pol at C, adaptive cross-pol cancellation
Ionospheric scintillation	~1 dB at Ku; worse at L/S	Frequency agile beams; diversity

Check Your Understanding 7

Using Fig. C, estimate the rain attenuation exceeded 0.1% of the time at Ka, and the corresponding annual minutes.
Why does site diversity work better for gateways than for individual consumer dishes?

Answers

A(0.1%) ≈ 18×0.12×(0.1)^(−0.657) ≈ 2.16×4.52 ≈ 9.8 dB; 0.1% of a year ≈ 8.8 hours (Ka needs ACM + site diversity to hold availability).
Gateways are few and can be connected by fiber to share traffic; consumer dishes are millions — diversity is economically impossible at the edge, so DTH relies on ACM alone.

Section 8 — Applications

Application	Why GEO dominates	Representative systems
DTH broadcast	One satellite → a continent of fixed 0.6 m dishes; no tracking	DVB-S2/S2X platforms (Astra, SES)
Fixed satellite services	Persistent trunk routes; simple ground infrastructure	Intelsat, SES, Eutelsat
VSAT enterprise	Hub-spoke economics; always-on remote sites	HughesNet, iDirect ecosystems
Meteorology	Continuous hemispheric stare — nowcasting needs it	GOES, Meteosat, Himawari (1/3 of GEO fleet by count)
MSS (L/S-band)	Regional beams + GEO persistence	Inmarsat (GEO heritage), Thuraya
Government/military	Assured wideband + protected payloads	WGS, Skynet
Hosted payloads	Long life, stable platform	WAAS/EGNOS navigation payloads (GEO transponders rebroadcast corrections)

GEO's quiet dominance: more than half of all operational GEO satellites are weather and data-relay craft — the "stare" applications where a fixed platform is worth more than low latency.

Section 9 — Trade-offs: Defending the GEO Niche

Axis	GEO	LEO/MEO challenge	GEO's counter
Latency	~250 ms RTT	LEO ~30 ms	PEPs; latency-insensitive bulk traffic (TV, trunk, IoT backhaul)
Terminal cost	Fixed dish, $50–500	Tracking/e-scan terminals $500–3 000	GEO remains cheapest per connected site
Coverage	38% of globe per sat, 3 sats ≈ global (−poles)	LEO needs hundreds–thousands	Fewest satellites for continental service
Rain	Ku/Ka fade severe	Lower path loss → more margin	C-band heritage + ACM + site diversity; GEO C-band is the all-weather king
Capacity density	Spot beams improving	Massive reuse	Digital payloads + Ka spot beams narrow the gap
Disaster resilience	Single-point regional failure	Constellation redundancy	GEO diversity (multiple operators/slots) — weaker argument, acknowledged

The verdict: GEO wins every mission whose value is persistence, simplicity, and coverage-per-satellite — broadcast, stare, trunk, remote-site connectivity. It loses where milliseconds matter. The modern satellite industry is not LEO or GEO but both: GEO for the steady backbone, LEO for the latency edge.

Section 10 — Tutorial Problems

Constants: μ = 3.986×10⁵ km³/s², R_E = 6 371 km, sidereal day = 86 164 s, c = 3×10⁵ km/s, k = R_E/(R_E+h_GEO) = 0.1511, FSPL = 92.45 + 20log f(GHz) + 20log d(km).

Derive h_GEO = 35 786 km from the sidereal-day condition.
For a ground station at Nairobi (φ = −1.29°, λ = 36.82°E) and a satellite at 26°E: compute γ, elevation, azimuth, and slant range.
Repeat Problem 2 for a station at 60°N, satellite at 26°E (station at 25°E). Comment on the operational consequences of the low elevation.
Compute the eclipse-season maximum daily eclipse duration: find the maximum time the satellite (35 786 km) spends in Earth's shadow, assuming the shadow cylinder radius R_E and the orbit plane passes through the shadow center (equinox condition).
Build a complete C-band uplink budget: terminal 2.4 m at 6 GHz (EIRP 55 dBW), d = 38 000 km, satellite G/T = −8 dB/K, misc 2 dB. Find C/N₀ and the C/N in a 36 MHz transponder with 5 dB output back-off (subtract back-off from the satellite-translated EIRP before computing the downlink; here simply report the uplink C/N₀ and comment on the back-off convention).
Using the ITU-R form with A_0.01 = 10 dB at 12 GHz, compute rain attenuation exceeded 0.01%, 0.1%, and 1% of the time. How many minutes per year is each?
A sun outage occurs when the sun passes within (HPBW/2 + 0.27°). Compute HPBW for 0.9 m and 2.4 m dishes at 12 GHz, and predict which dish suffers longer outages.
Three GEO satellites are nominally at 120° spacing. If satellite B drifts 0.1° east of its box center, quantify the change in the overlap region geometry at the equator (express as change in the minimum elevation at the former overlap boundary).
A national broadcaster must choose between (a) leasing two 36 MHz C-band transponders and (b) one 72 MHz Ku-band transponder with ACM. Write the decision memo: rain statistics, dish sizes for rural receive sites, interference environment, and cost structure (four paragraphs).
Orbit physics essay: "GEO is the only orbit where the satellite does the moving and the Earth does the waiting." Discuss with three quantitative examples (delay, station-keeping, eclipse geometry) — defend or refute.

Selected solutions

a = (μT²/4π²)^1/3 = (3.986×10⁵ × 86164²/39.478)^1/3 ≈ 42 164 km → h = 35 786 km. ✔
Δλ = 10.82°; cos γ = cos(−1.29°)cos(10.82°) = 0.9818 → γ = 10.95°; sin el = (0.9818−0.1511)/√(1.0228−0.2968) = 0.8307/0.8521 = 0.9749 → el ≈ 77.1°; d ≈ √(6371² sin²77.1 + 35786² + 2·6371·35786) − 6371 sin77.1 ≈ 36 070 km. Azimuth ≈ 270°+ (satellite west of station, roughly due west — compute via atan2 if required).
Δλ = 1°; cos γ = cos60°cos1° = 0.4999 → γ = 60.0°; sin el = (0.4999−0.1511)/√(1.0228−0.3022) = 0.3488/0.8488 = 0.411 → el ≈ 24.3°. Consequences: long tropospheric path → rain/noise penalties; terrain clearance critical; many European installations abandon below 10°.
Shadow half-angle: satellite travels through cylinder of radius R_E at radius a = 42 164: half-angle α = arcsin(R_E/a) = arcsin(0.1511) = 8.69°; time = (2×8.69/360)×sidereal day ≈ 70 min. ✔
L_FS = 92.45 + 15.6 + 91.6 = 199.7 dB; C/N₀ = 55 + (−8) − 199.7 − 2 + 228.6 = 73.9 dBHz uplink. With 5 dB OBO the satellite's downlink EIRP is quoted at back-off; uplink unaffected — but the composite C/N must combine uplink, downlink, and intermodulation C/Ns reciprocally.
Exponent −(0.546+0.043·log p): p = 0.01% → log p = −4 → 0.546−0.172 = 0.374 → A = 1.2×(10⁻⁴)^(−0.374) = 1.2×31.3 = 37.6 dB; p = 0.1% → exponent 0.417 → A = 1.2×17.8 = 21.4 dB; p = 1% → exponent 0.460 → A = 1.2×8.3 = 10.0 dB. Minutes/yr: 0.01% ≈ 53 min; 0.1% ≈ 8.8 h; 1% ≈ 3.65 days.
HPBW = 70λ/D: 0.9 m → 70×0.025/0.9 = 1.94°; 2.4 m → 0.73°. The larger dish has a narrower beam → shorter outages (fewer days) but deeper nulls when they occur; the small dish sees more, shallower events.
Overlap boundary sits where γ from A = γ from B = 60° nominally. Drift of B by 0.1° shifts the bisector by 0.05° in longitude; at the equator the minimum elevation at the new boundary (γ = 60.05° from each): el changes by approximately −(dγ)(k sinγ)/√(...) — a few hundredths of a degree; the real effect is at high latitudes where geometry compresses — quantify there: at the edge of coverage a 0.1° slot error shifts service contours ~11 km per 0.1° at the equator, compressed near the coverage rim.
Memo model answer: C-band — rain-immune, 1.8–2.4 m rural dishes cheap and forgiving, congested spectrum and interference from terrestrial links, transponder lease cheaper per MHz; Ku — 0.6–0.9 m dishes, ACM handles rain, cleaner interference, higher capacity per MHz. Rural distribution with harsh weather and unskilled installers → C-band; urban high-capacity → Ku.
Model essay: (1) delay — GEO's 250 ms vs LEO's 30 ms shows the satellite "fixed" but far; (2) station-keeping — the satellite burns propellant daily to stay fixed while LEO sats simply fall; (3) eclipse — GEO batteries, not the ground, absorb the midnight shadow; refute partially: everything orbits; "fixed" is a controlled illusion maintained at propellant cost.

Section 11 — Glossary of Key Terms

Analemma (figure-of-eight): Apparent daily path of a satellite with nonzero inclination or eccentricity; removed only at true GEO.
ACM: Adaptive Coding and Modulation: dynamically trades spectral efficiency against rain fade.
Bent-pipe: Frequency-translating transponder without demodulation.
Eclipse season: Weeks around each equinox when the satellite enters Earth's shadow daily (≤ ~72 min).
EPFD: Equivalent power flux-density: ITU limit protecting terrestrial services sharing Ka-band.
Graveyard orbit: Disposal orbit ~300 km above GEO, reached by final fuel expenditure.
Input/output back-off (IBO/OBO): Operating a multicarrier transponder below saturation for linearity; costs EIRP.
Orbital slot: A coordinated longitude (± station-keeping box) plus frequency filing; the GEO industry's real estate.
PEP: Performance-enhancing proxy: TCP acceleration gateway mitigating GEO latency.
Sidereal day: 86 164 s — Earth's rotation against the stars; the GEO period reference.
Station-keeping box: Allocated ±(0.05–0.1°) volume within which the satellite must remain.
Sun outage: Equinoctial noon-time link drop when the sun enters the antenna beamwidth.
TT&C: Telemetry, tracking & command: the satellite's control link, operated from diverse sites.
TWTA / SSPA: Tube vs solid-state power amplifiers; the linearity/efficiency trade at the heart of the payload.

Section 12 — References and Further Reading

T. Pratt, J. Bostian & J. Allnutt, Satellite Communications, 2nd ed., Wiley, 2003.
D. Roddy, Satellite Communications, 4th ed., McGraw-Hill, 2006.
B. G. Evans (ed.), Satellite Communication Systems, 3rd ed., IET, 1999.
ITU-R Recommendations P.618 (propagation) and P.837 (rain climate) — the engineering basis of Section 7.
SES/Astra and Eutelsat technical resources — DVB-S2 link budgets and spot-beam practice.

Exam guidance: GEO problems reward complete chains: Kepler → geometry (γ, el, d) → FSPL → C/N₀ → impairment margin → verdict. Always state the elevation mask and the back-off convention. The single most common exam error: quoting slant range as 35 786 km (that's the altitude) — compute it from elevation.