How Satellites Connect the Entire World

Thesis: Low Earth orbit constellations are a genuine engineering triumph and an unmanaged environmental commons. We are repeating a mistake we have made before.

Satellite internet has changed what connectivity means for ships, remote villages, and disaster zones, and it happened faster than most people noticed. I became interested after learning that thousands of satellites had been launched in only a few years. The engineering is remarkable. What troubles me is the pattern: a shared resource with no effective owner is being filled rapidly by parties who each benefit privately while the risks accumulate collectively. Orbit is behaving like the atmosphere and the oceans did before regulation, and electrical engineers are the people building the thing that is filling it.

The core technology

Geostationary satellites orbit at roughly 35,800 kilometers, appearing fixed above one point but imposing a round-trip signal delay of around half a second. Low Earth orbit satellites fly a few hundred kilometers up, reducing latency to tens of milliseconds, but each covers a small area and moves quickly, so continuous service requires a large constellation and rapid handovers. The enabling technology is the electronically steered phased array, which forms and redirects beams by adjusting the relative phase of many small antenna elements, allowing both satellite and user terminal to track each other without mechanical movement. Laser inter-satellite links now route traffic between spacecraft, reducing dependence on ground stations. User terminals rely on the same phased array technology, manufactured cheaply enough for consumer installation, which is arguably the harder achievement of the two.

Figure 1. A phased-array antenna steers its beam electronically by shifting the phase at each radiating element, letting satellite and ground terminal track each other with no moving parts.

Recent developments and real problems

Direct-to-cell service, in which an ordinary unmodified phone connects to a satellite, has moved from demonstration to limited commercial availability, which is a serious achievement given the tiny antenna and low transmit power of a handset. But the number of active satellites has risen several-fold in under a decade. Collision avoidance maneuvers are now routine, conjunction warnings are frequent, and the risk of a cascading debris chain reaction that renders useful orbits unusable is no longer theoretical. Astronomers report interference from both reflected sunlight and unintended radio emissions. Spectrum coordination is strained, and enforcement mechanisms are weak. Because satellites at these altitudes decay naturally within a few years, the constellations require continuous replacement launches, making the environmental accounting far less favorable than a single deployment suggests.

Figure 2. Tracked objects in orbit around Earth. The density of debris and active satellites illustrates the crowding problem this essay argues is an unmanaged environmental commons.

Outlook and solutions

Technical solutions exist and are underused: mandatory rapid deorbit after end of mission, propulsion redundancy so a failed satellite can still be removed, standardized transponders so every object can be tracked accurately, darkened surfaces and attitude control to reduce reflectivity, and shared conjunction data rather than proprietary tracking. Active debris removal demonstrations should be scaled. Most importantly, licensing should price the externality: an operator that leaves objects in orbit should bear a cost proportional to the risk it imposes on everyone else. International coordination is genuinely difficult here, but the alternative is discovering the limit experimentally, and that experiment cannot be reversed.

Conclusion

My view is that this generation of engineers will be judged less on whether we could build satellite mega constellations than on whether we managed the orbital environment while doing it. The insight I take away is that engineering competence is not the same as engineering responsibility. Building the system is the easier half. Designing the rules that let the system exist for a century is the part we keep postponing.

Sources

1. ESA (2025). ESA Space Environment Report — ~40,000 tracked objects, an estimated 1.2 million fragments above 1 cm, and debris growth even without further launches. https://www.esa.int/Space_Safety/Space_Debris/ESA_Space_Environment_Report_2025

2. ESA Space Debris Office. Annual Space Environment Report (latest edition, PDF). https://www.sdo.esoc.esa.int/environment_report/Space_Environment_Report_latest.pdf

3. ESA DISCOSweb. Space Environment Statistics (live catalogue figures). https://sdup.esoc.esa.int/discosweb/statistics/