The satellite industry has been telling itself — and its customers — a particular story about latency for decades. The story goes like this: geostationary satellites are 22,236 miles above Earth, so latency is unavoidable and applications sensitive to delay simply cannot use satellite communications.
But is this story still accurate? And more importantly, is it helpful?
Explore the December 2015 Issue
Check out more from this issue and find your next story to read.
The rise of LEO constellations has renewed attention to the latency question. OneWeb, SpaceX Starlink, and others promise latencies comparable to terrestrial broadband — in the range of 20-40 milliseconds rather than the 600+ milliseconds round-trip typical of GEO satellites.
However, the latency narrative has also been used as a crutch to avoid addressing legitimate performance issues in GEO satellite systems. TCP acceleration, web caching, and other performance enhancement proxies have made a real difference, but they’ve also allowed the industry to paper over underlying limitations rather than address them directly.
The truth is that for many applications — video streaming, web browsing, most business applications — GEO latency is not actually a problem. But the industry’s reflexive invocation of “latency” as an explanation for poor performance has done real damage to satellite’s reputation.VS
Maury Mechanick is a Washington-based attorney specializing in satellite and telecommunications law.
