N etworking in the 21st Century: The Sky's the Limit

	You Say GEO, I Say LEO, Oh My, Oh MEO It's a truism in technology that every solution presents an equal and opposite problem. Satellite systems are no exception. While a great to-do has been made about the well-known issues of GEOs and latency, it's also clear that many providers see enough benefit in GEOs to stake billions of dollars on next-generation GEO broadband. If LEOs live up to their promise, they'll certainly win in the end, but right now there are enough d oubts and trade-offs to make GEOs a sound business strategy. While Pioneer Consulting's Scott Clavenna predicts the launch of 350 broadband LEOs by the year 2010, he also envisions 144 broadband GEOs. That's a strong showing, given that worldwide coverage with LEOs requires about 48 birds and GEOs need only eight. Broadband MEOs, which cut the difference at about 20 satellites, are also skybound from companies like Hughes (Spaceway NGSO extension), TRW (Global EHF Satellite Network, GESN) and Orbital Sciences Corp. (OrbLink). Although rather rare, broadband MEOs may offer a way to reduce latency without paying the price of a full LEO constellation. Latency--that "L" word--is the trade-off with GEOs. Getting a satellite into a GEO orbit is typically more difficult and expensive than launching smaller, lighter and more easily manufactured LEOs into a lower orbit. But LEOs, with a life span of about five years, compared to about 10 for a GEO, are expected to burn out quicker in that orbit. That means more LEO s must be launched and additional spares must be readied. The orbit is also at an altitude where a LEO speeding along at 27,000 kilometers per hour (MEOs hit 19,000 km and GEOs 11,000 km) are more likely to be turned into molten plasma by a dislodged bolt of debris hurtling through space. While GEOs are always in sight of a terrestrial gateway by virtue of an orbit that matches the earth's rotation and MEOs are overhead for two to four hours, LEOs tend to be overhead for less than 15 minutes before having to perform complex airborne traffic handoffs to another satellite. That short span also means that LEO earth stations must use phased-array antennas that maintain an active link by keeping at least two satellites in view at all times. The antenna starts a new link before severing one with a satellite moving out of range--all of which adds to terminal complexity and, presumably, cost. On the other hand, because GEOs are further away, they must either incorporate more power--at more expense--in skyborne tr ansmitters, or require that bandwidth-hungry users rely on bulky, nontransportable dishes. GEO topology is touted as being simpler than that of terrestrial or LEO networks, but the more complex topology of LEOs means these constellations can tap greater bandwidth for reuse with their tightly focused spotbeams. GEOs generally are considered best for broadcasts and multicasts--applications where LEO mesh networks and transmission movement still prove problematic. Either option, however, is probably better than terrestrial-bound multicast. LEOs are expected to produce better results with interactive applications like voice and videoconferencing. Bob Egan, wireless research director for Gartner Group, adds that with caching, both LEOs and GEOs become "extremely viable" for one of the biggest markets they'll both serve--aggregating information for ISPs and businesses outside the United States.

For the Side Bar on LEOs Dance The Jitterbug GEOs Turn Up The Speed Of Light GEO-LEO Hybrids: Two Mints In One The Big Five Planned Constellations High Hopes, High Hoops Do LEOs Have Too Much Latency? A Q&A With Teledesic Mobile Voice Blazes an Astral Trail for Broadband Other Features Slicing Through the Hype of IP Switching By Joel Conover