Capacity would be a huge problem though, even with 5000 satellites, if you wanted to rival fixed-line or cellular wireless networks for a significant part of th

Capacity would be a huge problem though, even with 5000 satellites, if you wanted to rival fixed-line or cellular wireless networks for a significant part of the world's population... To put that in perspective, in Australia we have 13,400 cellular towers [1], with an extra 9,000 proposed to be built, to cover just 26.3 million subscribers [2]. This is to deliver 2.8% of the data downloaded over the last year (97.2% having been delivered over fixed line) [2]. So, with an LEO constallation, you might have a few dozen satellites over the populated parts of Australia at any one time, each of which has a finite amount of capacity (both uplink/downlink spectrum and capacity back to the ground station) shared between all the users. I think this network could see a massive benefit for rural areas, developing countries, maritime markets, etc. but to think that the network would have the capacity to deliver anything rivalling cellular networks in metro/suburban areas to most subscribers is very far fetched. 1. https://oztowers.com.au/Statistics https://oztowers.com.au/Statistics 2. http://www.abs.gov.au/ausstats/abs@.nsf/mf/8153.0 http://www.abs.gov.au/ausstats/abs@.nsf/mf/8153.0