-As we just saw with independent systems, they are not so independent. They need terrestrial infrastructures at least for a person using the satellite system to call someone on the terrestrial network. Interconnection with terrestrial in telephony will be immediate. Today, it evolves towards IP interconnection. The first and most general idea that comes to mind for this is to use the satellite system as an access network, just as I explained in the first sequence, for the standard telephone network. A mobile phone network, 2G, 3G, 4G, has its core network. I represented it in a very simple way. To interconnect with its core network, relays can be used. They directly interconnect with a given interface, or you can use the satellite gateway, the satellite transmitter. It uses the same interface than relay and it will be done in the same way except the part between the user and the relay. A specific interface is required when satellites are used between the user and its relay as well as a specific protocol stack that I will talk about in the last sequence of this lesson. This protocol stack will be dedicated to space communication called the air interface. There is a standard for geostationary satellites called GMR, meaning "GEO Mobile Radio". The GMR is linked to the terrestrial telephone infrastructure used. So it evolves according to it. It started with GSM, 2G. GMR-1 and GMR-2 were competing standards working in 2G. We then evolved towards 2G5, then 3G, and today we try to go toward 4G LTE. Many systems comply with this standard. In particular systems such as Thuraya and Inmarsat that are mainly supporting niche applications. There are also systems such as ICO, SkyTerra, TerreStar dedicated to the general public. I illustrated a small satellite... Not so small as it is probably the largest European satellite. The Alphasat satellite was launched in July 2013. It weighs 6.6 tons and has a huge capacity with a great amount of antennas to relay signals. The issue with these systems is when they offer an application for the general public because it is not so money-making. Telephony applications are too expensive for the general public and not adapted. The TerreStar system for example went bankrupt a year or two after it was launched before being acquired by the US government. Why does it not work? Because it is more expensive, because for geostationary satellites, the round-trip delay is long. If you have to wait one second before your interlocutor can talk back compared to 100 milliseconds with a terrestrial network, you have a lower-quality experience while paying more. Users that can mainly be found in cities, in rich countries, will not want to pay a subscription for this system. Regarding poor countries, the issue is that this system is expensive. If poor people have to pay expensive subscriptions, it will not work either even though these systems are needed since there are no terrestrial infrastructures. But there are some successes. Either niche applications that work well, or these systems can be bought by countries in need of infrastructures as it happened with TerreStar. These systems are then not directly used as an access network but as a backhaul network. Why so? As I will show it, costs are shared. Today, the objective that remains for telecommunications in the case of telephony but also IP, is to provide voice and Internet to the 3 billion users who do not have access to it yet. To do this, we will rather use backhauling. I explained it in sequence 1. Instead of setting the satellite as a direct access network for the user, we will use it to link the core network and the relay. The satellite will be used in a bidirectional way. The core network will resupply the relay and the relay will send the data back from user to the core. The benefit is that there is a group of users behind the relay, so the satellite link cost is shared between all the users. It is much more cost-effective. To end this part, I featured an example, O3B, Other 3 Billions. It represents the other 3 billion users. It is a project by one of the heads of Google who wants to send a constellation of MEOs, Medium Earth Orbit, instead of LEOs or GEOs. As of today, 8 have been sent but I believe that only 6 are operational. They aim at sending 16 of them to provide Internet and voice access to the 3 billion persons who do not have it yet.
