Starlink: Satellite Based Internet

… and yeah. Another year, another price increase. This time with more ..|. for users who dare to live in areas where it’s useful.

Huh.

LTE from Starlink satellites for low bandwidth communications. Seems potentially useful, similar to what Apple’s latest devices are doing with satellite comms to (something).

On the topic of LTE from orbit: https://spacenews.com/ast-spacemobile-secures-communications-with-prototype/ OMG THAT ANTENNA IS HUGE. This is BlueWalker3, the giant white reflector that has astronomers currently abuzz. Yes, I’m sure the next generation one will be darker, but this is the sort of tech you need (giant antennas) to talk LTE to low-earth-orbit with regular handsets (1/3 watt iirc). Personally, I don’t see how this can ever be profitable, but trillions will be wasted in the attempt.

Well, as long as we clutter space for random commercial purposes. We’re, what, 40 or 50 years from a solid Kesslering? Maybe less?

Do you realize how hard it is to deliver ads in the middle of nowhere right now? The driving force of the tech industry lately seems to be “But there are still times in your day we can’t deliver personalized custom advertising to you, how can we disrupt those times?”

I’ve no idea what it takes to talk ~340 miles or whatever Starlink’s orbital shell is from a regular cell phone. Not an area I’ve looked into, but I have to assume “very low data rate modulation” and, as you point out, “Really big antennas.”

IMHO less. China plans to launch its own starlink LEO competitor with 13k satellites. These two org’s aren’t going to cooperate, so expect a LEO collision within the next 5-10 years to kick off the process. With 120 starlink satellites per orbital plane and an orbital period of 90-100 minutes (or 47.5 seconds apart!), one collision in-plane could easily turn into 100 shredded satellites in a matter of hours. (read: targeted attack calculated to cast debris in-plane)

The BlueWalker 3 prototype / proof of concept satellite with its 693 sq ft antenna has already demonstrated 2G, 4G, 5G, and 14 Mbit downloads to an unmodified handset. China has its own prototype. I think they’ve opened it to the public for SMS send/receive so far, but I haven’t been following closely.

[research] Ah! 1/3 watt was the AVERAGE power. Modern phones transmit in short 1 watt bursts, so…
If you assume 1 watt radiated at 800 MHz, 350 miles “free space loss”, +50 dBi gain on the satellite, and you end up with a receiver signal of -70.75 dBm which is actually decent for an end-to-end system. but free space loss isn’t accurate, since it assumes you’re transmitting horizontally across the earth. I have no clue how that changes going vertically (i.e. through the ionosphere, decreasing atmosphere density), never looked into it.

but I also don’t know the subtle differences between SMS, LTE data, and LTE voice, so this is a armchair estimate.

Caveat, I’m not sure how to figure in the loss attributed to 1 watts being dispersed in all direction (omnidirectional antenna) vs something focused like a yagi. My gut is that it accounts for the need of another +10 dBi of gain.

My understanding is that “antenna gain” is literally “the difference between this and an omnidirectional radiator.” Just, in some particular direction or set of directions.

That is correct. To get more specific there are differences in measurement of dBi vs dBd, although I rarely see dBd outside of HF antenna design. The ‘i’ standing for isotropic, a hypothetical point-source of RF energy; and ‘d’ being an ‘ideal dipole’ for a given frequency, with its doughnut shaped radiation pattern.

dBi is really a mathematical construct, since you cannot have an infinitely small antenna and still have 2-poles, or have it made at resonant frequency to anything, but it works well as standard datum to calculate gain/loss from.

These can also be expressed as ERP and EIRP. Effective Radiated Power and Effective Isotropic Radiated Power, respectively.

In other words dBi is the calculated power of an RF source moving in a frictionless vacuum. By definition any antenna with a gain > 1 will not be omnidirectional.

Ah, that makes sense. Since cell phones are measured in radiated power, I can treat it like a black box. I am guessing that it’s mostly isotropic, since phones can be facing all manner of directions, though some may purposely design to not radiate towards the head due to various fcc and related international orgs’ rules. For simplicity, and armchair estimates, isotropic it is!