Hacker News Reader

Comment by jakeogh | Hacker News Reader

With the current setup, sure, but that's by design. The cellular modem could stay off until you decided to take the call if there was a nationwide page circuit listening, the user would get the ring, see the number the page sent, and if desired, answer, which powers on the modem, hits a tower and connects to a backend system that sent the page which took the incoming call.

Page messages are in-the clear, but that's fixable by (gasp) OTP.

View on HN

You want every single cell phone call in the world to send out a signal over every single cell tower?

No. But at a certain point, with the high speed modulations we have today, it is totally feasible to broadcast these passively to a multi-state region encompassing a radius of hundreds of miles.

There's not a legitimate engineering reason that the network needs to maintain constant fine-grained location data for each registered device at this point. The scope of the registration can be far more widely cast.

This would even have upsides for the devices and users. As check-ins to the network in which the device must transmit to the network would be far reduced, battery life improvements can be had.

Yes, this increases the amount of "broadcast" traffic, but honestly, even for some of the busiest telco switches in New York or LA, those data streams don't even approach the throughput requirements of a single HD Youtube stream...

/napkin overestimate using US 6B/calls/day with a nationwide 256B packet each, that's roughly a 100Mbps broadcast channel, which is ~5 digital TV channels, or one geostationary satellite's half-duplex bandwidth if it could see the entire US. As mdhardeman points out, it's easier than that, and there is plenty of room for re-transmission.

What is the passive bitrate of a tower->cell connection? LTE/GSM whatever.

With everyone's phones receiving and parsing that, batteries would die very quickly.

Pagers parse every single page. They only alert you when it's to your address. /napkin is just that, if you designed the protocol this would be very doable. The receiver can passively listen quite cheaply energy wise. This is nothing like decoding a video stream.

And ethernet and 802.11 receive every frame, whether it's addressed to the equipment or not.

Reply on news.ycombinator.com