I finally got tired of the 2021 theme and updated to the latest and greatest (?) WordPress theme, mostly so my post titles aren’t 96 points tall, and also so I can use the new features. But now this means the header links and post format will be a little bit off until I customize the 2023 template. Thankfully, I didn’t break my permalinks or anything seriously bad.
I wanted to give a little update on the evidence for my “2G GSM TDMA parameters were set up to mind control us” theory, or rather the lack of hard evidence. As I suspected, but verified on the 3GPP site, there was never any rationale given for why the TDMA slots and channels are the lengths that they are.
GSM Rec. 05.01, PHYSICAL LAYER ON THE RADIO PATH: GENERAL DESCRIPTION, shows up first as version 3.3.2 dated January 1991, with earlier versions apparently not kept under change control. Here’s a link to the latest 3GPP TS 05.01 Version 8.9.0 Release 99 in PDF (your other choice is a .zip’d Microsoft Word .doc file, which is the only format the older, pre-ETSI versions are available in). The relevant line, which shows up in the earliest version, as if tablets brought back from Mount Sinai:
“The time slot is a time interval of ≈ 576,9 μs (15/26 ms), that is 156,25 symbol duration, and its physical content is called a burst.”
You can tell this must be a very old spec, because the numbers are written European-style, with “,” and “.” swapped from US/UK-style, unlike every other 3GPP spec that I’ve ever read, as far as I can remember. One of the overview pages on the 3GPP site notes that the very old GSM specs haven’t always been updated to the current recommendations, style-wise.
There are 8 time slots per TDMA frame, which gives you a 216.66666… Hz TDMA pulse, assuming your phone transmits in 1 or more consecutive slots within the frame. You’ll get a square-wave interference pattern in any nearby audio equipment, and perhaps, for purposes of my theory, perturb any non-local consciousness effects in the vicinity of the transmitter, i.e. the 2G GSM phone talking to the base station.
Update: for completeness, I double-checked the specs of the Infineon PMB8876 S-GOLD 2™ used in the original 2007 iPhone, and it supports GPRS/EDGE up to multislot class 12, which gives you a max of 4 uplink slots per TDMA frame with a minimum of 1 non-transmitting slot in between them, which gives you as the potential user the possibility of seeing in your vicinity a 433-1/3 Hz harmonic with 2 uplink slots per frame, evenly spaced, or an 866-2/3 Hz harmonic with 4 uplink slots per frame, or an even dirtier noise signal with 2, 3, or 4 uplink slots irregularly spaced. This is only relevant for data, not voice. I really think the popularity of 432-Hz tuned music for “healing” has something to do with the 2G GSM interference, and the perhaps sinister reasoning behind choosing those specific timing values. Using a 13.000 MHz master clock is extra X-Files-y, so, thanks for that “clue”, evil European layer 1 protocol designers from 1991 who picked the magic numbers.
There are about 2 dozen different types of channels, each with a different number of bits and different error encoding schemes, because GSM has to be really unnecessarily complicated. The beauty, from an implementation standpoint, of 4G LTE (and 5G) is they finally jettisoned the circuit-switched protocol stack, which exists in parallel with the TCP/IP “packet-switched” protocol stack on 2G and 3G.
My contention is that there are a lot of questionable decisions made throughout the 2G protocol stack in terms of making it easy to spoof base stations and spy on people’s phone calls and texts outside of the “lawful intercept” wiretapping function, which is built into the network from Day 1, by the way. Some countries, like France and Germany, apparently didn’t want their citizens to have good encryption, so there’s a choice of no encryption at all (A5/0), which was only turned off circa 2007, if memory serves, and then the weakest encryption, A5/1, was slowly phased out after Ed Snowden’s leaks, but the carriers suspiciously drag their feet on turning these known-broken weak links off.
It’s only been less than a year since Android allowed some users to disable 2G connectivity on their phone, a feature I was trying to get added when I worked at Google on Android telephony 10 years ago, but I got literally nowhere advocating for it, with my manager, who always had some high priority task for me to work on and wasn’t interested whatsoever in security, or anyone else (the head of Android security had previously worked for NSA, and was only interested in application processor-side exploits and not the radio stack), and my impression was that every operator didn’t want their users to be able to disable 2G mode, even though we all knew it was both insecure and highly spoofable, and everyone in a position of authority could only be bothered preventing any blame for anything from landing on themselves. There was no promotion path for anyone, including myself, from trying to make an issue out of disagreeing with carriers demanding, perhaps via NDA guidelines, or perhaps through verbal warnings, I’m not sure, not to add a switch to disable 2G (until Jan. 2022, long after I stopped working for Google).
Lest you think Google is uniquely evil, I point to Apple’s continued refusal to add a “disable/enable 2G” toggle to the iPhone as proof that nobody in the tech industry cares about your privacy. Not really. Not unless they can use claims of having better security to get you to buy their brand of phone instead of the other guys’. They particularly don’t care if someone from the government or shadowy spy agencies is spying on you through the radio part of the path, which management considers spooky magic not to be thought about too much, rather than a landscape of potential threats.
Anyway, even if you think my “2G GSM 217-Hz noise has been messing with our brains all this time” theory is completely bizarre because you don’t believe in psychic resonant frequencies, there are about a zillion security and privacy reasons to turn off 2G in your phone right now, or switch to a CDMA/LTE/5G network like Verizon in the US, where 2G GSM isn’t even a possibility as a fallback option, except for roaming. 2G CDMA is insecure as hell, too, but Verizon already shut theirs off. 2G GSM is insecure by design, and 2G CDMA appears equally awful, from what little I’ve been forced to learn about it.
Leave a Reply