Writing

Software, technology, sysadmin war stories, and more. Feed
Thursday, October 27, 2011

Who's actually interested in listening to San Jose?

Using a software-defined radio means I can cast a wide net. Whereas most radios will pull in a single channel at a time (missing everything else), I can pull in a whole 4 MHz chunk all at once. It turns into a ridiculous amount of data streaming off the device, but a modern machine can keep up with it.

But enough with the numbers. What's the point of all of this? That's actually the easy part.

Let's say you want to listen to everything that San Jose PD does. All but one of their (known) channels lives between 460 and 461 MHz. That means you first have you capture that whole range. 1 MHz is no big deal. Then you just drop a software filter into the stream for every channel you want to hear. You set the center frequency and bandwidth appropriately, and feed the result to a FM demodulator.

Now you have audio, and you can do whatever you want with it. Recording WAVs or MP3s seems to be high on the list, with OGGs running a distant third in terms of codecs in use.

There is a catch, though: it chews a fair amount of CPU power to do this in the kind of massively-parallel manner which is required to get everything at the same time. It also starts consuming extra disk space to record all of these channels, but that's relatively minor in comparison.

Being the pragmatic type, I'm not about to stand up the capacity needed to catch all of San Jose, Sunnyvale, Milpitas, and the other nearby cities unless there's a reason to do it. That capacity could also easily be used to snag the air bands (SJC tower, ground, approach/departure) or anything else which is sufficiently interesting.

I want to hear from the listeners out there. Hit the contact link at the bottom of these posts to send me a message.