mismatched bandwidth pipes
Sunday, April 11 2010
I've been investigating my options with regard to controlling my main computer remotely, a capability that seems increasingly necessary given my reliance on low-tech remote interfaces (by which I mean cheap FM radios). Ideally I'd have a cheap remote control and a cheap FM radio that I could use in tandem to select digital audio and even get updates about things like email and weather. The FM radio is an ideal receiver for computer audio information; with a cheap Chinese transmitter attached to the sound card, I can receive whatever audio it happens to be producing at a range of up to four miles (though, for cheap portable radios powered by a single AA battery, the range is closer to a half mile). The advantage of this system over a conventional MP3 player is that I can quickly queue up a playlist at my computer using an uncompromised interface without any synchronization process. And I can also listen to audio from streaming internet sources. Best of all, a fully-analog FM radio uses almost no power; I find myself swapping the AA rechargeable battery about every other month.
But to make this remote interface two-way, I need some sort of radio transmitter to send commands back to my computer, and I need software on the computer to respond to these commands. The most obvious system to tap into is LIRC (particularly its Windows incarnation, WinLIRC), which allows computers to be controlled by infrared remotes. Of course, I want to control from a distance using radio waves, not infrared beams, but perhaps I could grab the signal before it reaches an infrared LED, broadcast it on a radio frequency, receive it, and send it on the computer as if it had come off a phototransistor.
So today I tried using an inexpensive 315 MHz transmitter/receiver pair I'd bought from SparkFun.com to broadcast a signal from an IR remote to a receiver attached through some additional electronics to one of my computer's serial ports, and then I monitored the data. At first I didn't see anything at all, and I eventually realized that the receiver was incapable of sourcing enough power to drive the circuit I was using, so I buffered it with a 74HCT04 and then it sort of work. But it didn't work very well; while the various signals sounded pretty much the same on the transmit and receive side when passed through a speaker, something was being lost along the way and digital data was getting scrambled. Still, there seemed to be a one-to-one relationship between what was being sent and what was being received, and for a system like WinLIRC (where the actual content of a transmission is less important than its consistency, and even then there is a certain amount of tolerance), it seemed like it might work. Unfortunately it did not. I installed WinLIRC and tried broadcasting the signals from various buttons on a universal remote, but WinLIRC kept reporting that they were inconsistent. I suspect the problem lay in the bandwidth of the transmitter/receiver pair, which is supposedly 4800 bps. But an IR remote sends its data modulated onto a 30-60 KHz carrier, and any such carrier sent through 4800 bps "pipe" would probably be mangled to the point where the superimposed data would be unrecoverable.
For linking purposes this article's URL is:feedback
previous | next