Advertisement

Analog Angle Article

IR is dead; long live IR

Every day we read that wireless is where the action is, and it's hard to argue with that. Yet I also see many announcements about new receivers and emitters for infrared (IR) links. Typically, these newer devices offer smaller size, higher sensitivity, and wider dynamic range. {IR is also wireless; but when engineers say wireless, they usually mean RF and not optical.)

What's going on? It's this: the IR link is a clear case of a solution that is matched to the problem. IR is a good example of a technology that has come a long way, and come down in cost and size while coming up in performance so that it is near-ubiquitous. While IrDA on PCs hasn't been a success, IR itself is very alive and well as the most common form of basic link out there.

Virtually every home appliance and device comes with it (even some coffee makers) since it adds only about a dollar to the BOM cost, consumes little power, and has much less design impact and subtleties than RF. Just as important, the software burden in memory space and coding complexity for a basic IR links is quite modest.

Furthermore, the IR format and protocols are relatively simple, since it is usually used in a point-to-point, non-networked application over a modest distance, with a pre-matched source and receiver pairing. (Sure, you can buy a “universal” remote control, but is still then normally configured to work with a limited set of peripherals.) This greatly simplifies all aspects of design. For most IR-based links, it's a near-perfect example of the well-known KISS design principle: keep it simple, Schweber. IR-based products usually work right out of the box, without complex setup, initialization, or problems. An IR link does one thing, does it well, and doesn't try to do everything.

There's only one problem with IR: the different schemes which vendors use for physical-level coding and modulation. I know of no good reason for this, other than to be different. Certainly, new and different peripherals may need additional codes and sequences, but do the most basic ones have to be different? And is there really a substantial benefit to one of the many IR modulation formats used, compared to the others? I'd call it the no-good-reason, un-KISS principle.

-x x x-

0 comments on “IR is dead; long live IR

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.