I’m currently working on a Smart Antenna for HDTV, since prior to the digital changeover, things worked well, and afterwords, I lost a bunch of channels. The design/build or buy decision is automatic, being there is nothing on the market. Yet, this is a non-rev project, not so much that there isnt demand, but beyond the marketing hassles, the regulatory part could turn into a real pain.
In fact, its interesting to note the examples given in the CEA-909 standard are specifically designed to avoid FCC part 15. Initially I went whoa, what is up with this. Why monostable multivibrators, and logic chips, thats 1970’s tech, and a small micro would be faster to develop, cheaper from a BOM pov, and require a much smaller footprint. Then I got thinking, antenna manufacturers operate in a specific mindset… as a general rule, most dont have microprocessor based products, even a lot of rotors are still analog oriented. Also, putting a microprocessor on a antenna mast subject to all sorts of weather conditions could lead to some interesting situations (not that the timing aspect of a rc based monostable multivibrator wont be subject to drift either). Fortunately I’ve got years of experience getting microprocessors to behave in adverse conditions so its no big deal… but the antenna firm that goes to a consultant who may or may not have such experience may be in for a lot more cost than expected.
That being said, FCC part 15 requires compliance testing for digital devices classified as unintentional radiators. If I could drop the clock rate to below 9Khz, be battery powered, and have no additional IO, I would be exempt. Sadly, there is no way around that for a smart antenna. Thus, if this were to be a commercial device, I’d either need to pay for compliance testing, or go analog. While I doubt ever taking the smart antenna design commercial, I could always develop in a microprocessor based domain, and convert to analog once I had all the specs and features wrung out. Thus there is an out to working around the regulatory part.
Another out, would be to make the smart antenna open source, and provide instructions, and support for those who want to build their own. Granted, there are some provisions for kits, and likely a smart antenna would fit under them for the most part, but for a low volume niche, selling kits is probably not financially prudent.
Also, the game plan is to have multiple features built into the smart antenna. The letter of the law would require testing of the worst case scenario, plus any significant variance in the design. More and more costs would ensue with that path.
Thus, the smart antenna remains non-rev, but there are some workarounds.