I've just come off a huge rant against patent trolls on another website (which, for political reasons, I probably can't link to here because it's a competing site) but after reading the actual patent in question, I think the radio companies may have some work cut out for them if they are not going to roll over for these companies.
The system they describe is very much like the current HD radio setup. It deals with a central voice channel being bundled with one or more data channels around it and sent from one "modem" to another. HD radio doesn't use traditional modems, of course, but the encoding hardware at the station and decoding hardware in your radio are really just fancy modulator/demodulators.
There may be some wiggle room for the stations to get this dismissed. Depending on how it's read, the patent seems to imply two way communications, while HD is one-way. It also appears to me to be a system for encoding data on the same channel as the (analog) voice, whereas HD broadcasts are actually completely separate and only combined at the antenna. They can even be transmitted from two different antennas in some cases.
The patent says that there needn't be an analog voice component, either; the main channel can also be data. That might apply to the multicast digital channels (HD-1, HD-2, etc.) but if that's the case then the patent, issued in the 90s, should be invalidated because digital multiplex broadcast methods have likely existed before then in labs, if not general use (i.e., prior art).
Bob Struble claims that this patent covers wireline transmission methods only, but I think he's mistaken because it was originally issued to AT&T. Nowhere did I see anything that specifically tied it to wired transmissions.
In the end, if the stations lose or just offer to license the patent, it's worth noting that one of the two in question expires this very month. The non-creative entity that is doing this is having its last-grasp of fresh air before it dies, so after this there won't be much they can go on. It's do or die time.