High Definition Radio is Here

nfranzen submits this story/advertisement: "Yesterday, I had the opportunity to buy the first High Definition (HD) Radio in the United States. HD Radio, invented by iBiquity Digital, adds a digital channel to the sidebands of an existing analog FM signal. The technology is still pretty new, but I can tell you first-hand that listening to my favorite local FM station in HD sounds just like I am listening to a CD. Well, except for the commercials (grin). Here are some links to local TV news coverage and a news release for more info. HD receivers will hit the open market following the Consumer Electronics Show next week in Vegas." We had an old story about the FCC approving these digital broadcasts in the FM radio bands.

Comments

Re:Solution looking for a problem

[iminplaya]

"So Local radio stations can compete against XM and Sirus."

There is no local radio anymore. It's all Clear Channel and...somebody else.

Re:What's the catch?

[IncohereD]

Encoding digital signals in a small amount of bandwidth has to come with a catch. What's this sound like if the signal strength is low? What kind of digital qaulity is this? Is there lossy compression used?

Keep in mind that digital signalling techniques weren't really invented at all until the 1940s. And that AM was deployed before than, and FM either before that or not much after.

Is it inconceivable to believe a brand new field has seen startingly gains in efficiency in 60 years time? Look at how much modems improved (56kps over the same line that once only supported 150bps...nearly a 400 times gain).

There is no catch. Telecommunications technology has just improved a hell of a lot in the last 100 years.

This is the reason why cell phone provides are so antsy to relaim all those 6 MHz wide UHF allocations....you can use that bandwidth so much more effectively with modern techniques, instead of throwing raw, uncompressed analog data out there.

Also witness the huge number of digital channels cable providers have packed into coax, despite the continued presence of regular TV stations, AND internet connections.

And this is the part where everyone should stop whining about taxes and having to give money to their local learning institution.

analog vs. digital

[wowbagger]

Well, there is a lot of analog out there, more than digital, but that's not really the problem - the problem is the "digital cliff" effect.

With AMPS, as the signal gets weaker, the audio noise floor comes up, and you get wideband static on the signal. Wideband static is fairly benign, in that humans aren't as offended by it (since it sounds like the surf). The user of the phone knows he is getting out of range well before the call drops, and so usually can terminate the call gracefully.

With digital, you get no real degradation of the signal so long as the channel bit error rate is less than the channel's error recovery capability. But when the BER gets above that threshold, then the quality drops dramatically. Moreover, the loss of quality is expressed as garbled vocoder output (I've always described it as "watery" - it sounds like you have water in your ears), or as complete failures of the vocoder (dropouts). Those are VERY offensive to the ear.

Also, the difference between a signal level that gives you a fully correctable BER and a signal level that gives you a BER bad enough the phone drops is almost nil - so just changing position can drop the call without warning.

Personally, if the phone makers would tie the received signal strength indicator (RSSI) into a variable noise generator, so that as the RSSI fell you started to get static, I think most people wouldn't bitch so badly about dropped calls.

There is also the problem that the usual vocoders for phone use are compressing the crap out of the signal - taking a 64 kb/second audio stream down to less than 4kb/sec. VSELP, IMBE and AMBE all do OK when fed voice in isolation, but put in any background noise and they get "confused" - they start making poor choices about the vectors they encode, and what comes out the other end is pretty rocky.

I had great fun feeding the first few seconds of Kansas's "Carry On Wayward Son" into an APCO-25 IMBE vocoder. While there is nothing but voice there, it is a chorus, and the poor vocoder just couldn't figure out what was going on.