RadioWorld 9/22 Reader's Forum

[dumber than a box of hair]

The Reader's Forum in the 9/22 RadioWorld magazine is a telling microcosm of the state of HD radio today. The first three letters were from people whose reputations in the radio business are beyond reproach: Ron Rackley, a well-known consulting engineer; Edd Monskie, the VP of engineering for Hall Communications; and Dick Pust, the GM of KGY in Washington state. All three echoed Tom Ray's experience of not being able to find HD radios for cars, but finding plenty of retail sales people who didn't have a clue what HD was.

Ron Rackley focused his letter on the constant dropouts in AM-HD on a signal of 1-2mV and said that a force-analog switch on the front panel would have avoided the dropouts.

Edd Monskie noted that the DOE of another company confided that when his HD2 goes off, no one complains. He was also getting "sketchy" AM-HD reception with a 2-3mV signal.

Dick Pust's listeners have been telling him that AM-HD radios have poor reception on analog AM...just what AM needs, right?...unless the listeners is within a mile of the tower. What came next in Pust's letter was either funny or pathetic: He called Roy Sampson at iBiquity and passed along the complaints. "He said it was the first complaint like that he'd heard and that he'd pass it along to his engineers."

You can't make this stuff up.

 

[Savage]

2 to 3 mv/m? Haw. I've been told by CE's of two major-market 50kw AMs that they have trouble getting reliable decode with a signal strength five times that. And of course, any kind of noise like lightning, a worn AC motor nearby or dirty power lines or a fluorescent light - "welcome back to analog." With some REALLY annoying mode-hopping to boot.

As far as Roy Sampson's ostrich game: gee, wonder why official pronouncements from iBiquity have approximately zero credibility any more? As long as the official company line is to lie at every opportunity....well, good luck selling your product to broadcasters...

 

[nd2023]

Not even the DJs of some stations even know their station is in HD. I called one station saying that their HD is out of sync with the analog. The DJ on the air was confused and didn't know what HD radio is.

 

[KB1OKL]

Dick Pust's listeners have been telling him that AM-HD radios have poor reception on analog AM...just what AM needs, right?...unless the listeners is within a mile of the tower.

I used to say stuff like that all the time here and get all kinds of boos and hisses, I live approx 10 miles from a 5 KW AMer and it never once came in on my Sony with a C Crane twin Coil ferrite tunable directional antenna which cost more than the tuner did and this station blasts in here all day and night, the IBOC during the day kills about 50 Khz of dial space.

What came next in Pust's letter was either funny or pathetic: He called Roy Sampson at iBiquity and passed along the complaints. "He said it was the first complaint like that he'd heard and that he'd pass it along to his engineers."

Now that's funny!!! hahaha!

You can't make this stuff up.

 

[tomservo]

Not even the DJs of some stations even know their station is in HD. I called one station saying that their HD is out of sync with the analog. The DJ on the air was confused and didn't know what HD radio is.

No offense to any DJ's out there, but I wouldn't expect them to know about anything beyond, "The station's off the air and on fire!"

Seriously, you wouldn't call the engineer and request Muskrat Love, would you?

A valid complaint is telling the engineers there's a sync issue and them summarily ignoring you. Or me, as I've been ignored twice now from Mississippi Public Broadcasting when I complained about the low audio levels on the HD feeds.

 

[KB1OKL]

Two comments, One, I wouldn't ask anyone to play Muskrat Love .
And two, a local AM IBOCer suddenly started keeping their 50 Khz wide whoosh machine on 24-7, I wrote an email to the engineer who checked his logs (he's only there for one afternoon a week) and found that during pattern change at local sunset the IBOC was not going off as it was supposed to and promptly fixed it, wished me happy DXing that night and invited me out to view the transimitters some night including a 1938 RCA which still functions.

 

[pianoplayer88key]

Ok... something totally does NOT compute here!!! Explain how the heck a digital signal can struggle with a field strength of upwards of 10mV/m, but an analog signal that's somewhere south of 50µV/m (and possibly much lower, assuming the radio is much more sensitive than my Tecsun PL-380 and Select-A-Tenna combination - I think I've seen specs of some communications receivers in the single digit microvolt range, and I'm assuming that's without an external antenna connected, unless the radio has no built-in antenna and you salvaged a poorly-made badly-misaligned 1" ferrite loopstick from a cheap pocket radio) can be nearly perfectly copyable? That just does NOT add up, to me. When digital was still mostly a thing of the future, but was being talked about, I had always thought that one of the advantages of going digital would be that you could perfectly decode a signal that on analog would be right at the atmospheric noise level.
Also, with these digital stations taking up upwards of 30kHz (and more) bandwidth, yet only offering 60kbps digital bandwidth, per one source I found. Last I checked, an AM channel was supposed to be about 10kHz wide (although you're allowed 20kHz to broadcast audio out to 10kHz). Assuming a possible spectral efficiency of 14 kbps / khz (per another wikipedia reference), Why shouldn't I expect a digital signal to be able to put a 140kbps signal in a 10kHz wide channel?
Also, shouldn't another advantage of digital be a steep dropoff on the transmitted signal once you're outside the spectral mask? For example, with a station broadcasting +/-5kHz (total 10kHz) bandwidth, shouldn't I expect to not find any trace of the signal 5,000.1 Hz away from the carrier, assuming I'm using an ultra-sensitive radio with a beverage antenna, very wide selectivity (about like a crystal set with only one cheap tuning capacitor, and an otherwise untuned antenna), and I'm at the 614.4V/m field strength contour of the digital signal?

Another thing seriously wrong with IBOC. Why can I sometimes hear the hiss from a station at night, but when I tune to the adjacent channel where their analog signal is supposed to be, all I hear is atmospheric noise?

 

[Savage]

60 kbp/s? Try 32 kbp/s for stereo, which diminishes still further when the signal downgrades to mono. Even giving the most generous allowance for the wonderfulness of IBOC's codec, there's just no way something with that low resolution is going to sound natural. And you get the COFDM sideband hiss long after the analog carrier is no longer readable because there are a total of 25 steady-state (constantly ON) digital carriers in each sideband. There's a log of energy in those digital carriers. That's why they're so obnoxious in analog, and skywave bounces them with a vengeance.

HD-AM is a cluster of compromises with each end-result being completely unacceptable even if compared with the 90-year old analog system it purports to replace. (Emphasize "cluster.")

 

[Play Freebird]

Ok... something totally does NOT compute here!!! Explain how the heck a digital signal can struggle with a field strength of upwards of 10mV/m, but an analog signal that's somewhere south of 50µV/m (and possibly much lower, assuming the radio is much more sensitive than my Tecsun PL-380 and Select-A-Tenna combination - I think I've seen specs of some communications receivers in the single digit microvolt range, and I'm assuming that's without an external antenna connected, unless the radio has no built-in antenna and you salvaged a poorly-made badly-misaligned 1" ferrite loopstick from a cheap pocket radio) can be nearly perfectly copyable? That just does NOT add up, to me. When digital was still mostly a thing of the future, but was being talked about, I had always thought that one of the advantages of going digital would be that you could perfectly decode a signal that on analog would be right at the atmospheric noise level.

The problem is that there's a necessary trade-off between channel bandwidth, data rate, and power. This relationship was explained in a 1948 paper by scientist Claude Shannon, and you can read more about it here, or just google "Shannon Limit":

http://www.physorg.com/news183117569.html

In order to squeeze the AM digital sidebands into the adjacent channels (which iBiquity was able to convince the FCC fall within the legal emission "mask"), it's necessary to use 64-state quadrature AM (64 QAM) which has poor noise immunity.

The IBOC coding scheme tries to overcome the noise problem with forward error correction (which detracts from the through data rate), but as Mr. Shannon pointed out, there's a physical limit to what you can do. This is just common sense; the more states you use, the harder it is for a receiver to distinguish them from the noise. His paper assumed Gaussian noise ("white noise") but in the medium wave band we also have to deal with power line hash, lightning noise, RFI from TV sets, etc.

Of course, more power would help (this is the approach they're trying with FM IBOC) but AM interference is already excessive at the present injection levels. So it's a no-win situation. Hello FCC, are you listening?

I should mention that it's indeed possible to decode a signal near the noise floor if the data rate is greatly reduced. This would make it possible to use bipolar phase-shift keying (BPSK), which, with only two states, is much more robust than 64 QAM. For instance, RDS uses this form of digital modulation. However, the data throughput wouldn't be high nearly enough for acceptable audio.



I don't always trust Wikipedia, but in this case there's a good article on QAM and its tradeoffs:

http://en.wikipedia.org/wiki/Quadrature_amplitude_modulation

 

[local oscillator]

When you get right down to it, AM IBOC is a cobbled together Rube Goldberg system that would allow iBiquity to promote a complete digital "solution" for radio -- AM would not be "left out" (would that it were!). FM IBOC was poorly thought out; AM not thought out at all. And . . . the emission masks were intended to catch the occasional bit of errant energy, not to be bulging at the seems with digital destruction.

 

[jhardis]

And . . . the emission masks were intended to catch the occasional bit of errant energy, not to be bulging at the seems with digital destruction.

There seems to be rampant confusion on this point. Speaking about the FM side (I haven't looked into the AM side), and on-the-record statements by petitioners not withstanding, no IBOC system tested or deployed in the last dozen years fits within the FCC's FM emissions mask (§ 73.317).

To state the true facts clearly, digital power levels of –20 dBc, –14 dBc, and –10 dBc are all greater than –25 dBc, the relevant emission permitted by § 73.317, by 5 dBc, 11 dBc, and 15 dBc, respectively. IBOC (at least for now) divides radiated power evenly between two spectral peaks (one each in the upper and lower first-adjacent channels), which therefore creates distinguishable emissions having individual powers of –23 dBc, –17 dBc, and –13 dBc, respectively. (Division by 2 and subtraction of 3 dBc are equivalent.) These distinguishable emissions also individually exceed the –25 dBc standard, by 2 dBc, 8 dBc, and 12 dBc, respectively.

Petitioners have shamelessly gotten away with some mathematical slight-of-hand. § 73.317 contains a specification of –25 (dBc), which is a measure of POWER. They then compare this number to other numbers with units of measure dBc/kHz ... that's dBc per kilohertz. The FM spectral mask doesn't say –25 dBc in each kilohertz interval along the spectrum, it says –25 dBc ... period. dBc/kHz are units of POWER SPECTRAL DENSITY. Comparing measures of these different things, with different units of measurement, is literally like comparing "apples and oranges."

For more detail, please see: http://fjallfoss.fcc.gov/ecfs/document/view?id=7020408278 and http://fjallfoss.fcc.gov/ecfs/document/view?id=7020417607.

- Jonathan

 

[Savage]

And the reason this nonsense gets blown past the Commission is: mostly the policymakers are not only no longer RF engineers, they're agenda-driven technically-clueless lawyers. Take, for example, Peter Doyle - Mass Media Bureau Chief whose degrees are in law and philosophy. He doesn't want to know about interference issues or poor performance of the HD system. All he knows is: "it's digital, so it's gotta be good."

As such Commission leadership is extremely susceptible to lobbyists and special interests. Even the pretense of critical analysis of the HD system's technical attributes was dispensed with.

The process was all about cobbling a rationale to justify HD. It was rammed through with all the cynicism of a banana republic junta. Once again, good policy is victimized by self-interested elites. At the end of the day, the majority of broadcasters and the general public are the losers.