You may recall that TV broadcasters got really enthusiastic about the transition to HDTV when early on the powers that be — the FCC Advisory Committee for Advanced Television Systems, the Association for Maximum Service Television and the NAB — promised reliable HDTV service that would replicate analog's.

I was a participant in various ACATS committee meetings and the word "replicate" is etched in my memory.

I knew then (and said so in the meetings) that real replication would be unattainable because the system engineering was based on unrealistic assumptions regarding rooftop antennas, an unrealistic receiver model (more on that shortly) and an unacceptable standard of reliability

So, if you have been following viewers' complaints from around the country and reading the articles describing the outcome of the shutdown in Wilmington, N.C., let me assure you: The reception problem is consistent with the system design. There should not be any surprise.

Broadcast and government engineers are aware of the reality, but their leaders are in denial.

Bloomberg quotes David Donovan, the president of MSTV, saying, "You should be able to receive the same number of digital signals that you receive in analog."

And TVNewsday quotes FCC Chairman Kevin Martin saying that the Wilmington test is "going smoothly," even though more than 10 percent of the over-the-air-only households called to complain about disruption of their service.

Is this what we want in New York City? Is this what we want for the rest of the country?

The ATSC digital broadcast system was designed to allow the DTV signal to fall off the "cliff edge," where pixellation appears and sound is lost, for as much as 10 percent of the time for up to 50 percent of the households.

Why was the system designed to provide such a poor standard of reliability? Because that was the standard of the NTSC analog system and, remember, ACATS had merely promised to replicate the analog service.

There is an important difference, however. In analog, the 10 percent means the picture quality and sound may be degraded, but not lost. In digital, it means either the picture or sound is lost.

It didn't have to be this way. The European DTV system allows service loss for no more than 1 percent of the time at a maximum of 25 percent of the locations.

If you think that the designed reliability of our service is not a handicap, think about the Emergency Alert System failing 10 percent of the time at 50 percent of the locations and whether that is adequate for homeland security.

Fortunately, there is a way to correct for the ATSC system's inherent shortcomings.

Basically, we need a new converter box.

that at least a 10 dB improvement in the signal-to-noise ratio at the receiver is required for "reliable" service indoors at UHF and Hi-VHF channels. You'll probably need more than that for Lo-VHF channels.

Reception is "reliable," by my definition, if the frequency of pixellation and lost sound does not compel a frustrated viewer to switch channels or sign up for cable.

You can't get the 10 dB boost on the transmission side, which has been fixed by the FCC's channel assignments, but you can get 10 dB improvement from a better receiver with an integrated active antenna.

The one that I have in mind would feature a high-gain dual bow-tie antenna, its preferred angle optimized for each channel by rotation during setup. The antenna would be electronically integrated with a dynamic channel impedance matcher preceding the tuner.

That's a bit technical for non-engineers, I know, but I am confident is would produce the needed 10 dB improvement in signal-to-noise performance.

This is the simple solution, not the solution. Space and polarization diversity reception could be added as well as automatic angular adjustment.

Over the years, the low reliability reception designed into the ATSC system was forgotten.

Now, with viewers starting to complain about their DTV reception problems, it all begins to come back.

An improved converter box may be too late for the millions who have to hook up their analog TV sets prior to Feb. 17, 2009. But it's worth working on so that they will have an improved second-generation choice in the years to come.

Oded Bendov is president of the TV Transmission Antenna Group, a Montclair, N.J.-based consulting engineering firm. Among its current projects is optimizing the broadcast antennas atop the Empire State Building. You may e-mail him at oded@tvantenna.tv.