DTV Reception Advances At NAB2002

Consumers have gotten used to easy NTSC reception. If the picture looks bad, move the indoor antenna around a bit and you can usually get a watchable picture. Even if the picture is poor, if the program is interesting enough, the viewer will usually continue watching it if the audio holds up...DTV is not as easy. Once reception is lost, it isn't always obvious how to get it back!
Publish date:
Social count:

Consumers have gotten used to easy NTSC reception. If the picture looks bad, move the indoor antenna around a bit and you can usually get a watchable picture. Even if the picture is poor, if the program is interesting enough, the viewer will usually continue watching it if the audio holds up.

DTV is not as easy. Once reception is lost, it isn't always obvious how to get it back! There are ways to deal with this problem. One is to make receivers less sensitive to degraded 8VSB signals. Another is to make it easier for the consumer to improve the signal.

Of course, the broadcaster can reduce viewer frustration by providing the best signal possible. These were all topics of discussion at NAB2002.


On the receiver improvement side, one of the hottest products at this year's NAB was Richard Citta's Linx "near optimal multipath combining receiver," on display in the Axcera booth. The Linx demodulator requires a bigger equalizer, which increases complexity, but as long as Moore's Law remains valid, the cost of processing power will soon decrease to the point it becomes available in consumer DTV sets. It would be nice to see consumer DTV receivers competing on their ability to receive more channels with less difficulty.

Charlie Rhodes discusses the Linx receiver in his column next month so I won't go into more detail here. You may also want to read the related story in this issue on p. 1.


Understanding the algorithms behind the Linx receiver or any of the more advanced 8VSB demodulators is not easy. I was pleased to find a technology that improves DTV reception that anyone who has played with an indoor TV antenna should be able to immediately grasp.

Broadcasters have recognized that reliable indoor reception is important for the future of terrestrial DTV. Advances in receiver technology will make it easier to receive DTV signals, but problems with indoor antenna location and orientation will continue to plague many viewers who want to be able to channel surf without getting up to readjust their antennas.

To make DTV reception easier, the Consumer Electronics Association has adopted a standard that allows the DTV tuner to control the position and gain of an antenna. A DTV receiver with an EIA/CEA-909-based interface can select any one of up to 16 antenna positions and up to four gain settings. The sixteen antenna positions are not specified - they could include eight horizontally polarized positions and eight vertically polarized positions or combinations of multiple antennas at different phase angles.

(click thumbnail)Photo 1. NxtWave demonstration setup
NxtWave had a demonstration of the interface in their suite at the Las Vegas Hilton Resort. Photo 1 shows the combination of antennas used for the demonstration - a conventional indoor set-top antenna, NxtWave's prototype Nxtenna, and a "Silver Sensor" log periodic antenna. I took some pictures of the spectrum analyzer displaying the received signal as the receiver stepped the antenna through the various options. In this case, the gain setting was not used and less than 16 positions were available.

(click thumbnail)Photo 2. VSB spectrum during antenna optimization

(click thumbnail)Photo 3. VSB spectrum after antenna optimization
Photo 2 from the spectrum analyzer shows one of the worst positions. Photo 3 shows the response selected by the tuner. These photos were taken during multiple scans, so ignore the times on the photos. The Nxtwave engineer showing me the system told me that in some cases, the Nxtenna had been able to find a signal even after manual attempts at aiming the antenna failed. This ususally happened when the best signal was in a direction that the human didn't anticipate.

The antenna doesn't have to include any special technology. A set of stacked directional antennas, which could be etched on a printed circuit board and an RF switch is sufficient. For more complex designs, some of the adaptive antenna technology used in the cellular radio industry may be useful. The trick will be in producing a TV antenna that won't dominate the top of the TV set.

If multiple antennas are used, they have to be spaced far enough apart to avoid interacting with each other. As you can see from the photo, the prototype antenna dominates the table top, but I think it looks neater than the rabbit ears. Given the size of many HDTV displays, there is another option - build the TV antenna system inside the TV set. Bring your DTV set home, plug it in, and start watching HDTV! Conceivably, the interface could be used with a set of outdoor antennas configured to allow reception of stations from different angles.

NxtWave is including the EIA/CEA-909 interface in 8VSB demodulator chips starting with its NXT2003 chip. While the concept is simple,`implementation is more complex. If the DTV tuner had to pause on each possible combination of the sixteen orientations and four gain settings and wait for sync lock, most viewers would lose patience before the search stopped. NxtWave's circuitry is able to determine the condition of the channel before sync lock is obtained, greatly reducing the search time.


Even with this smart antenna technology, there will still be cases where the signal level is too low or multipath distortion too high to allow reliable 8VSB reception. The ATSC receiver enhancement program described at last year's NAB engineering conference is making progress. During a panel discussion on the ATSC's 8VSB Enhancement Test Program, Victor Tawil, senior VP of MSTV, reported ten companies had submitted proposals and two types of enhancements were being evaluated: training signal enhancement proposals from Broadcom and Oren, and dual-rate/dual-stream proposals from NxtWave/Zenith and Philips. Tawil said the Consortium had not received equipment for testing from Broadcom or Oren, so testing is focused on the dual-rate/dual-stream proposals.

Zenith and NxtWave have combined their proposals for adding a robust lower data rate stream to the ATSC 8VSB signal to provide reliable reception under conditions where the full 19.39 Mbps 8VSB signal fails due to weak signal strength or certain multipath conditions. Philips' proposal adds a "pseudo" 2VSB-modulated robust data stream in place or in combination with the 8VSB modulated stream. (Please refer to my RF Technology columns from last year for more information on the Zenith, NxtWave and Philips proposals.)

During the panel discussion, the Philips representative said advanced receiver algorithms won't solve problems with insufficient signal strength or low CNR (carrier to noise ratio). A robust constellation, as proposed by Philips, is needed.

Matt Miller, president of NxtWave, pointed out there are several ways to tackle 8VSB reception problems, including optimizing the DTV transmitter's power and height; using a receiver with a high-gain/low-noise tuner; improving the VSB demodulator's equalization, timing and carrier recovery capability; enhanced VSB methods such as those proposed by Zenith and NxtWave; and adding the capability for receiver - antenna interaction. Normal technology evolution - improved algorithms and the IC gates to implement them - will drive improvement in VSB tuners. He added that there was much more cooperation inside the consumer electronics industry working towards improved VSB reception than was generally recognized.

MSTV and NAB joined together to form the VET (VSB Enhancement Testing) Consortium and conduct lab and field tests of the Philips and NxtWave/Zenith enhancement technologies. The tests will be comprehensive, with 1,500 to 1,800 lab tests, including backward compatibility tests with existing DTV sets with tuner and DTV set-top boxes. It first appeared the VETC would have to test 54 DTV sets and 46 set-top boxes, but after examination showed common circuitry among many of the sets, the number requiring compatibility tests dropped by two-thirds.

UHF DTV field tests are planned in Washington, using WUSA-DT Channel 34. High-band VHF DTV field tests are planned for New Haven, Conn. with WTNH-DT, Channel 10. The tests are expected to be completed in late June. A variety of data rates will be tested, ranging from 0.5 Mbps to 4.5 Mbps.

It isn't clear how much of the data from the tests will be made available to the public but broadcasters and manufacturers are invited to participate in the ATSC work. See for more details.

Next month, I'll look at improving DTV from the transmission side. Most transmitter manufacturers were showing low-power DTV transmitters at NAB this year. I'll describe some of the offerings and offer some tips on how to engineer a low-power DTV installation to reach the most viewers.

Comments are always welcome! Drop me a note at