Walter S. Ciciora
By Walter S. Ciciora, Ph.D., Recognized Industry Expert on Cable and Consumer Electronics Issues
When an interest arose in HDTV, the FCC put out its wish list. The dream included components that seemed impossible. The dream was for a system that was compatible with the existing analog NTSC system and which could be placed in the same 6 MHz as the NTSC signal, allowing channels to be closely spaced so that spectrum could be freed up for other purposes. After awhile, the dream faded, and its elements were abandoned, one by one. All that is left of the original dream is the desire to collapse channels to free up spectrum.

The original "impossible dream" was not far-fetched–technology now makes it possible. Only inertia, the unwillingness to recognize premature "standardization," and politics stand in the way of the dream.

Let's explore how the dream can come true. We need a way of including a large amount of data compatibly in the NTSC signal. Next, we need a video compression method that needs only a modest bit rate. The last element is a television receiver design that accommodates tightly packed channels. It turns out that all of these elements are now available.

The December 2001 issue of CED describes ways of embedding data in NTSC without damaging its television purpose. EnCamera Sciences developed one method, called dNTSC. EnCamera was sold to Dotcast Inc. in October 2000. dNTSC conveys 3.0 Mbps on a quadrature carrier and 1.5 Mbps as AM on the aural carrier. The second technique, championed by Chinook Inc., puts in 6 Mbps using spread spectrum. These techniques can be used simultaneously for up to 10 Mbps. Extend Networks of Israel claims 6 Mbps in NTSC. Not enough is known to determine if this method can be simultaneously used. So it appears that 10 to 16 Mbps can be included in the same 6 MHz channel as an NTSC signal.

This is about half of the data promised by the 8-VSB standard. Because significant advances have been made in video compression, Standard Definition Television (SDTV), MPEG-2 signals provide good results with 1 to 2 Mbps. The data rate that can be embedded harmlessly in NTSC supports several SDTV signals and other data services. If this full data rate is used, HDTV or near HDTV is possible. DemoGraFx has proposed a layered video compression. Microsoft has created a stir with its Corona compression that is said to be capable of providing HDTV video at between 3.5 Mbps and 9 Mbps. Work on MPEG-4 and alternative compression techniques such as wavelets is underway.

Consider the third part of the dream, the compaction of television signals so that spectrum can be freed for other purposes. The reason for blank channels was to accommodate cost-effective tuner design using vacuum tube technology from a half century ago. Tuner technology has improved tremendously. Much of that improvement is motivated by "cable-ready" television that must function in a fully loaded spectrum. While the broadcast environment is in some ways more difficult, modern tuner technology can easily meet this challenge.

One argument against filling most of the current blank channels with NTSC is that it would take years to replace television receivers with new models that have better tuners. This is a red herring. It would take longer to replace analog receivers with digital ones so that spectrum could be repurposed because digital receivers are much more expensive than NTSC receivers with better tuners. Most current receivers will do just fine with a compacted spectrum.

Finally, those television receivers that don't function acceptably can be helped with a simple converter including an adequately performing tuner, costing about $50. Very significantly, the worst-case scenario for the less-well-off is that their older receiver might suffer modest video quality degradation. This is in sharp contrast to their plight if NTSC is abandoned and replaced with digital. In that case, their existing television receivers become useless.

What about those who have already purchased a digital television receiver? The vast majority of "digital television products" sold have been large-screen displays that are "digital-ready" and are used for DVD home theaters. They will not lose any functionality. Very few 8-VSB digital receivers have been purchased by consumers. Fully integrated digital television (DTV) receivers have baseband inputs allowing them to be used with new set-top boxes. Only a trivial number of stand-alone DTV set-top converters have been sold. It is unlikely the less-well-off have purchased such an expensive product. The CBS color system was a standard, and some receivers were sold, but it was replaced by another standard. The number of full DTV products–not monitors with baseband inputs–which would be displaced is likely smaller than the number of CBS color wheel sets made obsolete.

Here's what is important to cable: 1) Digital must-carry is not necessary to save "free" broadcast TV if it steps back from obsolete technology and adopts the original dream which accommodates both analog and digital broadcasting in the same 6 MHz. 2) Cable needs to consider ways of including advanced video compression in its future. 3) Analog signals for the multiple older TV receivers in the home can carry massive amounts of data in the spectrum used for the analog signals. So now what?