DTV 2.0 is still being developed, but what’s next?

Jeffrey KraussThe television service we have today might be called Digital Television 1.0. The Advanced Television Systems Committee (ATSC) is in the final stage of designing the features of Digital Television 2.0. What about 3.0? People are thinking long-term thoughts about that, too.

One main feature of broadcast DTV 1.0 is channels having a bandwidth of 6 MHz, just like analog TV. For broadcast television, the modulation is 8VSB, and the data rate of the channel is around 19 Mbps. Each broadcast channel comes along with an FCC license, and each license authorizes the use of a 6 MHz channel. Another main feature of broadcast DTV 1.0 is one-way streaming of video. No video-on-demand. No delivery of programming for DVR storage and non-real-time viewing. No interactivity.

DTV 2.0 is still being developed, but a few features are becoming clear. It will have the ability to deliver programming for DVR storage and non-real-time viewing. It will have interactivity by using Internet access that is supplied by some non-broadcast technology (cable, DSL, Wi-Fi, etc.). Using this interactivity, it will support programming-related “second screen” applications on tablets.

And what about the 6 MHz channels? The 6 MHz channels will remain, so existing DTV receivers will continue to work just fine for the traditional broadcast programming part of DTV 2.0. But there is a question about the one-to-one correspondence between a channel and a licensee. The FCC wants to take back broadcast spectrum for use by mobile (cell phone) services. The FCC is planning a type of auction that would allow broadcasters to give up their full 6 MHz in return for cash proceeds from the auction, plus a share of another broadcaster’s channel. So the 6 MHz channelization will remain, but the one-to-one correspondence between channel and licensee will not.

Admittedly, that aspect is not part of the ATSC 2.0 planning. But if it occurs, it would take place in the same timeframe as 2.0: over the next 10 years. So my conception of DTV 2.0 includes shared channel ownership.

But DTV 3.0 will be very different. You can probably forget about 6 MHz channelization. The ATSC goal, for example, is to produce a substantial increase in spectral efficiency, which means the current 8VSB modulation won’t cut it. Whatever modulation and coding is chosen, existing DTV receivers won’t work.

It’s way too early to say what techniques DTV 3.0 will use to achieve the higher efficiencies, but one technology being proposed is LTE, the technology that underlies the 4G cell phone systems. LTE uses a modulation method called orthogonal frequency division multiplexing (OFDM) to allow multiple users to share a band of spectrum.

A more advanced version of LTE includes a technique called evolved Multimedia Broadcast Multicast Service (eMBMS), which is essentially a broadcast or multicast mode. And eMBMS was already part of LTE Release 8, but the folks that define the standards for LTE are now up to Release 10. Work is underway on Releases 11 and 12. Ericsson and Qualcomm demonstrated eMBMS in February at a big mobile communications show in Barcelona. Qualcomm plans to include eMBMS capability in its LTE chipsets, and commercial products are supposed to be available in 2014.

According to some reports, LTE has a number of advantages. OFDM tolerates multi-path interference better than 8VSB. I’ve seen a report claiming a spectral efficiency goal for LTE of 30 b/s/Hz downstream and 15 b/s/Hz upstream. That’s compared with a little more than 3 b/s/Hz downstream for 8VSB. Where 8VSB is limited to 6 MHz channels and 19 Mbps data rates, LTE could combine channels to achieve data rates up to 1.2 Gbps in 40 MHz downstream. LTE can also support upstream transmissions, with data rates up to 600 Mbps in a 40-MHz-wide channel block. Probably too good to be true, but you never know.

A DTV system using LTE would have additional benefits. LTE receivers are individually addressable and can support ratings information and targeted advertising. The system would use IP transport rather than MPEG transport. And the mass production of LTE chipsets for mobile communications would drive down receiver costs.

Would broadcasters be willing to go in that direction? It would mean giving up the individual spectrum rights they have now, because spectrum blocks would need to be aggregated and shared. The FCC licensing process would have to be changed. The whole business model for TV broadcasting would have to change. Maybe the spectrum would be licensed to a jointly owned cooperative (or maybe Verizon would acquire the licenses), and individual broadcasters would become merely programmers that lease spectrum capacity. It’s hard to know what to expect.

When will this all come about? Not anytime soon. But long-range planning is often worthwhile, even if it leads to unexpected results.