Advertisement

By Jeffrey Krauss,
PSIP Performer
and President of Telecommunications
and Technology Policy
Now that cable MSOs are getting serious about carrying digital TV broadcast stations, we need to examine some of the disagreements between broadcasters, TV set manufacturers and cable operators over the carriage of PSIP. PSIP (Program and System Information Protocol) is an industry standard for digital television that you can download from the ATSC Web site (www.atsc.org). PSIP consists of channel mapping data, program guide data, information about closed captions and content advisory ratings, and other data related to the current and future programs.

The cable industry doesn't use PSIP. A broadcast station's PSIP data is carried in-band, within the 6 MHz channel. A cable system carries this data out-of-band, in one high-speed data channel covering all of the channels on the cable system. The out-of-band channel is only accessible to cable set-top boxes that include conditional access (descrambling) circuitry.

PSIP was covered in the FCC's digital must-carry decision last year. Basically, the FCC said that cable systems must carry the part of PSIP that relates to the current program, but need not carry the part that provides the program guide for future programs. Of course, retransmission consent contracts can specify some other arrangement.

The channel mapping capability in PSIP allows the broadcaster to "brand" his digital channel with the same channel number as his analog channel, even though they are carried on different RF channels. So, for example, a broadcaster whose NTSC channel is 4 and digital channel is 28 can call the digital channel 4-1 if it carries a single HD program, or 4-1, 4-2, 4-3 and 4-4 if it carries four SD programs. In the case of the four SD programs, they are said to be carried on "virtual channels."

Some digital TV receivers may need the in-band PSIP data to navigate within each 6 MHz digital channel. Others, however, seem to work just fine without PSIP, or even when the PSIP data is wrong. We know this because some digital TV broadcast stations either haven't been sending PSIP, or have been sending incorrect data.

The cable industry recognized the role of PSIP in an agreement with the Consumer Electronics Association, and agreed to pass through any PSIP data that was delivered within unscrambled channels. But the cable industry did not agree to create PSIP data at each headend. Cable programmers are not expected to deliver PSIP data with cable programming. And that has become a point of dispute between the cable industry and the TV receiver manufacturers. TV manufacturers want to use PSIP guide data to create electronic program guides that compete with the guides that cable operators provide. They want access to the PSIP program guide data for each channel so that they can plug it into their own program guides. But if the programmers don't supply it, and the cable operator doesn't insert it, then it won't be there.

Then there is a PSIP-related dispute with the broadcasters, dealing with channel numbering. PSIP establishes the concept of virtual channels when multiple SD programs are carried in a single 6-MHz channel, and requires a broadcaster to use two-part channel numbering to designate them. Digital cable systems also have virtual channels, but cable operators assign a one-part number between 1 and 999 to each virtual channel. The out-of-band channel provides the map between the assigned channel number and the actual RF frequency where the virtual channel is carried. So the whole must-carry fight over channel positioning–whereby a broadcaster can request that his channel is carried on cable on the same channel number as off-air–becomes very murky, partly because cable systems don't support two-part channel numbers, and partly because the broadcaster can designate the first part of the two-part channel number to be the same as his analog channel.

Now, consider a digital TV receiver that does not have access to the out-of-band channel and tries to tune a digital broadcast channel. The first part of the two-part channel number carried within the broadcaster's PSIP data is likely to be wrong when that station is carried on cable, because under must-carry rules, the cable operator need not position the digital channel on its off-air channel number. Moreover, the cable operator can remultiplex several virtual channels from different broadcasters so they are carried in the same 6-MHz channel. Digital TV sets must be smart enough to deal with this.

So the channel number displayed on the TV will probably be different than the one that appears in the newspaper's program guide. At the headend, the cable operator can correct the broadcaster's PSIP data to show the same channel number as the program guide. But broadcasters don't like the idea of cable operators mucking around with their data.

These disputes don't affect a cable operator's ability to carry digital broadcast stations, but they do give broadcasters and TV set manufacturers an additional opportunity to blame the cable industry for the slow rollout of HDTV. So what else is new?

Advertisement
Advertisement