Editor’s note: This article is the first in a multipart series of reports examining proposals to recoup spectrum discussed in the FCC’s “Spectrum analysis: Options for Broadcast Spectrum OBI Technical Paper No. 3.”
An FCC analysis of TV spectrum released last month shows that — at least in theory — the agency could, from a technical point of view, meet its goal of recouping 120MHz of spectrum if it were to allow competing broadcasters to share a single 6MHz channel — even if both channels transmit HDTV signals.
The concept, one of many approaches aimed at freeing spectrum to meet the commission goals outlined in its National Broadband Plan for wireless Internet service, is detailed in the agency’s “Spectrum analysis: Options for Broadcast Spectrum OBI (Omnibus Broadband Initiative) Technical Paper No. 3.”
The analysis begins with an illustration of the dramatic spectrum savings potential of channel sharing with a discussion of six stations sharing one 6MHz channel to deliver their primary OTA signal in SD. The main disadvantage of this approach is image quality because with six broadcasters transmitting a 3Mb/s to 4Mb/s main SD channel, there would be no bits left for HD transmission. Not only would that cut out HD reception by an estimated 2.75 million OTA viewers who watch in high-def, but also it would mean broadcasters taking on the expense of a fiber-optic or microwave link to MVPD headends to deliver their HD signal. Ultimately, the analysis rejected the 6:1 channel sharing approach, but added that “up to six stations that do not broadcast in HD may choose to share a channel in a given market following a voluntary, incentive auction.”
However, the technical paper laid out the case for the viability of a 2:1 channel sharing strategy in which two channels’ HD streams occupy a single 6MHz channel. The authors based their conclusion that the strategy is viable on three factors: First, multicasting of two HD streams is happening today; second, advances in technology “promise to make broadcasting multiple HD streams more likely in the future;” and third, the influence of market factors — above technical factors — in determining the quality of an HD signal.
The authors of the paper pointed to two examples of ongoing 2:1 HD channel sharing: 10 ABC-owned stations transmitting their primary channel in HD and the Live Well HD Network launched in 2009 — both in 720p and WBOC in Salisbury, MD, which at the time the paper was written planned to begin transmitting a CBS affiliate in 1080i and a FOX affiliate in 720p on the same channel sometime in the second quarter of 2010. While FCC rules currently prohibit channel sharing as envisioned by the 2:1 approach, these examples of multicasting HD streams demonstrate the viability of the approach, it said.
According to the paper, technological advancements will make the possibility of broadcasting multiple HD streams in the same 6MHz channel more likely in the future. Specifically, progress in statistical multiplexing, up to 15 percent or greater improvement in the efficiency of MPEG-2 encoders and the possibility of a gradual transition to MPEG-4 AVC compression as Internet-enabled HDTVs enter the market jointly contribute to future viability of HD channel sharing, it said.
While there is a relationship between bitrate and image quality, ultimately it is market factors more than technical factors that determine HD signal quality, the paper said. To illustrate this point, the authors drew on a sampling of the bitrates devoted to the various multicast channels transmitted by the nine stations in the Washington, D.C., area. Taken over nine-minute periods every 30 seconds, the sample showed, as would be expected, that area broadcasters transmitting multiple channels devoted fewer bits to each. For example, the paper said, WTTG transmits only one HD channel with no multicast channels and thus devotes 14.7Mb/s to the channel, while public broadcaster WETA transmits three SD multicast channels and one HD stream at 8.3Mb/s.
The paper acknowledged the dynamic nature of assigning bits to multiple video streams based on the nature of the content being broadcast. “Specifically, underlying movement in the broadcast image generates higher bitrates,” it said. Two tactics —employing dynamic stat muxing “to borrow” bit capacity from one stream to devote to demanding sequences in another and strategic programming decisions that pair bitrate-intensive programming on one channel with less bitrate-demanding programming on another — help broadcasters deliver the quality of viewing experience they desire.
To determine if such strategic programming decisions could be used in the real world between competing broadcasters sharing a single 6MHz channel, the authors of the paper analyzed the programming of the nine stations in the Washington market. They examined the HD and SD main and multicast transmissions of each in 30-minute blocks for the period of one week (336 blocks). They further categorized the HD segments as “high-movement” and “low-movement.”
The analysis suggested “that these stations could potentially pair in ways that minimize the coincidence of high-movement HD programming.” In the week studied, “there were no instances of coincident high-movement HD programming among the hypothetical channel sharing partners,” it said.
The authors made no assertion that channel sharing is “the right approach” to use for every TV station, “but rather that it could be a valuable mechanism to enable voluntary, market-based decisions” by some broadcasters, the paper said.
