Mobile television (Part 2): Free, and location-free

Part 1 of this two-part series focused mostly on transmission protocols amenable—for technical reasons, business reasons, or both—to cellular handsets, including DVB-H (digital video broadcasting handheld), EVDO (evolution data optimized), UMTS (universal mobile telecommunications system), and MediaFlo (Reference 1). However, as Japan's 1seg and South Korea's T-DMB (terrestrial-digital-multimedia-broadcasting) networks suggest, it's possible to graft support for conventional digital-television broadcasts onto a mobile phone if the economics justify the surgery. And, as the earlier article points out, it's not a foregone conclusion that the cellular handset will be the predominant means by which viewers tune into mobile TV (see sidebar “More ahead”).

Any secondary application that drains the battery and precludes subsequent access to the unit's primary application—making and taking telephone calls, for example—will likely receive a cool reception by consumers, especially if it's also one that considerably adds to the unit's price tag. Mobile TV has both qualities. A significant disparity between all-in-one widget aspirations and feasible reality opens the door to system alternatives with more focused functions and to a lengthier list of alternative infrastructure approaches.

Before last month's 3GSM (Third Generation Groupe Spéciale Mobile) Congress conference in Barcelona, Spain, Michael Rayfield, handheld-business-unit general manager at Nvidia, agreed—at least from a technical standpoint—with what the material in Reference 1 points out: that generational improvements in DVB-T's (digital-video-broadcasting-terrestrial) power-consumption characteristics might obviate the need for a late-arriving DVB-H descendant. However, from a business perspective, subscription-supported cellular carriers control DVB-H spectrum, whereas DVB-T signals come from advertising-revenue-based local and nationwide television broadcasters.

According to Rayfield, no cellular-service provider and, therefore, no hardware partner will embed handset support for a feature unless, when a consumer uses the feature, it translates into tangible revenue for the carrier. “The last time the cellular-service providers added a low- to no-revenue feature was the camera phone.” Rayfield also pointed out that cellular phones don't tune in AM- or FM-radio transmissions. “And they didn't add the camera feature out of charity; they thought it'd be highly profitable, although it didn't turn out that way,” he said (Reference 2).

The cellular-versus-broadcast-service tug of war that Rayfield suggests is occurring in Europe is also likely to exist in any country in which revenue- and profit-fueled open-market conditions determine business success or failure—that is, when a government or another organization doesn't heavily subsidize the service. The struggle is occurring in the United States, where over-the-air-TV broadcasters have over time increasingly seen their direct-access audience shrink; consumers are instead indirectly viewing broadcasters' content using cable, IPTV (Internet Protocol television), satellite, and other service intermediaries. In such a scenario, lucrative advertising revenue flows directly to the intermediaries, not to the broadcasters, and content-licensing fees don't make up the shortfall.

Now, another set of potential delivery intermediaries, the cellular-service providers, is entering the picture. Several US broadcast-network and local-affiliate representatives paint the mobile-phone carriers as competitors, though none of them will go on record as saying so. This reluctance is no surprise, because the carriers are also these networks' and affiliates' partners, and a public statement critical of that relationship might adversely affect the licensing-fee-revenue flow. Despite that reluctance, the networks and affiliates do worry about the long-term fiscal impact of having only a supporting role in the mobile-TV market of the future. That concern explains the vigorous, ongoing industry development of the ATSC-M/H (Advanced Television Systems Committee-Mobile/Handset) specification.

The developers of 8-VSB (eight-level-vestigial-sideband)-based ATSC designed it with exclusively stationary reception in mind, unlike its COFDM (coded-orthogonal-frequency-division-multiplexing)-based peers. As a result, ATSC is prone to environmentally induced factors, such as multipath-signal distortion, which causes egregious mobile-TV reception. I was witness to an example of this bad reception during a Samsung-hosted bus ride in Las Vegas at last year's NAB (National Association of Broadcasters) Conference. Sinclair-owned TV station KVMY was broadcasting ATSC-compliant images, which in-vehicle electronics received and displayed on an LCD on the rear wall of the bus. Whenever the bus was moving, artifacts corrupted the images, and, when the bus speeded up, the images disappeared altogether, despite the bus-based receiver's cognizance of the SRS (supplementary-reference-sequence) bit stream and the use of multiple diversity-reception antennas.

Two LCDs next to that one, along with Samsung-supplied UMPC (ultramobile-personal-computer) handheld systems, showed no artifacts, no matter how fast the bus was moving. Samsung based them on A-VSB (advanced-VSB) technology, which Samsung and Rohde & Schwarz developed. A-VSB redirects a portion of the 19.2-Mbps ATSC digital-broadcast stream away from its traditional functions—carrying Dolby Digital audio and MPEG-2 video information—instead allocating it for two mobile purposes. One purpose is a 1-Mbps SRS bit stream to assist a receiver in remaining “locked” onto the broadcast signal while in the presence of dynamic interference. The other is a variable-bit-rate turbo code for EDAC (error detection and correction) and error concealment within the audio and video streams.

Samsung privately showed A-VSB at the 2006 NAB Conference, and first public exhibitions occurred at the January 2007 CES (Consumer Electronics Show). At NAB 2007, the company simultaneously demonstrated both quarter- and half-rate turbo-coding techniques. Quarter-rate mode targets mobile-TV reception in bullet trains moving at ultrahigh speeds. The NAB demo required 1.5 Mbps of EDAC bandwidth, and Samsung bundled it with a 0.5-Mbps H.264 audio-plus-video special-content stream of 320×240-pixel QVGA (quarter-video-graphics-array) resolution. The demonstrated 19.2-Mbps “wrapper” also contained a half-rate rebroadcast of the primary ATSC content for reception in vehicles moving at highway speeds. That broadcast comprised 1-Mbps of turbo code and 1 Mbps of H.264-encoded audio/video data.

Subtracting the SRS stream and two H.264-plus-turbo-code streams, all of which a traditional ATSC receiver ignores, left approximately 15 Mbps of conventional terrestrial-digital-broadcast data. Samsung officials believe that this reduced base bit rate is still sufficient for delivering high-quality audio and video, considering the mature state of Dolby Digital and MPEG-2 encoder technology. Even more bandwidth is available for the basic ATSC stream if an application requires only one mobile-tuned turbo-code-plus-content stream.

Samsung not only demonstrated an A-VSB-enhanced signal coming from Sinclair's broadcast antenna on Mount Potosi, 25 miles southwest of Las Vegas, but also partnered with Ion Media Networks to create a SFN (signal-frequency-network) rebroadcast of Sinclair's material with low-power transmitters on the Stratosphere and Paris hotels and at the Las Vegas Convention Center acting as the source (Figure 1). According to Samsung's documentation, without A-VSB-enabled synchronization, echoes would overwhelm a coordination of the three transmitters' signals. Also according to Samsung's literature, SFN “enables broadcasters to cover a service area with uniform signal strength, even in hilly or built-up terrain, improving service quality.”

The April 2007 NAB also featured initial public demonstrations of MPH (mobile-pedestrian-handheld) technology, a competitive ATSC-enhancement proposal, which partners LG Electronics and Harris champion. Both A-VSB and MPH, along with a rumored third approach that Micronas and Thomson are developing, are vying for the lucrative implementation-based royalty revenues from the developing ATSC-M/H specification. MPH's developers have publicly revealed relatively little about the technology. Press coverage at the technology's March 2007 unveiling noted, “According to Harris, the proposed format provides higher signal performance as well as the ability for broadcasters to add more of a payload to the mobile part of their spectrums than A-VSB, allowing the delivery of more channels to viewers and a more robust signal. At a recent press conference in New York, the company said that low-power testing of the format demonstrated performance at approximately a 7-dB greater signal threshold than A-VSB.” (Reference 3).

“The MPH physical layer offers a number of advantages [over A-VSB],” says Jay Andrick, vice president of broadcast technology at Harris, “including channel efficiency that scales across any number of MPH channels, full-time burst-mode transmission that translates to minimum receiver-battery drain, [and] absolute compliance and compatibility with ATSC A/110 Distributed Transmission Standard. And MPH receive devices do not rely on diversity antennas or tuner systems to achieve robust reception. Recent major-market field testing has proved that MPH offers extremely reliable service to the radio horizon using a variety of handheld receiver devices.”

Whether cellular carriers and their hardware-provider partners will embrace ATSC-M/H is unclear. Vendors are most likely to include the feature in “unlocked” phones and open-platform designs, such as OpenMoko's technology and Google's Android. Ever the optimists, however, the ATSC-M/H contenders claim that ATSC-enhanced broadcast TV complements rather than competes with the cellular providers' efforts in the mobile-TV arena. “A-VSB gives wireless-service providers the chance to free up crowded wireless-broadband spectrum by relying on TV spectrum as the most efficient way to deliver bandwidth-intensive broadcast video to mobile devices,” according to Samsung, “while OMA-BCAST [Open Mobile Alliance Mobile Broadcast Services Enabler Suite] applications delivered along with broadcast programs will entice consumers to use those same mobile devices to access interactive services carried over wireless broadband networks.”

A multiantenna SFN can compensate for terrain and other inconsistencies within a given desired reception area. However, the curvature of the earth, along with other variables, such as the broadcast signal's strength and frequency, limits the coverage of any terrestrial antenna-based system. To cover a large area, such as the United States, you need a single- or multiple-satellite-broadcast scheme, such as the one Sirius Satellite Radio uses. Sirius, in partnership with Chrysler, late last year rolled out its STMicroelectronics-powered Backseat TV add-on video service. Chrysler offers the service as an option on 2008 Dodge Grand Caravans, Chargers, and Magnums; Chrysler Town & Country and 300 models; and Jeep Commander and Grand Cherokees. The factory-installed SVC1 receiver costs $470, along with a $6.99/month increment beyond the basic Sirius Radio-service price (Figure 2). The aftermarket manufacturer's suggested retail price is $299.99, plus installation. Chrysler-installed units come with one year of service. Back-seat passengers can watch content from the Cartoon Network, Disney Channel and Nickelodeon, and front-seat occupants can simultaneously listen to Sirius Satellite Radio.

At the 2005 CES, Microsoft and Sirius announced that Backseat TV would employ Windows Media Video as its codec technology. However, according to a recent article, Sirius ended up using a proprietary variant of H.264, also known as MPEG-4 AVC and MPEG-4 Part 10 (Reference 4). A demonstration of Sirius Backseat TV at January's CES left me with an impression of lukewarm video quality. The company won't reveal the material's broadcast resolution, but I estimate that I was watching, at best, low-frame-rate QVGA content on the approximately 8-in. LCD.

In comparison with a conventional single-antenna Sirius Satellite Radio setup, a Backseat TV-enhanced installation employs dual antennas—one at the front of the vehicle and one at the rear—for reception-optimizing diversity selection. Enhanced EDAC algorithms also boost reception quality. Sirius' engineers squeezed three channels' worth of audio-plus-video material into the company's 12.5-MHz S-band spectrum on top of the existing 134 channels' worth of audio-only content. Pixel artifacts aside, the fact that they were able to meet this video goal without noticeably degrading the quality of the audio content represents an impressive technical achievement.

The commercial success of Backseat TV isn't a foregone conclusion, however. Considering the diversity of alternative video-entertainment sources, such as built-in and portable DVD players, video iPods, and the like, customers might resist paying for premium equipment and service in premium-vehicle variants just to get Sirius. And what of handheld Sirius video receivers, akin to today's S50 and Stiletto series of satellite-radio-only portable units? “Sirius has never talked about portable video, as in handheld-device video,” says Patrick Reilly, senior vice president of communications at Sirius. “But XM [Satellite Radio] has. There are clips on it.” Note that Sirius is pursuing a merger with XM. Perhaps, as is the case with today's portable satellite-radio receivers, the option of playing back captured material that you recorded with an ac-tethered device will mitigate power-consumption concerns. So stay tuned.