The views and opinions expressed in this article are those of the author and do not necessarily reflect the position of the Society of Motion Picture and Television Engineers - SMPTE.

This paper was originally presented by Thomas Edwards at the 2013 HPA Tech Retreat, February 18 – 22.

The past few years have seen pioneering work in the development and delivery of digital television signals to mobile devices. Broadcasters throughout the U.S. are using the broadcast spectrum to provide content, including live local and national news, sports and entertainment programming, to a growing number of portable devices. In establishing these services and supporting systems, broadcasters faced a variety of challenges.  The following paper discusses the lessons learned through this process and provides insight that may inform future launches of mobile DTV services.

Mobile DTV Development

The original specification for DTV in the U.S. was A/53, developed in 1993. This standard determined that a large amount of data, 19.39 Mbits/sec, fit into the 6 MHz then required by a single analog channel. To make this approach work, the standard presumed a model that used a stationary receiver— an antenna 30 ft high with a 6- to 10-dB gain—to provide content to a large cathode ray tube display. Given these elements, there was no sense that mobile reception would be a priority, much less a possibility. Thus, the DTV training sequences included in the standard were appropriate only for stationary multipath signals. There was no concept of mobile reception on a handheld device or with a small antenna on the ground or in motion.

Because neither the planning factors nor the training factors within A/53 accounted for a moving antenna, the standard simply didn’t work for mobile DTV services. The A/153 ATSC mobile DTV standard was created to address these and other issues unique to mobile services. With this standard, mobile data within the MPEG transport stream packets included an unreferenced packet ID (PID), typically the value of 0x1FF6, that legacy receivers and PSI treat as null packets and effectively ignore.
These IP packets, encapsulated within the transport stream, contain a payload of H.264 video with HE AAC-encoded audio, sent in realtime protocol (RTP) via a multicast user-data protocol (UDP). The mobile stream also incudes forward error correction (FEC) data necessary for reception by a smaller antenna. (Encryption data, a key enabler of mobile service delivery models, also may be included.) While a range of ratios is possible, the FEC coding used for mobile services typically is four times that required for legacy signals. A/153 standard includes additional training sequences to support mobile reception by antennas moving at speeds as great as 200 mph.
Figure 1 illustrates mobile IP multicast signals as shown by an analyzer. Time and date, service signaling channel, and the electronic service guide (ESG) are followed by the IP multicast (highlighted) carrying the audio and video data.
Figure 1.
Mobile data is transmitted only in specific time slots, which has the benefit of requiring receivers to be powered up only during those times. This mobile data is transmitted in 16 time slots of 12.1 ms each, the equivalent amount of time needed to transmit 156 transport stream packets. The number of groups (NoG) is the number of slots used for that mobile data. Each NoG costs 917 kbps of legacy signal and, in turn, provides about 240 kbps of HQQQ-encoded mobile payload.
Typically, a NoG=1 scheme is adequate for a kids’ animated program cartoon delivered at a 416 x 240 resolution and encoded at a baseline H.264 profile. A NoG=2 scheme is appropriate for live video with the same resolution and encoding parameters, and a NoG=3 scheme boosts image quality for live video, delivered at 640 x 368 and encoded to an H.264 main profile. Some stations use a NoG=4 scheme, which requires 3.6 Mbps of data from the 19.39 Mbps legacy 8VSB signal and, as a result of the 4:1 coding ratio for HQQQ, provides 970 kbps of usable payload for the mobile services.
Figure 2 illustrates the cadence of mobile data PIDs within the transport stream. This measurement in time, expressed from left to right and from top to bottom, of the MPEG transport stream shows, with each square, a packet colored according to its PID value. The bursts of mobile PIDs in specific time slots are clearly represented in lime green. The rest of the packets represent PIDs in the legacy signal.
Figure 2.
A Working Mobile Model
The advance of mobile technologies has enabled broadcasters to establish an effective and reliable transmission chain for mobile services. Figure 3 illustrates the final design now implemented at FOX O&O stations. For this model, Larcan provided the H.264 encoder, mobile exciter, mobile mux, GPS receiver, electronic service guide, and Ethernet switch. Triveni Digital software performs transport-stream monitoring.
At the front end of the signal flow within the typical mobile DTV service model, the legacy HD transport stream is fed to a NAVE encoder. This step must occur prior to mobile multiplexing to ensure that mobile PIDs remain in the correct place. The transport stream from the NAVE encoder then is split, with one output gong to the H.264 encoder and the other to the mobile mux. The H.264-encoded version of the main HD signal goes out over IP via an Ethernet switch. An ESG system and NTP generator also are attached, along with a probe that connects back through the switch and an IPsec VPN to the CA and monitoring provider.
Once the other copy of the legacy transport stream is fed from the NAVE encoder to the mobile mux, its transport stream output is sent to the station-to-transmitter link and, at the transmitter site, to a mobile modulator for RF transmission. The signal is locked to a GPS, which drives the time of day signal and ensures that the time slots remain coherent.
Off-air recording for mobile DTV requires a unique approach, as mobile data packets have non-systematic RS encoding in the exciter that a legacy A/53 receiver can’t decode. Consequently, it is not yet possible to record and pass the mobile signal in the simple way a facility can record and pass A/53 transport stream. Professional-grade A/153 receivers eventually will come to market and address the issue of off-air recording. In the meantime, recordings must be made right out of the mux, not over the air.
Mobile data must always be in the right place at the right time, and this presents a further challenge. The mobile mux and mobile exciter must be synchronized, typically by a GPS, to assure that there is no addition or dropping of transport stream packets between the mux and transmitter. Even null packets must be preserved, and this can cause problems when an ASI-to-SMPTE 310 converter drops null packets in order to match the imprecise rate of ASI to the far more precise SMPTE rate. Products designed specifically to address this requirement in mobile service delivery have begun shipping.

Figure 3.

Making Mobile TV Work

The success of mobile services depends on the availability of services, content, devices capable of receiving that content, and the technology to make viewing possible. To create a mobile ecosystem including these elements, broadcasters within the Mobile Content Venture (MCV) established mobile services and dedicated spectrum to them, and they also developed a legal framework that facilitates the use of broadcast content, which is not necessarily free and clear for mobile use.
Working with partners in manufacturing, the MCV helped to drive the release of new devices that support the Dyle Mobile TV application (and employ the requisite CA system). Candidate devices for Dyle certification are being tested to assure a high and uniform level of quality. Developed by MobiTV, the Dyle Mobile TV application is available free to consumers and operates on both Android and iOS operating systems. It decrypts MCV stations and plays “in the clear” stations, such as PBS stations, as well.
The solid mobile TV ecosystem built by MCV broadcasters provides the foundation for the growth of mobile TV across the U.S. As more and more networks participate, bringing a greater diversity of content to mobile services, more viewers will tune in and mobile TV will become even more valuable to the broadcast industry and the consumers it serves.
The views and opinions expressed in this article are those of the author and do not necessarily reflect the position of the Society of Motion Picture and Television Engineers - SMPTE.

Copyright 2014 the Society of Motion Picture and Television Engineers, Inc. (SMPTE). All rights reserved.