- Overview
- Broadcast Model
- Interactivity
- Data Delivery
- Authoring Content
- Packaging Content
- References
This chapter is from the book
This chapter is from the book
This chapter is from the book
3.2 Broadcast Model
3.2.1 MPEG Transport
MPEG transports are unbounded sequences of bits (i.e., bit-streams) grouped into 188-byte packets (see Figure 3.11). These bits are extracted from other layers including the physical layer modulation, such as the QAM and the QPSK.
)
Figure 3.11. Example packet sequences for two 8 Mbps channels carried in a 19.2 Mbps transport.
Each 188-byte packet is associated with a Packet ID (PID). This is not a unique identifier of the packet, but rather an identifier of the data stream the packet is part of. Generic MPEG equipment enables management at the PID level, namely is able to identify components of the transport by the PID values within the 188-byte packet headers. More specialized equipment may be able to manipulate the data within the PID, namely, manipulate the data carried within the collection of packets having a common PID.
In addition to transmission of content, packets containing MPEG Program Specific Information (PSI) must be transmitted. PSI packets include those carrying the Program Association Table (PAT), Program Map Table (PMT), and other tables. The PAT contains a data structure specifying which PIDs will carry the PMT, and the PMT contains a data structure specifying which PIDs will carry the video, audio and data stream that collectively constitute the broadcast program. Standards built on top of MPEG may specify various constraints on the bandwidth and frequency of appearance associated with transmission of MPEG PSI.
Multiplexers decide which transport packets to insert and when. Multiplexers are responsible for ensuring that the average bandwidth allocated to the PSI and each program element as specified by the applicable standards and content characteristics.
3.2.2 Virtual Channels
Whereas with analog TV there were channels, each associated with a frequency range, with iTV (being based on MPEG-2 transports) that channel can be divided into sub-channels called virtual channels. A virtual channel is labeled by two numbers, its major and minor channel numbers. For example, instead of Channel 7, with iTV we have Channel 7.1, 7.2, etc., where the major channel number is the analog number, 7, and the minor numbers follow the '.'.
An MPEG-2 transport bitstream (carried over a single continuous frequency band) is divided by assigning each component to a distinct PID. Each virtual channel is associated with a set of PIDs using dedicated tables encapsulated in separate dedicated PIDs (see following subsection and Figure 3.12). This means that virtual channels may be interleaved. For example, it is possible to follow packets of Channel 7.1 with packets of Channel 7.2, which can then be followed by packets of Channel 7.1 (see Figure 3.11).
)
Figure 3.12. An abstract depiction of the logical structure of an iTV MPEG-2 transport.
The virtual channel binding also associates the major and minor numbers with additional meta-data, such as network or channel name and Transport Stream ID (TSID), usable by electronic program guides or other receiver applications. Transmission of a new version of the virtual channel binding table changes the programs associated with each channel, and effectively changes the content of that channel.
3.2.3 Program Specific Information
In addition to transmission of content, also transmitted are packets containing MPEG Program Specific Information (PSI). PSI packets include those carrying the Program Association Table (PAT), Program Map Table (PMT), and other tables. The PAT contains a data structure specifying which PIDs will carry the PMT, and the PMT contains a data structure specifying which PIDs will carry the video, audio, and data stream that collectively constitute the broadcast program. Standards built on top of MPEG may specify various constraints on the bandwidth and frequency of appearance associated with transmission of MPEG PSI.
All MPEG-2 PSI tables should be broadcast repeatedly. A typical transport structure would contain in addition to the PSI, such as the PAT and the PMT required by MPEG-2, a table defining the mapping of virtual channels carried in that transport, such as the ATSC Master Guide Table (MGT) in conjunction with the Virtual Channel Table (VCT) or its DVB variants (see Figure 3.12). The PAT lists the MPEG-2 programs carried on the transport, and associates them with number identities unique within the transport, pointing to a PMT for each program.
Multiplexers decide which transport packets to insert and when. Multiplexers are responsible for ensuring that the average bandwidth allocated to the PSI and each program element are as specified by the applicable standards and content characteristics.
Whereas transport packets are the basic transport encapsulation, Packetized Elementary Stream (PES) is the basic streaming structure depicted in Figure 3.12; transport packets have segments of PES packets in their payload. PES is used for carrying both Video and audio frames in the payload of the packet. However, PES is rarely used for data carriage; Digital Storage Media Command and Control (DSM-CC) ISO/IEC 13818-6 is used to carry data (see Section 3.4 and Chapter 11 Transport File System). With PES, synchronization information, as well as other meta-data critical for the encoding and rendering of content, is carried in the header of a PES packet. A PES packet can be up to 65,542 bytes long, and thus its carriage may require up to 349 transport packets (each 188 bytes long).
3.2.4 Audio Content
Sound is pressure differences in air. When picked up by a microphone and fed through an amplifier, this becomes voltage levels. The voltage is sampled a number of times per second and converted into digital signals using analog-to-digital conversion. For CD quality audio there is a need to sample 44,100 times per second where each sample is a conversion of the analog pressure into a digital word with a resolution of 16 bits. When recording in stereo mode, this requires 1.4 Mbit per second, raising the need for compression. Various audio compression techniques (including AC-3 and MPEG AAF) are described in Chapter 9. When delivered using MPEG transports, audio data is broken into frames. The data bits representing the compressed audio frames are transmitted in the payload section of PES packets. It is synchronized utilizing a portion of the frame header called frame sync. See Chapter 9 for more details.
3.2.5 Video Content
Video consists of dynamically changing light patterns hitting a sensor. When picked up by a video camera, these patterns become voltage levels associated with location and time. The camera samples voltage levels to convert them into digital signals by dividing the camera viewport into a matrix of regions, called pixels, and recording light color and intensity for each pixel. Table 3.1 lists some video formats, their resolutions, and number of frames per second. For HDTV quality, a matrix of 1080 (DVD is 480) interlaced rows of 1920 pixels needs to be sampled 60 times per second (two fields of 540 lines, each sampled 30 times per second), where each sampled pixel contains 24 bits. Without appropriate compression, processing such data is not practical as it amounts to 540 × 1920 × 24 × 60 bits per second! State of the art sophisticated compression techniques are able to deliver HDTV video for less than 19.2 Mbps, a common DTV broadcast bandwidth.
Table 3.1. Summary of Video Format Standard
|
Standard |
Resolution |
Frames Per Second |
|---|---|---|
|
Film |
24/23.98 |
|
|
Analog NTSC |
720/704/640 × 480/486/512 |
29.97/30/59.94/60 |
|
Analog PAL |
720/704/640 × 576/612 |
50/25 |
|
Analog ITU-R BT.601-4 |
864/858/720 × 625i/525i/483i |
50/30/25 |
|
HDTV SMPTE 260M |
1920 × 1035i (Interleave) |
30/29.97 |
|
HDTV SMPTE 296M |
1280 × 720P (Progressive) |
60/59.94 |
|
HDTV SMPTE 274M |
1920 × 1080i (Interleave) |
60/59.94/30/29.97/25/24/23.98 |
|
HDTV SMPTE 274M |
1920 × 1080P (Progressive) |
30/29.97/25/24 |
The bit stream representing the compressed video data is carried in PES packets which are converted into 188 byte transport packets and multiplexed into an MPEG-2 transport. Video data essentially encodes a sequence of frames, using an explicit or implicit representation of frames. The explicit representation of frames, called I-frames, is used to initialize a motion vector predictor. The two types of implicit frames, P-Frames and B-Frames, are processed by a motion predictor. Typically, an I-Frame is followed by one or two B-Frames, followed by a P-Frame, followed by one or two B-Frames, followed again by a P-Frame.
Synchronization between audio and video, however, depends on the performance of the video and audio decoders. Smooth performance and consistent synchronization should be maintained as long as the transmitting encoder and receiving decoder are both compliant with the MPEG-2 delivery contract called the Transport System Target Decoder (T-STD) buffer model. For details, see the MPEG-2 specification for video decoding.
3.2.6 Data Content
Data differs from audio and video in that it is often not associated with sampling rates, decoding, and delivery contracts. For example, the broadcasting of an iTV application composed of a single HTML file may be achieved by transport packets that are distant, long time periods apart, constrained only by overall bandwidth allocation and without a requirement to meet any MPEG-defined time constraints. Further, the format by which the data is placed in the transport, called encapsulation, may vary significantly (see Chapter 11).
In contrast to data, a video component of an MPEG-2 program is placed in a single elementary stream, using a single PID. Similarly, each audio track associated with that video component is usually placed in a single elementary stream using a single PID. Data is different as it may be more complex than both audio and video. A single application may be carried in several PIDs, and several applications may share PIDs (see Figure 3.12). Data may be carried using a broadcast file system (see Chapter 11). The unique characterization of data broadcasting, also known as datacasting, has numerous implications, including challenges related to multiplexing and re-multiplexing, transport, and receiver design issues.