ISDB-S - Satellite Transmission
System for Advanced Multimedia Services Provided by Integrated Services
Digital Broadcasting |
(Courtesy Asia-Pacific Broadcasting Union.) |
Tomohiro Saito, Akinori Hashimoto, Fumiaki Minematsu,
Toshihiro Nomoto and Hajime Matsumura
NHK (Japan Broadcasting Corporation)
Integrated Services Digital Broadcasting (ISDB) is an emerging
digital broadcasting concept. With ISDB, everything is handled digitally.
The three kinds of systems, ISDB-S (Satellite), ISDB-T (Terrestrial),
and ISDB-C (Cable), were developed in Japan to provide flexibility,
expandability, and commonality for the multimedia broadcasting services
using each network.
Since digital satellite broadcasting began service using ISDB-S
in December of last year in Japan, we would like to introduce the
three ISDB systems mentioned above. The ISDB-S system is explained
in this article, the first in a series of three. This article is
a reprint from ABU Technical Review No.189. |
A new digital satellite broadcasting service started in Japan on December
1 2000. The technical standard for the transmission system was based on
ISDB (Integrated Services Digital Broadcasting), to broadcast "any
information accessible at any time" and provide not only video and
audio but also new multimedia services by flexibly integrating the information
to be broadcast. The system can handle multiple MPEG2-TSs
(transport streams) by employing a frame structure. Broadcasters have
their own MPEG2-TS in which they can integrate their services independently
of each other even when they share a satellite channel. It also provides
a large capacity sufficient to transmit two HDTV programs on one satellite
channel, robustness against heavy rainfall attenuation, and high operational
flexibility. One key feature of the system is hierarchical modulation,
in which multiple modulation schemes are applied simultaneously. An information
bit rate of 52 Mbps in one satellite channel is obtained by applying TC8PSK
(Trellis-Coded eight PSK) and a low C/N reception at 2 dB is achieved
by applying BPSK. The new ITU-R Recommendation which describes the ISDB-S
transmission system was approved as BO.1408 in 1999.
1. Present satellite broadcasting
in Japan |
|
|
Figure 1: Outline plan for BS digital broadcasting
|
Japan has assigned 12 channels for DBS (Direct Broadcasting Service) in
the 12 GHz BSS (Broadcasting Satellite Service) band. Eight channels were
assigned by the WARC-77(BS)Plan and an additional four channels by the
WRC-2000 Plan.
At present, three analog NTSC channels and one analog MUSE-HDTV channel
are in service. The number of households receiving analog satellite broadcasting,
including MUSE-HDTV, is more than 14 million (nearly one third of all
households in Japan).
A new digital satellite broadcasting service called BS Digital Broadcasting
started on December 1, 2000. Currently, BS digital broadcasting provides
various services on four channels. One of the channels is used for digital
simulcast with analog services; that is, the programs broadcast on the
analog channels are digitally coded, multiplexed, and simultaneously broadcast
on this channel in order to promote the smooth transition from analog
to digital. The other three channels are used for digital broadcasting,
giving a central role to digital HDTV, where each channel is shared by
two or more broadcasters. "10 million households in the first one thousand
days" is the target penetration rate. The MPT (Ministry of Post and Telecommunications),
the broadcasting industry and the electrical appliance industry are all
committed to achieving this goal.
2. Features of the transmission
system for BS digital broadcasting |
|
The transmission system for BS digital broadcasting is based on ISDB
(Integrated Services Digital Broadcasting) and is called satellite ISDB
(ISDB-S). ISDB is a new type of broadcasting for multimedia services.
The features of the ISDB-S system are summarized below and listed in
Table 1.
Table 1: Summary of system characteristics |
Moduclation
scheme |
TC8PSK/QPSK/BPSK |
Raised cosine roll-off factor
|
0.35(square
root) |
Transmission symbol rate
|
28.86
Mbaud |
Video coding
|
MPEG-2 |
MP@HL
for 1080i,720p |
MP@ML
for 480i |
MP@H14
for 480p |
Audio coding
|
MPEG-2
AAC |
FEC(Outer code)
|
Reed-Solomon(204,188)
|
FEC(Inner code)
|
Convolutional(constraint
length k=7) |
Inner code ratio
|
1/2
for BPSK |
1/2,2/3,3/4,5/6,7/8
for QPSK |
2/3
for TC8PSK |
Transport Layer
|
MPEG-2
systems |
Packet size
|
188
bytes |
Handling multiple MPEG-TSs
The system systematically integrates various kinds of digital contents,
each of which may include multi-program video from LDTV to HDTV, multi-program
audio, graphics, texts, and so on. Most of the digital contents are nowadays
encoded in the form of MPEG Transport Stream (MPEG-TS). The system therefore
has to cover a wide range of requirements that may differ from one service
to another.
The system can handle multiple MPEG-TSs (up to eight
TSs) to ensure the independence of broadcasters.
This may be regarded in terms of the number of logical channels corresponding
to that of broadcasters, although the physical satellite channels are
limited.
Large transmission capacity
The system employs hierarchical modulation schemes including TC8PSK, QPSK,
and BPSK. The maximum information bit rate of about 52 Mbps, which is
sufficient for transmitting two HDTV programs, is obtained in one satellite
channel by applying TC8PSK. The occupied bandwidth (99% energy bandwidth)
is enlarged to 34.5 MHz compared with 27 MHz of the conventional analog
system.
High service availability
The system is robust against rainfall attenuation. For example, BPSK,
with a coding rate of 1/2, achieves the required C/N of 2 dB.
Operational flexibility
Multiple modulation schemes can be used simultaneously to meet the requirements
of several services integrated on one satellite channel. For example,
audio, downloaded data, etc. can be transmitted by BPSK, which is robust
against rainfall attenuation, while video data can be transmitted by TC8PSK,
which provides the maximum transmission capacity. The information on the
modulation scheme is transmitted by a TMCC signal (Transmission and Multiplexing
Configuration Control) [2].
3. Video & audio coding
and multiplexing |
|
For source coding and multiplexing, the system employs MPEG2.
3.1 Video coding system
The system employs MPEG2 MP@HL for 1080i and 720p, MP@H14 for 480p,
and MP@ML for 480i, respectively. The required information bit rates
of video signals are listed in Table 2. For HDTV (1080i) and SDTV (480i),
more than 22 Mbps and more than 8 Mbps are needed, respectively. These
data rates are derived from picture quality tests conducted by ARIB
(Association of Radio Industries and Businesses) in Japan.
Table 2: Evaluated picture quality evaluation
test |
|
3.2 Audio coding system
We employed MPEG-2 AAC (Advanced Audio Coding) for the audio coding
system. This coding technique is suitable for CD quality multi-channel
audio transmission. The compression rate of MPEG-2 AAC is twice that of
MPEG2-BC (Backward Compatible), which has been widely employed for conventional
satellite digital broadcasting. CD quality stereo audio broadcasting can
therefore be transmitted at only about 128 kbps.
3.3 Video and audio multiplexing
The system employs MPEG2 systems for data multiplexing, in which different
kinds of information, such as audio, video, and data, are multiplexed.
Each broadcaster sends its signals in the form of MPEG2-TS, and they are
combined at an uplink station. Therefore, each broadcaster's MPEG2-TSs
are kept unchanged in order to guarantee flexible service integration
for each broadcaster sharing the satellite channel.
|
Figure 2: Block diagram of channel coding |
4.1 Outline of channel coding
Figure 2 is a block diagram of the channel coding. The system handles
three kinds of signals in order to transmit multiple MPEG-TSs
with various modulation schemes and achieve stable and easy reception.
The three signals are explained below:
- |
Main signal, which consists of multiple MPEG-TSs and carries the
program contents, |
- |
TMCC signal, which informs the receiver of the modulation schemes
applied, the identity of MPEG-TSs, etc. |
- |
Burst signal, which ensures stable carrier recovery at the receiver
under any reception condition (especially under low-CNR conditions) |
The input MPEG-TSs are combined into a single stream,
to which ordinary signal processes for satellite systems are applied (i.e.,
energy dispersal, interleaving, and inner coding). The control data, which
designate the modulation schemes, etc. for each TS packet, are encoded
into the TMCC signal, to which a series of channel coding is applied.
In most cases, some parts of the channel coding can be shared with the
main signal. The burst signal is energy-dispersed to avoid line spectra
in the transmission signal.
4.2 Frame Structure
To handle multiple MPEG-TSs and allow several modulation
schemes to be used simultaneously, a frame structure is introduced to
the main signal.
To combine the TSs, the error-protected 204-bytes
packets are assigned to the "slots" in a "data frame", as shown in Figure
3. The slot indicates the absolute position in the data frame and is used
as the unit that designates the modulation scheme and MPEG-TS identification.
The size of the slot (the number of bytes in the slot) is 204 bytes to
keep one-to-one correspondence between the slots and error-protected packets.
The data frame is composed of 48 slots.
|
Figure 3: Frame structure |
A modulation scheme can be assigned to each slot independently. Up to
four modulation schemes can be assigned in a frame.
Data in different slots may be mixed by intra-frame or convolutional interleaving.
It is technically inappropriate to interleave data to be modulated by
different schemes. A super-frame structure is employed in order to avoid
these inadequacies, as shown in Figure 4. That is, every frame within
the super-frame has an identical configuration and data are byte-interleaved
in the super-frame (inter-frame) direction.
|
Figure 4: Super-frame structure |
4.3 TMCC
TMCC carries the information of the modulation schemes and the MPEG2-TS
ID, which is assigned to the slots etc. The system can deal flexibly with
various requests on these transmission schemes by means of TMCC signals
[3]. The configuration for the receivers is automatically changed to demodulate
the appropriate signal format. The TMCC consists of 384-bit fixed-length
data.
Modulation scheme for transmission signal
The TMCC and burst signal for carrier recovery are provided by the fixed
modulation scheme (i.e., BPSK), but the modulation scheme for the transmission
signal can be selected arbitrarily, at least for a single slot. The information
on the modulation scheme of the transmission signal is provided to the
receiver by the TMCC signal.
Identification of MPEG-TS
When more than one MPEG2-TS is transmitted by one satellite transponder,
the information for the identification of each MPEG2-TS is provided by
the TMCC signal.
Control of transmitter and receiver
Information on whether the broadcaster is using special functions of the
transmission schemes (e.g., emergency alert broadcasting) is provided
to the receiver by the TMCC signal.
Frame synchronization
The frame synchronization and super-frame synchronization words are
contained in the TMCC signal.
The TMCC signal requires higher reliability than the main signal because
the TMCC signal is used to control the demodulation process at the home
receiver. To secure reliable transmission, the TMCC signal is outer coded
by RS(64,48). Finally, the TMCC is modulated by convolutional BPSK (r
= 1/2). These error correction techniques allow the TMCC signal to be
received until the reception C/N is around 0 dB.
4.4 Burst signal for carrier recovery
A burst signal is inserted to secure steady carrier recovery at the
receiver even in low C/N conditions. The burst signal is energy-dispersed
by BPSK modulation with a pseudo-random sequence. It is possible to carry
some information by modulating the burst with an information signal instead
of the pseudo-random sequence (although this is not included in the current
standard).
Figure 5 outlines the transmission signal.
|
Figure 5: Outline of the transmission signal |
4.5 Transmission capacity
The time for transmitting the amount of bits in a frame (48 slots) differs
by modulation scheme because of the different spectrum efficiencies of
the schemes. When QPSK (r=1/2) is used, for example, the time is twice
that of TC8PSK.
To keep the duration constant, we introduce dummy slots when modulation
schemes other than TC8PSK are assigned to some slots (Figure 6). Since
the bits in dummy slots are never transmitted, the actual number of bits
transmitted in a frame is reduced to the number of bits that can be transmitted
in a given frame duration by combining the designated modulation schemes.
|
Figure 6: Example of slot assignment |
When a single modulation scheme is applied throughout a frame, the transmissible
bit rate is given as,
|
(1) |
where TB: transmissible bit rate
SR: symbol rate (= 28.86 Mbaud)
SE: spectrum efficiency (3 for TC8PSK, 2 for QPSK, 1 for BPSK)
CR: coding rate (2/3 for TC8PSK, 1/2, 2/3,3/4, 5/6 and 7/8 for QPSK, 1/2
for BPSK)
The transmission signal contains parity bits for RS (204, 188) and the
burst signal for carrier recovery. 188/208 denotes the ratio of the time
required for transmitting the information bits by MPEG2-TS to the frame
duration. When several modulation schemes co-exist in a frame, the cumulative
transmissible bit rate is given as:
|
(2) |
where TBi denotes the transmissible bit rate for the i-th modulation scheme
as given by Eq. (1).
Ni is the number of slots, including dummy slots using the i-th modulation
scheme.
Note that, the value of Ni takes the number listed in Table 3 to keep
the frame duration constant.
Table 3: The number of slots (including dummy
slots) used for several modulation schemes |
|
The relation between large transmission capacity and high service availability
is a trade-off under the limits of the satellite emission power and channel
bandwidth. The transmission capacity and service availability therefore
need to be selected according to the service contents and the intentions
of the broadcaster.
In the digital broadcasting system, various modulation schemes - as shown
in Table 1 - can be employed. From among these schemes, broadcasters can
choose the most suitable modulation scheme for their services.
|
Figure 7: Service availability of available
modulation scheme |
Figure 7 summarizes the service availability of the system.
Here, the required C/N is evaluated with a BER below 10-11
as a quasi-error-free condition. To achieve this condition, it is theoretically
necessary to have 2x10-4 at the input to the RS decoder after sufficient
interleaving. For calculating the relationship, we assume reception with
a 45-cm-diameter satellite dish and a satellite EIRP (Equivalent Isotropic
Radiated Power) of 59 dBW.
|
Figure 8: Payload bit rate of available modulation
schemes |
Figure 8 shows the relationship between the required C/Ns and the payload
bit rates.
Conventional FM-NTSC, which transmits audio signals with a digital subcarrier,
begins to show impulsive noise on the screen at a C/N of around 9 dB,
and the audio signal collapses at a C/N of 6 dB. Viewers are expected
to continue to receive the program until the audio signal collapses, so
the C/N of 6 dB is regarded as the threshold of a conventional analog
system.
TC8PSK can obtain higher spectral efficiency than QPSK and BPSK, so it
is appropriate for providing two HDTV programs. However, the service availability
is insufficient in terms of continuity from a conventional analog system
to the new digital systems. On the other hand, BPSK and QPSK with a coding
efficiency of 1/2 can provide higher service availability than the conventional
analog system, (although the transmission bit rate is restricted [3]).
A new digital broadcasting service started in Japan in December 1, 2000.
The technical standard (called ISDB-S) for the transmission system was
established and ITU-R approved as Recommendation BO.1408 in 1999 [4].
The system uses the MPEG2 standard for video and audio coding and MPEG2
systems for multiplexing programs. The most appropriate transmission scheme
for each slot can be selected according to the broadcaster's requirements
or program content. The system therefore has large transmission capacity,
strong robustness against heavy rainfall attenuation, and high operational
flexibility.
The configuration of the receiver changes automatically in order to demodulate
the appropriate signal format, using the information of transmission scheme
derived from the TMCC signal.
Low-cost receivers have already been put in the market by many manufacturers
in Japan, so consumers can receive a variety of services inexpensively
via broadcasting satellite.
References
[1] |
K.Ohsaki, T.Kimura, N.Kawai : "Transmission Structure of Digital
Broadcasting", 1994, IEEE International Conference Communications
326.2(May,1994) |
[2] |
H.Katoh, A.Hashimoto, H.Matsumura, S.Yamazaki, O.Yamada : "A Flexible
Transmission Technique for the Satellite ISDB System", IEEE Trans.
on Broadcasting Vol.42, No.3, September 1996 |
[3] |
H.Katoh : "Digital Modulation Scheme for ISDB (Integrated Services
Digital Broadcasting) in the 12 GHz Band", Proc. AIAA 15th ICSSC
AIAA-94-1079, February 1994 |
[4] |
ITU-R Recommendation BO.1408 (1999) |
|