Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02379017 2004-10-28
22386-2703
Method for Preparation of Source-Coded Data, and Transmitter
and Receiver Therefor
Prior Art
The invention is based on a method for
transmitter-end or receiver-end preparation of source-coded
audio data of at least one useful signal source, in
particular for transmission via AM channels of a
predetermined channel raster.
For the transmission of digital audio data,
particularly via AM channels of a predetermined channel
raster with channels that are 9 or 10 kHz in width
(mediumwave in America) in medium- and longwave formats, as
well as in shortwave format, three different transmission
systems were developed within the DRM consortium (Digital
Radio Mondiale). All of these systems use a conventional AM
channel for transmission. In the T2M method, the digital
information can be linked via an auxiliary carrier to the
low-frequency input of the transmitter and can be
transmitted parallel to the AM analog signal (Funkschau
[Radio Show] no. 14, 1998, pp 44 to 46). The Skywave 2000
method uses a multiple carrier method with TCM modulation
(Trellis Code Modulation) in connection with QAM (Conference
Paper of the 51st Broadcast Engineering Conference, NAB 97,
pp. 27 to 48, Progress Towards the Development of Digital
Modulation in the Longwave, Mediumwave, and Shortwave Bands;
IBE, Transmission Engineering, March 1999, pp. 53 and 54).
Advantages of the Invention
According to one broad aspect, the invention
provides a method for transmission-end preparation of
source-coded audio data of at least one useful signal
1
CA 02379017 2004-10-28
22386-2703
source, in particular for transmission via AM channels of a
predetermined channel raster, with the following features:
the source-coded audio data of at least one useful signal
source are separated into a main data stream and at least
one auxiliary data stream, where the main data stream
contains at least the amount of information that is required
for a comprehensible reproduction of at least one useful
signal source and the auxiliary data stream contains
information for quality improvement, the main- and auxiliary
data streams are modulated and accommodated in respective
different channels of the predetermined channel raster.
The invention also provides a method for receiver-
end preparation of audio data, which are contained in main-
and auxiliary data streams, in particular, for reception via
AM channels of a predetermined channel raster, where
mutually associated main- and auxiliary data streams each
originate from at least one useful signal source and the
mutually associated main- and auxiliary data stream are
accommodated in respective different channels of the
predetermined channel raster, including the following steps:
a receiver with a low reproduction quality is used to
demodulate and decode only the main data stream, a receiver
with higher reproduction quality is intentionally used to
demodulate and decode only the main data stream or the main
data stream and at least one associated auxiliary data
stream are demodulated and decoded, where mutually
associated demodulated and decoded data streams are linked
to one another in such a way that an increase is achieved in
the reproduction quality for the at least one useful data
source.
In accordance with a further aspect of the
invention, there is provided a transmitter for the
2
CA 02379017 2004-10-28
22386-2703
preparation of source-coded audio data from at least one
useful signal source, in particular for transmission via AM
channels of a predetermined channel raster, with the
following features: a separation device for separating the
audio data of a useful signal source into a main data stream
and at least one associated auxiliary data stream, a
modulation unit modulating the main- and auxiliary data
streams, where this modulation unit can in particular be
supplied with carrier signals in such a way that mutually
associated main- and auxiliary data streams can be
transmitted in respective different channels of a
predetermined channel raster.
There is also provided a receiver for receiver-end
preparation of source-coded audio data, which are
accommodated in main- and auxiliary data streams, in
particular for reception via AM channels of a predetermined
channel raster, with the following features: a demodulation
unit and decoding unit for at least main data streams, an
evaluation unit for signaling and optionally encoding
additional information, where the signaling indicates which
channel contains an auxiliary data stream associated with a
main data stream and the optionally provided additional
information indicate what information the auxiliary data
stream contains and how the main data stream is to be
combined with the at least one auxiliary data stream on the
receiver end, a linkage unit for mutually associated main-
and auxiliary data streams, which can be controlled by the
evaluation unit.
In some embodiments, it is possible to achieve an
increase in reproduction quality, for example an improved
tonal quality, without having to deviate from the
predetermined channel raster, as is required in the methods
2a
CA 02379017 2004-10-28
22386-2703
mentioned at the beginning or which can only be achieved by
these methods without quality loss by means of an expensive
encoding. With a method according to an embodiment of the
invention, with a coupling of a main data stream and at
least one auxiliary data stream in different channels of the
channel raster, the receiver-end useful data rate can be
increased and therefore a quality improvement can be
achieved in comparison with conventional methods. With a
method according to an embodiment of the invention, it is
possible, through the use of simpler receivers, to
demodulate and decode only the main data stream, which
results in a comprehensible reproduction with a low bit rate
of approximately 24 kilobit/s. Receivers with a high
reproduction quality demodulate and decode both the main
data stream and also at least one auxiliary data stream of a
useful signal and link these two data streams so that a
higher reproduction quality is achieved.
In DVB signals, there is also a separation into a
base layer and an enhancement layer. However, these layers
are transmitted on the same channel. By contrast to the
invention, with this system, a simple receiver must receive
the entire data stream and can only execute a separation
afterward.
A method according to an embodiment of the permits
the achievement of numerous combinations for increasing
reproduction quality, for example in order to reduce the
amount of coding artifacts, increase audio bandwidth, or
increase the impression of three-dimensional sound, e.g. the
transitions from mono to stereo.
2b
CA 02379017 2004-10-28
22386-2703
Drawings
Exemplary embodiments of the invention will be
explained in detail in conjunction with the drawings.
Fig. 1 shows the transmitter-end and receiver-end
preparation of audio data according to an embodiment of the
invention,
Fig. 2 shows the AM channels within a
predetermined channel raster,
Fig. 3 shows the transmitter-end and receiver-end
preparation of audio data according to an embodiment of the
invention, with a receiver for high-quality reproduction,
2c
CA 02379017 2002-01-10
Fig. 4 shows the preparation of a stereo signal,
Fig. 5 shows combinations for separating and combining audio data in a base
layer and
an enhancement layer.
Description of the Exemplary Embodiments
In the embodiment of the invention according to Fig. 1, for example PCM data
of
a useful signal source 1 are encoded on the transmitter end by means of a
source encoder
2. In this case, the encoded signal is separated into a main data stream HD
(base layer)
and at least one auxiliary data stream ZD (enhancement layer), i.e. the source
encoder 2
in this exemplary embodiment functions simultaneously as a separation device
for the
audio data of the useful signal source 1. The main- and auxiliary data stream
are
modulated by means of a modulation unit 3 and are accommodated in respective
different
channels that are spaced apart by 9 kHz, for example the adjacent channels Kl
and K2
shown in Fig. 2 of the predetermined channel raster, e.g. of the AM middle
channel
raster. For the transmission of the main- and auxiliary data stream into the
different
channels K1 and K2, the respective carrier signals for these channels are
supplied to the
modulation unit 3. Naturally, these channels do not have to be adjacent, as
shown in Fig.
2, but can be accommodated at any location in the predetermined channel
raster.
Channels that can be used for the auxiliary data ZD can include, for example,
channels
that are freed by parallel program broadcasting due to the increased range of
digital
modulation or channels that have been or will be created through band
expansion as a
result of channels no longer being needed by other services (coastal radio,
marine radio,
aeronautical radio), for example expansion of the AM mediumwave range in the
USA
between 1600 and 1660 kHz or the shortwave range in the 31, 25, and 19 meter
band.
The data streams transmitted via separate channels are demodulated and decoded
on the receiver end. In the exemplary embodiment according to Fig. 1, a base
receiver 4
3
CA 02379017 2004-10-28
22386-2703
is provided, i.e. a receiver with low reproduction quality,
which demodulates and decodes only the main data stream HD
by means of the demodulator 5 and source decoder 6. This
is possible because the main data stream according to the
invention contains at least the amount of information from a
useful signal source that is required for a comprehensible
reproduction of the useful signal source. For example, the
main data stream HD contains just enough information from
the useful signal source for the reproduction quality not to
differ from the reproduction quality in the AM channels in
mediumwave, longwave, and shortwave formats, i.e. acceptable
speech comprehension, but with quality losses in music
transmissions.
In the exemplary embodiment according to Fig. 3,
on the transmitter end, the same signal preparation occurs
as in the exemplary embodiment according to Fig. 1, but on
the receiver end, a receiver 7 with high reproduction
quality, for example CD quality, is provided, which
demodulates and decodes both the main data stream HD and the
associated auxiliary data stream ZD by means of the
demodulation unit 8 and the decoding unit 9. In a linking
device, the main data stream HD and the associated auxiliary
data stream ZD are linked to each other in order to achieve
a quality improvement of the audio signal received. In the
exemplary embodiment according to Fig. 3, the source
decoding unit 9 simultaneously functions as a linkage unit.
For the correct linkage of a mutually associated main- and
auxiliary data streams, a signaling is incorporated into the
main data stream HD (base layer) on the transmitter end,
which indicates whether an auxiliary data stream ZD
(enhancement layer) is provided for the same useful signal
source (program source) and at what frequency, i.e. in what
channel, it is provided. Preferably additional information
4
CA 02379017 2004-10-28
22386-2703
is incorporated into the auxiliary data stream, which
indicates what information the auxiliary data stream ZD
contains and optionally, how the main data stream HD is to
be combined with the at least one associated auxiliary data
stream ZD. In order to evaluate the signaling and/or the
additional information, an evaluation unit 10 is provided
that is preferably associated with the demodulation unit.
This evaluation unit 10 controls the linkage unit and the
source decoder 9 in accordance with the evaluated signals so
that the linkage of associated main- and auxiliary data
streams occurs synchronously with one another.
4a
CA 02379017 2002-01-10
A receiver with high reproduction quality can naturally also execute an
intentional
demodulation and decoding of only the main data stream PID and can thus be
operated as
a base receiver.
Examples will be given below for separating out audio data from a useful
signal
source and possible combinations of the main data stream (base layer) and
auxiliary data
stream(s) (enhancement layer(s)). In the channel K1 according to Fig. 2, for
example, the
complete mono audio signal of a program source (useful signal source) can be
contained
in the main data stream with a low bit rate; channel K2 can contain an
auxiliary data
stream ZD with all additionally required data for a stereo program, with a
possibly higher
bit rate. In principle, the separation into the two data streams can be
produced with the
scalability of MPEG 4. First generation receivers as well as simple favorable
receivers
are intended to demodulate one channel and decode a monophonic signal.
Receivers with
higher reproduction quality are provided for the demodulation of both channels
Kl and
K2 and for the decoding of a stereophonic signal. Consequently, this
represents a sensible
transition scenario from the use of one channel to the use of two channels.
With the
introduction of DRM, receivers can be developed which decode only the base
layer. After
the second channel is subsequently brought on line, these receivers can also
receive the
base layer along with the enhancement layer. In addition to stereo
reproduction, through
the linkage of the main data stream with the at least one auxiliary data
stream, a quality
improvement can be achieved in the following directions: the auxiliary data
stream
reduces the amount of coding artifacts, the auxiliary data stream broadens the
audio
bandwidth.
Naturally, any combination of these quality improvement steps can also be made
with the addition of stereo reproduction.
Examples of separations between the base- and enhancement layers in
mono/stereo encoding are achieved as follows:
5
CA 02379017 2004-10-28
22386-2703
Various methods for encoding a stereo signal are
provided in the MPEG 4 standard. Of these, the following
methods 2 and 3 are suitable for the method according to the
invention:
1. Encoding of the right (R) and left (L) channel.
2. MS stereo encoding: based on the original
signal, a sum signal (mid) and a difference signal (side)
are produced before it is quantified. This is shown in
Fig. 4. The mid signal is transmitted in the base layer,
the side signal is transmitted in the enhancement layer.
3. Intensity stereo: the right and left channel
are transmitted separately. Only a main channel (base
layer) and an additional directional signal (enhancement
layer) are transmitted, from which a stereo signal is
produced.
Examples for the separation and combination of the
main data stream and auxiliary data stream(s) for the
production of a variable bit rate are shown in Fig. 5. The
first channel, for example channel K1, is used to transmit
an encoded signal with a bit rate x. The second channel,
for example channel K2, is used to transmit all of the
information required for achieving a higher bit rate. As
shown by Fig. 5, in addition to an additional bit rate in
the enhancement layer, a combination with stereo signals is
also possible. Fig. 5 also lists an example for different
encoder types, CELP encoders for the base layer and AAC
encoders (Advanced Audio Coding) for the enhancement layer.
The different separations below are also possible:
- the base layer contains two audio data streams
from different program sources. The enhancement layer
6
CA 02379017 2004-10-28
22386-2703
contains the data streams for increasing the useful data
from the audio program,
- two or more different base layers in different
channels each contain a respective audio data stream. The
enhancement layer of one channel contains the additional
useful data from two or more audio data streams.
6a
CA 02379017 2002-01-10
Numerous previously proposed methods are suitable for digital modulation, for
example the QAM, MPSK, or APSK methods.
7
