Field of the Invention
The present invention relates to lossless encoding of
parameters, and, in particular, to the generation and use
of an encoding rule for efficient compression of
parameters.
Background of the Invention and Prior Art
In recent times, the multi-channel audio reproduction
technique is becoming more and more important. This may be
due to the fact that audio compression/encoding techniques
such as the well-known mp3 technique have made it possible
„ to distribute audio records via the Internet or other
transmission channels having a limited bandwidth. The mp3
coding technique has become so famous because of the fact
that it allows distribution of all the records in a stereo
format, i.e., a digital representation of the audio record
, including a first or left stereo channel and a second or
right stereo channel.
Nevertheless, there are basic shortcomings of conventional
two-channel sound systems. Therefore, the surround
technique has been developed. A recommended multi-channel-
surround representation includes, in addition to the two
stereo channels L and R, an additional center channel C and
two surround channels Ls, Rs. This reference sound format
is also referred to as three/two-stereo, which means three
front channels and two surround channels. Generally, five
transmission channels are required. In a playback
environment, at least five speakers at five decent places
are needed to get an optimum sweet spot in a certain
distance of the five well-placed loudspeakers.

Several techniques are known in the art for reducing the
amount of data required for transmission of a multi-channel
audio signal. Such techniques are called joint stereo
techniques. To this end, reference is made to Fig. 9, which
shows a joint stereo device 60. This device can be a device
implementing e.g. intensity stereo (IS) or binaural cue
coding (BCC). Such a device generally receives - as an
input - at least two channels (CHI, CH2, ... CHn) , and
outputs at least a single carrier channel and parametric
data. The parametric data are defined such that, in a
decoder, an approximation of an original channel (CHI, CH2,
... CHn) can be calculated.
Normally, the carrier channel will include subband samples,
spectral coefficients, time domain samples etc., which
provide a comparatively fine representation of the
underlying signal, while the parametric data do not include
^.such samples of spectral coefficients but include control
parameters for controlling a certain reconstruction
algorithm such as weighting by multiplication, time
shifting, frequency shifting, phase shifting, etc. . The
parametric data, therefore, include only a comparatively
coarse representation of the signal or the associated
channel. Stated in numbers, the amount of data required by
a carrier channel will be in the range of 60 - 70 kbit/s,
while the amount of data required by parametric side
information for one channel will typically be in the range
of 1,5 - 2,5 kbit/s. An example for parametric data are the
well-known scale factors, intensity stereo information or
binaural cue parameters as will be described below.
The BCC Technique is for example described in the AES
convention paper 5574, "Binaural Cue Coding applied to
Stereo and Multi-Channel Audio Compression", C. Faller,
F. Baumgarte, May 2002, Munich, in the IEEE WASPAA Paper
"Efficient representation of spatial audio using perceptual
parametrization", October 2001, Mohonk, NY, in "Binaural

cue coding applied to audio compression with flexible
rendering", C. Faller and F. Baumgarte, AES 113th
Convention, Los Angeles, Preprint 5686, October 2002 and in
"Binaural cue coding - Part II: Schemes and applications",
"" C. Faller and F. Baumgarte, IEEE Trans, on Speech and Audio
Proc, volume level. 11, no. 6, Nov. 2003.
In BCC encoding, a number of audio input channels are
converted to a spectral representation using a DFT
' (Discrete Fourier Transform) based transform with
overlapping windows. The resulting uniform spectrum is
divided into non-overlapping partitions. Each partition
approximately has a bandwidth proportional to the
equivalent rectangular bandwidth (ERB). The BCC parameters
are then estimated between two channels for each partition.
These BCC parameters are normally given for each channel
with respect to a reference channel and are furthermore
quantized. The transmitted parameters are finally
calculated in accordance with prescribed formulas
(encoded), which may also depend on the specific partitions
of the signal to be processed.
A number of BCC parameters do exist. The ICLD parameter,
for example, describes the difference (ratio) of the
energies contained in 2 compared channels. The ICC
parameter (inter-channel coherence/correlation) describes
the correlation between the two channels, which can be
understood as the similarity of the waveforms of the two
channels. The ICTD parameter (inter-channel time
difference) describes a global time shift between the 2
channels whereas the IPD parameter (inter-channel phase
difference) describes the same with respect to the phases
of the signals.
One should be aware that, in a frame-wise processing of an
audio signal, the BCC analysis is also performed frame-
wise, i.e. time-varying, and also frequency-wise. This
means that, for each spectral band, the BCC parameters are

individually obtained. This further means that, in case a
audio filter bank decomposes the input signal into for
example 32 band pass signals, a BCC analysis block obtains
a set of BCC parameters for each of the 32 bands.
A related technique, also known as parametric stereo, is
described in J. Breebaart, S. van de Par, A. Kohlrausch, E.
Schuijers, "High-Quality Parametric Spatial Audio Coding at
Low Bitrates", AES 116th Convention, Berlin, Preprint 6072,
May 2004, and E. Schuijers, J. Breebaart, H. Purnhagen, J.
Engdegard, "Low Complexity Parametric Stereo Coding", AES
116th Convention, Berlin, Preprint 6073, May 2004.
Summarizing, recent approaches for parametric coding of
multi-channel audio signals ("Spatial Audio Coding",
"Binaural Cue Coding" (BCC) etc.) represent a multi-channel
audio signal by means of a downmix signal (could be
monophonic or comprise several channels) and parametric
side information ("spatial cues") characterizing its
perceived spatial sound stage. It is desirable to keep the
rate of side information as low as possible in order to
minimize overhead information and leave as much of the
available transmission capacity for the coding of the
downmix signals.
One way to keep the bit rate of the side information low is
to losslessly encode the side information of a spatial
audio scheme by applying, for example, entropy coding
algorithms to the side information.
Lossless coding has been extensively applied in general
audio coding in order to ensure an optimally compact
representation for quantized spectral coefficients and
other side information. Examples for appropriate encoding
schemes and methods are given within the ISO/IEC standards
MPEG1 part 3, MPEG2 part 7 and MPEG4 part 3.

These standards and, for example also the IEEE paper
"Noiseless Coding of Quantized Spectral Coefficients in
MPEG-2 Advanced Audio Coding", S. R. Quackenbush, J. D.
Johnston, IEEE WASPAA, Mohonk, NY, October 1997 describe
state of the art techniques that include the following
measures to losslessly encode quantized parameters:
• Multi-dimensional Huffman Coding of quantized spectral
coefficients
• Using a common (multi-dimensional) Huffman Codebook for
sets of coefficients
• Coding the value either as a hole or coding sign
information and magnitude information separately (i.e.
have only Huffman codebook entries for a given absolute
value which reduces the necessary codebook size,
"signed" vs. "unsigned" codebooks)
• Using alternative codebooks of different largest
absolute values (LAVs), i.e. different maximum absolute
values within the parameters to be encoded
• Using alternative codebooks of different statistical
distribution for each LAV
• Transmitting the choice of Huffman codebook as side
information to the decoder
• Using "sections" to define the range of application of
each selected Huffman codebook
• Differential encoding of scalefactors over frequency
and subsequent Huffman coding of the result
Another technique for the lossless encoding of coarsely
quantized values into a single PCM code is proposed within
the MPEG1 audio standard (called grouping within the

standard and used for layer 2) . This is explained in more
detail within the standard ISO/IEC 11172-3:93.
The publication "Binaural cue coding - Part II: Schemes and
applications", C. Faller and F. Baumgarte, IEEE Trans, on
Speech and Audio Proc, volume level. 11, no. 6, Nov. 2003
gives some information on coding of BCC parameters. It is
proposed, that quantized ICLD parameters are differentially
encoded
• over frequency and the result is subsequently Huffman
encoded (with a one-dimensional Huffman code)
• over time and the result is subsequently Huffman
encoded (with a one-dimensional Huffman code),
and that finally, the more efficient variant is selected as
the representation of an original audio signal.
Bosi M. et al.: "ISO/IEC MPEG-2 Advanced audio coding"
Journal of the Audio Engineering Society, vol. 45, no. 10,
pages 789-812 propose grouping and interleaving of
parameters, prior to their Huffman encoding.
The US application US 5,526,628 relates to variable-length
coding of input symbols using a plurality of variable
length-code tables. The same sequence of symbols is encoded
using different variable length coders or Huffman coders
having different code tables. The decision, which encoded
representation is to be transmitted is finally made,
depending on the resulting code length.
The European patent application 1 04 7 198 A3 proposes a
^-similar approach, additionally using only absolute-valued
codebooks.
Shen-Chuan Tai et al.: "An adaptive 3-D discrete cosine
transform coder for medical image compression" IEEE

Transactions on Information Technology in Biomedicine, vol.
4, no. 3, pages 259-263) propose a method of encoding two-
dimensional image data using a coding algorithm, originally
developed to encode... three-dimensional information. In the
proposed method, a segmentation technique based on the
local energy magnitude is used to segment subblocks of the
image into different energy levels. Those subblocks with
the same energy level are then gathered to form a three-
dimensional cuboid. Then, a three-DCT is employed to
compress the three-dimensional cuboid individually.
Faller C et al.: "Binaural Cue Coding - Part II: Schemes
and Applications", IEEE Transactions on Speech and Audio
Processing, Vol. 11, no. 6, 6 October 2003, pages 520-531,
propose a differential coding in the time and the frequency
direction prior to Huffman coding. Finally, the
differential representation resulting in the lower bit rate
is chosen.
As mentioned above, it has been proposed to optimize
compression performance by applying differential coding
over frequency and, alternatively, over time and select the
more efficient variant. The selected variant is then
„ signaled to a decoder via some side information.
There has been quite some effort made to reduce the size of
a downmix audio channel and the corresponding side
information. Nonetheless the achievable bit rates are still
- too high to allow for every possible application. For
example, streaming of audio and video content to mobile
phones requires the least possible bit rates and therefore
a more efficient encoding of the content.

Summary of the invention
It is the object of the present invention to provide an
/^improveci coding concept achieving a lossless compression of
parameter values with higher efficiency.

In accordance with the first aspect of the present
invention, this object is achieved by a compression unit
for compression of parameters, the parameters including a
""""'first set of parameters including a representation of a
first portion of an original signal, the parameters further
including a second set of parameters including a
representation of a second portion of the original signal,
the second portion neighboring the first portion,
■""comprising: a supplier for supplying a first tuple and a
second tuple, each tuple having at least two parameters,
the first tuple having two parameters from the first set of
parameters and the second tuple having one parameter from
the first set of parameters and one parameter from the
second set of parameters; a bit estimator for estimating a
number of bits necessary to encode the sets of parameters
using a first sequence of tuples including the first tuple
and to encode the sets of parameters using a second
sequence of tuples including the second tuple, based on an
encoding rule; and a provider for providing encoded blocks,
the provider being operative to provide the encoded blocks
using the sequence of tuples resulting in a lower number of
bits, and for providing a sequence indication indicating
the sequence of tuples from which the encoded blocks are
derived.
In accordance with the second aspect of the present
invention, this object is achieved by a decoder for
decoding encoded blocks of parameters, the parameters
including a first set of parameters including a
representation of a first portion of an original signal,
the parameters further including a second set of parameters
including a representation of a second portion of the
original signal, the second portion neighboring the first
portion, and for processing a sequence indication,
comprising: a decompressor, the decompressor being
operative to decompress, using a decoding rule depending on
an encoding rule used for encoding sequences of tuples, an

encoded block of parameters to derive a sequence of tuples
of parameters, each tuple having at least two parameters;
and a frame builder for receiving the sequence indication,
the sequence indication indicating a used sequence of
tuples from a number of different sequences underlying the
encoded block, and for building the sets of parameters
using the information of the used sequence of tuples.
In accordance with the third aspect of the present
invention, this object is achieved by a method for
compression of parameters, the parameters including a first
set of parameters including a representation of a first
portion of an original signal, the parameters further
including a second set of parameters including a
representation of a second portion of the original signal,
the second portion neighboring the first portion.
In accordance with the fourth aspect of the present
invention, this object is achieved by a computer program
implementing the above method, when running on a computer.
In accordance with the fifth aspect of the present
invention, this object is achieved by a method for decoding
encoded blocks of parameters, the parameters including a
first set of parameters including a representation of a
first portion of an original signal, the parameters further
including a second set of parameters including a
representation of a second portion of the original signal,
the second portion neighboring the first portion, and for
processing a sequence indication.
In accordance with the sixth aspect of the present
invention, this object is achieved by a computer program
implementing the above method, when running on a computer.
In accordance with the seventh aspect of the present
invention, this object is achieved by a compressed
representation of parameters, the parameters including a

first set of parameters including a representation of a
first portion of an original signal, the parameters further
including a second set of parameters including a
representation of a second portion of the original signal,
the second portion neighboring the first portion of the
original signal, comprising: an encoded block of parameters
representing a used sequence of tuples; and a sequence
indication indicating the used sequence of tuples of a
first or a second sequence underlying the encoded block of
parameters, where the first sequence includes a first tuple
having two parameters from the first set of parameters and
where the second sequence includes a second tuple having
one parameter from the first set of parameters and one
parameter from the second set of parameters.
The present invention is based on the finding that
parameters including a first set of parameters of a
representation of a first portion of an original signal and
including a second set of parameters of a representation of
a second portion of the original signal can be efficiently
encoded, when the parameters are arranged in a first
sequence of tuples and in a second sequence of tuples,
wherein the first sequence of tuples comprises tuples of
parameters having two parameters from a single portion of
the original signal and wherein the second sequence of
tuples comprises tuples of parameters having one parameter
from the first portion and one parameter from the second
portion of the original signal. An efficient encoding can
, be achieved using a bit estimator to estimate the number of
necessary bits to encode the first and the second sequence
of tuples, wherein only the sequence of tuples is encoded,
that results in the lower number of bits.
The basic principle therefore is, that one rearranges the
parameters to be encoded, for example in time and in
frequency, and finally uses the one arrangement( sequence
of tuples) of the parameters for the compression, that

results in the lower number of bits for the compressed
parameters.
In one embodiment of the present invention, two sets of
spectral parameters, describing the spectral representation
of two consecutive time portions of an original signal are
adaptively grouped in pairs of two parameters to enhance
the coding efficiency. Therefore, on the one hand a
sequence of tuples is generated using tuples of parameters
consisting of two neighboring frequency parameters from the
same time portion. On the other hand, a second sequence of
tuples is generated using tuples, that are built using a
first parameter from the first time portion and the
corresponding parameter from the second time portion of the
original signal. Then, both sequences of tuples are encoded
using a two-dimensional Huffman code. The two encoded
sequences of tuples are compared in their sizes and the
tuple resulting in the lower number of bits is finally
selected to be transmitted. The information, which kind of
tuples has been used to build the encoded data is
transmitted to a decoder as additional side information.
One advantage of the previously described inventive encoder
is, that due to the grouping of parameters into tuples
consisting of two parameters, a two-dimensional Huffman
code can be applied for the compression, which generally
results in a lower bit rate.
second advantage is, that the adaptive grouping, i.e. the
concept to dynamically decide between two possible grouping
strategies during the encoding process, yields a further
decrease in the bit rate of the side information.
, Deciding between the two grouping strategies only once for
a set of two consecutive frames additionally reduces the
amount of required side information, since the indication,
which grouping strategy has been used during the encoding,

has to be transmitted only once for a set of two complete
consecutive time frames.
In a further embodiment of the present invention an
inventive compression unit additionally comprises a
differential encoder, that differentially encodes the
parameters either in time or in frequency prior to the
adaptive grouping. That differential encoding together with
the adaptive grouping and an appropriate Huffman codebook
further reduces the size of the side information to be
transmitted. The two differential encoding possibilities
together with the two grouping strategies result in a total
number of four possible combinations, further increasing
the probability of finding an encoding rule, that results
in a low side information bit rate.
In a further embodiment of the present invention, the
inventive concept is used for a decompression unit,
allowing to decode encoded blocks of parameters and to
rebuild the original frames based on a side information
signaling the grouping scheme underlying the encoded blocks
of parameters. In an advantageous modification the
inventive decoder also allows the decoding of data that has
not been adaptively grouped, therefore a compatibility of
the inventive decoder with existing equipment can be
achieved.
Preferred embodiments of the present invention are
subsequently described by referring to the enclosed
drawings, wherein:
,Fig. 1 shows an inventive compression unit;
Fig. 2 shows two possibilities of adaptively grouping
parameters;

Fig. 3 shows some additional possibilities of adaptively
grouping parameters;
Fig. 4 shows differential encoding schemes;
Fig. 5 shows an inventive decoder;
Fig. 6 shows a prior art multi-channel encoder.
Detailed description of preferred embodiments
Fig. 1 shows an inventive compression unit 90, comprising a
supplier 100, a bit estimator 102 and a provider 104.
The supplier 100 supplies a first sequence of tuples 106a
and a second sequence of tuples 106b at two data outputs.
The provider 104 receives the tuples 106a and 106b on two
of his data inputs 108a and 108b. The bit estimator
receives the two tuples on his data inputs 110a and 110b.
The bit estimator 102 estimates the number of bits that
result from applying an encoding rule to the two tuples
106a and 106b. The bit estimator 102 chooses the tuple
resulting in the lower number of bits and signals via a
signaling output 112a, whether the tuple 106a or 106b will
result in the lower number of bits.
Based on the decision of the bit estimator 102, the tuple
resulting in the lower number of bits is finally encoded
into encoded blocks 118, that are provided via output 120a
of the provider 104, wherein the provider further signals a
sequence indication at his signaling line 120b, indicating
what original sequence of tuples (106a or 106b) was encoded
to derive the encoded blocks 118.
In an alternative embodiment, the same functionality can be
achieved, when the dashed connections 122a and 122b between
the supplier 100 and the provider 104 are omitted. In this

alternative scenario the bit estimator 102 would encode the
sequence of tuples 106a and 106b, and would transfer two
different encoded blocks 124a and 124b to the provider 104,
where the provider additionally signals from which of the
original sequences of tuples 106a and 106b the encoded
blocks 124a and 124b are derived. To this end, the
signaling output 112a of the bit estimator 102 can be used
or the signaling can be derived by the provider 104
implicitly.
In this alternative embodiment the provider 104 would
simply forward the encoded block with the lower number of
bits to its output 120a, additionally providing the
sequence indication.
Fig. 2 shows an example of two adaptive grouping schemes
that are used to derive a sequence of tuples to be encoded.
To explain the principle of the inventive adaptive
grouping, four subsequent time frames 130a to 130d of an
original signal are shown, wherein each of the frames is
having a set of five spectral parameters 132a to 132e.
According to the present invention, the spectral parameters
of two consecutive frames are grouped either in frequency,
as illustrated by the tuples 134a and 134b or in time, as
illustrated by the tuples 136a and 136b to build the
sequences of tuples. The grouping in time results in a
first sequence of tuples 138, whereas the grouping in
frequency results in the second sequence of tuples 140.
The sequences of tuples 138 and 140 are encoded using for
example a Huffman codebook, resulting into two different
sequences of code words 142 and 144. According to the
present invention, the sequence of code words requiring the
fewer number of bits, is finally transmitted to a decoder,
that has to additionally receive a sequence indication,
signaling whether time grouping or frequency grouping is
underlying the sequence of code words. As can be seen in

>
Fig. 2, for the shown example of adaptive grouping of pairs/
of parameters (2-dimensional), the sequence indication can
consist of only one single bit.
^ Fig. 3 shows some alternative grouping strategies, that can
be used to implement the inventive adaptive grouping,
allowing for Huffman codes with dimensions bigger than 2.
Fig. 3 shows a grouping strategy for a two-dimensional
Huffman code 146a, for a three-dimensional Huffman code
146b and a for a four-dimensional Huffman code 146c. For
each of the strategies two consecutive time frames are
illustrated, wherein the parameters belonging to the same
tuple are represented by the same capital letters.
In the case of the two-dimensional Huffman code the
grouping is done as already illustrated in Fig. 2, building
two-dimensional tuples in frequency 148a and in time 14 8b.
In case of building tuples consisting of three parameters,
the frequency tuples 158a are such, that three neighboring
frequency parameters within one frame are grouped together
to form a tuple. The time tuples 150b can be built such,
that two neighboring parameters from one frame are combined
with one parameter from the other frame, as is shown in
""" Fig. 3.
Four-dimensional frequency grouped tuples 152a are built
corresponding to the other frequency tuples by grouping
four neighboring parameters of one frame into one tuple.
The time grouping tuples 152b are built such, that two
neighboring parameters of one frame are combined with two
neighboring parameters of the other frame, wherein the
parameter pairs of the single frames are describing the
same spectral property of the two consecutive time frames.
Allowing different grouping schemes, as illustrated in Fig.
3, one can significantly reduce the bit rate of the side
information, for example if one uses a variety of

predefined Huffman codebooks of different dimensions, the
dimension of the grouping can be varied within the encoding
process such, that the representation resulting in the
lowest bit rate can be used at any time within the encoding
process.
Fig. 4 shows, how an inventive compression unit, that
additionally comprises a differential encoder, can be used
to further decrease the side information, by applying some
differential encoding before the Huffman encoding process.
To illustrate the differential encoding in time and
frequency or in time and frequency, the same absolute
representation of parameters 160, that was already shown in
Fig. 2, is used as a basis for the various differential
encoding steps. The first possibility is to differentially
encode the parameters of the absolute representation 160 in
frequency, resulting in the differentially encoded
parameters 162. As can be seen in Fig. 4, to differentially
encode the absolute representation 160, the first parameter
of each frame is left unchanged, whereas the second
parameter is replaced by the difference of the second
parameter and the first parameter of the absolute
representation 160. The other parameters within the
differentially encoded representation are built following
the same rule.
Another possibility is the differential coding in time,
yielding the representation 164. This representation is
"~ built by leaving the complete first frame unchanged,
whereas the parameters of the following frames are replaced
by the difference of the parameter of the absolute
representation and the same parameter of the previous
frame, as can be seen in Fig. 4.
A third possibility is to first encode differentially in
frequency, followed by a differential encoding in time or
vice versa. both resulting in the same encoded

representation 166, that is differentially encoded in time
and frequency.
It is to be noted, that one has the chance to use those
four different representations of the original signal as
input to the adaptive grouping. Having a look at the
different representations 160 to 166 of the given example
of parameters, one can clearly see how the differential
encoding has impact on the transmitted rate of side
information. Looking at the absolute representation 160,
one recognizes, that neither a grouping in time nor in
frequency would result in tuples having the same content.
Therefore no appropriate Huffman codebook is constructable,
that would assign the shortest code words to the tuples
occurring most.
The case is different looking at the differentially in
frequency encoded representation 162, where one could
construct a Huffman codebook that only needs to have four
entries to cover the full representation, and where either
the tuple (1, 1) or the tuple (2, 2) would be assigned the
code word with minimum length, to achieve a compact side
information.
The advantage is less obvious in the representation being
differentially encoded in time 164. Nonetheless one can
gain also here, grouping in frequency and making use of the
numerous tuples (5, 5) and (10, 10) .
For the representation that is differentially encoded in
time and in frequency 166, one would even achieve a further
reduction of the side information bit rate than in the
representation 162, since a grouping in time would result
in a high multiplicity of the tuple (1, 0), as indicated in
the figure, allowing to construct a Huffman codebook, that
would assign the shortest code word to the previous tuple.

As can be clearly seen in Fig. 4, the high flexibility of
the inventive concept making use of adaptive grouping and
of differential encoding allows to choose the strategy that
fits the original audio signal most, thus allowing to keep
the side information bit rate low.
To summarize, in one preferred embodiment the quantized
parameter values are first differentially encoded over time
(variant 1) and differentially over frequency (variant 2) .
The resulting parameters can then be grouped adaptively
over time (variant a) and frequency (variant b) . As a
result, four combinations are available (la, lb, 2a, 2b)
from which the best is selected and signaled to the
decoder. This could be done by a 2 bit information only,
representing the variants la, lb, 2a, 2b by, for example,
the bit combination 00, 01, 10, 11.
Fig. 5 shows a decoder according to the current invention,
to decode encoded blocks of parameters, wherein the block
of parameters includes a first frame having a set of first
spectral parameters and a second frame having a set of
second spectral parameters.
_ The decoder 200 comprises a decompressor 202 and a frame
builder 204. The decompressor receives on an input an
encoded block of parameters 206. The decompressor derives,
using a decoding rule, a sequence of tuples of parameters
208 from the encoded block of parameters 206. This sequence
of tuples of parameters 208 is input into the frame builder
204.
The frame builder additionally receives a sequence
indication 210, indicating what sequence of tuples have
^ been used by the encoder to build the encoded block of
parameters.
The frame builder 204 then reorders the sequence of tuples
208 steered by the sequence indication 210 to reconstruct

the first frame 112a and the second frame 112b from the
sequence of tuples of parameters 208.
Preferred embodiments of the present invention described
above achieve a further enhancement of the coding
efficiency by introducing adaptive grouping of values to be
coded using a multi-dimensional Huffman code. As an
example, both, two-dimensional grouping of values over
frequency can be done as well as two-dimensional grouping
of values over time. The encoding scheme would then do both
types of encoding and choose the more advantageous one
(i.e. the variant which requires less bits). This decision
is signaled to the decoder via side information.
In further examples, as illustrated in Fig. 3, it is also
possible to build higher-dimensional Huffman codes,
applying different grouping strategies to build the tuples.
The given examples show grouping strategies that build the
tuples by grouping together parameters from two consecutive
frames only. It is also possible to do the grouping using
parameters from three or more consecutive frames, doing the
grouping in a straight forward way.
In a modification of the inventive encoder, it is also
possible to combine the differential grouping and the
differential encoding strategies with the use of different
Huffman codebooks to derive the shortest possible
representation of the side information. This could further
reduce the side information bit rate of an encoded audio
signal at the cost of having additional side information
parameters, signaling the Huffman codebook used for the
encoding.
The described preferred embodiments of the present
invention show the inventive concept for examples, where
the grouping strategy does not change within two
consecutive time frames. In a modification of the present
invention it is of course also possible, to have multiple

changes between the grouping in time and in frequency
within a set of two frames, which would imply that the
sequence indication is also supplied within the frames, to
signal the change of grouping strategy.
In the given examples, the parameters are differentially
encoded before being Huffman encoded. Of course every other
lossless encoding rule is also possible prior to the
Huffman encoding of the parameters, the aim of the encoding
being to derive as much tuples with the same content as
possible.
There are four different possible parameter representations
given in Fig. 4, namely the absolute representation, the
differential representation in frequency, the differential
representation in time and the differential representation
in time and frequency. To choose between four
representations, the side information signaling which
representation has been used, has to be at least two bits
in size, as indicated in Fig. 4. To balance the win of a
possible efficiency gain of the coding versus the
additional spectral representation indication, one could of
course also decide to principally allow only two possible
representations, reducing the spectral representation
indication to the length of a single bit.
As an example of an inventive decoder, Fig. 5 shows a
decoder 200 that receives in addition to the encoded block
of parameters 206 some side information. The side
information steering the frame builder 204 only comprises a
sequence indication 210 in the given example. A decoder
according to the present invention can of course process
any other side information required, especially a spectral
representation indication, indicating the spectral
representation that has been used to encode original
frames.

Depending on certain implementation requirements of the
inventive methods, the inventive methods can be implemented
in hardware or in software. The implementation can be
performed using a digital storage medium, in particular a
disk, DVD or a CD having electronically readable control
signals stored thereon, which cooperate with a programmable
computer system such that the inventive methods are
performed. Generally, the present invention is, therefore,
a computer program product with a program code stored on a
"~ machine readable carrier, the program code being operative
for performing the inventive methods when the computer
program product runs on a computer. In other words, the
inventive methods are, therefore, a computer program having
a program code for performing at least one of the inventive
methods when the computer program runs on a computer.
While the foregoing has been particularly shown and
described with reference to particular embodiments thereof,
it will be understood by those skilled in the art that
*~ various other changes in the form and details may be made
without departing from the spirit and scope thereof. It is
to be understood that various changes may be made in
adapting to different embodiments without departing from
the broader concepts disclosed herein and comprehended by
~~~ the claims that follow.

WE CLAIM :
1. A compression unit for compression of parameters, the parameters
comprising a first set of parameters comprising a representation of a
first portion of an original signal, the parameters additionally
comprising a second set of parameters having a representation of a
second portion of the original signal, the second portion neighboring
the first portion, comprising :
a supplier operative to supply a first group of tuples and a second
group of tuples, the tuples within the groups of tuples having
parameters from one of the following representations :
wherein the first group of tuples is having tuples with two parameters
from the first set of parameters and wherein the second group of
tuples is having one parameter from the first and one parameter from
the second set of parameters; and
the first group of tuple consisting of an odd number of neighboring
parameters of the representation of the original signal from one set of
parameters; and
the second tuple consisting of a majority of neighboring parameters of
the representation of the original signal from one set of parameters
and of a minority of neighboring parameters of the representation of
the original signal from the other set of parameters, wherein the
majority of parameters is alternatingly taken from the first and from
the second set of parameters for consecutive tuples of the second
sequence of tuples.

a bit estimator operative to estimate the number of bits necessary to
encode sequences of tuples for at least eight sequences of tuples,
each sequence having one of the tuples of the groups of tuples; and
a provider operative to signal, in addition to the sequence indication,
an encoded representation indication indicating the representation
resulting in the lower number of bits.
2. The compression unit as claimed in claim 1 wherein the first and the
second set of parameters are comprising a spectral representation of
the first and the second portion of the original signal as
representation.
3. The compression unit as claimed in claim 1, wherein the first portion
and the second portion of the original signal are neighbored in time or
in space.
4. The compression unit as claimed in claim 1, wherein the first set of
parameters comprises a representation of a first frame of an original
audio or video signal and in which the second set of parameters
comprises a representation of a second frame of the original audio or
video signal.
5. The compression unit as claimed in claim 1, wherein
the supplier being operative to supply the first tuple and all remaining
tuples of the. first sequence of tuples using only tuples having
parameters from one single set of parameters, and to supply the first
tuple and all remaining tuples of the second sequence of tuples using

only tuples having at least one parameter from the first set of
parameters and at least one parameter from the second set of
parameters; and wherein the provider is operative to provide a single
sequence indication for one encoded block.
6. The compression unit as claimed in claim 1, wherein the supplier is
operative to supply.
the first tuple consisting of two parameters from a single set of
parameters, the two parameters being neighbored parameters within
the representation of the original signal; and the second tuple
consisting of a first parameter from the first set of parameters and a
second parameter from the second set of parameters, the first and the
second parameter describing an identical property of the original signal
within the representation.
7. The compression unit as claimed in claim 1, wherein the supplier is
operative to supply
the first tuple consisting of an integer multiple of two parameters from
a single set of parameters, the parameters being neighbored
parameters within the representation of the original signal; and
the second tuple consisting of an two or more parameters from the
first set of parameters, the parameters being neighbored within the
representation of the original signal and of the same number of
parameters from the second set of parameters, the parameters being
neighbored within the representation of the original signal, the

parameters from the first and the second set of parameters describing
an identical property of the original signal within the representation.
8. The compression unit as claimed in claim 1, wherein the bit estimator
is operative to encode, using the encoding rule, the first sequence of
tuples to derive a first encoded block and to encode the second
sequence of tuples to derive a second encoded block, and to estimate
the necessary number of bits by counting the bits of the first and of
the second encoded block, and wherein the provider is operative to
forward the first or the second block, choosing the block with the
lower number of bits.
9. The compression unit as claimed in claim 1, wherein the bit estimator
is operative to estimate the bits necessary to encode the sequences of
tuples according to the encoding rule.
10. The compression unit as claimed in claim 1, wherein the sets of
parameters are comprising a parametric representation of a video
signal or of an audio signal.
11. The compression unit as claimed in claim 1, wherein the encoding rule
allows encoding tuples having the same length but different
parameters and result in codewords of different length.
12. The compression unit as claimed in claim 1, wherein the parameters
comprise BCC parameters describing a spatial interrelation between a
first and a second original audio channel and in which the BCC
parameters are selected from the following list of BCC parameters;

interchannel coherence/correlation (ICC)
interchannel level difference (ICLD)
interchannel time difference (ICTD)
interchannel phase difference (IPD).
13. The compression unit as claimed in claim 1, comprising a differential
encoder for processing the original signal such that the first set of
parameters and the second set of parameters comprise a difference
representation, the difference being a difference in one of time,
frequency or time and frequency.
14. The compression unit as claimed in claim 1, wherein the provider is
operational to signal the sequence indication and the encoded
representation indication using a binary data word having at least
three bits.
15. The compression unit as claimed in claim 1, wherein the encoding rule
in such that each tuple has associated therewith a signal code word of
a group of different code words.
16. The compression unit as claimed in claim 1, wherein the encoding rule
comprises a Huffman codebook.
17. A decoder for decoding encoded blocks of parameters, the parameters
comprising a first set of parameters having a representation of a first
portion of an original signal, the parameters additionally comprising a
second set of parameters having a representation of a second portion

of the original signal, the second portion neighboring the first
portion, and for processing a sequence indication, comprising:
a decompressor, the decompressor being operative to decompress,
using a decoding rule depending on an encoding rule used for
encoding sequences of tuples, an encoded block of parameters to
derive a sequence of tuples of parameters, each tuple having the
same length and at least two parameters; and a frame builder for
receiving the sequence indication, the sequence indication indicating a
used sequence of tuples from a number of different sequences
underlying the encoded block, and for building the sets of parameters
using the information of the used sequence of tuples, wherein the
first and the second sets of parameters are having a spectral
representation of the first and the second portion of the original signal
as representation.
18. The decoder as claimed in claim 17, wherein the first portion and the
second portion of the original signal are neighbored in time or in
space.
19. The decoder as claimed in claim 17, wherein the first set of
parameters comprises a representation of a first frame of an original
audio or video signal and in which the second set of parameters
comprises a representation of a second frame of the original audio or
video signal.

20. The decoder as claimed in claim 17, wherein the frame builder is
operative to receive a single sequence indication for one encoded
block of parameters, and to build the sets of parameters by using a
sequence of tuples comprising only tuples of the type indicdated by
the sequence indication.
21. The decoder as claimed in claim 17, wherein the frame builder is
operative to build sets of parameters comprising a parametric
representation of a video signal or of an audio signal.
22. The decoder as claimed in claim 17, wherein the frame builder is
operative to build sets of parameters comprising BCC parameters
describing a spatial interrelation between a first and a second audio
channel and in which the BCC parameters are chosen form the
following list of BCC parameters :
interchannel coherence/correlation (ICC)
interchannel level difference (ICLD)
interchannel time difference (ICTD)
interchannel phase difference (IPD).
23. The decoder as claimed in claim 15, comprising a differential decoder
for receiving a representation indication and for processing the sets of
parameters such, that the parameters are derived from a difference
spectral representation, the difference being a difference in time, in
frequency or in time and in frequency, wherein the differential decoder
is operative to process the first and the second set of parameters
depending on the representation indication.

24. The decoder as claimed in claim 23, wherein the frame builder and the
differential decoder are operative to receive a sequence indication and
a representation indication as a binary data word having at least three
bits.
25. The decoder as claimed in claim 17, wherein the decoding rule in such
that each code word within the encoded block of parameters has
associated therewith a single tuple of a group of different tuples.
26. A method for compression of parameters, the parameters including a
first set of parameters comprising a representation of a first portion of
an original signal, the parameters additionally comprising a second set
of parameters comprising a representation of a second portion of the
original signal, the second portion neighboring the first portion, the
method comprising :
supplying a first group of tuples and a second group of tuples of equal
length, each tuple having at least two parameters, the tuples within
the groups of tuples having parameters from one of the following
representations :
a differential representation in time, a differential representation in
frequency, a differential representation in time and in frequency, and
an absolute representation.
estimating a number of bits necessary to encode the sets of
parameters using a first sequence of tuples comprising the first tuple

and to encode the frames using a second sequence of tuples
comprising the second tuple, based on an encoding rule;
providing encoded blocks using the sequence of tuples resulting in a
resulting in a lower number of bits; and
providing a sequence indication indicating the sequence of tuples from
which the encoded blocks are derived.
27. A method for decoding encoded blocks of parameters, the parameters
comprising as first set of parameters comprising a representation of a
first portion of an original signal, the parameters additionally
comprising a second set of parameters having a representation of a
second portion of the original signal, the second portion neighboring
the first portion, and for processing a sequence indication, the method
comprising:
decompressing, using a decoding rule depending on an encoding rule
used for encoding sequences of tuples, the encoded block of
parameters to derive a sequence of tuples of parameters, each tuple
having the same length and at least two parameters;
receiving the sequence indication indicating a used sequence of tuples
from a number of different sequences underlying the encoded block;
and

building the sets of parameters using the information of the used
sequence of tuples.

ABSTRACT

TITLE "A COMPRESSION UNIT AND A METHOD FOR COMPRESSION
OF PARAMETERS TO ENHANCE COOLING EFFICIENCY"
The invention relates to a compression unit for compression of parameters,
the parameters comprising a first set of parameters comprising a
representation of a first portion of an original signal, the parameters
additionally comprising a second set of parameters having a representation of
a second portion of the original signal, the second portion neighboring the
first portion, comprising : a supplier operative to supply a first group of tuples
and a second group of tuples, the tuples within the groups of tuples having
parameters from one of the following representations : wherein the first
group of tuples is having tuples with two parameters from the first set of
parameters and wherein the second group of tuples is having one parameter
nfrom the first and one parameter from the second set of parameters; and the
first group of tuple consisting of an odd number of neighboring parameters of
the representation of the original signal from one set of parameters; and the
second tuple consisting of a majority of neighboring parameters of the
representation of the original signal from one set of parameters and of a
minority of neighboring parameters of the representation of the original signal
from the other set of parameters, wherein the majority of parameters is
alternatingly taken from the first and from the second set of parameters for
consecutive tuples of the second sequence of tuples, a bit estimator operative
to estimate the number of bits necessary to encode sequences of tuples for
at least eight sequences of tuples, each sequence having one of the tuples of
the groups of tuples; and a provider operative to signal, in addition to the
sequence indication, an encoded representation indication indicating the
representation resulting in the lower number of bits.