CN101160725A - Lossless information encoding for maximum bitrate - Google Patents

Abstract

When comparing a first coding rule that produces a coded representation of an information value with a variable length with a second coding rule that produces a coded representation of an information value with a fixed length, and when the coded representation that results in requiring a smaller number of information units is selected A tightly encoded representation of information values that does not exceed a predetermined size can be derived when the encoding rules of . Thus, it can be ensured that the maximum bit rate is at least the maximum bit rate of the second coding rule deriving the second coded representation. The selection of an encoding rule via a rule message is sent together with the encoded representation of the information value, after which the correct information value can be derived at the decoder using a decoding rule adapted to the encoding rule used during encoding.

Description

Lossless information encoding for maximum bitrate

technical field

The invention relates to lossless coding of information values, and in particular to the concept of ensuring a maximum bit rate for the coded representation of information values.

Background technique

Recently, multi-channel audio reproduction techniques have gained importance, perhaps due to the fact that audio compression/encoding techniques, such as the now well-known mp3 technology, allow the distribution of audio over networks or other transmission channels with limited bandwidth record possible. The reason why this mp3 encoding technique has become so famous is due to the fact that it can distribute recordings in stereo format, that is, as a digital representation of an audio recording containing a first or left stereo channel and a second or right stereo channel .

However, the traditional two-channel system has its fundamental shortcomings, therefore, surround sound technology was developed. A proposed 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 known as three/two stereo, meaning that it has three front channels and two surround channels. Generally speaking, five transmission channels are necessary. In a playback environment, at least five speakers need to be placed in five suitable positions to obtain the sweet spot of sweet sound at a specific distance from the five speakers that are already properly placed.

There are already many well-known techniques that can be used to reduce the amount of data required to transmit multi-channel audio signals, these techniques are called joint stereophony techniques. For this purpose, please refer to Figure 9, which shows a joint stereo device 60. The device may be a device for implementing intensity stereo (abbreviated as IS for intensity stereo) or stereo cue coding (abbreviated as BCC for binauralcue coding). Such a device typically receives at least two channels (CH1, CH2, ... CHn) as input and outputs at least a single carrier channel and parameter data. This parameter data is defined such that an approximation of the original channels (CH1, CH2, . . . CHn) can be calculated in the decoder.

Generally speaking, the carrier channel will contain sub-band samples, spectral coefficients, time-domain samples, etc., so that a relatively good representation of the underlying signal can be provided, while the parameter data does not contain samples of the spectral coefficients, but includes Control parameters to control specific reconstruction algorithms such as multiplicative weighting, time shifting, frequency shifting, phase shifting, etc. The parameter data therefore only contain a relatively coarse representation of the signal or the channel associated with it. If expressed in numbers, the amount of data required by the carrier channel is approximately in the range of 60kbit/s to 70kbit/s. However, the amount of data required for the parametric auxiliary information of one channel is typically in the range of 1.5 kbit/s to 2.5 kbit/s. Several well-known parameter data include: scaling factor, intensity stereo information or stereo cue parameters, as described below.

For example, BCC technology is described in the following papers: AES convention paper5574, "Binaural Cue Coding applied to Stereo and Multi-ChannelAudio Compression", C. Faller, F. Baumgarte, May 2002, Munich; IEEEWASPAA Paper "Efficient representation of spatial audio using perceptual parametrization", October 2001, Mohonk, NY; "Binauralcue coding applied to audio compression with flexible rendering", C. Faller and F. Baumgarte, AES 113 ^th Convention, Los Angeles, Preprint 5686, October 2002; and "Binauralcue cuecoding-Part II: Schemes and applications", C. Faller and F. Baumgarte, IEEE Trans. on Speech and Audio Proc., volumelevel.11, no.6, Nov.2003.

In the BCC encoding method, firstly, multiple audio input channels are converted into spectral representations using a discrete Fourier transform (DFT)-based transformation with overlapping windows. The uniform spectrum obtained by the above method is divided into non-overlapping parts, and the bandwidth of each part is approximately proportional to the Equivalent Rectangular Bandwidth (ERB for short). Then for each part, the BCC parameters are estimated between the two channels. In general, the BCC parameters for each channel are given relative to a reference channel and further quantized. The transmission parameters are finally calculated according to specified equations (encoded), which may also depend on the specific part of the signal to be processed.

Multiple BCC parameters do exist. For example, the ICLD parameter is used to describe the difference (ratio) in the energy contained in two compared channels. The inter-channel coherence/correlation (ICC) parameter is used to describe the correlation between two channels, which can be understood as the similarity of the two channel waveforms. The inter-channel time difference (ICTD) parameter is used to describe the global time shift between two channels, while the inter-channel phase difference (IPD) parameter is used to describe the signal phase difference between them.

It should be noted that in frame-wise processing of audio signals, BCC analysis is also performed frame-wise, ie time-varying, but also frequency-wise. This means that, for each spectral band, the BCC parameters are obtained separately. This also means that if the input signal is decomposed into eg 32 bandpass signals with an acoustic filter bank, the BCC analysis block obtains a set of BCC parameters for each of these 32 bands.

A related technique, so-called 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 116 ^th Convention, Berlin, Preprint 6072, May 2004; and E. Schuijers, J. Breebaart, H. Purnhagen, J. Engdegard, "LowComplexity Parametric Stereo Coding", AES 116 ^th Convention, Berlin, Preprint 6073, May 2004 .

In summary, recent approaches to parametric coding of multi-channel audio signals (spatial audio coding, stereo cue coding, etc.) rely on downmixed signals (which can be single-channel or include several channels) and parametric side information (spatial audio coding, etc.) hint) to represent a multi-channel audio signal, the parametric side information characterizes the perceived spatial sound stage. It is generally desirable to keep the data rate of side information as low as possible in order to minimize overhead information and to leave as much available transmission capacity as possible for encoding of the downmix signal.

One way to keep the bit rate of the side information low is to losslessly code the side information of the spatial audio scheme, eg by applying an entropy coding algorithm to the side information.

Lossless coding has been widely used in general audio coding to ensure an optimal compact representation of quantized spectral coefficients and side information. Examples of suitable coding schemes and methods can be found in ISO/IEC standards MPEG1 Part 3, MPEG2 Part 7 and MPEG4 Part 3.

These standards and, for example, 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 techniques in the state of the art, including by means of Losslessly encode quantized parameters:

●Multidimensional Huffman (Huffman) coding of quantized spectral coefficients

● For coefficient groups, use a common (multidimensional) Huffman codebook

Encode the value as a hole, or encode the sign information as well as the quantity information separately (i.e. have only Huffman codebook entries for a given absolute value, which reduces the required codebook size, " Signed" vs. "unsigned" codebooks)

Use alternative codebooks with different maximum absolute values, ie different maximum absolute values among the parameters to be coded

● For each LAV, use a surrogate codebook with a different statistical distribution

●Send the Huffman codebook selection to the decoder in the form of auxiliary information

● Use "sections" to define the scope of application of each selected Huffman codebook

Differential encoding of the frequency scaling factors, and subsequent Huffman encoding of the result

In the MPEG1 audio standard, another technique (called grouping in the standard and used for layer 2) of encoding coarse quantized values into a single PCM code in a lossless manner is proposed, which is described in ISO/ It is explained in more detail in IEC standard 11172-3:93.

The publication "Binaural cue coding-Part II: Schemes and applications", C.Faller and F.Baumgarte, IEEE Trans.on Speechand Audio Proc., volume level.11, no.6, Nov.2003 mentioned some relevant BCC parameter coding information. It proposes to differentially encode the quantized ICLD parameters in the following manner:

- Differentially encode in frequency, and then Huffman encode the result (using 1D Huffman codes)

- Differentially encode in time, and then Huffman encode the result (using 1D Huffman codes)

Finally, a more efficient variant is chosen as the representation of the original audio signal.

As mentioned before, it has been proposed to optimize compression performance by differentially encoding in frequency (and optionally time) and then selecting the more efficient variant. Then, the selected variant is signaled to the decoder by some side information.

The prior art techniques described above serve to reduce the amount of data that has to be transmitted eg in audio or video streams. Lossless encoding techniques using the previously described entropy-based encoding schemes typically result in a non-constant bit-rate bitstream.

Although these prior art techniques are suitable for significantly reducing the size of the data to be transferred, they all have a fundamental drawback. Since entropy coding mainly compresses information values that are considered to appear frequently in the data set to be compressed, a large number of rare parameters that appear continuously will lead to a large code length. Since this combination of parameters may sometimes appear in a complex data stream to be encoded, the generated bit stream generally has a section with a relatively high bit rate.

In these partitions, if the bit rate exceeds the maximum feasible bit rate of the transport medium, such as the maximum net data rate of a wireless link in streaming applications, the transmission of encoded data may stall Or even be interrupted, which is of course the worst.

Contents of the invention

The object of the present invention is to provide a concept for lossless coding of information values while ensuring a lower bit rate.

According to a first aspect of the invention, this object is achieved by an encoder for encoding an information value described by more than one bit to derive an encoded representation of the information value, said encoder comprising: a bit estimator adapted to estimate the number of information units required to encode an information value using a first encoding rule such that when encoding an information value, the would result in coded representations with a different number of information units, said second coding rule is such that, when coding an information value, would result in a coded representation with the same number of information units, wherein said coded representation starts with at least derived from an information value combination of two combined information values; and a provider adapted to provide a coded representation derived using coding rules that result in a smaller number of information units for the coded representation, and to provide an indication of the coded representation on which the coded representation is based The rule information of the encoding rules.

According to a second aspect of the invention, this object may be achieved by a rule for decoding a coded representation of an information value described by more than one bit and for processing a code indicating the code used to code the information value information, the decoder includes: a receiver for receiving the coded representation and the rule information; and a decompressor for decoding the coded representation, the decompressor depending on Deriving said information value based on said information rule using a first decoding rule such that said information value is derived from encoded representations having a different number of information elements, and using A second decoding rule, said second decoding rule being such that said information value is derived from an encoded representation having the same number of information values, wherein said information value has at least two combined information values from said encoded representation Exported from the combination of information values.

According to a third aspect of the invention, this object is achieved by a method for encoding an information value described by more than one bit to derive an encoded representation of said information value, said method comprising: using A first encoding rule and a second encoding rule to estimate the number of information units required to encode said information value, said first encoding rule being such that, when encoding an information value, will result in a different number of An encoded representation of an information unit, said second encoding rule being such that, when encoding an information value, will result in an encoded representation with the same number of information units, wherein said encoded representation is derived from an information value having at least two combinations and providing an encoded representation derived using an encoding rule that results in a smaller number of information units for the encoded representation, and providing rule information indicating the encoding rule on which the encoded representation is based.

According to a fourth aspect of the invention, this object is achieved by a computer program which, when run on a computer, performs the method described above.

According to a fifth aspect of the invention, this object may be provided by a rule for decoding a coded representation of an information value described by more than one bit and for processing a code indicating the code used to code the information value information, the method comprising: receiving the coded representation and the rule information; and depending on the rule information, decoding the coded representation using a first decoding rule or a second decoding rule, the A first decoding rule such that said information value is derived from coded representations with a different number of information elements and a second decoding rule such that said information value is derived from coded representations with the same number of information values The information value, wherein the information value is derived from a combination of information values in the encoded representation having at least two combined information values.

According to a sixth aspect of the present invention, the object is achieved by a computer program which, when run on a computer, performs the method described above.

According to a seventh aspect of the present invention, this object is achieved by an encoded representation of an information value, wherein said encoded representation comprises: a first part generated using a first encoding rule such that, when for encoding an information value would result in an encoded representation having a different number of information units; the second part is generated using a second encoding rule such that, when encoding an information value, would result in an encoded representation having an encoded representation of the same number of information units, wherein the encoded representation is derived from a combination of information values having at least two combined information values; and rule information indicating the encoding rule used.

The present invention is based on the discovery that when comparing a first coding rule that produces a coded representation of an information value with a variable length with a second coding rule that generates a coded representation of an information value with a fixed length, and when the choice results in the need When encoding rules for encoding representations of a small number of information units, a densely encoded representation of information values that does not exceed a predetermined size can be derived. Thus, it can be ensured that the maximum bit rate is at least the maximum bit rate of the second coding rule deriving the second coded representation. The selection of an encoding rule via a rule message is sent together with the encoded representation of the information value, after which the correct information value can be derived at the decoder using a decoding rule adapted to the encoding rule used during encoding.

The principle will be outlined in more detail in the following paragraphs, assuming that a properly designed variable length code matches the statistics of the information value to be encoded.

When applying entropy coding to quantized values, the actual requirements needed to represent a data set depend on what is known about the values to be encoded. In general, the more similar the values, the fewer bits will be consumed. Conversely, very dissimilar data sets require high bit rates. It can occur in this way that certain data blocks require a very high data rate, which is disadvantageous, for example in the case of a transmission channel with a limited transmission capacity.

The proposed method ensures a known upper bound on the bit requirements for encoding entropy-encoded datasets even at very rare values. In particular, the method ensures that the bit requirements do not exceed those using PCM codes. The encoding method can be summarized as follows:

• Encode the dataset using a conventional entropy (eg Huffman) encoding process. The resulting bit requirements are stored.

• Calculate bit requirements for PCM representation. Note that this is just multiplying the number of values to be encoded by the PCM code length, or a part of the PCM code length, so it is easy to calculate.

• If the bit requirement of the entropy coding exceeds that of the PCM coding, the PCM coding is selected and the selected PCM coding is signaled to the decoder with appropriate side information.

The decoding stage works in a corresponding manner.

In a preferred embodiment of the invention, entropy coding schemes are compared to PCM codes for coding quantized values.

In the above-described embodiments of the present invention, the maximum bit rate is limited by the word length of the PCM code. Therefore, in the case of knowing the word length, it is possible to advantageously design a system including an encoder, a transmission medium and a decoder, by selecting the transmission medium so that its transmission capacity exceeds the maximum bit rate defined by the PCM code, ensuring security operation.

In a second preferred embodiment, based on the previous embodiments of the invention, several information values are additionally combined into a single value that can be represented more efficiently using PCM coding, i.e., the single value has a value close to a power of 2 scope. Grouping is illustrated in more detail by the following example:

Quantized variable values in the range 0...4 (i.e. 5 possible different values) cannot be efficiently represented in PCM codes, because the minimum possible code length of 3 bits wastes 3 out of 2^3=8 possible values . Combining 3 of these variables (thus 5^3=125 possible combinations) into a single code of 7 bits length can significantly reduce the redundancy, since 5^3=125 is close to 2^7=128.

Therefore, a combined implementation of the method and the proposed concept for constraining an upper bound on bit requirements would use packet PCM coding (and fall-back coding) for determining an upper bound on the data rate as an alternative to PCM .

This combined implementation has the obvious advantage of further reducing the maximum bit rate.

Description of drawings

Preferred embodiments of the present invention are described below with reference to the accompanying drawings, wherein:

Fig. 1 shows the encoder of the present invention;

Figure 2 shows an example of bit estimation according to the inventive concept;

Figure 3a shows grouping of two information values prior to PCM encoding;

Figure 3b shows a grouping of three information values;

Fig. 4 shows the decoder of the present invention; and

Fig. 5 shows a multi-channel audio encoder according to the prior art.

Detailed ways

Figure 1 shows a block diagram of an inventive encoder for encoding information values, or deriving an encoded representation of an information value, ensuring a fixed maximum bit rate. The encoder 100 includes a bit estimator 102 and a provider 104 .

An information value 106 to be coded is input to a bit estimator 102 and a provider (104). In a possible implementation manner, the bit estimator 102 uses the first encoding rule and the second encoding rule to estimate the number of required information units. Information is available to provider 104 via rule data link 108 which encoding rule will result in an encoded representation requiring a smaller number of information units. The provider 104 then encodes the information value 106 using the obtained encoding rule and transmits at its output an encoded representation 110 together with rule information 112 indicating the encoding rule used.

In a modification of the previously described embodiment of the invention, bit estimator 102 encodes information value 106 using first and second encoding rules. Then, the bit estimator 102 calculates the information units required for the two coded representations, and transmits the coded representation with the smaller number of information units and the rule information to the provider 104 . The dashed data link 114 in Figure 1 represents a possible transfer of the encoded representation from the bit estimator to the provider. The provider 104 then forwards only the encoded representation to its output and additionally transmits the rules information 112 .

Figure 2 shows how the bit estimator 102 estimates the number of bits required to derive the encoded representation by comparing Huffman codes to PCM codes.

The Huffman codebook 120 is used to assign integer values 122 to codewords 124 represented by sequences of bits. It should be noted here that the Huffman codebook as simple as possible is selected here, so as to focus on the basic idea of the concept of the present invention.

The PCM codes used for comparison and to ensure a maximum constant bit rate include PCM codewords of length 4 bits, allowing 16 possible codewords, as shown in PCM specification 126.

In the simple example shown here, the information value 128 to be encoded is represented by 6 consecutive integers (011256), which means that there are only ten possible settings for each information value. Information value 128 is input to bit estimator 102, which uses a Huffman codebook (shown in Huffman partition 130 of bit estimator 102) and a PCM representation (shown in PCM partition 132 ), deriving the number of bits required to construct the encoded representation. It can be seen from Figure 2 that the entropy coded representation of information values requires 22 bits, while the PCM coded representation requires 24 bits, that is, the number of information values multiplied by the bit length of a single PCM codeword. In the case of Figure 2, the encoder of the present invention would take an entropy-coded representation of the information value and emit appropriate regular information output along with the entropy-coded representation.

Figures 3a and 3b illustrate the possibility of further reducing the maximum bit rate by advantageously grouping the information values 128 to form PCM-coded groups of information values.

Next, the same information value 128 as in Figure 2 is used to emphasize the impact of PCM grouping on the inventive concept of encoding an information value.

Also, since there are only ten possible settings for a single information value, it may be advantageous to group two consecutive information values into information value groups 140a to 140c before building the PCM representation of the combined value. This is possible because a 7-bit PCM code allows 128 different combinations, whereas a group comprising two arbitrary information values can only create 100 different combinations.

A single 7-bit PCM codeword 142a-142c is now assigned to each of the groups of information values 140a-140c. As can be seen in Figure 3a, applying the grouping strategy before building the PCM representation results in an encoded representation of the information value 128 with only 21 bits, comparable to the 24 bits required for the ungrouped PCM representation in Figure 2. In the above grouping strategy, each information value in the data stream consumes an average of 3.5 bits (7 bits/2 information values).

As shown in Figure 3b, the efficiency of grouping can be further increased by grouping 3 values into information value groups 146a to 146b. In this way, 1000 possible combinations can be formed, and can be covered by 10-bit PCM codes, as shown by PCM code words 148a and 148b in FIG. 3 . Therefore, the PCM representation requires only 20 bits, further reducing the average number of bits per information value to 3.33 (10/3).

It can be clearly seen that by grouping the values, there is a clear benefit to the bit rate required for encoding, since the example given in Figures 3a and 3b reduces the maximum bit rate by 12.5% (16.7%). Furthermore, applying grouping to the example in FIG. 2 would even cause the bit estimator 102 to make a different decision and signal a PCM code that produces a coded representation that requires a lower number of bits.

Fig. 4 shows a block diagram of a decoder according to the invention. The decoder 160 comprises a decompressor 162 and a receiver 163 for providing an encoded representation 110 and rule information 112 indicating the encoding rule used to encode the information value.

The decompressor 162 processes the rule information 112 to derive decoding rules suitable for deriving the information value 106 from the encoded information 110 .

Decompressor 162 then decompresses encoded representation 110 using the decoding rules and provides information value 106 at its output.

In the description of the preceding paragraphs, the concept of the present invention has been explained in detail by comparing an entropy coding scheme which produces variable bit length codes with a PCM coding scheme which produces constant bit length codes. The inventive concept is by no means limited to these kinds of codes that are compared during encoding. Basically, any combination of two or more codes is suitable for comparison and leads to an encoded representation of the information value that is as compact as possible, especially more compact than when only one code is used.

The invention is described in the context of audio coding, wherein parameters describing eg the spatial properties of an audio signal are encoded and decoded according to the concept of the invention. The concept of the invention, which ensures a maximum bit rate for encoded content, can also be advantageously used for any other parameter representation or information value.

It is especially suitable for implementations that entropy encode previously quantized parameters, since then the expected coding efficiency is higher. However, direct spectral representations of audio or video signals can also be used as input to the coding scheme of the present invention. In particular, when a signal is described by different parts of the signal that follow each other in time, where the time parts are described by parameters including the frequency representation of the signal, the above-mentioned encoding measures can be used both in frequency and in time. PCM grouping can also be applied to group parameters in time or frequency.

While the decoder of the invention described above derives information about which decoding rules to use to decode an encoded representation by means of rule information that signals the decoder to indicate the rules, in an alternative embodiment, the decoder 160 can also directly derive from the coded representation 110 which decoding rule to use, for example by identifying a particular bit sequence in the coded representation, with the advantage that auxiliary information signaling rule information can be omitted.

Depending on the specific implementation requirements of the method of the invention, the method of the invention can be implemented in hardware or software. The implementation may be performed using a digital storage medium, in particular a disc, DVD or CD, having stored thereon electrically readable control signals, operating with a programmable computer system to perform the method of the invention. In general, therefore, the present invention is a computer program product having program code stored on a machine-readable carrier, which program code can be used to carry out the method of the present invention when the computer program product is run on a computer. In other words, the method of the present invention is a computer program with a program code for performing at least one of the methods of the present invention when the computer program is run on a computer.

Although the above has been specifically shown and described with reference to specific embodiments, it will be understood by those skilled in the art that various other changes in form and details may be made without departing from the spirit and scope of the present invention. Change. It should be understood that various changes may be made to accommodate different embodiments without departing from the broader concepts disclosed herein and encompassed by the appended claims.

Claims (20)

1. encoder is used for by encoding greater than the described value of information of one bit, and to derive the coded representation of this value of information, described encoder comprises:

Bit estimator is suitable for using first coding rule and second coding rule to estimate the encode quantity of required information unit of the value of information,

Described first coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of varying number;

Described second coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of equal number, and wherein, described coded representation is derived from the value of information combination of the value of information with at least two combinations; And

Device is provided, is suitable for providing using causing the coded representation of deriving at the coding rule of the information unit of the lesser amt of coded representation, and provide the described coded representation of indication based on the Rule Information of coding rule.

2. encoder as claimed in claim 1, wherein, described first coding rule makes the quantity of the information unit that the coded representation of the fixed qty value of information is required depend on the Back ground Information value.

3. encoder as claimed in claim 1, wherein, described first coding rule makes that deriving coded representation comprises use Huffman code book.

4. encoder as claimed in claim 1, wherein, described second coding rule makes that deriving coded representation comprises to the single PCM code word of value of information combination distribution.

5. encoder as claimed in claim 1, wherein, described first coding rule make to be derived coded representation and is comprised that the differential coding of derived information value represents.

6. encoder as claimed in claim 5, wherein, described differential coding is represented to be in time or to carry out differential coding on the frequency.

7. encoder as claimed in claim 1, wherein, the described value of information comprises the parameter of describing frame of video or audio signal frame.

8. encoder as claimed in claim 1, wherein, the described value of information comprises the BCC parameter of the spatial property of describing multi-channel audio signal, and described BCC parameter is selected from following BCC parameter list:

ICC (inter-channel coherence/correlation)

ICLD (interchannel level difference)

ICTD (interchannel time difference)

IPD (inter-channel phase difference).

9. encoder as claimed in claim 1, wherein, described information unit is a bit.

10. decoder is used for being decoded by the coded representation greater than one the described value of information of bit, and is used to handle indication to the encode Rule Information of employed coding rule of this value of information, and described decoder comprises:

Receiver is used to receive described coded representation and described Rule Information; And

Decompressor is used for described coded representation is decoded, and described decompressor depends on described rule information and use the first decoding rule or the second decoding rule to derive the described value of information,

The described first decoding rule makes, derives the described value of information from the coded representation of information unit with varying number, and uses the second decoding rule;

The described second decoding rule makes, derives the described value of information from the coded representation of the value of information with equal number, and wherein, the described value of information has from described coded representation in the value of information combination of the value of information of two combinations at least to be derived.

11. decoder as claimed in claim 10, wherein, the described first decoding rule makes that deriving the described value of information comprises use Huffman code book.

12. decoder as claimed in claim 10, wherein, the described second decoding rule makes that deriving the described value of information comprises the value of information combination that has the value of information of at least two combinations to the PCM of described coded representation code assignment.

13. decoder as claimed in claim 10, wherein, the coded representation of the value of information is based on the value of information that comprises the parameter of describing single frame of video or audio signal frame.

14. decoder as claimed in claim 10, wherein, the coded representation of the value of information is based on the value of information of the BCC parameter that comprises the spatial character of describing multi-channel audio signal, and wherein said BCC parameter is selected from following BCC parameter list:

ICC (inter-channel coherence/correlation)

ICLD (interchannel level difference)

ICTD (interchannel time difference)

IPD (inter-channel phase difference).

15. one kind be used for to by encode greater than the described value of information of one bit, with the method for the coded representation of deriving the described value of information, described method comprises:

Use first coding rule and second coding rule to estimate to the encode quantity of required information unit of the described value of information,

Described first coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of varying number;

Described second coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of equal number, and wherein, described coded representation is derived from the value of information combination of the value of information with at least two combinations; And

Provide to use to cause the coded representation of deriving at the coding rule of the information unit of the lesser amt of coded representation, and provide the described coded representation of indication based on the Rule Information of coding rule.

16. one kind is used for by decoding greater than the coded representation of one the described value of information of bit and being used to handle indication to the encode method of Rule Information of employed coding rule of this value of information, described method comprises:

Receive described coded representation and described Rule Information; And

Depend on described Rule Information, use the first decoding rule or the second decoding rule that described coded representation is decoded,

The described first decoding rule makes, derives the described value of information from the coded representation of information unit with varying number, and uses the second decoding rule;

The described second decoding rule makes, derives the described value of information from the coded representation of the value of information with equal number, and wherein, the described value of information has from described coded representation in the value of information combination of the value of information of two combinations at least to be derived.

17. computer program with program code, when described program code moves on computers, carry out a kind of be used for to by encode greater than the described value of information of one bit, with the method for the coded representation of deriving the described value of information, described method comprises:

Use first coding rule and second coding rule to estimate to the encode quantity of required information unit of the described value of information,

Described first coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of varying number;

Described second coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of equal number, and wherein, described coded representation is derived from the value of information combination of the value of information with at least two combinations; And

Provide to use to cause the coded representation of deriving at the coding rule of the information unit of the lesser amt of coded representation, and provide the described coded representation of indication based on the Rule Information of coding rule.

18. computer program with program code, when described program code moves on computers, carry out a kind of being used for by decoding greater than the coded representation of one the described value of information of bit and being used to handle indication to the encode method of Rule Information of employed coding rule of this value of information, described method comprises:

Receive described coded representation and described Rule Information; And

Depend on described Rule Information, use the first decoding rule or the second decoding rule that described coded representation is decoded,

The described first decoding rule makes, derives the described value of information from the coded representation of information unit with varying number, and uses the second decoding rule;

The described second decoding rule makes, derives the described value of information from the coded representation of the value of information with equal number, and wherein, the described value of information has from described coded representation in the value of information combination of the value of information of two combinations at least to be derived.

19. the coded representation of a value of information, wherein, described coded representation comprises:

The first that uses first coding rule to produce, described first coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of varying number;

The second portion that uses second coding rule to produce, described second coding rule makes, when the value of information is encoded, will cause having the coded representation of the information unit of equal number, wherein, described coded representation is derived from the value of information combination of the value of information with at least two combinations; And

Rule Information is indicated employed coding rule.

20. coded representation as claimed in claim 19, described coded representation is stored on the computer-readable medium.

Images