MXPA00008636A - Efficient coding of side information in a lossless encoder - Google Patents

Abstract

For"Super Audio CD"(SACD) the DSD signals are losslessly coded, using framing, prediction and entropy coding. Besides the efficiently encoded signals, a large number of parameters, i.e. the side-information, has to be stored on the SACD too. The side information comprises the prediction filter coefficients and the probability table which have been used for encoding, and which should be transmitted to the decoder. The smaller the storage capacity that is required for the side-information, the better the overall coding gain is. Therefore coding techniques are applied to the side-information too so as to compress the amount of data of the side information. The frames can be segmented, each segment having its own set of filter coefficients and probability table.

Description

EFFICIENT CODING OF SIDE INFORMATION IN AN ENFORCED ENCODER FIELD OF THE INVENTION The invention relates to an apparatus for encoding without loss a digital information signal, for a lossless coding method, with an apparatus for decoding and with a record carrier.

BACKGROUND OF THE INVENTION For the "" DC of Superaudio (Compact Disc of Superaudio) (DCSA)) the SDI (Digital Information Signals) signals are encoded without losses, using coding by group selection, prediction and entropy. of the efficiently encoded signals, a large number of parameters, ie the lateral information, has to be stored in the DCSA as well.The smaller the storage capacity that is required for lateral information, the better the total coding gain. therefore, coding techniques are applied to the lateral information as well.A description of the lossless coding of SDI signals is given in the publication "Improved lossles coding of 1-bit audio signals", by F. Bruekers et al., preprinted 4563 (1-6) presented at the 103rd AES Convention, September 26-29, 1997 in New York.

BRIEF DESCRIPTION OF THE INVENTION The main purpose of the invention is to provide methods that can be used for example in DCSA to save the number of bits that have to be used to store lateral information. In the following description those methods will be presented.

BRIEF DESCRIPTION OF THE FIGURES These and other aspects of the invention will be better explained here later in the description of the figures, in which Figure 1 shows the diagram of a circuit of a lossless encoder and Figure Ib shows the diagram of a circuit of a corresponding decoder, which uses linear prediction and arithmetic coding, Figure 2 shows subsequent tables of a multi-channel information signal, • Figure 3 shows the segmentation of the time equivalent frames of the information signal of multiple channels, and Figure 4 shows the contents of a frame of the output signal of the coding apparatus.

DETAILED DESCRIPTION OF THE INVENTION The lossless coding and decoding process, for example of 1-bit oversampled audio signals, will be briefly explained below by means of Figure 1, which shows a mode of the coding apparatus in Figure 1 and shows one embodiment of the decoding apparatus in Figure Ib. The lossless coding in the apparatus of Figure 1 is performed on isolated parts (frames) of the audio signal. A typical length of such a frame is 37632 bits. The two possible bit values of the input signal F A 'and A' represent the sample values +1 and -1 respectively. By frame, the set of coefficients for the prediction filter zAA (z), denoted by 4, is determined in a unit "generating the coefficient of the filter 12, for example by the autocorrelation method." The sign of the output signal of the filter, Z, determines the value of the estimated bit Fp, while the magnitude of the filter output signal, Z, is an indication of the probability that the prediction is correct.After quantizing the output signal of the filter Z in a quantizer 10, an estimated input signal Fp is obtained, which is exorated in a combination unit 2, resulting in a residual signal E. A correct prediction, F = Fp, is equivalent to E = 0 in the signal residual E. The content of the probability table, p (|. |), is designated by frame, so that by possible value of Z, p0 is the probability that E = 0. For small values of | Z | the probability for a correct prediction is close to 0.5 and for large values from | Z | the probability for a correct prediction is close to 1.0. Clearly, the probability for an incorrect prediction, F? Fp or E = 1, is P ^ l-po. The probability tables for the tables (or segments, to be described later) are determined by unit 13. Using this probability table, provided by unit 13 to unit 8, unit 8 generates a probability value Po in response to its input signal, which is the Z signal. The arithmetic encoder (Arithmetic Coding CA) in the apparatus of Figure a, denoted by 6, encodes the bit sequence of E so that the code (D) requires fewer bits. For this, the arithmetic coder uses the probability that the bit n of the signal E, E [n], has a particular value. The number of bits to encode the bits E [n] = 0 is dn = -2log (p0) + M (bits) (Eq. 1) which is practically no greater than 1 bit, since p0 > 1/2 The number of bits to encode the E [n] = l bit is dn = -2log (p?) + M = -2log (l-p0) + M (bits) (Eq. 2) which is not less than 1 bit. The M in both equations represents the non-optimal behavior of the arithmetic coder, but can be neglected in practice. A correct prediction (E [n] = 0) results in less than 1 bit and an incorrect prediction (E [n] = l) results in more than one bit in the code (D). The probability table is designed so that on the average for the complete table, the number of bits for code D is minimal. In addition to the code D, also the coefficients of the prediction filter 4, generated by the coefficient generating unit 12, and the content of the probability table, generated by the unit that determines the probability table, 13 have to be transmitted from the encoder to the decoder. For that purpose, the coding apparatus comprises a multiplexer unit 14, which receives the output signal from the encoder 6, as well as the lateral information of the generating units 12 and 13. This lateral information comprises the coefficients of the prediction filter and the table of probability. The multiplexer unit 14 provides the serial data stream of information to a transmission medium, such as a record carrier.

In the decoding apparatus of Figure Ib, the inverse process of the encoder is implemented exactly, thereby creating a lossless coding system. The demultiplexing unit 20 receives a serial data stream comprising the data D and the lateral information. It retrieves the data D from it and provides the data D to an arithmetic decoder 22. The arithmetic decoder (DA Arithmetic Decoder) is provided with identical probabilities a. those provided to the arithmetic decoder, to recover the correct values of the signal E. Therefore, the demultiplexing unit retrieves the same coefficients of the prediction filter and probability table used by the coder of the received serial data stream and provides the coefficients from the prediction filter to the prediction filter 24 and the probability table to the generating unit of the probability value 26. The circuit constructions shown in Figure 1 are for the encoding / decoding of a flow of unique information data. The encoding / decoding of a multi-channel information signal, such as a multi-channel digital audio signal, requires that the processing described above be carried out with reference to Figure 1 at multiple times by the circuits of Figure 1 , or can be carried out by a plurality of such circuits. Another solution can be found in the international patent application IB 99/00313, which corresponds to no. Serial Number 09/268252 (PHN 16.805). It should be noted here that according to the invention, the coding apparatus may be devoid of quantizer Q and combination unit 2. Reference is made to the first patent publications discussing this. In the DCSA the 1-bit audio channels are divided into frames of constant length and the optimal per-frame strategy for coding will be used. The frames can be decoded independently of the neighboring frames. Thus, we can discuss the data structure within a single frame. Figure 1 shows equivalent frames of time B of two channel signals, such as the left and right components of the signal of a stereophonic digital audio signal, indicated by, B (l, ml), B (l, m), B (l, m + 1), for the "left component of the signal and for, B (r, ml), B (r, m), B (r, m + 1), for the right component of the signal The tables can be segmented, as will be explained later, if they are not segmented, the tables will be coded in their entirety, by a set of filter coefficients and a probability table for the whole table. a table can have its own set of filter coefficients and probability table.In addition, the segmentation in a table by the filter coefficients does not need to be the same for the probability tables.As an example, Figure 3 shows the two tables time equivalents B (l, m) and B (r, m) of the two channel signals that are being segmented. Table B (l, m) has been segmented into three segments fs (l, l), fs (l, 2) and fs (l, 3) to carry out three different prediction filtrations in the table. It should be noted, however, that filtering in two segments, such as the segments fs (l, l) and fs (l, 3) can be the same. Table B (l, m) has been further segmented into two segments ps (l, l) and ps (l, 2) to have two different probability tables for those segments. Picture. B (r, m) has been segmented into three segments fs (r, l), fs (r, 2) and fs (r, 3) to carry out three different prediction filtering in the table. However, it should be noted again that the filtering in two segments, such as the segments fs (r, l) and fs (r, 3) can be the same. Table B (r, m) has been further segmented into four segments ps (r, l), ps (r, 2), ps (r, 3) and ps (r, 4) to have four different probability tables for those segments. Again, it should be noted that some of the segments may have the same probability table.

The decision to have the same probability table for different segments can be taken in advance by the user of the device, after having carried out a signal analysis on the signals in the segments. Or the devices may be able to carry out this signal analysis and decide in response to this. In some situations, a signal analysis carried out on two segments can result in probability tables that differ only slightly. In such a situation, it can be decided to have one and the same probability table for both segments. This unique probability table could be equal to one of the two probability tables established for the two segments, or it could be an average version of both tables. An equivalent reasoning is valid for the sets of filter coefficients in the different segments. To summarize: to encode a small portion of audio in an audio channel signal, the coding algorithm in the DCSA requires both a prediction filter (the filter) and a probability table (the table). To improve the coding gain it can be efficient to use different filters in different channels. But also within the same channel it can be beneficial to use different filters. This is the reason why the concept of segmentation was introduced. A channel is divided into two segments and a particular filter is used in a segment. Several segments, also from other channels, can use the same or a different filter. Besides the storage of the filters that are used, also the information about the segments (segmentation) and information about which filter is used in which segment (projection) has to be stored. For the tables, the same idea is applicable, however, the segmentation and projection may be different from the projection and segmentation and projection for the filters. In the case of equal segmentation for both the filter and the table this is indicated. The same idea is used for the projection. If the segmentation for the filters is the same for the channels, this is also indicated. The same idea is used for the projection. First, it will give a description of the content of a frame of a transmission signal comprising the coded channel signals and the corresponding lateral information. Figure 4 shows a schematic drawing of the table. In addition to the synchronization information (not shown) the table comprises the two words wi and w2, followed by segmentation information on the prediction filters. Next, a word w3 is present followed by segmentation information on the probability tables. Then follow two words 4 and w5, followed by projection information on the prediction filters. Then follow a word 6, followed by projection information on the probability tables. Then follow the filter coefficients and the probability tables, provided by the generating units 12 and 13, respectively. The table ends with the data D, provided by the arithmetic coder 6. The word wi is in this example a bit in length and may have the value of "0" or "1", and indicates whether the segment information for the Filter coefficients and probability tables are the same ("1"), or not ("0"). The word w4 is in this example one bit long and may have the value of "0" or "1", and indicates whether the projection information for the filter coefficients and the probability tables are the same ("1" ), or not ("0"). Word 2, again one bit long, can have the value of "0" or "1", and indicates whether the channel signals have the same segmentation information for the projection filter coefficients ("1") , or not ("0"). The word w3 (one bit long) may have the value of "0" or "1", and indicates whether the channel signals have the same segmentation information for the probability tables are the same ("1"), or not ("0"). The word w5 may have the value of "0" or "1", and indicates whether the channel signals have the same projection information for the coefficients of the prediction filter ("1"), or not ("0"). The word w6 may have the value of "0" or "1", and indicates whether the channel signals have the same projection information for the probability tables ("1"), or not ("0"). First, the representation of the total number of S segments in a table will be described. To encode a number, for example the total number of segments in a frame in a particular channel signal,. a type of execution length coding is applied. It is important that the code is short for small values of S. Since the number of segments in the channel S >; 1, S = 0 does not need to be encoded. In the DCSA, the following codes were used. Table 1: Observation: Here the "1" was used as a delimiter. It is clear that in general the role of "0" or "1" can be exchanged. The basic idea of the delimiter is that a certain sequence is violated; the sequence "0" is violated by "1". An alternative is for example to "invert" the next symbol and the "no inversion" is used as a delimiter. In this way, long constant sequences are avoided. An example of investment sequences that begins with a "1" is (not used in the DCSA): In the second place, the representation of the segment sizes will be described. The length of a segment will be expressed in the number of bytes of the channel signal. The B bytes in a frame of a channel signal are divided into segments S. For the first S-l segments the number of bytes of each segment has to be specified. For the 5th segment, the number of bytes is implicitly specified, this is the remaining number of bytes in the channel. The number of bytes in segment i is equal to B so that the number of bytes in the last segment is: Bs-? = B- B-i Since the number of bytes in the first Sl segments is a multiple of R, the resolution R = 1, we will define it as: Bi = b ± R and consequently: Bs-? ~ = B b-iR The values of Sl of A are stored and R is stored in a channel only if S > 1 and when it is not already stored for another channel. The number of bits that are required to store for b depends on their possible values. 0 < I < bi? max with, for example bi, max = so that the number of bits required to store i is: #bit s (jbi) = L2log (Jbl f ma?) J + 1 This has the advantage that the required number of bits for segment length may decrease for segments at the end of the frame. If restrictions are imposed, for example, on the minimum length of a segment, the calculation of the number of bits can be adapted accordingly. The number of bits to store the resolution R is: # bits (R) In the third place, the representation in the serial data flow will be described. The representations given above will be used under table 1. This will be illustrated by some examples. To distinguish between filters and probability tables, the subscripts f and t were used. To distinguish between segments in different channels, a double argument was used: (channel number, segment number). Next, a first example is given. For a case of 2 channels, we have different segmentations for the filters and. probability tables, and different segmentations for both channels. The following table shows the parameters in the flow.

In the previous table 2, the first combination (Yi r Y2) r equal to (0,1) is the code of the code word (S) in table 1 above, and indicates that in the channel signal numbered 0 as the The table is divided into two segments for purposes of prediction filtering. In addition, the combination (y?, Y2, y3) equal to (0,0,1) is the code of the code word (S) in table 1 above, and indicates that in the channel signal numbered 1 as the box is divided into three segments for the purpose of prediction filtering. Next, we find a combination (yi) equal to (1), which is the first code word in table 1, which indicates that channel signal numbered 0, the table is not divided by the probability table. Finally, we find a combination (y?, Y2, y3) equal to (0,0,1), which indicates, that the box of the second channel signal is divided into three segments, each with a corresponding probability table . Next, follow another example for a case of five channels. It was assumed that for this case of five channels, we have a segmentation for filters and tables, and that segmentation is the same for all channels.

Remark: The unique bits of the code (S) interspersed in the segmentation information can be interpreted as "another segment will be specified" in the case of "0" or "no more segments will be specified in the case of" 1". , the projection will be described For each of the segments, all the segments of all the channels are considered together, it must be specified which filter or table is used.; first the segments of channel 0 followed by the segments of channel 1 and so on. The filter or table number for segments s, N (s) is defined as: N? Mn () with Nma? (s), being the maximum number allowed for a given segment, defined as: Nmax (s) = l + max (N (i)) with 0 < i < s The number of bits that are required to store N (s) and equal to: The number of bits that is required to store a filter or table number according to this method depends on the set of numbers that has already been assigned. If the tables use the same projection as the filters, which is not always possible, this is indicated. Also when all channels use the same projection this is indicated. With two examples the idea will be illustrated.

Example 3 Assume that in total we have 7 segments (0 up 6), some segments use the same filter and some use a unique filter. Also assume that tables use the same projection specification as filters.

Number Number Number Number of #bits channel segment filter filter possible 0 or 1 0 or 1 0, 1 or 2 0, 1, 2, or 3 0, 1, 2, 3 or 4 0, 1, 2, 3 or 4 #bits Total 12 Segment number 0 uses filter number 0 by definition, so bits are not required for this specification. Segment number 1 can use a filter assigned to the beginning (0) or the next major filter (1) not yet designated, so 1 bit is required for this specification. Segment number 1 uses filter number 0 in this example. Segment number 2 can use a filter assigned to the start (0) or the next major filter (1) not yet assigned, so 1 bit is required for this specification. Segment number 2 uses filter number 1 in this example. Segment number 3 can use a filter assigned at the beginning (0 or 1) or the next highest filter (2) not yet assigned, so 2 bits are required for this specification. Segment 3 uses filter number 2 in this example. Segment number 4 can use a filter designated at the beginning (0, 1 or 2) or the next highest filter (3) not yet assigned, so 3 bits are required for this specification. Segment number 4 uses filter number 3 in this example. Segment number 5 can use a filter assigned at the beginning (0, 1, 2, or 3) or the next largest filter (4) not yet assigned, so 3 bits are required for this specification. Segment number 5 uses filter number 3 in this example. Segment number 6 can use a filter assigned to the start (0, 1, 2 or 3) or the next larger filter (4) not yet assigned, so 3 bits are required for this specification. Segment number 6 uses filter number 1 in this example. In total 12 bits are required to store the projection. The total number of segments (7 segments in this example) is known at this point in the flow.

Another example. Assume that in total we have 6 channels each with a segment and each segment uses the same prediction filter and the same probability table.

In total 2 bits are required to store the full projection. Remark: One reason to give the indication that the following specification is also used for another application (for example, for the tables the same segmentation is used as for the filters) is that this simplifies the decoder. Although the invention has been described with reference to preferred embodiments thereof, it should be understood that these are not limiting examples. Thus, different modifications may become apparent to those skilled in the art without departing from the scope of the invention as defined in the claims. As an example, the invention could also have been incorporated in a modality in which signal blocks and time equivalents are encoded, without making use of segmentation. In such an embodiment, the serial data stream obtained, as well as the data stream of Figure 4, would be devoid of the segmentation information described here for filters and probability tables, as well as some of the indicator words, such as the words wi, w2 and w3. In addition, the invention underlies each and every one of the novel features and combinations of such features.

Claims (41)

1. CHAPTER CLAIMEDICATORÍO Having described the invention, it is considered as a novelty and, therefore, the content is claimed in the following CLAIMS: I .- An apparatus for encoding a digital information signal, such as a digital audio signal of n channels, where n is an integer greater than 1, characterized in that it comprises: input means for receiving the digital information signal, - A coding means for encoding the digital information signal to obtain a digital encoded information signal, the encoding means is adapted to encode each of the channel signals of the n-channel digital audio signal to obtain a channel signal encoded by each of the channel signals in response to probability values for each of the channel signals, prediction filter means for carrying out a prediction filtering on each of the channel signals of the audio signal digital of n channels in response to a set of prediction filter coefficients for each channel signal to obtain a channel signal from the filter d e prediction of each of the channel signals, - means for determining the coefficient of the prediction filter to generate a set of coefficients of the prediction filter for each of the channel signals, means for determining the probability value for generating values of probability for each of the channel signals in response to a probability table for each of the channel signals and the channel signal filtered by corresponding prediction for each of the channel signals, - means for determining the table of probability to generate the probability tables for each of the channel signals, conversion means for generating the first projection information and a probability of m sets of prediction filter coefficients, where m is an integer for which have that 1 < m < n, the first projection information and the m sets of the projection filter coefficients are representative of the n sets of prediction filter coefficients or each n channels, and to generate the second projection information and a plurality of p tables of probability, where p is an integer for which we have 1 < p < n, the second projection information and the p probability tables are representative of the n probability tables for the n channels, combining means for combining the compressed digital information signal, the first and second projection information signals, the plurality of m sets of prediction filter coefficients and the plurality of p probability tables in a composite information signal, output means for transmitting the composite information signal.
2. 2. An apparatus for encoding a digital information signal, such as a n-channel digital audio signal, where n 'is an integer greater than 1, characterized in that it comprises: input means for receiving the digital information signal, - encoding means for encoding the digital information signal to obtain a digital encoded information signal, the encoding means is adapted to encode time-equivalent signal blocks of each of the channel signals of the digital audio signal of n channels by dividing the signal blocks equivalent in time in M segments, and by coding the signal portions of the channel signals in all the M segments in the block of time equivalent signals, to obtain a portion of the signal coded by each of the portions of the signal in the M segments in response to probability values for each of the portions of? = pl the signal, do nde M = ¿LsP? Y sP is e ^ number of segments in / = 0 the block of signals equivalent in time of the ith channel signal, - means that determine the probability value to generate probability values for each of the M portions of the signal in response to a probability table for each of the M portions of the signal,. means to determine the probability table to generate the probability tables for each of the M portions of the signal, conversion means to convert the information about the length and location of the M segments in the n channel signals of the first information of the segment, and to generate the first projection information and a probability of m probability tables, where m is an integer for which we have 1 < m < M, the first projection information and the m probability tables are representative of the M probability tables, means for combining the portion of the encoded digital information signal comprised in blocks of time equivalent signals, the first segment information, the first projection information signal and the plurality of m probability tables in a composite information signal, output means for transmitting the composite information signal. The apparatus according to claim 2, characterized in that it also comprises prediction filter means for carrying out a prediction filtering on the digital information signal to obtain a digital information signal filtered by prediction, the filter means of prediction are adapted to filter by prediction blocks of time-equivalent signals of each of the channel signals of the digital audio signal of n channels by dividing the blocks of equivalent signals in time into segments, and filter by prediction the portions of the Signal of the channel signals in all P segments in the blocks of time-equivalent signals, to obtain a portion of the filtered signal by prediction of each of the P portions in response to a set of prediction filter coefficients for each one of the portions of the signal, where P = and sf ± is the number of segments in the block of signals equivalent in time of the ith channel signal, - means to determine the coefficient of the prediction filter to generate a set of coefficients of the prediction filter for each One of the P portions of the signal, the conversion means are further adapted to convert the information about the length and location of the P segments into the n channel signals into a second segment information, and to generate the second segment information. information of the projection and a plurality of p sets of coefficients of the projection filter, where p is an integer for which we have 1 < p < P, the second projection information and the P sets of prediction filter coefficients are representative of the P sets of prediction filter coefficients, the combining means are further adapted to combine the second information of the segment, the second signal of projection information and the plurality of p sets of predictive filter coefficients in the composite information signal. 4. The apparatus according to claim 3, characterized in that the conversion means are generated to indicate a first indicator word (wi) of a first value, which indicates that the segmentation of the blocks of signals equivalent in time for the probability tables is different from the segmentation of the blocks of time-equivalent signals for the sets of coefficients of the prediction filter and of a second value that indicates that the segmentation of the blocks of time-equivalent signals for the probability tables is the same as for the prediction filter coefficients, and to provide only one of the first or second segment information in the latter case, the combining means are adapted to combine the first indicator word and only one of the first segment information or the second segment information. segment in the composite information signal, in the event that the first pal Indicator open has the second value. The apparatus according to claim 4, characterized in that the conversion means are adapted to generate the one of the first or second information of the segment in the event that the first indicator word has the second value. The apparatus according to claim 3, characterized in that the conversion means are adapted to generate a second indicator word (w2) of a third value indicating that the blocks of time equivalent signals all have the same segmentation for the sets of coefficients of the prediction filter and are adapted to generate a second word indicating a fourth value indicating that the blocks of time equivalent signals each have a different segmentation for the sets of coefficients of the prediction filter, so that the conversion means are adapted to generate the second segment information for only one block of time equivalent signals in case the second indicator word has the third value and are adapted to generate the second segment information for each of the signal blocks equivalent in time in case the second indicator word has the fourth value, and that the combining means are further adapted to combine the second indicator word in the composite information signal. The apparatus according to claim 2, characterized in that the conversion means are adapted to generate a third indicator word (w3) of a fifth value indicating that the blocks of time equivalent signals all have the same segmentation for the probability tables and are adapted to generate a third word indicating a sixth value indicating that the blocks of time-equivalent signals each have a different segmentation for the probability tables, that the conversion means are adapted to generate the first information of the segment for only one block of time-equivalent signals in the event that the third indicator word has the fifth value and are adapted to generate the first segment information for each of the blocks, of time equivalent signals in the case of that the third indicator word has the sixth value, and that the means of combination they are further adapted to combine the third indicator word in the composite information signal. The apparatus according to claim 3, characterized in that the conversion means are adapted to generate a fourth indicator word (w4) of a seventh value, which indicates that the projection information for the probability tables is different from the information of projection for the coefficients of the prediction filter and of an eighth value indicating that the projection information for the probability tables is the same as for the prediction filter coefficients, and for providing the first and second projection information only in In the latter case, the combining means are adapted to combine the fourth indicator word and the first or second projection information only in the composite information signal, in case the fourth indicator word has the eighth value. The apparatus according to claim 3, characterized in that the conversion means are adapted to generate a fifth indicator word (w5) of a ninth value indicating that the blocks of signals, equivalent in time, all have the same projection information for the sets of the prediction filter coefficients and are adapted to generate a fifth indicator word of a tenth value indicating that the blocks of time equivalent signals each have different projection information for the sets of filter coefficients of prediction, because the conversion means are adapted to generate the second projection information only for a block of time equivalent signals in case the fifth indicator word has the ninth value and are adapted to generate the second projection information for each one of the signal blocks equivalent in time in the case that the fifth indicator word has the tenth value, and because the combination means are further adapted to combine the fifth indicator word in the composite information signal. 10. The apparatus according to claim 2 or 3, characterized in that the conversion means are further adapted to convert information related to the number of segments in a block of time-equivalent signals of a channel signal in a numerical code, the Combination means are further adapted to combine the numerical code into the composite information signal. 1, The apparatus according to claim 10, characterized in that the numerical code satisfies the following table where S is the number of segments in a block of time-equivalent signals of a channel signal. 12. The apparatus according to claim 3, characterized in that the first set of prediction filter coefficients is assigned to the first of the P segments, the second projection information is devoid of the projection information for projecting the first set of filter coefficients. of projection for the first segment of P segments, (a) the first bit in the second projection information .indicates whether the set of coefficients of the prediction filter for the second segment is the first set of coefficients of the prediction set or a second set of coefficients of the prediction filter, (bl) if the first set of coefficients of the prediction filter is also the set of filter coefficients for the second segment, then the second bit in the second projection information indicates whether the set of coefficients of the prediction filter for the third segment is the first set of coefficient s of the prediction filter or the second set of coefficients of the prediction filter, (b2) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second segment, then the next two bits in the second information of projection indicate whether the set of coefficients of the prediction filter for the third segment is the first, second or third set of coefficients of the prediction filter, (cl) if the first set of coefficients of the prediction filter is the set of coefficients of the prediction filter. filter for the second and third segments, then the third bit of the second projection information indicates whether the set of coefficients of the prediction filter for the fourth segment is the first or second sets of coefficients of the prediction filter, (c2) if the first set of coefficients of the prediction filter is the set of filter coefficients for the second segment and the s The second set of coefficients of the prediction filter is the set of filter coefficients for the third segment, then the third and fourth bits of the second projection information indicate whether the set of coefficients of the prediction filter for the fourth segment is the first, second or the third set of coefficients of the prediction filter, (c3) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second segment, and the first or second set of filter coefficients is the set give filter coefficients for the third segment, then the third and fifth bits in the second projection information indicate whether the set of coefficients of the prediction filter for the fourth segment is the first, second or third set of coefficients of the prediction filter , (c4) if the second set of coefficients of the prediction filter is the set of filter coefficients or for the second segment and the third set of filter coefficients is the set of coefficients of the prediction filter for the third segment, then the fourth and fifth bits in the second information indicate whether the set of coefficients of the prediction filter for the fourth segment is the first, second, third or fourth sets of coefficients of the prediction filter. 13. The apparatus according to claim 2, characterized in that the first probability table is assigned to the first of M segments, the first protection information is devoid of the projection information for projecting the first probability table into the first segment of the M segments, (a) the first bit in the first projection information indicates whether the probability table for the second segment is the first probability table or a second probability table, (bl) if the first probability table is also the probability table for the second segment, then the second bit of the first projection information indicates whether the probability table for the third segment is the first probability table or the second probability table, (b2) if the second probability table is the probability table for the second segment, then the next two bits in the first projection information indicate whether the probability table for the third segment is the first, second or third probability table, (cl) if the first probability table is the probability table for the second and third segments, then the third bit in the first forecast information indicates whether the probability table for the fourth segment is the first or second probability table, (c2) if the first probability table is the probability table for the second segment and the second probability table is the probability table for the third segment, then the third and fourth bits in the first projection information indicate whether the probability table for the fourth segment is the first, second or third probability table,, (c3) if the second probability table is the probability table for the second segment, and the first or second probability table is the probability table for the third segment, then the fourth and fifth bits in the first injection information indicate whether a probability table for the fourth segment is the first, second or third probability table, (c4) if the second probability table is the probability table for the second segment, and the third probability table is the probability table for the third segment, then the fourth and fifth bits in the first projection information indicate whether the probability table for the fourth segment is the first, second, third or fourth probability table. The apparatus according to any of the preceding claims, characterized in that the output means comprises writing means for writing the composite information signal on a record carrier. 15. The apparatus according to claim 14, characterized in that the output means further comprises channel coding means and / or error correction decoding to carry out a channel coding step and / or a step encoding error correction on the composite information signal before writing the composite information signal on the record carrier. 16. A method, characterized in that it carries out the coding of a digital information signal, such as a digital audio signal, in an apparatus according to any of claims 1 to 15. 17. The method according to claim 16, characterized in that it also comprises the step of writing the composite information signal on a record carrier. 18. A record carrier, characterized in that it comprises the composite information signal generated by the apparatus according to any of claims 1 to 15, in a track on a record carrier. 19. An apparatus for decoding an encoded composite information signal comprising encoded data of a n-channel digital information signal, such as a n-channel digital audio signal, where n is an integer greater than 1, and lateral information that has a relationship with the _ encoded digital information signal, the apparatus is characterized by comprising input means for receiving a composite information signal, - recovery means for retrieving encoded data information and lateral information from the composite information signal, decoding means for decoding encoded data information to obtain channel signals in response to a set of probability values for each of the channel signals, prediction filter means to carry out the prediction filtering on each of the channel signals of the digital audio signal of n channels in response to sets of prediction filter coefficients, one set for each of the channel signals, to obtain a filtered channel signal by predicting each of the channel signals, the sets of coefficients of the prediction filter are derived from lateral information, means generating the probability value to generate n conju In addition to the probability values, one for each of the channel signals in response to a channel signal filtered by corresponding prediction and the corresponding probability table, the n probability tables, one for each of the channel signals, are derived from the lateral information, the recovery means are further adapted to recover the first and second projection information, a plurality of n sets of prediction filter coefficients and a plurality of p probability tables of the lateral information, means for reconversion to convert the first projection information and the m sets of prediction filter coefficients into n sets of prediction filter coefficients, one for each of the channel signals, where m is an integer for which one has what 1 <; m < n, and to convert the second projection information and the p probability tables into n probability tables, one for each of the channel signals, where p is a whole number for which one has to 1 < p < n, output means for transmitting the n channel signals. 20. An apparatus for decoding an encoded composite information signal comprising encoded data of a n-channel digital information signal, such as a n-channel digital audio signal, where n is an integer greater than 1, and the information side has a relation to the encoded digital information signal, the apparatus is characterized in that it comprises: input means for receiving a composite information signal, • - recovery means for recovering coded data information in the lateral information of the signal composite information, decoding means for decoding the data information encoded in M signal proportions in response to the corresponding sets of probability values, one for each of the M portions of the signal, where M = Yip (and spx is the number of segments in the / = 0 block of equivalent signals in time of the ith channel signal, - means of generating the probability value to generate M sets of probability values, one for each of the M portions of the signal in response to a corresponding probability table, the M probability tables, one for each of the portions of the signal, are derived from the lateral information, the means of recovery are adapted, in addition, to retrieve the first information of the segment and the first projection information and a plurality of M probability tables of the lateral information, where M is an integer for which we have 1 <m <M, means of reconversion to convert the first projection information and m probability tables into "" M probability tables, one for each of the portions of the signal, and to reconvert the first information of the segment in information about the length and location of the M segments in the n channel signals, to obtain the blocks of time equivalent signals in the n channel signals, means of output to transmit the blocks of signals equivalent in time of the n channel signals. 21. The apparatus according to claim 20, characterized in that it further comprises prediction filter means for carrying out prediction filtering on the signal blocks equivalent in time of each of the channel signals of the signal of digital information of n channels by dividing the blocks of equivalent signals in time in segments, and a filtering by prediction of the portions of the signal of the channel signals in all the P segments in the blocks of signals equivalent in time and for all the n channel signals, to obtain a portion of the signal filtered by prediction by each of the P portions of the signal in response to a set of prediction filter coefficients for each of the portions of the signal, where P ? ^ / T And yes? is e ^ number of segments in the block of equivalent signals in time of the same channel signal, the recovery means are also adapted to recover the second segment information, the second projection information and the P sets of the lateral information prediction filter coefficients, where P is a whole number for which we have 1 < p < P, the reconversion means are further adapted to convert the second segment information into information about the length and location of the P segments in the n channel signals and to retrieve the p sets of prediction filter coefficients in P sets of prediction filter coefficients, one for each of the P portions of the signal, using the second projection information. 22. The apparatus according to claim 21, characterized in that the recovery means are adapted to retrieve a first indicator word (wi) of the lateral information., the first indicator word, when it is of a first value, indicates that the segmentation of the blocks of time equivalent signals for the probability tables is different to the segmentation of the blocks of time equivalent signals for the prediction filter coefficients , and when it is of a second value, indicates that the segmentation of the blocks of time-equivalent signals for the probability tables is the same as for the coefficients of the prediction filter, and to retrieve a segment information only for lateral information in In the latter case, the recovery means are also adapted to copy the segment information to obtain the first and second segment information, in the latter case. 23. The apparatus according to claim 21, characterized in that the recovery means are adapted to recover a second indicator word (w2) from the lateral information, the second indicator word, when it is a third value, indicates that the equivalent signal blocks in time they all have the same segmentation for the prediction filter coefficients and, when it is a fourth value, it indicates that the blocks of time-equivalent signals each have a different segmentation for the prediction filter coefficients, the means of recovery they are further adapted to recover the second segment information for only one block of time-equivalent signals from the lateral information in the case where the second indicator word has the third value and are adapted to retrieve the second segment information by each of the signal blocks equivalent in time in the case that the second indicator word has the fourth value, and the reconversion means are further adapted to copy the second segment information nl times to obtain the P segments of the signal blocks equivalent in time for all n channel signals, in the if the second indicator word has the third value. 24. The apparatus according to claim 21, characterized in that the recovery means are adapted to recover a third indicator word (w3) from the lateral information, the third indicator word, when it is a fifth value, indicates that the equivalent signal blocks in time they all have the same segmentation for the probability tables, and when it is of a sixth value, it indicates that the blocks of signals equivalent in time each have a different segmentation for the probability tables, the means of recovery are adapted, in addition , to retrieve the first segmentation information for only one block of time equivalent signals in the event that the third indicator word has the fifth value and are adapted to retrieve the first segment information for each of the equivalent signal blocks in time in the event that the third indicator word has the sixth value, and that The conversion means are also adapted to copy the first segment information of a block of equivalent signals in time n times to obtain the M segments of the blocks of signals equivalent in time for all n channel signals, in the case that the third indicator word has the fifth value. 25. The apparatus according to claim 21, characterized in that the recovery means are adapted to retrieve a fourth indicator word (w4) from the lateral information, the fourth indicator word is from a seventh value, indicating that the projection information for the probability tables is different from the projection information for the prediction filter coefficient sets and, when it is an eighth value, it indicates that the projection information for the probability tables is the same as for the prediction filter coefficients , that the recovery means are also adapted to recover only a projection information of the lateral information in the latter case, the recovery means are also adapted to -copy the recovered projection information in the case that, the fourth indicator word has the eighth value. 26. The apparatus according to claim 21, characterized in that the recovery means is adapted to retrieve a fifth indicator word (5) from the lateral information, the fifth indicator word, when it is a ninth value, indicates that the blocks of Equivalent signals in time all have the same projection information for the prediction filter coefficients and, when it is of a tenth value, it indicates that the blocks of signals equivalent in time each have different projection information for the coefficients of the prediction filter, that the recovery means are also adapted to recover the second projection information for only a block of time equivalent signals in case the fifth indicator word has the fifth value and are adapted to recover the second projection information of each of the blocks of time equivalent signals in the case of the fifth word indicator has the tenth value. 27. The apparatus according to claim 20 or 21, characterized in that the conversion means are further adapted to convert recovered information to a numerical code for a block of equivalent signals 4 in time of lateral information, the numerical code represents a number of segments in the signal block equivalent in time. The apparatus according to claim 27, characterized in that the numerical code satisfies the following table where S is the number of segments in a block of time-equivalent signals of a channel signal. The apparatus according to claim 1, characterized in that the recovery means are adapted to recover a plurality of sets of coefficients of the lateral information prediction filter and to recover a bit array of the second projection information, the The apparatus further comprises allocation means for assigning the first set of prediction coefficients to the first of the P segments, (a) the allocation means are further adapted to assign the first set of coefficients of the prediction filter to the second segment in response to the first bit in the bit array that is of a first binary value and is adapted to assign the second set of filter coefficients projection to the second segment in response to the first bit that is of the second binary value, (bl) if the first set of coefficients is also the set of filter coefficients for the second segment, then the allocation means are adapted, in addition, to assign the first set of prediction filter coefficients to the third segment in response to the second bit in the array of bits that are of a first binary value and are adapted to assign the second set of coefficients of the prediction filter to the third segment in response to the second bit that is of the second binary value, (b2) if the second set of coefficients is the conj In addition to filter coefficients for the second segments, then the allocation means are further adapted to assign either the first or the second or the third set of prediction filter coefficients to the third segment in response to the values of the following two bits in the array of bits, (cl) if the first set of coefficients is the set of filter coefficients for the second and third segments, then the allocation means are further adapted to assign either the first or the second set of filter coefficients to the fourth segment in response to the value of the third bit in the bit array, (c2) if the first set of eoefficients of the prediction filter is the set of filter coefficients for the second segment and the second set of coefficients filter is the set of filter coefficients for the third segment, then the allocation means are also adapted to assign any of the first or the second or the third set of coefficients of the prediction filter to the four segments in response to the values of the third and fourth bits in the bit array, (c3) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second segment and the first or second sets of filter coefficients is the set of filters for the third segment, then the allocation means are adapted to assign any of the first, second or third set of coefficients of the filter to the four segments in response to the values of the fourth and fifth bits. in the bit array, (c4) if the second set of prediction filter coefficients is the set of filter coefficients for the second segment, and the third set of filter coefficients is the set of filters for the third segment, then the allocation means are adapted to assign any of the p first, or the second or third or fourth set of filter coefficients to the fourth segment in response to the fourth and fifth in the bit array. 30. The apparatus according to claim 20, characterized in that the recovery means are adapted to retrieve a plurality of probability tables from the lateral information and to recover an array of bits from the first projection information, the apparatus further comprises means of assignment to assign the first probability table to the first of the M segments, (a) the allocation means are further adapted to assign the first probability table to the second segment in response to the first bit in the bit array that is of a first binary value and are adapted to assign the second probability table of the second segment in response to the first bit that is of the second binary value, (bl) 'if the first probability table is also the probability table for the second segment, then the allocation means are adapted to assign the first probability table to the third segment in response to the second bit in the array of bits that are of a first binary value and are adapted to assign the second probability table to the third segment in response to the second bit that is of a second binary value, (b2) if the second probability table is the probability table for a second segment, then the allocation means are also adapted to assign any of the first or second or third probability table to the third segment in response to the values of the next two bits of the bit array, (cl) if the first probability table is the probability table for the second and third segments, then the allocation means are further adapted to assign any of the first or the second probability table to the fourth segment in response to the value of the third bit in the bit array, (c2) if the first probability table is the probability table for the second segment and the second probability table is the probability table for the third segment, then the allocation means are further adapted to assign either the first, or the second or- the third probability table to the fourth segment in response to the values of the third and fourth bits in the bit array, (c3) if the second probability table is the probability table for the second segment, and the first q the second probability table is the probability table for the third segment, then the allocation means are adapted to assign either the first, or the second or third probability table to the fourth segment in response to the values of four and fifth bits in the bit array, (c4) if the second probability table is the probability table for the second segment, and the third probability table is the probability table for the third segment, then the allocation means are adapted to assign any of the first, or the second or third or fourth probability table to the fourth segment in response to the fourth and fifth bits in the bit array. The apparatus according to any of claims 19 or 30, characterized in that the input means comprises reading means for reading the information signal composed of a record carrier. 32. The apparatus according to claim 31, characterized in that the input means further comprise means for decoding and / or correcting errors to carry out a channel decoding step and / or a correction step of errors on the composite information signal before providing , the composite information signal to the means of recovery. 33. The apparatus according to claim 1, characterized in that the coding means are adapted to encode blocks of signals equivalent in time of each of the channel signals of the information signal of n channels, to obtain blocks of equivalent signals in time coded by each of the signal blocks in response to probability values for each of the signal blocks, the prediction filter means are adapted to carry out the prediction filtering on each of the blocks of Equivalent signals in time in response to the n sets of prediction filter coefficients, one for each block of time-equivalent signals, the means they determine. the probability table are adapted to generate the n probability tables, one for each block of time equivalent signals.
3. 3 . The apparatus according to claim 33, characterized in that the conversion means are adapted to generate a first indicator word (W4) of a first value, which indicates that the projection information for the probability tables is different from the information of projection for the prediction filter coefficients and a second value indicating that the projection information for the probability tables is the same as for the prediction filter coefficients, and for providing the first or second projection information only in the In the latter case, the combining means are adapted to combine the first indicator word and the first information or the second projection information only to the compound information signal, in the event that the first indicator word has the second value. 35. The apparatus according to claim 33, characterized in that the conversion means are adapted to generate a second indicator word (w5) of a third value indicating that the blocks of time equivalent signals all have the same projection information for the sets of the prediction filter coefficients and are adapted to generate a second word indicating a fourth value indicating that the blocks of signals are equal. Before each time have a different projection information to that of the sets of filter coefficients of prediction, that the. conversion means are adapted to generate the second projection information only for a block of equivalent signals, in time in case the second indicator word has the third value and are adapted to generate the second projection information for each of the signal blocks equivalent in time to the value of the second indicator word having the fourth value, and that the combining means are further adapted to combine the second indicator word in the composite information signal. 36. The apparatus according to claim 33, characterized in that the first set of prediction filter coefficients is assigned to the first of the n signal blocks equivalent in time, the second projection information is devoid of the projection information to project the first set of coefficients of the projection filter to the first block of time equivalent signals of the n time-equivalent signal blocks, (a) the first bit in the second projection information .indicates the set of coefficients of the prediction filter for the second block of time-equivalent signals is the first set of filter coefficients of 'prediction or a second set of coefficients of the prediction filter, (bl) if the first set of coefficients of the prediction filter is also the set of coefficients of the prediction filter for the second block of time equivalent signals, then is the second bit in the second projection information indicates whether the set of coefficients of the prediction filter for the third block of time-equivalent signals is the first set of coefficients of the prediction filter or the second set of coefficients of the prediction filter, (b2) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second block of time equivalent signals, then the next two bits in the second projection information indicate whether the set of filter coefficients The prediction for the third block of time-equivalent signals is the first, second or third set of coefficients of the prediction filter, (cl) if the first set of coefficients of the prediction filter is the set of filter coefficients for the second and third block of time equivalent signals, then the third bit of the second projection information indicates whether the set of coefficients of the prediction filter for the fourth block of time-equivalent signals is the first or second sets of coefficients of the prediction filter , (c2) if the first set of coefficients of the prediction filter is the set of coefficients d the filter for the second block of time equivalent signals and the second set of coefficients of the prediction filter is the set of filter coefficients for the third block of time equivalent signals, then the third and fourth bits in the second projection information indicates whether the set of coefficients of the prediction filter for the fourth block of time-equivalent signals is the first, second or third set of coefficients of the prediction filter, (c3) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second block of time-equivalent signals, and the first or second sets of filter coefficients is the set of filter coefficients for the third block of time-equivalent signals, then the fourth and fifth bits in the second projection information indicate whether the set of prediction filter coefficients for the fourth block of time-equivalent signals is the first, second or third set of coefficients of the prediction filter, (c4) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second block of equivalent signals in time and the third set of filter coefficients is the set of coefficients of the prediction filter for the third block of time equivalent signals, then the fourth and fifth bits in the second projection information indicate whether the set of coefficients of the prediction filter for the fourth block of time equivalent signals is the first, second, third or fourth sets of coefficients of the prediction filter. 37. The apparatus according to claim 33, characterized in that the first probability table is assigned to the first of n signal blocks equivalent in time, the first projection information is devoid of the projection information for projecting the first probability table. to the first block of signals equivalent in time of the n blocks of signals equivalent in time, (a) the first bit in the first projection information indicates whether the probability table for the second block of signals equivalent in time is the first table of probability or a second probability table, (bl) if the first probability table is also the probability table for the second signal block equivalent in time, then the second bit in the first projection information indicates whether the probability table for the third block of signals equivalent in time is the first probability table or the second table of probability, (b2) if the second probability table is the probability table for the second signal block equivalent in time, then the next two bits in the first projection information indicate whether the probability table for the third block of signals equivalent in time is the first, second or third probability table, (cl) if the first probability table is the probability table for the second and third signal blocks equivalent in time, then the third bit of the first projection information indicates whether the probability table for the fourth signal block equivalent in time is the first or second probability table, (c2) if the first probability table is the probability table for the second signal block equivalent in time and the second probability table is the probability table for the third block of signals equivalent in time, then the third and fourth bits in the first projection information indicate whether the probability table for the fourth signal block equivalent in time is the first, second or third table of probability, (c3) if the second probability table is the probability table for the second signal block equivalent in time, and the first or the second probability table is the probability table for the third signal block equivalent in time, then the fourth and fifth bits in the first projection information indicate §i the probability table for the fourth signal block equivalent in time is the first, second or third probability table, (c4) if the second probability table is the probability table for the second time signal cloque, and the third probability table is the probability table for the third time block signals equivalent in time, then the fourth and fifth bits in the first projection information indicate whether the probability table for the fourth signal block equivalent in time is the first, second, third or fourth probability table. 38. The apparatus according to claim 19, characterized in that the decoding means is adapted to decode the encoded data information into n blocks of time equivalent signals, one for each of the n channel signals, the means- recovery are adapted to retrieve a first indicator word (w4) from the lateral information, the first indicator word is from a first value, 'indicating that the projection information for the probability tables is different from the projection information for the whole of prediction filter coefficients and, when it is of a second value, indicates that the projection information for the probability tables is the same as for the prediction filter coefficients, that the recovery means are also adapted to recover only a projection information of the lateral information in the latter case, the means of recovery are adapted, moreover, to copy the recovered projection information in the event that the first indicator word has the eighth value. 39. The apparatus according to claim 19, characterized in that the decoding means is adapted to decode the coded data information into n blocks of time equivalent signals, one for each of the n channel signals, the recovery means. they are adapted to retrieve a second indicator word (w) from the lateral information, the second indicator word, when it is a third value, indicates that the blocks of equivalent signals in time all have the same projection information for the filter coefficients of prediction and, when they are of a fourth value, indicate that the blocks of signals equivalent in time each have a different projection information to that of the coefficients of the prediction filter, that the means of recovery are adapted, in addition, to retrieve the second projection information for only one block of time equivalent signals in the case d e that the fifth indicator word has the third value and are adapted to recover the second projection information of each of the signal blocks equivalent in time in case the fifth indicator word has the fourth value. 40. The apparatus according to claim 19, characterized in that the decoding means are further adapted to decode the coded data information into n blocks of time equivalent signals, one for each of the n channel signals, the means of retrievals are adapted to recover a plurality of sets of coefficients of the lateral information prediction filter and to recover a bit array of the second projection information, the apparatus further comprises allocation means for assigning the first set of coefficients of prediction to the first of the n blocks of time equivalent signals, (a) the allocation means are further adapted to assign the first set of coefficients of the prediction filter to the second block of time equivalent signals in response to the first bit in the array of bits that is of a first binary value and are adapted to assign the sec a set of coefficients of the projection filter to the second block of time-equivalent signals in response to the first bit that is of a second binary value, (bl) if the first set of coefficients is also the set of coefficients of the filter for the second block of time-equivalent signals, then the allocation means are further adapted to assign the first set of coefficients of the prediction filter to the third block of time-equivalent signals in response to the second bit in the array ^ bits which are of a first binary value and are adapted to assign the second set of coefficients of the prediction filter to the third block of time equivalent signals in response to the second bit that is of the second binary value, (b2) if the second set of coefficients is the set of filter coefficients for the second signal block equivalent in time, then the allocation means are adapted further, to assign any of the first or second or third set of prediction filter coefficients to the third signal block equivalent in time in response to the values of the next two bits in the bit array, (cl) if the first set of coefficients is the set of filter coefficients for the second and third signal blocks equivalent in time, then the allocation means are further adapted to assign any of the first or second set of filter coefficients to the fourth block of signals equivalent in time in response to the value of the third bit in the array of bits, (c2) if the first set of coefficients of the prediction filter is the set of filter coefficients for the second block of signals equivalent in time and the second set of filter coefficients is the set of filter coefficients for the third block of signals equivalent in time, then the allocation edios are further adapted to assign any of the first or second or third set of coefficients of the prediction filter to the four block of signals equivalent in time in response to the values of the third and fourth bits. in the bit array, (c3) if the second set of coefficients of the prediction filter is the set of filter coefficients for the second signal block equivalent in time and the first or second sets of filter coefficients is the set of filters for the third block of signals equivalent in time, then the allocation means are adapted to assign any of the first, second or third set of filter coefficients to the four block of signals equivalent in time in response to the values of the fourth and fifth bits in the bit array, (c4) if the second set of prediction filter coefficients is the set of filter coefficients for the second signal blog equivalent in time, and the third set of filter coefficients' is the set of filters for the third block of signals equivalent in time, then the allocation means are adapted to assign any of the first, or the second or the third or the fourth set of filter coefficients to the four signal block equivalent in time in response to the fourth and fifth bits in the bit array. 41. The apparatus according to claim 19, characterized in that the decoding means is adapted to decode the information, encoded data in n blocks of time equivalent signals, one for each of the n channel signals, the means of retrievals are adapted to retrieve a plurality of probability tables from the lateral information and to retrieve an array of bits from the first projection information, the apparatus further comprises allocation means for assigning the first probability table to the first of ios n cloques of time-equivalent signals, (a) the allocation means are further adapted to assign the first probability table to the second block of signals equivalent in time in response to the first bit in the array of bits that is from a first binary value and are adapted to assign the second probability table of the second block of signals equivalent in time in response to the first bit that is of 'a second binary value, (bl) if the first probability table is also the probability table for the second block of signals equivalent in time, then the allocation means are adapted to assign the first probability table to the third block of signals equivalent in time in response to the second bit in the array of bits that is of a first binary value and are adapted to assign the second probability table to the third block of equivalent signals, in time in response to second bit that is of a second binary value, (b2) if the second probability table is the probability table for a second signal block equivalent in time, then the allocation means are also adapted to assign any of the first or 'the' second or third probability table to the third signal block equivalent in time in response to the values of the next two bits of the array of bits, (cl) if the first probability table is the probability table for the second and third block of signals equivalent in time, then the allocation means are also adapted to assign any of the first or] second probability table to fourth signal block equivalent in time in response to the value of the third bit in the bit array, (c2) if the first probability table is the probability table for the second block of signals equivalent in time and the second probability table is the probability table for the third signal block equivalent in time, then the allocation means are further adapted to assign any of the first, or the second or third probability table to the fourth block of probability. signals equivalent in time in response to the values, of the third and fourth bits in the array of bits, (c3) if the second probability table is the probability table for In the same block of signals equivalent in time, and the first or second probability table is the probability table for the third signal block equivalent in time, then the allocation means' are adapted to assign any of the first, or the second or third probability table to the fourth signal block equivalent in time in response to the values of the fourth and fifth bits in the bit array, (c4) if the second probability table is the probability table for the second block of signals equivalent in time, and the third probability table is the probability table for the third signal block equivalent in time, then the allocation means are adapted to assign any of the first, or the second or third or fourth probability table to the fourth signal block equivalent in time in response to the fourth and fifth bits in the bit array.