Information technology MPEG audio technologies Part 1: MPEG Surround

Transcription

1 INTERNATIONAL STANDARD ISO/IEC :007 TECHNICAL CORRIGENDUM Published INTERNATIONAL ORGANIZATION FOR STANDARDIZATION МЕЖДУНАРОДНАЯ ОРГАНИЗАЦИЯ ПО СТАНДАРТИЗАЦИИ ORGANISATION INTERNATIONALE DE NORMALISATION INTERNATIONAL ELECTROTECHNICAL COMMISSION МЕЖДУНАРОДНАЯ ЭЛЕКТРОТЕХНИЧЕСКАЯ КОМИССИЯ COMMISSION ÉLECTROTECHNIQUE INTERNATIONALE Inforation technology MPEG audio technologies Part 1: MPEG Surround TECHNICAL CORRIGENDUM Technologies de l'inforation Technologies audio MPEG Partie 1: Abiance MPEG RECTIFICATIF TECHNIQUE Technical Corrigendu to ISO/IEC :007 was prepared by Joint Technical Coittee ISO/IEC JTC 1, Inforation technology, Subcoittee SC 9, Coding of audio, picture, ultiedia and hyperedia inforation. Throughout this Technical Corrigendu, changes to existing text and tables are highlighted using a grey background. In the first paragraph of 3.5, replace: a ( l aliasing condition vector defined for every paraeter tie slot l and all subbands that are the last subband (highest in frequency within a paraeter band. a ( l aliasing condition vector defined for every paraeter set l and all subbands that are the last subband (highest in frequency within a paraeter band. ICS Ref. No. ISO/IEC :007/Cor.:009(E ISO/IEC 009 All rights reserved Published in Switzerland

2 In the first paragraph of 3.5, replace: ( l r weighted correlation su based on the input downix signa defined for ( l every paraeter tie slot l and all subbands that have an adjoining paraeter border, used for Low Power MPEG surround. r weighted correlation su based on the input downix signa defined for every paraeter set l and all subbands that have an adjoining paraeter border, used for Low Power MPEG surround. In 4.3., replace the title of Table : Table Outline of difference between the High Quality and the Low Power MEG Surround syste Table Outline of difference between the High Quality and the Low Power MPEG Surround syste In 4.5, replace the third paragraph and the caption of Figure 11: If the MPEG Surround is connected with an arbitrary downix coder (including High Efficiency AAC via the tie doain, as shown in MPEG Surround, spatial paraeters MPEG Surround AAC SBR T /F Delay cos to exp analysis MPEG Surround exp to cos F/T tie- doain output, sapling rate Fs 96 HQ: 181 LP: 01 Delay, in tie - doain saples at F s Figure 11, the additional delay introduced by the MPEG Surround decoding process will be as outlined above. MPEG Surround, spatial paraeters MPEG Surround AAC SBR T /F Delay cos to exp analysis MPEG Surround exp to cos F/T tie- doain output, sapling rate Fs 96 HQ: 181 LP: 01 Delay, in tie - doain saples at F s Figure 11 Delay when connecting MPEG Surround in the tie-doain for arbitrary core codec (including HE-AAC ISO/IEC 009 All rights reserved

3 If the MPEG Surround is connected with an arbitrary downix coder, the additional delay introduced by the MPEG Surround decoding process will be as outlined above. The connection of MPEG Surround with a High Efficiency AAC downix coder in the tie doain is shown in MPEG Surround, spatial paraeters MPEG Surround AAC SBR T/F Delay cos to exp analysis MPEG Surround exp to cos F/ T tie- doain output, sapling rate Fs 96 HQ: 181 LP: 01 Delay, in tie - doain saples at F s Figure 11, where the HE-AAC coprises AAC and SBR decoding. MPEG Surround, spatial paraeters MPEG Surround AAC SBR T/F Delay cos to exp analysis MPEG Surround exp to cos F/ T tie-doain output, sapling rate Fs 96 HQ: 181 LP: 01 Delay, in tie-doain saples at F s Figure 11 Delay when connecting MPEG Surround with HE-AAC in the tie doain In 4.5, replace the title of Figure 1: Figure 1 Delay when connecting MPEG Surround with HE-AAC in the -doain Figure 1 Delay when connecting MPEG Surround with HE-AAC in the doain In 4.5, replace the fifth paragraph: Transission of MPEG Surround side inforation with respect to transission of the coded downix signal is done in such a anner that there is no need to delay the downix signal before it is processed by the MPEG Surround. This eans that MPEG Surround data is conveyed such that it is available when needed by the MPEG Surround decoding process. The teporal relationship between downix data and spatial data is defined in Clause 6. Note that special consideration is required if an MPEG Surround and an HE-AAC are connected in the tie doain while a connection in the doain would have been possible according to subclause 4.4. In this case, the spatial paraeters have to be delayed by 961 tie saples, which is the su of 57 saples for HE-AAC and 704 saples for MPEG Surround and Nyquist analysis. ISO/IEC 009 All rights reserved 3

4 Transission of MPEG Surround side inforation with respect to transission of the coded downix signal is preferably done in such a anner that there is no need to delay the downix signal before it is processed by the MPEG Surround. This eans that MPEG Surround data is preferably conveyed such that it is available when needed by the MPEG Surround decoding process. The teporal relationship between downix data and spatial data is defined in Clause 7. For the Baseline MPEG Surround Profile defined in 4.7., restrictions for the teporal relationship are specified. Note that special consideration is required if an MPEG Surround and an HE-AAC are connected in the tie doain while a connection in the doain would have been possible according to 4.4. In this case, the spatial paraeters have to be delayed by 961 tie saples, which is the su of 57 saples for HE-AAC and 704 saples for MPEG Surround and Nyquist analysis. In 4.7., Note 3 at the botto of Table 4, replace: Note 3: A low power utilizes only residual coding data for the first 8 bands, corresponding to approxiately.7 khz bandwidth. Note 3: A low power utilizes only residual coding data for the first 7 bands, corresponding to approxiately.4 khz bandwidth at a sapling frequency of 44.1 khz. At the end of 4.7., add the following new paragraphs and table: If a Baseline MPEG Surround Profile is used in cobination with one of the MPEG downix coders listed in Table 4A, the following additional restrictions apply. The MPEG Surround frae length ust be an integer ultiple of the downix coder frae length, i.e. bsfraelength ust have one of the allowed values listed in Table 4A (see for details of downsapled or upsapled operation of MPEG Surround. Furtherore, in the case that MPEG Surround data is ebedded in a downix bitstrea (as defined in 7..3 and 7..4, the teporal relationship between downix data and spatial data ust be such that the sactiealign paraeter (defined in 7..5 has the value 0. In the case that MPEG Surround data and downix data are conveyed in separate streas (as defined in 7.., the teporal relationship ust be such that the tie stap of an MPEG Surround access unit ust be the sae as the tie stap of an downix access unit. 4 ISO/IEC 009 All rights reserved

5 Table 4A Allowed values for bsfraelength in the Baseline MPEG Surround Profile when used in cobination with MPEG downix coders Downix coder AAC with upsapled MPEG Surround AAC with upsapled MPEG Surround HE-AAC 104/048 - with upsapled MPEG Surround HE-AAC 960/190 - with upsapled MPEG Surround BSAC - with upsapled MPEG Surround BSAC with SBR - with upsapled MPEG Surround AAC LD 51 AAC ELD 51 AAC ELD with SBR 51/104 - with upsapled MPEG Surround MPEG1/ Layer II MPEG1/ Layer III Downix coder frae length ( saples Allowed values for bsfraelength 31, 63 7, 15, 3, 31, 39, 47, 55, 63, 71 14, 9, 44, 59 9, 59 14, 9, 44, 59 31, , , 9, 44, 59 31, 63 7, 15, 3, 31, 39, 47, 55, 63, 71 31, , 15, 3, 31, 39, 47, 55, 63, 71 7, 15, 3, 31, 39, 47, 55, 63, 71 31, 63 7, 15, 3, 31, 39, 47, 55, 63, 71 17, 35, 53, 71 35, 71 17, 35, 53, 71 35, 71 If a Baseline MPEG Surround Profile is used to decode MPEG Surround data conveyed as buried data in a PCM downix signal (as defined in 7.3, the teporal relationship between downix data and spatial data ust be such that the sactiealign paraeter (defined in 7..5 either has the value 0 or has the value -N*(bsFraeLength+1, where N is the nuber of bands used in MPEG Surround, i.e. N=64 for noral operation, N=3 for downsapled operation, or N=18 for upsapled operation (see In the first paragraph of , replace: case DIFF_TIME: if ( (pg > 0 (ixedtiepairxxx[pi][setidx] { case DIFF_TIME: if ( (pg > 0!(ixedTiePairXXX[pi][setIdx] { ISO/IEC 009 All rights reserved 5

6 In the first paragraph of , replace: paraeter tie slot l paraeter set l In , replace: The allowed values for bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex depend on bsfraelength, bssaplingfrequencyindex and bsresidualfraesperspatialfrae or bsarbitrarydownixresidualfraesperspatialfrae, respectively, as shown in Table 88. Table 88 Allowed cobinations of bssaplingfrequencyindex and bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex (bsfraelength+1/ (bsresidualfraesperspatialfrae+1 or (bsfraelength+1/ (bsarbitrarydownixresidualfraespersp atialfrae+1 Allowed cobinations of {bssaplingfrequencyindex, bsresidualsaplingfrequencyindex} or {bssaplingfrequencyindex, bsarbitrarydownixresidualsaplingfrequencyindex} 15 {0x0, 0x0}, {0x1, 0x1}, {0x, 0x}, {0x3, 0x3}, {0x4, 0x4}, {0x5, 0x5}, {0x6, 0x6}, {0x7, 0x7}, {0x8, 0x8}, {0x9, 0x9}, {0xa, 0xa}, and {0xb, 0xb} {0x0, 0x0}, {0x1, 0x1}, {0x, 0x}, {0x3, 0x3}, {0x4, 0x4}, {0x5, 0x5}, {0x6, 0x6}, {0x7, 0x7}, {0x8, 0x8}, {0x9, 0x9}, {0xa, 0xa}, and {0xb, 0xb} 18 {0x0, 0x}, {0x1, 0x}, {0x, 0x3}, {0x3, 0x5}, {0x4, 0x5}, {0x5, 0x6}, {0x6, 0x8}, {0x7, 0x8}, {0x8, 0x9}, {0x9, 0xb}, and {0xa, 0xb} 4 {0x0, 0x3}, {0x1, 0x3}, {0x, 0x5}, {0x3, 0x5}, {0x4, 0x5}, {0x5, 0x8}, {0x6, 0x9}, {0x7, 0x9}, and {0x8, 0xb} 30 {0x0, 0x3}, {0x1, 0x3}, {0x, 0x5}, {0x3, 0x7}, {0x4, 0x7}, {0x5, 0x8}, {0x6, 0x9}, {0x7, 0x9}, and {0x8, 0xb} 3 {0x0, 0x3}, {0x1, 0x4}, {0x, 0x5}, {0x3, 0x6}, {0x4, 0x7}, {0x5, 0x8}, {0x6, 0x9}, {0x7, 0xa}, {0x8, 0xb}, {0x9, 0xb}, {0xa, 0xb}, and {0xb, 0xb} The allowed values for bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex are shown in Table 88. Table 88 Allowed values of bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex Paraeter bsresidualsaplingfrequencyindex bsarbitrarydownixresidualsaplingfrequencyindex Allowed values 0x3, 0x4, 0x5, 0x6, 0x7, 0x8, 0x9, 0xa, 0xb 0x3, 0x4, 0x5, 0x6, 0x7, 0x8, 0x9, 0xa, 0xb 6 ISO/IEC 009 All rights reserved

7 In 6.3.3, replace: pi exp i ( k + 0.5( n, 0 k < 18, 0 n< pi 0.5 exp i ( k+ 0.5( n, 0 k < 18,0 n < and replace: using a 180 saple version of the window function c[i] where the additional interediate saples are obtained by linear interpolation of neighboring saples of the original 640 saple window function specified in subclause 4.A.6.Table 4.A.87. replacing the window function ci [] by a 180 saple version c 18 [] i, which is obtained fro the original 640 saple window function specified in ISO/IEC :009, Table 4.A.87 according to: ci [ / ] if ieven ( c[( i 1/ ] c[( i+ 1 / ] / if i {55,511,767,103} c18[] i = c[( i 1 / ]/ if i= 179 ( c[( i 1/] + c[( i+ 1/] / else and replace: pi exp i ( k + 0.5( n 510, 0 k < 18, 0 n< pi exp i ( k+ 0.5 ( n 510, 0 k< 18,0 n < In the first paragraph of , replace: configuration 6 rows and 3 coluns, according to: configuration 6 rows and 5 coluns, according to: In the first paragraph of , replace: configuration 8 rows and 3 coluns, according to: configuration 8 rows and 5 coluns, according to: ISO/IEC 009 All rights reserved 7

8 In , replace: L qf 104 ax 64, ceil N qf,long = 18 ax 64, ceil N qf,short for long windows for short windows. L qf 104 in 64, ceil N qf,long = 18 in 64, ceil N qf,short for long windows for short windows. In , replace: M r M r ( k,n ( k,n ( ( π k n 19 0 k < 64 = cos, 18 0 n < 18 ( ( π k+ 0.5 n 19 0 k < 64 = cos, 18 0 n < 18 In , replace: M r M r ( k,n ( k,n and replace: ( ( π k n k < 18 = cos, 56 0 n < 56 ( ( π k+ 0.5 n k < 18 = 0.5 cos, 56 0 n < 56 using a 180 saple version of the window function c[i] where the additional interediate saples are obtained by linear interpolation of neighboring saples of the original 640 saple window function specified in subclause 4.A.6.Table 4.A.87. replacing the window function ci [] by the 180 saple version c [] 18 i, which is defined in ISO/IEC 009 All rights reserved

9 In , replace: l, l, l, ( σl l, ( σls where wl =, w = l (, l σ + ( σ, l L Ls (, l σ + ( σ, L Ls down ix gain for the centre channel. Ls, w l, ( σr = l (, l σ + ( σ, l, R, R Rs l, l, ( σrs wrs = and where g l (, l c is the σ + ( σ, R Rs where w R ( σ L ( PR,L ( σ Ls ( PR,Ls wl = ( ( ( σ P ( P, wls L R,L + σ Ls R,Ls ( ( ( σ P ( P L R,L + σ Ls R,Ls ( σ R ( PL,R ( σ Rs ( PL,Rs w ( ( ( σ P + σ ( P Rs = ( ( ( σ P + σ ( P. =, R L,R Rs L,Rs R L,R Rs L,Rs =, In 7..1, replace: The spatial frae length is preferred to be an integer ultiple of the frae length of the underlying downix coder. Asynchronous fraing of spatial data and the downix data (i.e., different frae lengths is possible. However, in this case, additional buffering of the spatial data in the ight be needed. In general spatial data is conveyed in such a anner that it is available to the MPEG Surround in tie when it is required to process the decoded downix signals, assuing the ost efficient connection of downix to the MPEG Surround. This is a direct connection of HE-AAC and MPEG Surround in the doain in case of MPEG Surround using noral operation (as opposed to upsapled or downsapled operation as defined in subclause 6.3.3, and a connection in the PCM tie doain in all other cases. When HE-AAC and MPEG Surround are connected in the tie doain even though the ost efficient connection would have been in the doain, the spatial paraeters have to be delayed accordingly in order to aintain the tie alignent between spatial data and downix data. Inforation about this delay is given in subclause In the case that the spatial data is ebedded in the downix data strea (see subclause 7.., 7..3, and 7..4, the teporal relationship between spatial fraes and downix fraes is indicated by the value of sactiealign (see subclause In the case that the downix data and the spatial data are conveyed in separate streas, the teporal relationship between spatial fraes and downix fraes is indicated by the tie staps of the corresponding streas. If the transport layer does not provide tie staps (as e.g. in case of LATM, the transport layer needs to define the teporal relationship between the data of these both streas by other eans. The spatial frae length is preferred to be an integer ultiple of the frae length of the underlying downix coder. Asynchronous fraing of spatial data and the downix data (i.e. different frae lengths is possible. However, in this case, additional buffering of the spatial data in the ight be needed. In genera spatial data is preferably conveyed in such a anner that it is available to the MPEG Surround in tie when it is required to process the decoded downix signals, assuing the ost efficient connection of downix to the MPEG Surround. This is a direct connection of HE-AAC and MPEG Surround in the doain in the case that both use the sae nuber of OMF bands (see 4.4, and a connection in the PCM tie doain in all other cases. When HE-AAC and MPEG Surround are connected in the tie doain even though the ost efficient connection would have been in the doain, the spatial paraeters have to be delayed accordingly in order to aintain the tie alignent between spatial data and downix data. Inforation about this delay is given in 4.5. In the case that the spatial data is ebedded in the downix data strea (see 7.., 7..3, and 7..4, the teporal relationship between spatial fraes and downix fraes is indicated by the value of sactiealign (see If sactiealign has the value 0, this indicates that the spatial data is conveyed in the preferred anner outlined above. In the case that the downix data and the spatial data are conveyed in separate streas, the teporal relationship between spatial fraes and downix fraes is indicated by the tie staps of the access units of the corresponding streas. If a downix coder other than HE-AAC is used, the tie stap of an access unit carrying an SAC frae identifies the first PCM saple of the corresponding tie doain downix signal ISO/IEC 009 All rights reserved 9

10 frae that is input to the MPEG Surround. If HE-AAC is used as downix coder, the tie stap of the SAC frae identifies the first PCM saple of the corresponding tie doain downix signal frae at the output of the AAC core. If the transport layer does not provide tie staps, the teporal relationship between the data of these both streas needs to be defined by other eans. In case of LATM (see ISO/IEC , the first MPEG Surround access unit and the first downix coder access unit in an AudioMuxEleent( are considered to have the sae tie stap. In 7.., replace: The transission of spatial audio data requires a spatial eleentary strea that depends on the eleentary strea containing the related coded audio downix data. The actual spatial data is either conveyed in the spatial eleentary strea or ultiplexed into the downix data stored in the eleentary strea upon which the spatial eleentary strea depends. The latter is specified for MPEG-/4 AAC payloads (see subclause 7..3 and for MPEG-1/ Layer I/II/III payloads (see subclause Backwards copatibility with s that can decode the coded audio downix data but not the spatial audio data is achieved in both scenarios. If the downix signal is encoded with the cobination of a ono AAC downix coder and SBR bandwidth extension, it is possible that both data for the MPEG-4 Paraetric Stereo (PS tool as well as data for MPEG Surround in a configuration is present siultaneously in the bitstrea conveying the downix signal. Such a bitstrea can be decoded into a channel stereo signal in accordance with the MPEG-4 HE-AAC v Profile, whereby the MPEG Surround data is ignored. When such a bitstrea is used in cobination with an MPEG Surround, the PS data in the bitstrea is ignored and the downix bitstrea is decoded in accordance with the MPEG-4 HE-AAC profile. This provides a ono downix signal in the doain that is used as input to the subsequent MPEG Surround decoding process for a configuration as described in subclause The interface to ISO/IEC is in line with the specification given in ISO/IEC subclause 1.6. An eleentary strea carrying spatial audio data is identified by the Audio Object Type MPEG Surround (Object Type ID 30. The AudioSpecificConfig( for this object carries the SpatialSpecificConfig( data and a sacpayloadebedding flag that indicates whether the SpatialFrae( payload is conveyed as an eleentary strea or ebedded into the downix data, as defined in ISO/IEC subclause The signaling of the availability of spatial audio data is possible either by eans of a spatial eleentary strea that depends on the eleentary strea containing the related coded audio downix data (as, e.g., indicated by the dependson_es_id field defined in ISO/IEC :004, 6.5. or by eans of including the SpatialSpecificConfig( at the end of the AudioSpecificConfig( of the downix eleentary strea in a backward copatible way (as defined in ISO/IEC :009, 1.6. The actual spatial data is either conveyed in the spatial eleentary strea or ultiplexed into the downix data stored in the eleentary strea upon which the spatial eleentary strea, if present, depends. The latter is specified for MPEG-/4 AAC payloads (see 7..3 and for MPEG-1/ Layer I/II/III payloads (see Backwards copatibility with s that can decode the coded audio downix data but not the spatial audio data is achieved in all these scenarios. If the downix signal is encoded with the cobination of a ono AAC downix coder and SBR bandwidth extension, it is possible that both data for the MPEG-4 Paraetric Stereo (PS tool as well as data for MPEG Surround in a configuration is present siultaneously in the bitstrea conveying the downix signal. Such a bitstrea can be decoded into a channel stereo signal in accordance with the MPEG-4 HE-AAC v Profile, whereby the MPEG Surround data is ignored. When such a bitstrea is used in cobination with an MPEG Surround, the PS data in the bitstrea is ignored and the downix bitstrea is decoded in accordance with the MPEG-4 HE-AAC profile. This provides a ono downix signal in the doain that is used as input to the subsequent MPEG Surround decoding process for a configuration as described in The interface to ISO/IEC is in line with the specification given in 1.6 of ISO/IEC An eleentary strea carrying spatial audio data is identified by the Audio Object Type MPEG Surround (Object Type ID 30. The AudioSpecificConfig( for this object carries the SpatialSpecificConfig( data and a sacpayloadebedding flag that indicates whether the SpatialFrae( payload is conveyed as an eleentary strea or ebedded into the downix data, as defined in of ISO/IEC :009, ISO/IEC 009 All rights reserved

11 In 7..3, replace: Spatial audio data can be conveyed in the AAC extension_payload( echanis using extension_type EXT_SAC_DATA ( 1100, as defined in ISO/IEC subclause 8.8 and ISO/IEC subclause The extension_payload( for type EXT_SAC_DATA is used to carry a SacDataFrae(, coplete or split into several fragents, using the sae syntax eleents anctype, ancstart, and ancstop as defined in the next subclause. Spatial audio data can be conveyed in the AAC extension_payload( echanis using extension_type EXT_SAC_DATA ( 1100, as defined in ISO/IEC :006, 8.8 and ISO/IEC :009, The extension_payload( for type EXT_SAC_DATA coprises the sac_extension_data(, as defined in ISO/IEC :006, 6.3 and ISO/IEC :009, , which is used to carry a SacDataFrae(, coplete or split into several fragents, using the sae syntax eleents anctype, ancstart, ancstop, and ancdatasegentbyte as defined in 7..4, and where in the seantics of the syntax eleent ancdatasegentbyte, the ter AncDataEleent is to be replaced by sac_extension_data. In 7..5, replace: sactiealign sactiealign Identifies the PCM saple in the output frae of the downix that corresponds to the beginning of the present SAC frae. The position of the first saple of the output frae is represented as 0. The present SAC frae is the first SAC frae that is copleted (i.e., ancstop==1 in the present downix frae. Identifies the PCM saple in the tie doain output frae of the downix that corresponds to the beginning of the present SAC frae (i.e. the first saple of the tie doain input signal that is consued by the MPEG Surround decoding process for the present SAC frae. The position of the first saple of the output frae is represented as 0. The present SAC frae is the first SAC frae that is copleted (i.e. ancstop==1 in the present downix frae. If HE-AAC is used as downix coder, the tie doain output frae of the AAC core (delay-free upsapled by a factor of two in case of noral operation of MPEG Surround with 64 bands is considered here. In 7.3.3, replace Table 114 with the following Table 114 bsbdtype bsbdtype Type of data 0 MPEG Surround frae, i.e. SacDataFrae(0 1 MPEG Surround header+frae, i.e. SacDataFrae( reserved reserved for use by ISO/IEC user specific 7 padding ISO/IEC 009 All rights reserved 11

12 At the end of F.6.1, add the following new paragraph and table: Suitable values for bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex can depend on bsfraelength, bssaplingfrequencyindex and bsresidualfraesperspatialfrae or bsarbitrarydownixresidualfraesperspatialfrae, respectively, as shown in Table F.0A. Table F.0A Suitable cobinations of bssaplingfrequencyindex and bsresidualsaplingfrequencyindex or bsarbitrarydownixresidualsaplingfrequencyindex (bsfraelength+1/ (bsresidualfraesperspatialfrae+1 or (bsfraelength+1/ (bsarbitrarydownixresidualfraespersp atialfrae+1 Allowed cobinations of {bssaplingfrequencyindex, bsresidualsaplingfrequencyindex} or {bssaplingfrequencyindex, bsarbitrarydownixresidualsaplingfrequencyindex} 15 {0x0, 0x3}, {0x1, 0x3}, {0x, 0x5}, {0x3, 0x3}, {0x4, 0x3}, {0x5, 0x5}, {0x6, 0x3}, {0x7, 0x3}, {0x8, 0x5}, {0x9, 0x6}, {0xa, 0x6} and {0xb, 0x8} {0x0, 0x3}, {0x1, 0x4}, {0x, 0x5}, {0x3, 0x3}, {0x4, 0x4}, {0x5, 0x5}, {0x6, 0x3}, {0x7, 0x4}, {0x8, 0x5}, {0x9, 0x6}, {0xa, 0x7} and {0xb, 0x8} 18 {0x0, 0x4}, {0x1, 0x4}, {0x, 0x5}, {0x3, 0x4}, {0x4, 0x4}, {0x5, 0x5}, {0x6, 0x4}, {0x7, 0x4}, {0x8, 0x5}, {0x9, 0x7}, {0xa, 0x7} and {0xb, 0x8} 4 {0x0, 0x5}, {0x1, 0x5}, {0x, 0x7}, {0x3, 0x5}, {0x4, 0x5}, {0x5, 0x7}, {0x6, 0x5}, {0x7, 0x5}, {0x8, 0x7}, {0x9, 0x8}, {0xa, 0x8} and {0xb, 0xa} 30 {0x0, 0x6}, {0x1, 0x6}, {0x, 0x8}, {0x3, 0x6}, {0x4, 0x6}, {0x5, 0x8}, {0x6, 0x6}, {0x7, 0x6}, {0x8, 0x8}, {0x9, 0x9}, {0xa, 0x9} and {0xb, 0xb} 3 {0x0, 0x6}, {0x1, 0x7}, {0x, 0x8}, {0x3, 0x6}, {0x4, 0x7}, {0x5, 0x8}, {0x6, 0x6}, {0x7, 0x7}, {0x8, 0x8}, {0x9, 0x9}, {0xa, 0xa} and {0xb, 0xb} 1 ISO/IEC 009 All rights reserved