# US 5,706,396 A · Error protection system for a sub-band coder suitable for use in an audio signal processor > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 5,706,396 A - Document ID: US5706396A - Status: Granted - Kind: A - Application number: 08/281,734 - Filed: 1994-07-27 - Granted: 1998-01-06 - Earliest claimed: 1992-01-27 - Assignees: Deutsche Thomson-Brandt GmbH - Inventors: Ernst F. Schroder; Georg Plenge; Leon Maria van de Kerkhof - Web page: https://patentgrep.com/patent/US5706396A ## CPC - H04B 1/667: {using a division in frequency subbands (for TV signals H04N19/63)} - G10L 19/005: Correction of errors induced by the transmission channel, if related to the coding algorithm - H04B: TRANSMISSION - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract In a digital audio-signal transmission process, digital audio signals are coded in sub-bands. One or more scaling factors are determined for each sub-band. The coded signals and scaling factors are transmitted in data blocks, then decoded and reproduced at a receiver. The data block contains several types of information including a header, error processing information, a CRC (cyclic redundancy check) protection word, audio information including a bit allocation table, scaling factor selecting information, one or several scaling factors per transmitted sub-band, filler bits, and supplementary information with audio-signal-dependent supplementary information. To improve the protection of the above-described sub-band coded audio signals against transmission errors, supplementary error protection information or an error protection bit is added to the data block instead of filler bits, in order to protect the scaling factors. Acoustically uncomfortable play-back disturbances caused by scaling factor errors are thus eliminated. The invention may be used in digital audio broadcast (DAB), DAB-receivers, recorders and play-back systems. ## Claims ### Claim 1 A method of processing digital audio signals comprising the steps of: sub-band coding said audio signals, generating a scaling factor for sub-band coded audio signal components; and generating data blocks comprising: (a) a header including control information, (b) a first error processing code for said header, (c) an audio data field including said scaling factors, (d) an ancillary information field including ancillary information relating to said audio signals, and (e) a second error processing code for protecting said scaling factors, said second code being inserted near the end of said data blocks in a region that incorporates items of said ancillary information. ### Claim 2 (depends on 1) A method according to claim 1, wherein said step of generating said data blocks includes the step of deriving said second code from the most significant bits of said scaling factors. ### Claim 3 (depends on 1) A method according to claim 1, wherein said second error processing code of a given data block is associated with scaling factors of a following data block. ### Claim 4 (depends on 1) A method according to claim 1, wherein said step of generating data blocks includes the step of forming groups of scaling factors within a data block, a word of said second error processing code being associated respectively with each of said groups. ### Claim 5 (depends on 1) A method according to claim 1, wherein said step of generating data blocks includes the step of jointly protecting first and second channels of a two channel signal by said second error processing code. ### Claim 6 (depends on 1) A method according to claim 1, wherein said step of generating data blocks comprises the step of replacing stuffing bits with said second error processing code. ### Claim 7 (depends on 1) A method according to claim 1, wherein said header includes at least one of the following types of information: synchronising information, information relating to bit rate, information relating to sampling rate. ### Claim 8 (depends on 1) A method according to claim 1, wherein said first and second error processing codes are cyclic redundancy check codes (CRC). ### Claim 9 A method of decoding digital audio information comprising sub-band audio signals; a scaling factor for sub-band coded audio signal components; and data blocks comprising (a) a header including control information, (b) a first error processing code relative to said header, (c) an audio data field including said scaling factors, (d) an ancillary information field including ancillary information relating to said audio signals and (e) a second error processing code for protecting said scaling factors, said code being inserted near the end of said data blocks in a region that incorporates items of said ancillary information, wherein said method of decoding comprises thesteps of: detecting a scaling factor level change, comparing said change with a statistically expected level change, said detected level change being considered erroneous if a deviation from said expected change is detected, and compensating for an erroneous scale factor change. ### Claim 10 (depends on 9) A method according to claim 9, wherein said step of detecting said level change comprises the step of comparing a change in amplitude of said scaling factor to a threshold. ### Claim 11 (depends on 9) A method according to claim 9, wherein said step of comparing said detected level change with a statistically expected level change comprises the step of determining if said detected change is followed by a level change of opposite direction. ### Claim 12 Apparatus for decoding digital audio information comprising sub-band audio signals; a scaling factor for sub-band coded audio signal components; and data blocks comprising (a) a header including control information, (b) a first error processing code relative to said header, (c) an audio data field including said scaling factors, (d) an ancillary information field including ancillary information relating to said audio signals and (e) a second error processing code for the protecting said scaling factors, said second code being inserted near the end of said data blocks in a region that incorporates items of said ancillary information, said apparatus comprising: a source of said data blocks; means coupled to said source for extracting said second code and said scaling factors from said data blocks; means responsive to said scaling factors and to said second code for compensating for erroneous scaling factors using said second code; and means responsive to said scaling factors for decoding said coded audio signals. ### Claim 13 (depends on 12) Apparatus according to claim 12, and further comprising: means for detecting significant level transitions of said scaling factors, wherein a significant level transition is considered erroneous if followed by a second transition in an opposite direction. ### Claim 14 A medium storing method for producing a coded digital audio signal, said method comprising the steps of: sub-band coding said audio signals, generating a scaling factor for sub-band coded audio signal components; and generating data blocks comprising: (a) a header including control information, (b) a first error processing code relative to said header, (c) an audio data field including said scaling factors, (d) an ancillary information field including ancillary information relating to said audio signals and (e) a second error processing code for protecting said scaling factors, said second code being inserted near the end of said data blocks in a region that incorporates items of said ancillary information. ### Claim 15 A method of communicating a digital audio signal, comprising the steps of: (a) sub-band coding said audio signal, wherein at least one scale factor is associated with a sub-band; (b) transmitting said sub-band coded audio signal in successive data frames, wherein a data frame includes: (a1) a header including control information; (a2) an audio data code including: a bit allocation table; scale factor selection information; a scale factor; and sampling values associated with said sub-bands; (a3) error detection data including a first error protection code, wherein said first protection code is a Cyclic Redundancy Check (CRC) code for detection of transmission errors in said header, said scale factor selection information and said bit allocation table; and (a4) an ancillary information field containing information concerning said audio data signal and including a second CRC protection code for protecting the MSB's of said scale factor, wherein said second protection code is inserted at the end of said data frame in a region of said data frame that incorporates items of said ancillary information. ### Claim 16 (depends on 15) A method according to claim 15, wherein said header control information includes a synchronizing word and at least one of: bit rate information; sampling information; and mode information. ### Claim 17 (depends on 15) A method according to claim 15, wherein said error detection data further includes an error correction code. ### Claim 18 A method of receiving a sub-band coded digital audio signal organized in successive data frames, wherein at least one scale factor is associated with a sub-band, said method comprising the steps of: (a) decoding information in a data frame of said sub-band coded audio signal, including: (a1) a header including control information; (a2) an audio data code including: a bit allocation table; scale factor selection information; a scale factor; and sampling values associated with said sub-bands; (a3) error detection data including a Cyclic Redundancy Check (CRC) first error protection code for detecting transmission errors in said header, said scale factor selection information and said bit allocation table; and (a4) an ancillary information field containing information concerning said audio data signal and including a second CRC protection code for protecting the MSB's of said scale factor, wherein said second protection code is inserted at the end of said data frame in a region of said data frame that incorporates items of said ancillary information; and (b) reproducing audio information using an item of decoded data frame information produced in step (a). ### Claim 19 (depends on 18) A method according to claim 18, wherein said header control information includes a synchronizing word and at least one of: bit rate information; sampling information; and mode information. ### Claim 20 (depends on 18) A method according to claim 18, wherein said error detection data further includes an error correction code.