Acquisition and error recovery of audio data carried in a packetized data stream

• US 5,703,877 A
• Filed: 11/22/1995
• Issued: 12/30/1997
• Est. Priority Date: 11/22/1995
• Status: Expired due to Term

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: General Instrument Corporation (CommScope Holding Company, Inc.)
• Inventors: Moroney, Paul, Walker, G. Kent, Nuber, Ray
• Primary Examiner(s): HSU, ALPUS
• Application Number:
  US08/562,611
• Time in Patent Office: 769 Days
• Field of Search: 370/389, 370/395, 370/503, 370/509, 370/510, 370/514, 370/516, 375/362, 375/365, 375/366, 375/368, 375/371, 348/423, 348/461, 348/462, 348/464, 348/466, 348/467
• US Class Current: 370/395.64
• CPC Class Codes: G10L 19/005 Correction of errors induced by the transmission channel, if related to the coding algorithm
• CPC Class Codes: H04N 19/895 in combination with error concealment
• CPC Class Codes: H04N 21/4305 Synchronising client clock from received content stream, e.g. locking decoder clock with encoder clock, extraction of the PCR packets arrangements for synchronising receiver with transmitter by comparing receiver clock with transmitter clock H04L7/0012; arrangements for synchronising receiver with transmitter wherein the receiver takes measures against momentary loss of synchronisation H04L7/0083
• CPC Class Codes: H04N 21/43072 of multiple content streams on the same device
• CPC Class Codes: H04N 21/4341 Demultiplexing of audio and video streams
• CPC Class Codes: H04N 21/4392 involving audio buffer management
• CPC Class Codes: H04N 21/8106 involving special audio data, e.g. different tracks for different languages

Abstract

Audio data is processed from a packetized data stream carrying digital television information in a succession of fixed length transport packets. Some of the packets contain a presentation time stamp (PTS) indicative of a time for commencing the output of associated audio data. After the audio data stream has been acquired, the detected audio packets are monitored to locate subsequent PTS'"'"'s for adjusting the timing at which audio data is output, thereby providing proper lip synchronization with associated video. Errors in the audio data are processed in a manner which attempts to maintain synchronization of the audio data stream while masking the errors. In the event that the synchronization condition cannot be maintained, for example in the presence of errors over more than one audio frame, the audio data stream is reacquired while the audio output is concealed. An error condition is signaled to the audio decoder by altering the audio synchronization word associated with the audio frame in which the error has occurred.

Claims

1. A method for processing digital audio data from a packetized data stream carrying digital television information in a succession of fixed length transport packets, each of said packets including a packet identifier (PID), some of said packets containing a program clock reference (PCR) value for synchronizing a decoder system time clock (STC), and some of said packets containing a presentation time stamp (PTS) indicative of a time for commencing the output of associated data for use in reconstructing a television signal, said method comprising the steps of:

monitoring the PID'"'"'s for the packets carried in said data stream to detect audio packets, some of said audio packets carrying an audio PTS;

storing audio data from the detected audio packets in a buffer for subsequent output;

monitoring the detected audio packets to locate audio PTS'"'"'s;

comparing a time derived from said STC with a time derived from the located audio PTS'"'"'s to determine whether said audio packets are too early to decode, too late to decode, or ready to be decoded; and

adjusting the time at which said stored audio data is output from said buffer on an ongoing basis in response to said comparing step.

2. A method in accordance with claim 1 wherein a PTS pointer is provided to maintain a current PTS value and an address of said buffer identifying where a portion of audio data referred to by said current PTS is stored, said timing adjustment being provided by the further steps of:

replacing said PTS value in said PTS pointer with a new current PTS value after data stored at said address has been output from said buffer;

replacing said address in said PTS pointer with a new address corresponding to a portion of audio data referred to by said new current PTS value;

suspending the output of data from said buffer when said new address is reached; and

recommencing the output of data from said buffer when said decoder system time clock reaches a presentation time derived from said new current PTS value.

3. A method in accordance with claim 2 wherein said presentation time is determined from the sum of said new current PTS value and an offset value that provides proper lip synchronization by accounting for a video signal processing delay.

4. A method in accordance with claim 1 wherein the time at which the audio data is output from said buffer is dependent on an offset value added to said PTS for providing proper lip synchronization by accounting for a video signal processing delay.

5. A method in accordance with claim 1 comprising the further steps of:

examining the detected audio packets to locate the occurrence of at least one audio synchronization word therein for use in achieving a synchronization condition prior to locating said audio PTS'"'"'s;

commencing a reacquisition of said synchronization condition if said comparing step determines that said audio packets are too late to decode.

6. A method in accordance with claim 5 wherein two consecutive audio synchronization words with a correct number of audio data bytes in between define an audio frame, said audio frame including only one of said two consecutive audio synchronization words, said method comprising the further steps of:

detecting the occurrence of errors in said audio packets;

upon detecting a first audio packet of a current audio frame containing an error, advancing a write pointer for said buffer by the maximum number of payload bytes (N) contained in one of said fixed length transport packets and designating said current audio frame as being in error;

monitoring the detected audio packets of said current audio frame for the next audio synchronization word after said error has been detected, and if said synchronization word is not received where expected in the audio stream, discarding subsequent audio data while searching for said synchronization word rather than storing the subsequent audio data into said buffer;

resuming the storage of audio data in said buffer upon detection of said next audio synchronization word if said next audio synchronization word is located within N bytes after the commencement of the search therefor; and

if said next audio synchronization word is not located within said N bytes after the commencement of the search therefor, commencing a reacquisition of said synchronization condition.

7. A method in accordance with claim 6 comprising the further step of concealing television audio errors whenever the audio data from which said television audio is being reconstructed is in error.

8. A method in accordance with claim 7 wherein:

a current audio frame is designated as being in error by altering the audio synchronization word for that frame; and

said concealing step is responsive to an altered synchronization word for concealing audio associated with the corresponding audio frame.

9. A method for processing digital audio data from a packetized data stream carrying digital television information in a succession of transport packets having a fixed length of N bytes, each of said packets including a packet identifier (PID), some of said packets containing a program clock reference (PCR) value for synchronizing a decoder system time clock, and some of said packets containing a presentation time stamp (PTS) indicative of a time for commencing the output of associated data for use in reconstructing a television signal, said method comprising the steps of:

monitoring the PID'"'"'s for the packets carried in said data stream to detect audio packets;

examining the detected audio packets to locate the occurrence of audio synchronization words for use in achieving a synchronization condition, each two consecutive audio synchronization words defining an audio frame therebetween;

monitoring the detected audio packets after said synchronization condition has been achieved to locate an audio PTS;

searching the detected audio packets after locating said audio PTS to locate the next audio synchronization word;

storing audio data following said next audio synchronization word in a buffer;

detecting the occurrence of errors in said audio packets;

upon detecting a first audio packet of a current audio frame containing an error, advancing a write pointer for said buffer by N bytes and designating said current audio frame as being in error;

monitoring the detected audio packets of said current audio frame for the next audio synchronization word after said error has been detected, and if said synchronization word is not received where expected in the audio stream, discarding subsequent audio data while searching for said synchronization word rather than storing the subsequent audio data into said buffer;

resuming the storage of audio data in said buffer upon detection of said next audio synchronization word if said next audio synchronization word is located within N bytes after the commencement of the search therefor; and

if said next audio synchronization word is not located within said N bytes after the commencement of the search therefor, commencing a reacquisition of said synchronization condition.

10. A method in accordance with claim 9 comprising the further step of concealing television audio errors whenever the audio data from which said television audio is being reconstructed is in error.

11. A method in accordance with claim 10 wherein:

a current audio frame is designated as being in error by altering the audio synchronization word for that frame; and

said concealing step is responsive to an altered synchronization word for concealing audio associated with the corresponding audio frame.

12. A method in accordance with claim 9 wherein said audio data includes information indicative of an audio sample rate and audio bit rate, at least one of said audio sample rate and audio bit rate being variable, said method comprising the further step of attempting to maintain synchronization of said audio packets during a rate change indicated by said audio data by:

ignoring a rate change indicated by said audio data on the assumption that the rate has not actually changed;

concealing the audio frame containing the data indicative of an audio sample rate change while attempting to maintain said synchronization condition; and

commencing a reacquisition of said synchronization condition if said condition cannot be maintained.

13. A method in accordance with claim 9 wherein said audio data includes information indicative of an audio sample rate and audio bit rate, at least one of said audio sample rate and audio bit rate being variable, said method comprising the further step of attempting to maintain synchronization of said audio packets during a rate change indicated by said audio data by:

processing said audio data in accordance with a new rate indicated by said audio data in the absence of an error indication pertaining to the audio frame containing the new rate, while attempting to maintain said synchronization condition;

processing said audio data without changing the rate if an error indication pertains to the audio frame containing the new rate, while concealing the audio frame to which said error condition pertains and attempting to maintain said synchronization condition; and

commencing a reacquisition of said synchronization condition if said condition cannot be maintained.

14. Apparatus for acquiring audio information carried by a packetized data stream and processing errors therein, comprising:

means for detecting audio transport packets in said data stream;

means for recovering audio data from said detected audio transport packets for storage in a buffer;

means for locating an audio presentation time stamp (PTS) in said detected audio transport packets;

means responsive to said PTS for commencing the output of audio data from said buffer at a specified time;

means for monitoring the detected audio transport packets after the output of audio data from said buffer has commenced, to locate subsequent audio PTS'"'"'s;

means for comparing a time derived from a decoder system time clock (STC) to a time derived from the subsequent audio PTS'"'"'s to determine whether audio data stored in said buffer is too early to decode, too late to decode, or ready to be decoded; and

means responsive to said comparing means for adjusting the time at which said stored audio data is output from said buffer.

15. Apparatus in accordance with claim 14 further comprising:

means for maintaining a PTS pointer with a current PTS value and an address of said buffer identifying where a portion of audio data referred to by said current PTS is stored;

means for replacing said PTS value in said PTS pointer with a new current PTS value after data stored at said address has been output from said buffer, and for replacing said address in said PTS pointer with a new address corresponding to a portion of audio data referred to by said new current PTS value;

means responsive to said PTS pointer for suspending the output of data from said buffer when said new address is reached; and

means for recommencing the output of data from said buffer at a time derived from said new current PTS value.

16. Apparatus in accordance with claim 15 further comprising:

means for concealing error in an audio signal reproduced from data output from said buffer and reestablishing the detection of said audio transport packets if the time derived from said new current PTS value is outside a predetermined range.

17. Apparatus in accordance with claim 14 wherein said audio transport packets each contain a fixed number N of payload bytes, said packets being arranged into successive audio frames commencing with an audio synchronization word, said apparatus further comprising:

means for detecting the occurrence of errors in said audio packets;

means for advancing a write pointer for said buffer by N bytes and designating a current audio frame as being in error upon detecting an error in an audio transport packet of said current audio frame;

means for monitoring the detected audio transport packets of said current audio frame for the next audio synchronization word after said error has been detected, and if said synchronization word is not received where expected in the audio stream, discarding subsequent audio data while searching for said synchronization word rather than storing the subsequent audio data into said buffer;

means for resuming the storage of audio data in said buffer upon detection of said next audio synchronization word if said next audio synchronization word is located within said fixed number N of bytes after the commencement of the search therefor; and

means for reestablishing the detection of said audio transport packets if said next audio synchronization word is not located within said fixed number N of bytes after the commencement of the search therefor.

18. Apparatus in accordance with claim 17 further comprising:

means for concealing error in an audio signal reproduced from data output from said buffer when the data output from said buffer is in error.

19. Apparatus in accordance with claim 18 further comprising:

means for altering the audio synchronization word associated with a current audio frame to designate that frame as being in error;

wherein said concealing means are responsive to altered synchronization words for concealing errors in audio associated with the corresponding audio frame.

20. Apparatus for acquiring audio information carried by a packetized data stream and processing errors therein, comprising:

means for detecting audio transport packets in said data stream, said packets being arranged into successive audio frames commencing with an audio synchronization word;

means responsive to said synchronization words for obtaining a synchronization condition enabling the recovery of audio data from said detected audio transport packets for storage in a buffer;

means for detecting the presence of errors in said audio data;

means responsive to said error detecting means for controlling the flow of data through said buffer when an error is present, to attempt to maintain said synchronization condition while masking said error; and

means for reestablishing the detection of said audio transport packets if said controlling means cannot maintain said synchronization condition.

21. Apparatus in accordance with claim 20 wherein said audio transport packets each contain a fixed number N of payload bytes, and said means responsive to said error detecting means comprise:

means for advancing a write pointer for said buffer by said fixed number N of bytes and designating a current audio frame as being in error upon the detection of an error in an audio transport packet thereof;

means for monitoring the detected audio transport packets of said current audio frame for the next audio synchronization word after said error has been detected, and if said synchronization word is not received where expected in the audio stream, discarding subsequent audio data while searching for said synchronization word rather than storing the subsequent audio data into said buffer; and

means for resuming the storage of audio data in said buffer upon detection of said next audio synchronization word if said next audio synchronization word is located within said fixed number N of bytes after the commencement of the search therefor.

22. Apparatus in accordance with claim 20 further comprising:

means for concealing error in an audio signal reproduced from data output from said buffer when the data output from said buffer is in error.

23. Apparatus in accordance with claim 22 further comprising:

means for altering the audio synchronization word associated with an audio frame containing a data error to designate that frame as being in error;

wherein said concealing means are responsive to altered synchronization words for concealing errors in audio associated with the corresponding audio frame.

24. A method for managing errors in data received in bursts from a packetized data stream carrying digital information in a succession of fixed length transport packets, at least some of said packets containing a presentation time stamp (PTS) indicative of a time for commencing the fixed rate presentation of presentation units from a buffer into which they are temporarily stored upon receipt, said method comprising the steps of:

monitoring received packets to locate associated PTS'"'"'s, said received packets carrying presentation units to be presented;

synchronizing the presentation of said presentation units from said buffer to a system time clock (STC) associated with the packetized data stream using timing information derived from the PTS'"'"'s located in said monitoring step; and

identifying discontinuity errors resulting from a loss of one or more transmitted packets between successive ones of the received packets and, if a discontinuity of no more than one packet is identified, advancing a write pointer of said buffer by a suitable number of bits to compensate for the discontinuity, while maintaining the synchronization of said presentation with respect to said STC.

25. A method in accordance with claim 24 wherein said transport packets each contain a fixed number N of payload bytes, said method comprising the further steps of:

advancing said write pointer by said fixed number N of bytes upon the detection of a discontinuity error;

continuing said monitoring step after said discontinuity error has been detected in order to search for a synchronization word, and if said synchronization word is not located where expected, discarding subsequent presentation units while searching for said synchronization word rather than storing said subsequent presentation units in said buffer; and

resuming the storage of presentation units in said buffer upon the detection of said synchronization word if said synchronization word is located within said fixed number N of bytes after the commencement of the search therefor.