Program server for interactive television system

• US 5,442,389 A
• Filed: 12/28/1992
• Issued: 08/15/1995
• Est. Priority Date: 12/28/1992
• Status: Expired due to Term

• Current Assignee: AT&T IPM Corp. (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Blahut, Donald E., Szurkowski, Edward S.
• Primary Examiner(s): Groody, James J.
• Assistant Examiner(s): GRANT, CHRISTOPHER C
• Application Number:
  US07/997,985
• Time in Patent Office: 960 Days
• Field of Search: 358/86, 455/4.2, 455/5.1, 455/6.1, 370/85.1, 370/85.3, 370/73, 370/71, 370/124, 370/110.1, 370/94.1, 370/94.2
• US Class Current: 725/92
• CPC Class Codes: H04N 21/2747 Remote storage of video programs received via the downstream path, e.g. from the server
• CPC Class Codes: H04N 21/4722 for requesting additional data associated with the content
• CPC Class Codes: H04N 21/812 involving advertisement data advertising per se G06Q30/02
• CPC Class Codes: H04N 7/17318 Direct or substantially direct transmission and handling of requests
• CPC Class Codes: H04N 7/17336 Handling of requests in head-ends

Abstract

The invention relates to a system for transmitting stored programs, such as movies and musical works, to customers via a distribution network, such as a cable television system. The programs are stored in compressed form in a program library, such as a tape cartridge library. On receipt of a request for a program, a request processor sends control messages causing a data block comprising the requested program to be read from the program library at high speed and stored in a large dynamic random access memory (DRAM) in a server. The server then sends the program from DRAM over the distribution system to a customer as a series of digital packets. Each instance of sending a program is managed by a separate command word. The customer can request such operations as "fast forward" by sending control messages that change pointers in the command word. Command words can be grouped to send multiple audio and/or video overlays simultaneously and linked to send program sequences. The request processor, server, program library and interfaces to distribution networks can be at different locations.

Claims

1. A server for sending one or more data blocks from a data memory in specified channels over a distribution system at controlled time intervals, which comprises:

a command-word memory for storing a command word for each said data block to be sent, each command word including at least the identity of the channel for said data block, the address in said data memory of the next portion of said data block to be sent and a time indication;

first processing means for receiving requests to send data blocks and for storing said command words in said command memory in response to said requests;

second processing means for periodically processing each command word in said command-word memory, said second processing means comprising;

means for updating the time indication in said command word;

means operative when said time indication reaches a threshold value for retrieving the portion of said data block from the address in said data memory specified in said command word and updating said address; and

means for sending said retrieved portion in said identified channel over said distribution system.

2. The server of claim 1 wherein each said command word includes an incremental value related to the time desired between the sending of successive portions of its associated data block and said means for updating said time indication further comprises:

means for incrementing said time indication by the amount of said incremental value.

3. The server of claim 1 wherein each said command word includes a quantity representing the number of portions of its associated data block still to be transmitted and at least one command flag, a first one of said command flags being asserted when all portions of the associated data block have not been sent, and said second processing means further comprises:

means for modifying said quantity when each said portion is sent;

means responsive to said quantity for negating said first command flag when all portions of the associated data block have been sent; and

means, responsive to said first command flag for substituting a new command word from said first processing means for a command word processed by said second processing means when said first command flag is not asserted.

4. The server of claim 3 wherein each command word also includes a first identity specifying a first group of one or more data blocks to which its associated data block belongs, a second identity specifying a second group of one or more data blocks and wherein a second one of said command flags is asserted to indicate an abort state, which further comprises:

first and second token memories for storing token words at addresses corresponding to said identities, each said token word including at least one token flag, and wherein said second processing means further comprises;

means operative during odd passes through said command words for clearing said first token memory and, for each command word, reading a token word from the address in said second token memory corresponding to said first identity and, when either the first command flag or the second command flag is asserted in the processed command word, storing a token word at the address in said first token memory corresponding to said second identity;

means operative during even passes through said command words for clearing said second token memory and, for each command word, reading a token word from the address in said first token memory corresponding to said first identity and, when either the first command flag or the second command flag in the processed command word is set, storing a token word at the address in said second token memory corresponding to said second identity;

the states of said first and second token flags in said stored token words corresponding, respectively, to the states of said first and second command flags in said processed command word.

5. The server of claim 4 wherein a third one of said command flags is asserted to indicate a pause state and said second processing means further comprises:

means for asserting said third command flag in the command word being processed when the first token flag in the token word retrieved for said command word is asserted; and

means for inhibiting the sending of portions of the data block associated with a command word when the third command flag in said command word is asserted.

6. The server of claim 5 which further comprises:

a third token memory for storing token words at addresses corresponding to said identities, each said token word in said third token memory including said first and second token flags and a third token flag;

means connected to said first processing means for writing token words into said third token memory and wherein said second processing means further comprises;

means operable during the reading of a token word from an address in either said first token memory or said second token memory for reading the token word from the same address in said third token memory and creating a combined token word wherein the first and second token flags correspond to the logical-OR of such flags in both token words read and wherein said third token flag corresponds to the third token flag from said third token memory.

7. The server of claim 6 wherein said second processing means further comprises:

means operable during the processing of a command word for negating the first command flag and asserting the second command flag in said command word when the second token flag is asserted in the token word retrieved for said command word.

8. The server of claim 6 wherein said second processing means further comprises:

means operable after the processing of a command word for sending said command word to said first processing means when the third token flag is set in the token word retrieved for said command word.

9. The server of claim 1 wherein said first processing means also receives requests to send control messages over identified channels and wherein said second processing means further comprises:

first buffer means connected to said second processing means for storing the addresses of said data-block portions to be sent and the channel identities for said portions;

second buffer means connected to said first processing means for storing control messages to be sent and the channel identities for said control messages, and wherein said means for retrieving and means for sending further comprise;

multiplexing means connected to said first and second buffer means for retrieving said addresses and associated channel identities in sequence from said first buffer means and, for each said address, reading said portion from said address in said data memory and sending said retrieved portion in said identified channel over said distribution system and, when said first buffer means is empty, retrieving each said control message and associated channel identity in sequence from said second buffer means and sending said control message in said identified channel.

10. The server of claim 1 or claim 9 in which each retrieved data-block portion and its associated channel identity and each control message and its associated channel identity are sent over said distribution system as digital packets and wherein said server is adapted for sending digital packets over one or more distribution links, which further comprises:

a channel memory associated with each distribution link for indicating particular channels to be sent over said distribution link and wherein said means for sending further comprises;

means for sending a digital packet over said distribution link when the channel identity in said digital packet corresponds to a channel indicated in said channel memory.

11. The server of claim 10 in which control messages are sent over at least one dedicated channel, certain control messages specify channels to be sent on each distribution link and wherein each channel memory further comprises:

means responsive to said certain control messages for storing in said channel memory the identities of the channels to be sent on the distribution link associated with said channel memory.

12. The server of claim 10 in which each distribution link further comprises:

means for buffering the data items to be sent over the distribution link and sending such data items over said distribution link at the appropriate transmission speed for said distribution link.

13. The server of claim 1 in which said data memory is a dynamic random access memory, each said command word contains a fourth command flag, said command-word memory contains at least one command word having said fourth command flag asserted and said second processing means further comprises:

means responsive to said fourth command flag in said command words for refreshing said data memory when said fourth command flag is asserted.

14. The server of claim 1 which further comprises:

means for writing said data blocks into said data memory from an external source.

15. The server of claim 14 in which said data memory is divided into at least two segments, said second processing means reads from one segment of data memory at a time and said means for writing further comprises:

means for inhibiting writing into a segment of data memory when said second processing means is reading from the same segment.

16. The server of claim 14 in which said data memory is a dynamic random access memory, said second processing means periodically refreshes said data memory and said means for writing further comprises:

means for inhibiting writing into said data memory when said second processing means is refreshing said data memory.

17. The server of claim 14 in which said means for writing further comprises:

means for receiving said data blocks from said external source as a series of packets, each said packet containing one of said portions of said data block and the address in said data memory at which said portion is to be written.

18. A method of sending portions of one or more data blocks from a data memory in specified channels over a distribution system at controlled time intervals, which further comprises:

creating and storing a command word for each said data block to be sent, each command word including at least the identity of the channel for said data block, the address in said data memory of the next portion to be sent and a time indication;

periodically processing each stored command word, including the steps of;

updating the time indication;

when said time indication reaches a threshold value, retrieving the portion of said data block from the address in said data memory specified in said command word and updating said address; and

sending said retrieved portion in said identified channel over said distribution system.

19. The method of claim 18 wherein each command word includes an incremental value related to the time desired between the sending of successive portions of its associated data block and said step of updating the time indication comprises:

incrementing said time indication by the amount of said incremental value.

20. The method of claim 19 wherein each said command word includes a quantity representing the number of portions of its associated data block still to be transmitted and at least one command flag, a first one of said command flags being asserted when all portions of the associated data block have not been sent, which further comprises:

modifying said quantity when each said portion is sent;

negating said first command flag when all portions of the associated data block have been sent; and

substituting a new command word for a processed command word when said first command flag is not asserted in said processed command word.

21. The method of claim 20 wherein each command word includes a first identity specifying a first group of one or more data blocks to which its associated data block belongs, a second identity specifying a second group of one or more data blocks and a second one of said command flags is asserted to indicate an abort state, which further comprises:

during odd passes through said command words, clearing a first token memory and, for each command word, retrieving a token word from an address in a second token memory corresponding to said first identity and, when either the first command flag or the second command flag is asserted in the processed command word, storing a token word at the address in said first token memory corresponding to said second identity; and

during even passes through said command words, clearing said second token memory and, for each command word, retrieving a token word from an address in said first token memory corresponding to said first identity and, when either the first command flag or the second command flag is asserted in the processed command word, storing a token word at the address in said second token memory corresponding to said second identity;

the states of first and second token flags in a token word stored for a command word corresponding, respectively, to the states of said first and second command flags in said command word.

22. The method of claim 21 wherein a third one of said command flags is asserted to indicate a pause state, which further comprises:

asserting said third command flag in the command word being processed when the first token flag in the token word retrieved for said command word is asserted; and

inhibiting the sending of portions of the data block associated with a command word when the third command flag in said command word is asserted.

23. The method of claim 22 which further comprises:

storing token words in a third token memory at addresses corresponding to said identifies, each said token word in said third token memory including said first and second token flags and a third token flag;

during the reading of a token word from an address in either said first token memory or said second token memory, reading the token word from the same address in said third token memory and creating a combined token word wherein the first and second token flags correspond the logical-OR of such flags in both token words read and wherein said third token flag corresponds to the third token flag from said third token memory.

24. The method of claim 23 which further comprises:

during the processing of a command word, negating the first command flag and asserting the second command flag in said command word when the second token flag is asserted in the token word read for said command word.

25. The method of claim 24 in which control messages are sent in addition to said portions of data blocks, which further comprises:

storing the addresses of data-block portions to be read from said data memory, together with the channel identities for such portions, in a first buffer;

storing said control messages, together with the channel identities for such control messages, in a second buffer;

retrieving said addresses and associated channel identities in sequence from said first buffer and, for each said address, reading said data block portion from said address in said dam memory and sending said retrieved portion in said identified channel over said distribution system; and

when said first buffer means is empty, retrieving each said control message and associated channel identity in sequence from said second buffer and sending said control message in said identified channel.

26. The method of claim 18 or claim 25 wherein each retrieved data-block portion and its associated channel identity and each control message and its associated channel identity are sent over said distribution system as digital packets and wherein said digital packets are sent over one or more distribution links, which further comprises:

indicating particular channels to be sent over a distribution link in a channel memory associated with said distribution link; and

sending a digital packet over each distribution link when the channel identity in said digital packet corresponds to a channel indicated in said channel memory.

27. The method of claim 18 which further comprises:

writing said data blocks into said data memory from an external source.

28. The method of claim 27 wherein said data memory is divided into at least two segments, which further comprises:

inhibiting writing into a segment of said data memory when said second processing means is reading from the same segment.

29. A method of sending one or more data blocks from a storage library at controlled rates to one or more users over specified channels in a distribution system, which comprises:

upon receipt of a request for a data block;

retrieving said data block from said storage library and storing said data block in a random-access memory;

creating and storing a command word for each such request, each said command word containing at least the identity of the channel for said data block, the address in said data memory of the next portion to be sent and a time indication;

periodically processing each stored command word, including the steps of;

updating the time indication;

when said time indication reaches a threshold value, retrieving the portion of said data block from the address in said random-access memory specified in said command word and updating said address; and

sending said retrieved portion in said identified channel over said distribution system.

30. The method of claim 29 wherein said data blocks represent programs such as audio or audio-visual works and said method is adapted to receive "pause" and "play" requests, which further comprises:

on receipt of a "pause" request for a program, asserting a pause flag in the command word for said program;

on receipt of a "play" request for a program, negating said pause flag; and

wherein said processing step further comprises;

when said pause flag is asserted, inhibiting the sending of data-block portions for said program.

31. The method of claim 30 adapted for sending sequences of programs and wherein each said command word also contains a first identity for the program associated with said command word and a second identity for the next program to be sent, which further comprises:

assigning a token flag to each possible identity; and

wherein said processing step further comprises;

asserting said token flag assigned to said second identity in the command word being processed; and

if the token flag assigned to said first identity in said command word was asserted during the previous periodic processing of said command words, asserting the pause flag in said command word to place the program associated with said command word in the pause state;

whereby all programs in said sequence except the first are maintained in the pause state.

32. The method of claim 31 adapted to receive "fast forward" requests wherein each said command word also contains a quantity representing an amount of the program to be played, which further comprises:

storing a list of subprogram addresses, each said subprogram address identifying the start of a subprogram within the current program;

on receipt of a "fast forward" request for said current program;

retrieving the present address for said current program from its command word;

determining the subprogram following said present address from said list;

creating and storing a new command word to play a portion of said next subprogram, beginning at the address stored for said subprogram, said new command word having as its first identity the second identity from the command word for the current program;

deleting the command word for the current program;

repeating said step of creating and storing a new command word for each subprogram in sequence, each said new command word having as its first identity the second identity for the previous command word; and

upon receipt of a "play" request for said current program;

creating and storing a new command word to play the remainder of said current program, beginning at the address stored in said list for the next subprogram, said new command word having as its first identity the second identity for the previous command word.

33. The method of claim 31 adapted to receive "fast reverse" requests wherein each said command word also contains a quantity representing the amount of a program to be played, which further comprises:

storing a list of subprogram addresses, each said subprogram address identifying the start of a subprogram within the current program;

on receipt of a "fast reverse" request for said current program;

retrieving the present address for said current program from its command word;

determining the address of the subprogram preceding said present address from said list;

creating and storing a new command word to play a portion of said preceding subprogram, beginning at the address stored for said subprogram in said list, said new command word having as its first identity the second identity from the command word for the current program;

deleting the command word for the current program;

repeating said step of creating and storing a new command word for each subprogram in reverse sequence, each said new command word having as its first identity the second identity for the previous command word; and

upon receipt of a "play" request for said current program;

creating and storing a new command word to play the remainder of said current program, beginning at the address of the preceding subprogram, said new command word having as its first identity the second identity for the previous command word.

34. The method of claim 31 wherein said program is an audio/visual work comprising one or more audio and one or more video tracks to be played simultaneously, each said command word also contains an incremental value relating to the time desired between the sending of successive portions of its associated data block and said creating and storing step further comprises:

creating and storing a command word for each audio track and each video track, the first and second identities being the same in all said command words, said channel identities and said addresses in data memory being different in each said command word and said incremental values in said command words being selected appropriately for each respective track.

35. The method of claim 34 wherein each program in a sequence comprises a group of one or more audio and/or video tracks to be played simultaneously, wherein said creating and storing step further comprises:

creating and storing the command words for all the audio and/or video tracks in each group with the same first and second identities.

36. The method of claim 34 adapted to show motion pictures in different languages in accordance with received requests, wherein said retrieving and storing step further comprises:

upon receipt of a request for a motion picture in a specified language, retrieving and storing a first data block for the video track of said motion picture and a second data block for the audio track of said motion picture in said specified language.

37. The method of claim 34 adapted to show motion pictures with subtitles in different languages in accordance with received requests, wherein said retrieving and storing step further comprises:

upon receipt of a request for a motion picture with subtitles in a specified language, retrieving and storing a first data block for the video track of said motion picture, a second dam block for the audio track of said motion picture and a third data block for the subtitles in the requested language.

38. The method of claim 31 adapted to show an abridged version of a program, which further comprises:

creating a list of scene addresses for the scenes in said abridged version, each scene address identifying the start of a scene in said program;

wherein said retrieving and storing step further comprises;

upon receipt of a request for said abridged version;

retrieving the data block for said program from said storage library and storing said data block in said random-access memory; and

wherein said creating and storing step further comprises;

sequentially creating and storing command words for each scene in said list of scene addresses.

39. The method of claim 29 adapted to change programs in accordance with a received request, which further comprises:

upon receipt of a request to change programs from a first version currently being sent to a second version;

retrieving the present address for said first version from the command word for the data block for said first version;

retrieving from said program library and storing in said data memory at least the portion of the data block for the second version corresponding to the remainder of the program;

creating and storing a new command word for the remainder of said second version and

deleting the command word for said first version.

40. The method of claim 39 in which said first version is a motion picture soundtrack in a first language and said second version is the soundtrack for said motion picture in a second language.

41. The method of claim 39 in which said first version is a motion picture having a first rating and said second version is said motion picture having a second rating.