JP4460159B2 - Preprocessed information embedding system - Google Patents

Description

[0001]
  The present invention provides copy protection data.insideIt relates to a method and apparatus for providing embedded digital or analog content such as audio or video.
[0002]
The ability to communicate copyrighted entertainment directly to typical consumers is increasing rapidly. This is the case for cable TV networks and other means when using the Internet, a viewing system with payment. This increased capacity, along with it, brings many benefits to consumers. For example, consumers can sample content (eg, audio or video) while online and purchase content at any time of the day. However, one obvious drawback of this system is that it provides each consumer with a high-quality master copy of the content. In many circumstances, this is nothing more than providing consumers with high-quality entertainment. But unfortunately, this gives conscientious consumers (eg, pirates) a means of making illegal copies of content with little effort.
[0003]
  In particular, there is the potential problem of making unauthorized copies and widespread distribution, for example, by computer networks such as the Internet. Improper dubbing of storage media such as compact discs, digital video discs or magnetic tapes, distribution of multiple copiesUConventional ones are also a problem. In all cases, it is desirable to include information in the first copy sent to consumers who specify a specific copy belonging to a specific recipient. Informing recipients of the existence of this embedded information will discourage potential pirates from using content illegally. Many copies by the authoritiesOriginalWill be able to track.
[0004]
Even this new capacity must be able to perform its function economically. In other words, it is not feasible to implement marking measures that cost more than suppressing pirates. On the other hand, if a low-cost solution is possible, security can be improved and the problem can be substantially reduced.
[0005]
It is therefore desirable to provide a system for marking content in a cost effective manner.
[0006]
There are many techniques for embedding information in content. Each has advantages and disadvantages, and their common use is that some computation is required. Processing hardware must be able to perform the necessary calculations quickly. If the hardware is not fast enough, for example in response to a user downloading data from the network, there will be an undesired latency in delivery time. In addition, there is a potential limitation on the overall throughput of the transmission system, which limits the number of users who can download data or access the network simultaneously. Furthermore, it would not be possible and legally feasible for a legitimate online distributor to get faster hardware.
[0007]
Therefore, a system that reduces real-time computing conditions to embed copy protection data in digital or analog content (eg, audio, video, computer games, stock price and weather data information services, online shopping or e-commerce data, etc.) It is desired to provide.
[0008]
  Pre-processed a selected number of copies of the same content, then properly encoded, eg suitable for downloading by the userdataIt would be desirable to provide a system that dynamically selects these preprocessed copies to generate a signal.
[0009]
  The system must provide the ability to distribute pre-processed content to multiple users simultaneously, but encodeddataThe signal is generated at the user location according to the ID value given to the user.
[0010]
It is desirable to provide multiple layers that embed data.
[0011]
It is desirable to provide binary or multi-level, non-binary data embedding.
[0012]
It would be desirable to provide a technique that provides a smooth transition between two data streams.
[0013]
  Reduces delivery delay and network transmission throughputTheIt would be desirable to provide an improved online distribution system.
[0014]
For example, it should be possible for content to be processed on an offline basis by an online distributor using available hardware.
[0015]
  In this system, for example, the content is sent to the user by e-mail,OfferIt must also be suitable for the off-line distribution system provided.
[0016]
In this case, the content may be stored on a compact disk (CD), digital video disk (DVD), computer floppy disk, or the like.
[0017]
The present invention provides a system having the above and other advantages.
[0018]
Summary of invention
There are many applications that depend on the ability to transmit content (eg, audio, video and / or other data). Increasingly, in order to protect the intellectual property rights of copyright holders, including authors, performers, etc., such transmissions need to be marked in a way that identifies specific copies to belong to a specific recipient. There is. Preferably, this marking is provided in a secure manner. The most available marking solutions embed information in the content, thereby reducing the possibility of exchanging or deleting marking information.
[0019]
The marking is particularly important for e.g. music, video or other digital or analog work that is downloaded by a computer such as the Internet, cable or satellite television network, telephone network. Typically, users pay a fee to download content, but some are free for content samples or other promotional ones.
[0020]
However, while the ability to download content is conveniently provided to many legitimate users, non-licensers, such as pirates, use various techniques to illegally copy and distribute content. can do. This causes significant losses for content providers and online distributors.
[0021]
In order to facilitate tracking this illegal distribution, information identifying the recipient (eg, account number, social security code, or other unique identifier) is embedded in this content.
[0022]
The presence of identification information can be advertised to potential pirates and can facilitate tracking pirates without warning.
[0023]
The present invention particularly relates to an online music distribution system (such as an Internet website via a computer network for a user to purchase a generally distributed audio program for retailing compact discs or magnetic tapes. Can be used to access any distribution site). The present invention is also suitable for use with video, images, or other content to which embedded information is applied. For example, a two-way cable network allows viewers to download digital audio or video content.
[0024]
  The use of the present invention by online music distributors is a pirate of recorded music.EditionIs particularly relevant because it causes significant losses in the music industry.
[0025]
There are several options for embedded information for online music distribution and other applications.
[0026]
For example, pre-embedded copies can be stored with sufficient quality without delaying download requests. For downloaded audio data such as popular songs, this requires that many uniquely identified copies result in increased costs associated with increased storage at the server.
[0027]
Also, information can be embedded during the transaction (eg, US Pat. No. 5,687,191 (title “Post Compress Hidden Data Transport”) or US Pat. No. 5,822,360 (title “Method and Apparatus for Transporting Auxiliary Data in Audio Signals”). ). The approach described in US Pat. No. 5,822,360 is by additional computer processing, but only the necessary copies are processed and no additional server space (eg, memory) is required.
[0028]
Another option disclosed here requires that two copies of the content be pre-processed. The copy is stored at the server (in this case a unique copy consists of two pre-processed copies and is typically provided to a single user when a download is requested.
[0029]
  For example, two server disks can be used to store each of the preprocessed copies of the audio. The first disk contains all embedded copies with “0” and the second disk contains all embedded copies with “1”. Each server is delivered to the userData signalIs connected to a selector function that selects one of the servers for each segment of content.
[0030]
Based on the account number or other unique identifier to be embedded, the selector function selects a segment from each server for each segment. The output of the selector function is a copy to be delivered to the consumer.
[0031]
Alternatively, two copies of the content may be distributed to one or more users, in which case the user is given the appropriate processing power to make up a unique copy. If cryptographic precautions are employed, the user cannot reliably access the two copies before embedding the identification data. These copies can be distributed to many users at the same time to broadcast concerts or other live events over the Internet.
[0032]
  In accordance with the present invention,Consists of multiple segments,ConsecutiveA method for providing a data signal composed of a content signal having embedded logic values pre-processes a first copy of a plurality of segments of the content signal to embed information representing the first logic value; Providing at least a first preprocessed segment and preprocessing a second copy of the plurality of segments of the content signal to embed information representing a second logical value different from the first logical value; Providing at least a second preprocessed segment.The first and second preprocessed segments are optionally stored, for example, on the online distributor's server.
[0033]
  Specify consecutive logical valuesChoiceGiven a signal,ChoiceIn response to the signal, the corresponding first and second specific pre-processed segmentAre combinedTheA data signal in which successive logical values are embedded is given.
[0034]
The first and second logical values may consist of binary bits (eg, the first and second values can each indicate zero or one).
[0035]
  dataIf the signal segment contains audio data,datasignalInEmbeddingMaretaThe signal may be provided at a desired audible level.dataWhen the signal segment contains video data,datasignalInEmbeddingMaretaInformation may be provided at a desired visual level.
[0036]
  Consecutive logical values, like online distributors,dataSignalSourceYou may identify.
[0037]
  further,ChoiceThe signal is based on user request,To retrieve the data signalGiven (in this case, successive logical values can identify the user).
[0038]
Successive logic values may be provided in a scrambled sequence that is encrypted and scrambled, for example, to stop operation by a pirate.
[0039]
  In the scrambling process, the corresponding first and second specific pre-processed segments are:Data in which consecutive logical values are embeddedTo give a signal,ChoiceTime multiplexed in response to the signal.
[0040]
  dataThe signal may be digital or analog.
[0041]
  Optional but embedded informationMulti-layered dataIt may be given to the signal.
[0042]
  In addition, the transition between embedded segments is optionalfunctionIt becomes smooth according to.
[0043]
  A data segment that has become a smooth transition in advancedataIt may be given to the signal.
[0044]
  Binary or multi-level (M> 2) logical value isdataIt may be given to the signal.
[0045]
Corresponding devices and data signals are also provided.
[0046]
Detailed Description of the Invention
The present invention provides a method and apparatus for embedding information in content in a manner that minimizes computational conditions when embedding. The content in this case is in the form or combination of digital or analog audio, video, images or other media.
[0047]
In accordance with the present invention, pre-processing is performed prior to final processing so that on-the-fly (eg, real-time) embedding can be performed by selecting from specially prepared segments and assembling the entire stream. Can be done. The present invention is particularly suitable for an online distribution model (where content is distributed via a communication channel in response to a user request).
[0048]
  FIG. 1 shows a conventional data embedding system 100. The content in which the data is embedded is subdivided into N frames, with M samples per frame. For example, content is a sign110As shown in FIG. 4, frames C (N−1),... C (1), C (0) are included. For example, user data identifying a user is represented by a data package module 140 (where the data is generally indicated by reference numeral 150 and is represented by frames U (N−1),. Is converted to binary user data having Module 140 may optionally add error correction code, modulation and packet header / trailer to the user data.
[0049]
The data embedding module 120 aligns the packaged data (as indicated by U (0), U (1), etc.) with the respective content frame (C (0), C (1), etc.). , The i-th packetized data bit U (i) 170 is embedded in the corresponding i-th content frame C (i) 160 to give the i-th embedded data frame 180. A continuous frame of embedded content is indicated at 130. The embedding process can use known techniques, including additional techniques such as spread spectrum modulation, as well as techniques that modulate signal parameters or the characteristics of the content itself.
[0050]
The data packaging module 140 typically uses relatively few processing cycles that are comparable to the data embedding module 120.
[0051]
FIG. 2 shows a first embodiment of a data embedding system 200 according to the present invention.
[0052]
  The pre-processing data embedding system according to the present invention is a conventional system that can be used in two steps:embeddedAnd (2) generation of target content.
[0053]
  Data embedding module200Receives the content stream 110 but has two routes220 and 225(One is for generating a binary ‘0’ embedded content stream 230 and the other is a binary ‘1’ embedded content stream.235For generating Two pretreatmentsWasThe content stream is mux250 and corresponds to binary user data itself or from the data packetization module 140ChoiceDepending on the signal, it can be multiplexed into each target embedded stream 230. If U (i) is a non-binary value, the preprocessing stage increases by having more than one output path.
[0054]
This means that with minimal processing power (and also by the data packaging module), the generation of multiple, uniquely identified content streams (for example, if the encoder 200 is transacting or Internet multicasting is applied) To be equal).
[0055]
FIG. 3 illustrates a pre-processing module and online distribution system for distributing content in accordance with the present invention. In this example, assume that a user communicates with online distributor 350 via, for example, a two-way Internet including communication channel 385 and nodes 380 and 390. The online distributor 350 is also a pre-processing module 310 (typically in common with the online distributor 350, but communicates with the online distributor via a communication path. ).
[0056]
The pre-processing module 310 operates on an offline basis, for example, before an online distributor executes an order from a user to download digital content. Optionally, the pre-processing module 310 operates on a real-time basis, such as when live events are received over the communication path 318 and processed immediately for multicast to the user population.
[0057]
In general, new content stored in the content function 315 can be processed immediately after receipt or when convenient. Content (referred to as “first data” or “host waveform”) is provided to the data embedding module 210 to have a logical value (eg, binary zero or one) embedded therein. The content is divided into a plurality of segments at a time, and the logical value is embedded in each segment (described with reference to FIG. 4).
[0058]
The data embedding module 210 can use known techniques to embed data into a ready-made signal. For example, the following US patents and filing techniques incorporated herein by reference may also be used: US Pat. No. 5,822,360 (“Method and Apparatus for Transporting Auxiliary Data in Audio Signals”), US, filed Dec. 6, 1996. Application No. 08 / 764,096 (“Method and Apparatus for Embedding Auxiliary Data in a Primary Data Signal”), US Pat. No. 5,687,191 (“Post Compression Hidden Data Transport”), US Application No. 08 filed Aug. 18, 1997 / 919,434 (“Post Compression Hidden Data Transport for Video”), US Pat. No. 5,719,937 (“Multi-Media Copy Management System”), US Application No. 08 / 919,434 (“Multi-Media” filed on November 25, 1997) Media Copy Management System ”), US application entitled“ Digital Hidden Data Transport ”. Other systems for providing embedded information may also be suitable for use with the present invention.
[0059]
In general, the term “embedded” means that accompanying or auxiliary data is provided to the host waveform, primary data signal, without substantially interfering with the primary data signal. For example, embedded data should not be heard when embedded in an audio signal. Typically, the data rate of embedded data is much lower than that of the primary data signal.
[0060]
For example, US Pat. No. 5,822,360 discloses a technique for embedding data by modulating a spread spectrum signal. Spread spectrum signals have relatively low noise power, but can be recovered at a particular decoder by correlating the received signal with the pseudo-noise (PN) sequence used for spreading at the encoder .
[0061]
Techniques for embedding data often require time domain or frequency domain analysis that is computationally efficient, taking advantage of human auditory and visual characteristics, such that the data is embedded in audio and video data, respectively. The embedded data thus basically disturbs while establishing a useful hidden data channel within the primary data signal.
[0062]
  In the present invention, sophisticated computer robust embedding technology is used when content must be immediately available for delivery to the user.ofBefore, a pretreatment process can be taken. Logical value is,It can be embedded using the most sophisticated technology available, but the content is immediately available for download or broadcast to the user. In memory 322, content segments with embedded binary zeros are stored in memory portion 325, while content segments with embedded binary 1 are stored in memory portion 330. When two or more logical values are embeddedTo come, It will be appreciated that a memory portion is provided for each value. Further, in practice, a library of content may be stored in functions 325 and 330, including, for example, songs, movies, computer games, and the like.
[0063]
The memory 322 may be associated with a web service, for example.
[0064]
The memory 322 acts only as a buffer when it is desired to pass the content received (via path 318) in the preprocessing module 310 to the user without sending a delay or without delay. For example, the data rate change is made uniform, and the data embedding module 210 processes the time.
[0065]
Online distributor 350 can maintain a database 360 that includes an available identification number 362 and a user record 364. Available identification numbers may simply be sequential numbers or other codes. In practice, the available identification number function 362 can maintain a current order (or user) number (which is then incremented for each new order (or user)). User record function 364 maintains a record of the identification number associated with each user or order. The term “order” means fulfilling user requests for free samples, promotional programs, content, etc. as well as paid content.
[0066]
  When a user request is received at Control 365 or control data is received via communication path 352 (which may be the same as or different from path 318), an identification number or code is associated with the user or orderDateThe record is written by the user recording function 364. As a result, if an illegally copied content is found, it is possible to locate the user who first obtained the content.
[0067]
  The identification number or code may also identify an online distributor or other entity, such as the copyright holder of the content, or give a registration number to, for example, an industry policing organization.YeahAlso good.
[0068]
Copy protection is also a primary object of the present invention, but it is understood that embedded data can be used for any essential purpose, especially when the user has a receiver that can read the data. Like. For example, the embedded data can be used to identify online distributors or other entities, give electronic coupons and deliver prizes to users. Corresponding receivers can read the embedded data, for example displaying corresponding information to the user.
[0069]
At the user's premises 395 or elsewhere, the user can order and / or receive content using a personal computer (PC) 396, a television set top box 397, or any other available means.
[0070]
  In response to a user request or control data via path 352, control 365 passes to selector 370.Like selection signalGive identification number or code, Based on time multiplexFor example, a multiplexer (MUX) 375 is used to select different segments with embedded logic values from functions 325 and 330. The selector isdataLike a signal, it acts like a switch that allows a user to deliver consecutive segments from functions 325, 330 (not both at the same time).
[0071]
The term “consecutive” as used herein is understood to include both adjacent and non-adjacent segments that follow.
[0072]
The selection signal may be a binary identification signal or a signal derived from the identification signal. Encryption techniques may be used for this purpose (eg, to convert the identification signal to a selection signal or vice versa).
[0073]
Appropriate capabilities are provided in the user premises to decrypt the received data.
[0074]
  If the control 365 has bidirectional capabilities, it can also create a record indicating that the content was delivered without error. Also, the delivery time and data can be recorded in the user record 364, for example.dataInformation embedded in the signal may also indicate delivery time and data.
[0075]
The delivered content is given to the user for storage in the PC 396 or the set top box 397, for example. The same or different communication channels can be used for upstream request signals and downstream delivery. As an example of using different channels, upstream requests are provided over the telephone network and downstream delivery is provided over the television network.
[0076]
If sufficient bandwidth is available, the content can be delivered simultaneously to multiple users using separate signals embedded with a unique identification number. If the bandwidth is limited and the user number is large, the system of FIG. 6 described below can be used.
[0077]
FIG. 4 illustrates the selection of a data segment with embedded binary data in accordance with the present invention. The first copy 400 of content includes sequential segments (eg, segment 1 (405), segment 2 (410), segment 3 (415),..., Segment N (420)). Each segment has an embedded logical value (in this example binary 0). Each segment need not have a value embedded in it. In practice, another security factor can be achieved with the present invention by selecting only certain segments for the embedded data, for example according to a pseudo-random signal such as a PN sequence. Also, the embedded values are given in a scrambled order according to any known encryption technique to prevent data manipulation by an attacker. Corresponding information must be given to the decoder to recover scramble or encryption.
[0078]
A second copy 450 of the same content is a continuous segment corresponding to the first copy 400 (eg, segment 1 (455), segment 2 (460), segment 3 (465),..., Segment N (470)). Each segment has a logical value (binary 1 in this example) embedded in it.
[0079]
For example, if the content is an audio track, each segment may be composed of a specific period of the track corresponding to one or more frames of data, for example.
[0080]
  dataSignal 480 is formed by selecting segments from the first and second copies according to the desired embedded bit pattern. For example, if the desired embedded bit pattern is 101 ... 1, segment 1 (455) from copy 2 (450), followed by segment 2 (410) from copy 1 (450), Segment 3 (465) from copy 2 (450), ..., segment N (470) from copy 2 (450) must be selected.dataThe signal 480 thus has the desired bit pattern 101 ... 1 embedded in it.
[0081]
The final composite copy is thus constructed by selecting the previously created segment from either the first or second copy of the audio data with embedded binary information.
[0082]
  As described above, the binary data sequence may identify a user downloading over the network or provide other information. in this case,dataThe signal can be assembled in response to the purchase of content by the user.
[0083]
  The overall effect is that the last copy received is perceptually identical to the same content from the other copy that does not have embedded data.dataThe consumer does not know the state of the information embedded in the signal. For audio data, this result is not audible when the audio data is recovered and playedData embedded at power levelAchieved by giving The embedded data can also be spectrally shaped according to the audio spectrum to further enhance the hidden area.
[0084]
  For video data, the embedded data can be seen when the video data is recovered and displayed.AbsentCan be given at power level.
[0085]
FIG. 5 illustrates content selection according to the identification signal in accordance with the present invention. Copies of content with embedded binary 0's and 1's are denoted by the symbols 500 'and 550', respectively. The processing system includes a selector 500 that selects a segment from a copy according to a unique identification signal, as described in connection with the selector 370 of FIG.
[0086]
  dataSignals can include audio, video (moving pictures, still pictures), computer games or other content. The advantage of using a binary signal is that there are only two logical values and only two copies of the content need to be stored. An M (ary) signaling scheme is used, such as M-level pulse amplitude modulation (PAM), which provides a copy of M content with embedded data.
[0087]
The content data rate can be used for embedded data. This is given for more general product brands (but not system specific restrictions). By using this convention, binary can be applied to specific parts of audio or other content (eg, segments) in the form of a specific number of audio samples in each segment.
[0088]
For example, digital audio found on a compact disc (CD) operates at a rate of 44100 samples per second. In this case, for example, 1000 samples per segment of audio may be used for each binary number of embedded information. That is, one bit of embedded information is distributed over 1000 audio samples. Each segment as described in connection with FIG. 4 thus consists of at least 1000 samples. This means that a desired audio copy can carry approximately 44 bits of embedded (eg, auxiliary) information every second.
[0089]
The first and second copies of the audio data are encoded with “0” and “1” every 1000 sample segments, respectively. Alternatively, the copy has the same audio content. That is, the same audio data is given to the corresponding segment.
[0090]
  FIG. 6 shows a second embodiment of the data embedding system according to the present invention. This example is for example for users for live events.Group ofIt is particularly suitable for multicast transmission.
[0091]
Preprocessors (eg, “0” embedding module 220 and “1” embedding module 225) are on the content server / distributor / provider side 610 and processing systems (eg, data embedding modules) are on the client / user side 650 . The “0” and “1” streams from modules 220 and 225 are respectively sent to a number of user terminals 660-A, 660-B,..., 660-X by a stream encryptor / multiplexer unit 630. “Packed” into a single stream to be distributed. Alternatively, the “0” and “1” streams can be provided to the user terminal in separate data streams.
[0092]
For example, the terminal is a set-top box (eg, a decoder) or a personal computer that is coupled to a cable television network and receives content having a television or other signal.
[0093]
Content with embedded logical values is encrypted at function 630 according to the stream key provided by stream access control function 615. The use of encryption keys will be apparent to those skilled in the art, and therefore details thereof are omitted.
[0094]
Only a legitimate user can obtain the stream key from the stream access control unit 615, and two streams in each decryptor / demultiplexer 662-A, 662-B, ... 662-X. Remember again. The user also receives an identification (ID) value from the user key generation unit 620. In each of the data embedding modules 210-A, 210-B,..., 210-X, the ID is embedded in the content as described with reference to FIGS. Both the stream key and ID value are installed on the terminal 660-A, 660-B, ..., 660-X, for example, when the terminal is manufactured, or using a smart card (renewed periodically) As described above, various means can be given by installing in the terminal at that time, or by reliably transmitting to the terminal (using the same or different communication path as in multicast content).
[0095]
  The ID value is embedded in real time by the data embedding modules 210-A, 210-B,..., 210-X on the user side in order to generate user specific content. To prevent copyright infringement, various mechanisms include decryptor / demultiplexer and data embedding module.InA stream of "1" and "0" similar to the ID value cannot be clearly accessed on the client / user side 650, which can be used to ensure a structure that is reliably coupled. For example, "1" and "0" streams are pseudo-randomly exchanged at frame boundaries. Also, the ID value is scrambled beforehand in a corresponding manner, and the correct ID value is encoded by the data embedding module without showing the ID value itself. This ensures the security of the stream being stored or distributed and at the same time preserved even after the ID value has been decrypted.
[0096]
  This equipment can be used by content providers, for example, in cable networks.Paid viewing systemIt is particularly suitable for applications that want to minimize the use of distribution channel bandwidth while content is distributed using. Cable operators need only allocate bandwidth for "1" and "0" streams. The user terminals 660-A, 660-B,..., 660-X generate content that is uniquely marked by the corresponding data embedding modules 210-A, 210-B,. This is the content of the userillegalCopy andillegalHelps prevent redistribution.
[0097]
  In a further improvement, the bandwidth does not always have to be exchanged for the two streams. For example, a distributor can select a specific segment of content, send both a "0" and "1" stream to the user, and send only a "0" or "1" stream at another time. , Or unmarked contentCan even send.
[0098]
Figures 7 and 8 show the multilayer data embedded according to the present invention. Multiple independent streams of data referred to as the data layer are embedded in the same content. The present invention can be adapted for use with multilayer embedding schemes such as those disclosed in US Pat. No. 5,822,360 above. Here, embedding different user data (user data X and Y) has an advantage in using the same technique such as frame synchronization, but a technique of embedding the same or dissimilar data can be used.
[0099]
Multi-layer data applications include providing separate user data data streams for tracking, hyperlinks or electronic coupons, for example.
[0100]
As shown in FIG. 7, a first stream of user data (eg, user data X) is processed by the data packaging module 140 and frames X (N-1),..., X (1), X The corresponding binary user data indicated by the code 710-X with (0) is given.
[0101]
In the embedding module 710-X, the “0” embedding module 220 and the “1” embedding module 225 are used as described above to provide the target streams 230-X, 235-X. The streams 230-X and 235-X are given to the MUX 250 to obtain a content stream 730 having embedded user data X.
[0102]
Similarly, as shown in FIG. 8, a second stream of user data (eg, user data Y) is processed by the data packaging module 140 ′, and frames Y (N−1),. 1), corresponding binary user data indicated by reference numeral 710-Y having Y (0) is given.
[0103]
In the embedding module 710-Y, the “0” embedding module 220 ′ and the “1” embedding module 225 ′ are used as described above to provide the target streams 230-Y, 235-Y. The streams 230-Y and 235-Y are given to the MUX 250 ′ to obtain a content stream 735 having embedded user data Y.
[0104]
  FIG., FIG., Streams 730 and 735 are combined by adder 740 and scaled by scaler 745 to provide a data stream 760 with multi-layer embedded data. For example, a scale factor of 0.5 can be used when there are two streams with different user data. Scaler 745 provides the size of the content and user data in stream 760 at the same level as that of streams 730 and 735.
[0105]
Here, if more than one layer of embedded user data is used, the scaler 745 is adjusted according to the number of layers used.
[0106]
Figures 9 and 10 illustrate multi-level non-binary data embedding according to the present invention. The system can be extended to embed multilevel (non-binary) value data. At the M level, the system is called M-ary (M ≧ 2). For the purpose of illustration, four (M = 4) logical values (0, 1, 2, 3) are given to distinguish them. It should be understood that all multi-level changes can be implemented as a parallel combination of binary data embedding.
[0107]
M = 4 level data embedding module 810 embeds “0” embedding module 220 for embedding logical value “0”, “1” embedding module 225 for embedding logical value “1”, and logical value “2”. The "2" embedding module 840 and the "3" embedding module 850 for embedding the logical value "3" are provided, and the respective data streams 830, 835, 845, and 855 are provided. Data streams 830, 835, 845, and 855 are provided to MUX 250 to provide a content data stream 860 with embedded multi-level user data. As described above, the MUX 250 outputs a continuous frame of data having a desired logical value embedded under the control of the user data stream 150, for example.
[0108]
FIG. 11 shows a multiplexer with a transition / fade function according to the present invention. To ensure a smooth transition at the frame boundary, a clear window / fade in / fade out can be performed at multiplexer 250 ". Multiplexer 250" can be used, for example, instead of multiplexers 250, 250 'described above. .
[0109]
When splicing frames (eg, temporal multiplexers) from different data streams, the content signal cannot be continuous at the boundary between frames. This results in artifacts (eg, audible artifacts for audio content or visual artifacts for video content). Transition periods are provided as described in connection with FIGS. 11, 12 and 13 to avoid these effects.
[0110]
The transition period is typically shorter than the frame length. For example, if the frame length is 2000 samples, the transition length may be 100 to 200 samples.
[0111]
  transitionfunctionIs a linear ramp or exponentfunctionFixed like a decreasefunctionOr adapt to dynamically adjust characteristics based on host signalfunctionIt is. The objective is to ensure that the transition does not generate any artifact that affects the quality of the target content.
[0112]
  Target embedded streams 230 and 235 with embedded logical values 0 and 1 transition in multipliers 930 and 940functionMultiply by 910 and 920. Transition shownfunctionThe 910 ramps from 0 to 1 (in two steps). transitionfunctionWhen 910 reaches 1, transitionfunctionThe 920 begins to ramp from 1 to 0 (in two steps). The outputs of multipliers 930 and 940 are combined by adder 950 to give embedded content 960.
[0113]
  The illustrated embedded content 960 includes a first frame C (1) and a second frame C (0). transitionfunctionThe effects of the transition regions 910 and 920 are shown in regions 965 and 968, respectively.
[0114]
  12 and 13 show a system for transition control between two streams according to the present invention. Another way to ensure a smooth transition at the frame boundary is to provide an additional transition stream. This is shown in Fig. 11.functionTransitions like 910 and 920functionTo eliminate the need for MUX 250 "in Fig. 11 and to assemble the final target stream, the pre-processed frames are, for example, Using MUX 250 or 250 ′, temporal multiplexing is performed as described above.
[0115]
User data frames U (N-1),..., U (1 → 2), U (1), U (0 → 1), U (0) are given. U (1 → 2) represents the transition frame between frames U (1) and U (2), and U (0 → 1) represents the transition between frames U (0) and U (1) Represents a frame.
[0116]
The “0” data embedding module 220 gives a content frame 1030 in which a logical value “0” is embedded, and the “1” data embedding module 225 gives a content frame 1035 in which a logical value “1” is embedded.
[0117]
The first and second transition streams 1050 and 1055 are generated by the embedded modules T0 → 1 (1035) and T1 → 0 (1045).
[0118]
Here, the embedded frame marked with 'X' does not need to be generated since it is not selected for target content generation. This is true because transitions are always limited to transition frames (eg, C (1 → 2), C (0 → 1), etc.).
[0119]
The final target content stream 960 including the transition frames C (1 → 2) and C (0 → 1) is output from the MUX 250 based on the selection signal from the user data stream 1020.
[0120]
Various other modifications are possible to the invention, including:
• Use pre-processed embedded systems with compressed content. In other words, embedding binary data in compressed audio, video or other digital data.
Use unequal length segments or variable data rates.
• Base segment boundaries on error correction boundaries, packet boundaries, or other signal specific structures.
• Configure unique identification (ID) values or add extras (eg, error correction or error checking) to prevent attacker collusion. As an example, instead of the binary phase shift key (BPSK) that uses one PN used for both "0" and "1", one PN sequence used for "0" and the other PN sequence used for "1" Is used.
[0121]
Thus, it can be seen that the present invention provides a system in which a composite data signal is provided to a user with embedded user identification information.
[0122]
  In an embodiment particularly suitable for online distribution of content, two copies of the content can be preprocessed and stored, for example, on a server used by the online distributor. Each copy has data embedded in a continuous segment indicating a logical value such as binary 0 or 1. Segments are assembled according to the desired identification signal and distributed to the userdataGive a signal. If an unauthorized copy or distribution of content occurs, the original user is therefore detected from the unauthorized copy.
[0123]
Optionally, instead of preprocessing and storing two copies of the content, the content is processed and sent to the user on a real-time basis. This is especially desirable when the content is a live event.
[0124]
In another modification that is particularly suitable for multicast transmission to the user terminal, a data embedding module is provided at the user terminal.
[0125]
Normally, the embedding process is performed in the digital domain (however, information is once embedded), but is carried to a host signal in a digital or analog form.
[0126]
Although the invention has been described in connection with various specific embodiments, it will be understood that various adaptations and modifications can be made without departing from the spirit and scope of the invention as set forth in the claims. Those skilled in the art will understand.
[Brief description of the drawings]
FIG. 1 shows a conventional data embedding system.
FIG. 2 shows a first embodiment of a data embedding system according to the present invention.
FIG. 3 illustrates a pre-processing module and online distribution for distributing digital or analog content in accordance with the present invention.
FIG. 4 illustrates the selection of data segments with embedded binary data according to the present invention.
FIG. 5 shows content selection according to an identification signal in accordance with the present invention.
FIG. 6 shows a second embodiment of the data embedding system according to the present invention.
FIG. 7 illustrates multi-layer data embedding according to the present invention.
FIG. 8 illustrates multi-layer data embedding according to the present invention.
FIG. 9 illustrates multi-level, non-binary data embedding according to the present invention.
FIG. 10 illustrates multi-level, non-binary data embedding according to the present invention.
FIG. 10 shows a transition according to the present invention.functionA multiplexer with
FIG. 12 shows a system for transition control between two streams according to the present invention.
FIG. 13 shows a system for transition control between two streams according to the present invention.

Claims (27)

A method for providing a data signal composed of a content signal composed of a plurality of segments and embedded with continuous logical values,
Pre-processing a first copy of the plurality of segments of the content signal to embed information representing a first logical value to provide at least a first pre-processed segment;
Pre-processing a second copy of the plurality of segments of the content signal to embed information representing a second logic value to provide at least a second pre-processed segment;
According to the selection signal for specifying the sequential logic value, the step of combining a specific segment of the first pre-treated segment and said second preprocessed segments providing said data signal,
Including methods. The method of claim 1, wherein the first and second logical values comprise binary bits. The segment of the data signal comprises an audio signal;
Information embedded in the data signal, when the data signal is reproduced, given as inaudible method according to claim 1 or 2. The segment of the data signal comprises video data;
Information embedded in the data signal, when said data signal is displayed, is provided so that it can not be seen, the method according to any one of claims 1 to 3. The contiguous logical value is one that identifies the source of the data signal, method according to any one of claims 1 to 4. The method according to claim 1, wherein the consecutive logical values are given in an encrypted manner. In the step of providing a data signal by combining the said particular segment, the are time multiplexed to provide a data signal, method according to any one of claims 1 to 6. 8. A method according to any preceding claim, wherein the data signal comprises digital data. 8. A method according to any preceding claim, wherein the data signal comprises analog data. further,
Storing the first and second preprocessed segments in a storage device prior to providing the combined data signal ;
Providing the data signal in combination by providing a specific segment of the first and second preprocessed segments from the storage device in response to the selection signal ; Item 10. The method according to any one of Items 1 to 9. The selection signal is provided in accordance with a user request to receive the data signal;
The contiguous logical value is one for identifying the user,
The method according to claim 1. further,
Said first and second preprocessed segments from distributor of the content delivery network comprises the step of communicating at least one user terminal in the network,
The user terminal is located away from the distributor;
Wherein the step of providing the data signals by combining, in the user terminal, Ru is performed after the first and second pre-processing segment is received, the method according to any one of claims 1 to 11. further,
Including providing an identification value to the user terminal;
The contiguous logical value is determined according to the identification value, The method of claim 12. The contiguous logical value A method according to any one of the data signal identifying a user who is given, claims 1 to 13. The method of claim 1, wherein embedded information is provided in multiple layers in the data signal . 16. A method according to any one of the preceding claims, wherein transition regions between the particular segments are smoothed according to a transition function . 17. A method according to any of the preceding claims, wherein an additional data stream consisting of preprocessed transition regions is provided corresponding to the boundaries of the first and second preprocessed segments . Wherein the plurality of segments, said data signal, M> to provide a multi-level logic value with two levels, pre is processed, the method according to any one of claims 1 to 17. 19. A method as claimed in any preceding claim , wherein the second copy of the plurality of segments is preprocessed to embed information representing a second logical value. It said first and second pre-processed segments are stored, the assembly process is performed at the distributor of the content delivery network, The method according to any one of claims 1 to 19. The pre-processing is executed by a pre-processing module of the content distribution network,
The assembly process is performed at a network distributor,
Said first and second pre-treated segment, in response to a request signal from the user terminal of the network for the distribution is given from the pre-processing module to the distributor, one of the claims 1 to 20 The method of crab. The method according to any one of claims 1 to 21, wherein the plurality of segments comprises compressed content. further,
Storing the first and second preprocessed segments in a storage device prior to combining and providing a data signal ;
Receiving a request from a plurality of users at a distributor provided in the network ;
For each of the plurality of users, by reading and combining at least a particular segment of the first and second preprocessed segments from the storage device according to each selection signal designating a continuous logical value Assembling customized data signals in communication with each of the plurality of users according to each user's request ;
The method according to claim 1, comprising: 24. The method of claim 23, wherein each of the successive logical values provided in the customized data signal provides information associated with each user. further,
25. A method according to any preceding claim, comprising multicasting the data signal to a plurality of users on the net. further,
Receiving a request from a plurality of users of the network at a distributor provided in the network;
Each user, according to the request, a step of assembling a customized data signals to be distributed,
26. A method according to any of claims 1 to 25, comprising: A device for providing a data signal comprising a content signal composed of a plurality of segments and embedded with continuous logic,
A first copy of the plurality of segments of the content signal is pre-processed to embed information representing a first logical value to provide at least a first pre-processed segment; Means for pre-processing a second copy of the plurality of segments to embed information representative of a second logical value to provide at least a second pre-processed segment;
Means for combining said first preprocessed segment and said specific segment of said second preprocessed segment to provide said data signal according to a selection signal designating said successive logic values;
Including the device.

Images