Patent Grant
8160249
This invention relates to digital signal processing, and more particularly to a method and a system for encoding at least one digital watermark into a signal as a means of conveying information relating to the signal and also protecting against unauthorized manipulation or use of the signal.
Many methods and protocols are known for transmitting data in digital form for multimedia applications (including computer applications delivered over public networks such as the interne or World Wide Web (“WWW”). These methods may include protocols for compression of data, such that it may more readily and quickly be delivered over limited bandwidth data lines. Among standard protocols for data compression of digital files may be mentioned the MPEG compression standards for audio and video digital compression, promulgated by the Moving Picture Experts Group. Numerous standard reference works and patents discuss such compression and transmission standards for digitized information.
Digital watermarks help to authenticate the content of digitized multimedia information, and can also discourage piracy. Because piracy is clearly a disincentive to the digital distribution of copyrighted content, establishment of responsibility for copies and derivative copies of such works is invaluable. In considering the various forms of multimedia content, whether “master,” stereo, NTSC video, audio tape or compact disc, tolerance of quality will vary with individuals and affect the underlying commercial and aesthetic value of the content. It is desirable to tie copyrights, ownership rights, purchaser information or some combination of these and related data into the content in such a manner that the content must undergo damage, and therefore reduction of its value, with subsequent, unauthorized distribution, commercial or otherwise. Digital watermarks address many of these concerns. A general discussion of digital watermarking as it has been applied in the art may be found in U.S. Pat. No. 5,687,236 (whose specification is incorporated in whole herein by reference).
Such prior art applications have been drawn to providing basic digital watermarking functionality. For instance, it has been known to provide an apparatus or method for encoding or decoding independent information, including a digital watermark, represented as a series of data bits into or out of a series of digitized samples, wherein the apparatus contained:
Further applications of basic digital watermarking functionality have also been developed. Examples of such applications are shown in U.S. Pat. No. 5,889,868 (whose specification is incorporated in whole herein by reference). Such applications have been drawn, for instance, to implementations of digital watermarks that were deemed most suited to particular transmissions, or particular distribution and storage mediums, given the nature of digitally sampled audio, video, and other multimedia works. There have also been developed techniques for adapting watermark application parameters to the individual characteristics of a given digital sample stream. and for implementation of digital watermarks that are feature-based—i.e., a system in which watermark information is not carried in individual samples, but is carried in the relationships between multiple samples, such as in a waveform shape. For instance, natural extensions may be added to digital watermarks that may also separate frequencies (color or audio), channels in 3D while utilizing discreteness in feature-based encoding only known to those with pseudorandom keys (i.e., cryptographic keys) or possibly tools to access such information, which may one day exist on a quantum level.
A matter of general weakness in digital watermark technology relates directly to the manner of implementation of the watermark. Many approaches to digital watermarking leave detection and decode control with the implementing party of the digital watermark, not the creator of the work to be protected. This weakness removes proper economic incentives for improvement of the technology. One specific form of exploitation mostly regards efforts to obscure subsequent watermark detection. Others regard successful over encoding using the same watermarking process at a subsequent time. Yet another way to perform secure digital watermark implementation is through “key-based” approaches.
This paper draws a distinction between a “forensic watermark,” based on provably-secure methods, and a “copy control” or “universal” watermark which is intended to be low cost and easily implemented into any general computing or consumer electronic device. A watermark can be forensic if it can identify the source of the data from which a copy was made. For example, assume that digital data are stored on a disk and provided to “Company A” (the “A disk”). Company A makes an unauthorized copy and delivers the copy to “Company B” (the “B disk”). A forensic watermark, if present in the digital data stored on the “A disk,” would identify the “B disk” as having been copied from the “A disk.”
On the other hand, a copy control or universal watermark is an embedded signal which is governed by a “key” which may be changed (a “session key”) to increase security, or one that is easily accessible to devices that may offer less than strict cryptographic security. The “universal” nature of the watermark is the computationally inexpensive means for accessing or other associating the watermark with operations that can include playback, recording or manipulations of the media in which it is embedded.
A fundamental difference is that the universality of a copy control mechanism, which must be redundant enough to survive many signal manipulations to eliminate most casual piracy, is at odds with the far greater problem of establishing responsibility for a given instance of a suspected copying of a copyrighted media work. The more dedicated pirates must be dealt with by encouraging third party authentication with “forensic watermarks” or those that constitute “transactional watermarks” (which are encoded in a given copy of said content to be watermarked as per the given transaction).
The goal of a digital watermark system is to insert a given information signal or signals in such a manner as to leave little or no evidence of the presence of the information signal in the underlying content signal. A separate but equal goal is maximizing the digital watermark's encoding level and “location sensitivity” in the underlying content signal such that the watermark cannot be removed without damage to the content signal.
One means of implementing a digital watermark is to use key-based security. A predetermined or random key can be generated as a map to access the hidden information signal. A key pair may also be used. With a typical key pair, a party possesses a public and a private key. The private key is maintained in confidence by the owner of the key, while the owner's public key is disseminated to those persons in the public with whom the owner would regularly communicate. Messages being communicated, for example by the owner to another, are encrypted with the private key and can only be read by another person who possesses the corresponding public key. Similarly, a message encrypted with the person's public key can only be decrypted with the corresponding private key. Of course, the keys or key pairs may be processed in separate software or hardware devices handling the watermarked data.
Two conventional techniques for providing key-based confidentiality and/or authentication currently in use involve reciprocal and non-reciprocal encrypting. Both systems use non-secret algorithms to provide encryption and decryption, and keys that are used by the algorithm.
In reciprocal algorithm systems, such as DES, the same key and algorithm is used both to encrypt and decrypt a message. To assure confidentiality and authenticity, the key should be known only to the sending and receiving computers, and were traditionally provided to the systems by “secure” communication, such as courier.
In the prior art there have been developed systems wherein a common key may be developed by the sender and receiver using non-secure communications. In such systems, as described in U.S. Pat. Nos. 4,200,770, 5,375,169 and 5,583,939, each party to a communication generates a numerical sequence, operates on the sequence and transfers the result to the other party. By further operation using the transferred result and the locally generated sequence, each party can develop the identical enciphering key, which cannot be obtained from the transferred results alone.
As implemented for use over the interne, the most common prior art encryption systems are those denoted by the Secure Socket Layer (SSL) and IPSEC protocols.
In non-reciprocal systems, such as described in U.S. Pat. No. 4,218,582, a first party to a communication generates a numerical sequence and uses that sequence to generate non-reciprocal and different encrypting and decrypting keys. The encrypting key is then transferred to a second party in a non-secure communication. The second party uses the encrypting key (called a public key because it is no longer secure) to encrypt a message that can only be de-crypted by the decrypting key retained by the first party. The key generation algorithm is arranged such that the decrypting key cannot be derived from the public encrypting key. Similar methods are known for using non-reciprocal keys for authentication of a transmission. In this application, the non-secure “public” key is used to a message that has been encrypted using a secure “private” key known only to the originating party. In this method the receiving party has assurance that the origination of the message is the party who has supplied the “public” decrypting key. Prior art systems for key generation have often relied upon supposedly-random or quasi-random numbers generated by a fixed mathematical algorithm.
Adaptations of key systems specifically used in conjunction with digital watermarking have been developed, as disclosed in, for example, U.S. Pat. No. 5,822,432 (which is incorporated in whole herein by reference). Such adaptations have included, for instance, providing methods for the human-assisted generation and application of pseudorandom keys for the purpose of encoding and decoding digital watermarks to and from a digitized data stream. In such methods, a pseudorandom key and key application “envelope” may be generated and stored using guideline parameters input by a human engineer interacting with a graphical representation of the digitized data stream. Key “envelope” information is permanently associated with the pseudo-random binary string comprising the key. Key and “envelope” information may then be applied in a digital watermark system to the encoding and decoding of digital watermarks. Such a method can improve encoding and decoding with digital watermarks by providing: separation of the encoder from the decoder; increased information capacity (relative to spread spectrum methods); destruction or degradation of content when attempts to erase watermarks take place; detection of presence of watermarks without ability to access watermark information; multi-channel watermark capability; use of various classes of keys for watermark access control; support for alternative encoding, decoding, or other component algorithms; and/or use of a digital notary to authenticate and time stamp watermark certificates.
While, as described above, various prior art approaches do exist for implementation of digital watermarking (though not necessarily for forensic or copy control use), there are additional desirable features for digital watermarking systems that are not currently believed to be available. For instance, it would be desirable to be able to secure a data signal by using data reduction techniques to reduce the data signal into a reduced data signal; in conjunction with cryptographic techniques, so that an output signal can reliably and efficiently be securely delivered.
It would further be advantageous to user remainder signals (produced by data reduction techniques) as a vehicle for performing encryption upon and using in conjunction with encrypting/decrypting of a data signal to be secured.
It would likewise be desirable to combine data reduction techniques to reduce a data signal into a reduced data signal; produce a remainder signal from the data signal; and then embed complementary watermarks in reduced data signal and the remainder signal, for effective and secure delivery of an output signal.
It would still further be desirable to combine scrambling techniques in conjunction with data reduction techniques such that data signals can be reduced and transmitted on a secured basis.
It would likewise be desirable to provide cost-efficient and universal systems for digital watermarking, and to provide systems adaptable both to copy protection and forensic tracing of “pirated” data signals to detect and deter unauthorized copyists thereof.
It would also be desirable to provide a system of digital watermarking that is highly compatible with known and future methods for compression of data used in conjunction with electronic transmission thereof. It would further be desirable to provide digital watermarking techniques in conjunction with known and effective “key” systems for cryptography and data signal protection.
The prior art does not meet these needs.
The present invention provides a method of securing a data signal which comprises the steps of: applying a data reduction technique to reduce the data signal into a reduced data signal; embedding a first watermark into said reduced data signal to produce a watermarked, reduced data signal; and adding said watermarked, reduced data signal to said remainder signal to produce an output signal.
The present invention also provides a method of securing a data signal which comprises the steps of: applying a data reduction technique to reduce the data signal into a reduced data signal; subtracting said reduced data signal from the data signal to produce a remainder signal; embedding a first watermark into said reduced data signal to produce a watermarked, reduced data signal; embedding a second watermark into said remainder signal; to produce a watermarked remainder signal; and adding said watermarked, reduced data signal to said watermarked remainder signal to produce an output signal.
The present invention also provides a method of securing a data signal which comprises the steps of: applying a data reduction technique to reduce the data signal into a reduced data signal; subtracting said reduced data signal from the data signal to produce a remainder signal; using a first scrambling technique to scramble said reduced data signal to produce a scrambled, reduced data signal; using a second scrambling technique to scramble said remainder signal to produce a scrambled remainder signal; and adding said scrambled, reduced data signal to said scrambled remainder signal to produce an output signal.
The present invention also provides a method of securing a data signal which comprises the steps of: applying a data reduction technique to reduce the data signal into a reduced data signal; subtracting said reduced data signal from the data signal to produce a remainder signal; using a first cryptographic technique to encrypt the reduced data signal to produce an encrypted, reduced data signal; using a second cryptographic technique to encrypt the remainder signal to produce an encrypted remainder signal; and adding said encrypted, reduced data signal to said encrypted remainder signal to produce an output signal.
The present invention also supplies a system for securing a data signal which comprises: means to apply a data reduction technique to reduce the data signal into a reduced data signal; means to subtract said reduced data signal from the data signal to produce a remainder signal; means to apply a first cryptographic technique to encrypt the reduced data signal to produce an encrypted, reduced data signal; means to apply a second cryptographic technique to encrypt the remainder signal to produce an encrypted remainder signal; and means to add said encrypted, reduced data signal to said encrypted remainder signal to produce an output signal.
The present invention also supplies a system for securing a data signal which comprises: (a) a computer processor; (b) at least one computer memory; (c) a data reduction algorithm; and (d) at least one digital watermarking algorithm, wherein said computer processor is supplied with programming in conjunction with said computer memory: (I) to apply said data reduction algorithm to the data signal to yield a reduced data signal; and to subtract said reduced data signal from the data signal to produce a remainder signal; (II) to embed a first watermark into said reduced data signal by application of said at least one digital watermarking algorithm to produce a watermarked remainder signal; and (IV) to add said watermarked, reduced data signal to said watermarked remainder signal to produce an output signal.
The present invention also provides a method of securing a data signal which comprises the steps of: evaluating the data signal to determine its characteristics and reducibility; selecting at least one appropriate data reduction technique for the data signal based on the data signal's characteristics; applying said at least one appropriate data reduction technique to the data signal to produce a reduced data signal; embedding at least one digital watermark in the reduced data signal; and supplying an output signal corresponding to the data signal, said output signal comprising said watermark and said reduced data signal.
The present invention also supplies a method for the protection of a data signal, comprising the steps of: (a) defining and analyzing a plurality of data substreams within the data signal; (b) associating at least one key or key pair with data reduction digital watermarking for at least one of said data substreams; (c) employing said at least one key or key pairs for at least one step selected from the group of: (i) identifying at least one associated watermark; (ii) encoding at least one associated watermark; (iii) detecting at least one associated watermark; or (iv) decoding at least one associated watermark.
A method for protected distribution of a data file is also provided, which method comprises: (a) embedding one or more digital watermarks in the data file using data reduction techniques in creating said digital watermark; (b) and distributing the digitally watermarked file to an end user.
Also provided is a method for analyzing a data signal that has been embedded with at least one digital watermark using a data reduction technique, said method comprising: receiving the data signal; processing the data signal to detect information relative to the digital watermark; analyzing the detected information to determine if the output of the data signal is authorized; and outputting said data signal if the detected information establishes that output is authorized.
Also provided is a device for analyzing a data signal that has been embedded with at least one digital watermark using a data reduction technique, said device comprising: an interface for receiving the data signal; a detector for processing the data signal to detect information relative to the at least one digital watermark; an analyzer to analyze the detected information to determine if output of the data signal is authorized or unauthorized; and an signal generator to output data if the detected information establishes that output is authorized.
There are two design goals in an overall digital watermarking system's low cost, and universality. Ideally, a method for encoding and decoding digital watermarks in digitized media for copy control purposes should be inexpensive and universal. This is essential in preventing casual piracy. On the other hand, a more secure form of protection, such as a “forensic watermarks,” can afford to be computationally intensive to decode, but must be unaffected by repeated re-encoding of a copy control watermark. An ideal method for achieving these results would separate the signal into different areas, each of which can be accessed independently. The embedded signal or may simply be “watermark bits” or “executable binary code,” depending on the application and type of security sought. Improvements to separation have been made possible by enhancing more of the underlying design to meet a number of clearly problematic issues.
The present invention interprets the signal as a stream which may be split into separate streams of digitized samples or may undergo data reduction (including both lossy and lossless compression, such as MPEG lossy compression and Meridian's lossless compression, down sampling, common to many studio operations, or any related data reduction process). The stream of data can be digital in nature, or may also be an analog waveform (such as an image, audio, video, or multimedia content). One example of digital data is executable binary code. When applied to computer code, the present invention allows for more efficient, secure, copyright protection when handling functionality and associations with predetermined keys and key pairs in software applications or the machine readable versions of such code in microchips and hardware devices. Text may also be a candidate for authentication or higher levels of security when coupled with secure key exchange or asymmetric key generation between parties. The subsets of the data stream combine meaningful and meaningless bits of data which may be mapped or transferred depending on the application intended by the implementing party. The present invention utilizes data reduction to allow better performance in watermarking as well as cryptographic methods concerning binary executable code, its machine readable form, text and other functionality-based or communication-related applications. Some differences may simply be in the structure of the key itself, a pseudo random or random number string or one which also includes additional security with special one way functions or signatures saved to the key. The key may also be made into key pairs, as is discussed in other disclosures and patents referenced herein. The present invention contemplates watermarks as a plurality of digitized sample streams, even if the digitized streams originate from the analog waveform itself. The present invention also contemplates that the methods disclosed herein can be applied to non-digitized content. Universally, data reduction adheres to some means of “understanding” the reduction. This disclosure contemplates data reduction which may include down sampling, lossy compression, summarization or any means of data reduction as a novel means to speed up watermarking encode and decode operations. Many forms of data reduction rely upon sampling of a data signal, for instance frequency or time sampling of a digital audio or video signal. For example, a signal may be sampled on a regular basis every x fractions of a second, where x is arbitrarily chosen, such that representative data slices of the signal are obtained. Other data reduction techniques include bit depth reduction. Bit depth reduction relies on the fact that when measuring items, scales of different degrees of precision can be used. For example, one can measure things on a scale with three division marks (zero to two), or on a scale of the same magnitude with ten division marks (zero to nine). Scales with more divisions are of higher precision than scales with fewer divisions. On a computer, because of processing and storage limitations, numerical values (e.g., numerical values relating to a digitized signal) are also represented with varying degrees of precision. For example, one can use two bits (a scale of zero to three) to represent a numerical value or use five bits (a scale of zero to thirty-one) to represent the same numerical value. The number of bits used to represent a numerical value is generally referred to as the “bit depth.” Numerical data may be reduced for storage or transmission by reduction of the bit depth scale.
While any of a number of different data reduction techniques can be used in conjunction with the present invention, essentially a lossy method for data reduction yields the best results for encode and decode operations. Data reduction methods should be appropriately chosen with an eye toward the particular type of data signal being reduced. Some data signals may more readily be reduced than others. For instance, when the data reduction technique chosen is a compression technique, it will be realized that not all data signals or files are equally compressible. For example, there are limits to the degree to which aesthetic information (such as music or video signals) may be compressed without losing their aesthetic or informational value. Thus, in practicing the present invention, techniques can be applied for intelligent selection of data reduction, and differential data reduction techniques can be selected for differential substreams of an aggregate data stream. For example, a computer processor implementing the present invention for protection of a data signal stream comprising, say, both video and text portions, can be programmed to “split” the aggregate data stream into video and text signal substreams, and to apply a first data reduction algorithm most suitable for video data to the first substream, while applying a second data reduction algorithm most suitable for text data to the second substream.
It is desirable to have both copy control and forensic watermarks in the same signal to address the needs of the hardware, computer, and software industries while also providing for appropriate security to the owners of the copyrights. This will become clearer with further explanation of the sample embodiments discussed herein.
The present invention also contemplates the use of data reduction for purposes of speedier and more tiered forms of security, including combinations of these methods with transfer function functions. In many applications, transfer functions (e.g., scrambling), rather than mapping functions (e.g., watermarking), are preferable or can be used in conjunction with mapping. With “scrambling,” predetermined keys are associated with transfer functions instead of mapping functions, although those skilled in the art may recognize that a transfer function is simply a subset of mask sets encompassing mapping functions. It is possible that tiered scrambling with data reduction or combinations of tiered data reduction with watermarking and scrambling may indeed increase overall security to many applications.
The use of data reduction can improve the security of both scrambling and watermarking applications. All data reduction methods include coefficients which affect the reduction process. For example, when a digital signal with a time or space component is down sampled, the coefficient would be the ratio of the new sample rate to the original sample rate. Any coefficients that are used in the data reduction can be randomized using the key, or key pair, making the system more resistant to analysis. Association to a predetermined key or key pair and additional measure of security may include biometric devices, tamper proofing of any device utilizing the invention, or other security measures.
Tests have shown that the use of data reduction in connection with digital watermarking schemes significantly reduces the time required to decode the watermarks, permitting increases in operational efficiency.
Particular implementations of the present invention, which have yielded extremely fast and inexpensive digital watermarking systems, will now be described. These systems may be easily adapted to consumer electronic devices, general purpose computers, software and hardware. The exchange of predetermined keys or key pairs may facilitate a given level of security. Additionally, the complementary increase in security for those implementations where transfer functions are used to “scramble” data, is also disclosed.
For a more complete understanding of the invention and some advantages thereof, reference is now made to the following descriptions taken in connection with the accompanying drawings in which:
The embodiments of the present invention and its advantages are best understood by referring to the drawings, like numerals being used for like and corresponding parts of the various drawings.
An Overview
A system for achieving multiple levels of data reduction is illustrated in
First data reduced signal 20 is subjected to a second data reduction technique 101 to generate a second data reduced signal 21. Second data reduced signal 21 is then subtracted from first data reduced signal 20 to generate a second remainder signal 31.
Each of the successive data reduced signals is, in turn, interactively subjected to data reduction techniques to generate a further data reduced signal, which, in turn, is subtracted from its respective parent signal to generate another remainder signal. This process is generically described as follows. An (n−1) data reduced signal 28 (i.e, a signal that has been data reduced n−1 times) is subjected to an nth data reduction technique 109 to generate an nth data reduced signal 29. The nth data reduced signal 29 is then subtracted from the (n−1) data reduced signal 28 to produce an nth remainder signal 39.
An output signal can be generated from the system illustrated in
Each level may be used to represent a particular data density. E.g., if the reduction method is down-sampling, for a DVD audio signal the first row would represent data sampled at 96 kHz, the second at 44.1 kHz., the third at 6 kHz., etc. There is only an issue of deciding what performance or security needs are contemplated when undertaking the data reduction process and choice of which types of keys or key pairs should be associated with the signal or data to be reduced. Further security can be increased by including block ciphers, special one way functions, one time stamps or even biometric devices in the software or hardware devices that can be embodied. Passwords or biometric data are able to assist in the determination of the identity of the user or owner of the data, or some relevant identifying information.
A variety of keys may advantageously be chosen. Additionally, any key or keys employed need not remain static over time but may be changed from time to time. For instance, the key may be changed in real time, or upon detection of a “marker” signal within the data signal stream. The key can also be a ciphered key. As is known in the art, the key or keys may be generated by any of a variety of effective methods, including steganographic cipher, symmetric cryptographic cipher, and asymmetric cryptographic cipher. Keys may be derived (in whole or in part) from the signal stream itself or may be derived from sources completely external to the signal stream.
Additionally, and given that information signals may comprise a variety of forms of information (e.g., audio, still image, video, computer code, or text), it is appreciated that a single multimedia information signal stream may be divided into multiple substreams based on the various constituent information forms in the multimedia information stream. It could be advantageous, in such a substreamed context, to associate predetermined discrete, and particular, forms or instances of key or key pair to particular information substreams—for instance, a predetermined first key or key pair could be assigned for association and use with a video substream whereas a predetermined second key or key pair could be assigned for association and use with a text substream. Thus, complex watermarking of a multi-substream data signal may be flexibly accomplished. Such complexity may contribute, inter alia, to more effective watermarking and security as multiple watermarks would have to be compromised in order to compromise the entire aggregate information stream or set of substreams. Keys and key pairs are understood to be multifunctional, insofar as they are useful for both the encoding and decoding of watermarks.
An example of a real world application is helpful here. Given the predominant concern, at present, of MPEG 1 Layer 3, or MP3, a perceptual lossy compression audio data format, which has contributed to a dramatic re-evaluation of the distribution of music, a digital watermark system must be able to handle casual and more dedicated piracy in a consistent manner. The present invention contemplates compatibility with MP3, as well as any perceptual coding technique that is technically similar. One issue, is to enable a universal copy control “key” detect a watermark as quickly as possible from a huge range of perceptual quality measures. For instance, DVD 24 bit 96 kHz, encoded watermarks, should be detected in at least “real time,” even after the signal has been down sampled, to say 12 kHz of the 96 kHz originally referenced. By delineating and starting with less data, since the data-reduced signal is obviously smaller though still related perceptually to the original DVD signal, dramatic increases in the speed and survival of the universal copy control bits can be achieved. The present invention also permits the ability to separate any other bits which may be associated with other more secure predetermined keys or key pairs.
Where the data stream is executable computer code, the present invention contemplates breaking the code into objects or similar units of functionality and allowing for determination of what is functionally important. This may be more apparent to the developer or users of the software or related hardware device. Data reduction through the use of a subset of the functional objects related to the overall functionality of the software or executable code in hardware or microchips, increase the copyright protection or security sought, based on reducing the overall data to be associated with predetermined keys or key pairs. Similarly, instead of mapping functions, transfer functions, so-called “scrambling,” appear better candidates for this type of security although both mapping and transferring may be used in the same system. By layering the security, the associated keys and key pairs can be used to substantially improve the security and to offer easier methods for changing which functional “pieces” of executable computer code are associated with which predetermined keys. These keys may take the form of time-sensitive session keys, as with transactions or identification cards, or more sophisticated asymmetric public key pairs which may be changed periodically to ensure the security of the parties' private keys. These keys may also be associated with passwords or biometric applications to further increase the overall security of any potential implementation.
An example for text message exchange is less sophisticated but, if it is a time sensitive event, e.g., a secure communication between two persons, benefits may also be encountered here. Security may also be sought in military communications. The ability to associate the securely exchanged keys or key pairs while performing data reduction to enhance the detection or decoding performance, while not compromising the level of security, is important. Though a steganographic approach to security, the present invention more particularly addresses the ability to have data reduction to increase speed, security, and performance of a given steganographic system. Additionally, data reduction affords a more layered approach when associating individual keys or key pairs with individual watermark bits, or digital signature bits, which may not be possible without reduction because of considerations of time or the payload of what can be carried by the overall data “covertext” being transmitted.
Layering through data reduction offers many advantages to those who seek privacy and copyright protection. Serialization of the detection chips or software would allow for more secure and less “universal” keys, but the interests of the copyright owners are not always aligned with those of hardware or software providers. Similarly, privacy concerns limit the amount of watermarking that can be achieved for any given application. The addition of a pre-determined and cryptographic key-based “forensic” watermark, in software or hardware, allows for 3rd party authentication and provides protection against more sophisticated attacks on the copy control bits. Creating a “key pair” from the “predetermined” key is also possible.
Separation of the watermarks also relates to separate design goals. A copy control mechanism should ideally be inexpensive and easily implemented, for example, a form of “streamed watermark detection.” Separating the watermark also may assist more consistent application in broadcast monitoring efforts which are time-sensitive and ideally optimized for quick detection of watermarks. In some methods, the structure of the key itself, in addition to the design of the “copy control” watermark, will allow for few false positive results when seeking to monitor radio, television, or other streamed broadcasts (including, for example, Internet) of copyrighted material. As well, inadvertent tampering with the embedded signal proposed by others in the field can be avoided more satisfactorily. Simply, a universal copy control watermark may be universal in consumer electronic and general computing software and hardware implementations, but less universal when the key structure is changed to assist in being able to log streaming, performance, or downloads, of copyrighted content. The embedded bits may actually be paired with keys in a decode device to assure accurate broadcast monitoring and tamper proofing, while not requiring a watermark to exceed the payload available in an inaudible embedding process. E.g., A full identification of the song, versus time-based digital signature bits, embedded into a broadcast signal, may not be recovered or may be easily over encoded without the use of block ciphers, special one way functions or one time pads, during the encoding process, prior to broadcast. Data reduction as herein disclosed makes this operation more efficient at higher speeds.
A forensic watermark is not time sensitive, is file-based, and does not require the same speed demands as a streamed or broadcast-based detection mechanism for copy control use. Indeed, a forensic watermark detection process may require additional tools to aid in ensuring that the signal to be analyzed is in appropriate scale or size, ensuring signal characteristics and heuristic methods help in appropriate recovery of the digital watermark. Simply, all aspects of the underlying content signal should be considered in the embedding process because the watermarking process must take into account all such aspects, including for example, any dimensional or size of the underlying content signal. The dimensions of the content signal may be saved with the key or key pair, without enabling reproduction of the unwatermarked signal. Heuristic methods may be used to ensure the signal is in proper dimensions for a thorough and accurate detection authentication and retrieval of the embedded watermark bits. Data reduction can assist in increasing operations of this nature as well, since the data reduction process may include information about the original signal, for example, signal characteristics, signal abstracts, differences between samples, signal patterns, and related work in restoring any given analog waveform.
The present invention provides benefits, not only because of the key-based approach to the watermarking, but the vast increase in performance and security afforded the implementations of the present invention over the performance of other systems.
The architecture of key and key-pair based watermarking is superior to statistical approaches for watermark detection because the first method meets an evidentiary level of quality and are mathematically provable. By incorporating a level of data reduction, key and key paired based watermarking is further improved. Such levels of security are plainly necessary if digital watermarks are expected to establish responsibility for copies of copyrighted works in evidentiary proceedings. More sophisticated measures of trust are necessary for use in areas which exceed the scope of copyright but are more factually based in legal proceedings. These areas may include text authentication or software protection (extending into the realm of securing microchip designs and compiled hardware as well) in the examples provided above and are not contemplated by any disclosure or work in the art.
The present invention may be implemented with a variety of cryptographic protocols to increase both confidence and security in the underlying system. A predetermined key is described as a set of masks: a plurality of mask sets. These masks may include primary, convolution and message delimiters but may extend into additional domains. In previous disclosures, the functionality of these masks is defined solely for mapping. Public and private keys may be used as key pairs to further increase the unlikeliness that a key may be compromised. Examples of public key cryptosystems may be found in the following U.S. Pat. Nos. 4,200,770; 4,218,582; 4,405,829; and 4,424,414, which examples are incorporated herein by reference. Prior to encoding, the masks described above are generated by a cryptographically secure random generation process. Mask sets may be limited only by the number of dimensions and amount of error correction or concealment sought, as has been previously disclosed.
A block cipher, such as DES, in combination with a sufficiently random seed value emulates a cryptographically secure random bit generator. These keys, or key pairs, will be saved along with information matching them to the sample stream in question in a database for use in subsequent detection or decode operation. These same cryptographic protocols may be combined with the embodiments of the present invention in administering streamed content that requires authorized keys to correctly display or play said streamed content in an unscrambled manner. As with digital watermarking, symmetric or asymmetric public key pairs may be used in a variety of implementations. Additionally, the need for certification authorities to maintain authentic key-pairs becomes a consideration for greater security beyond symmetric key implementations, where transmission security is a concern.
Signal Processing in a Multi-Watermark System (a Plurality of Streams May be Watermarked)
Watermarking process step 300 may be chosen from among various watermarking processes known in the art. As an example, a digital audio data signal may be represented, for purpose of watermarking, by a series of samples in 1 dimension. {S1, S2, S3 . . . Sn}. This series is also referred to as a sample stream. The sample stream approximates an analog waveform of sound amplitude over time. Each sample represents an estimate of the wave amplitude at the instant of time the sample is recorded. For monaural audio, there is one such sample stream. Stereo audio is comprised of two sample streams, one representing the right channel, and the other representing the left. Each stream is used to drive a corresponding speaker to reproduce the stereo sound. What is referred to as CD quality audio is characterized by 16 bit (2 byte) stereo samples, recorded at 44.1 Khz, or 44,100 samples per second in each channel. The dynamic range of sound reproduction is directly proportional to the number of bits per sample. Some lower quality recordings are done at 8 bits. A CD audio recording can be stored using any scheme for containing the 2 sample streams in their entirety. When these streams are played back at the same frequency they were recorded at, the sound recorded is reproduced to a high degree of accuracy. The sample stream is processed in order from first sample to last. For the purpose of the invention disclosed, the stream is separated into sample windows, each of which has a fixed number of consecutive samples from the stream, and where windows do not overlap in the sample stream. Windows may be contiguous in the sample stream. For illustration, assume each window contains 128 samples, and that windows are contiguous. Thus, the windows within the stream look like
{>S1, S2, S3 . . . S128!, >S129, S130, S131 . . . S256! . . . >Sn-128 . . . Sn!}
wherein the bracketed set > . . . ! denotes each window and any odd samples at the end of the stream which do not completely fill a window can be ignored, and simply passed through the system unmodified.
These windows will be used as input for the discrete Fast Fourier Transform (and its inverse) operation. Briefly, Fourier Transform methods are based on the principle that a complex waveform, expressed as amplitude over time and represented by a sample stream, is really the sum of a number of simple waveforms, each of which oscillates at different frequencies. By complex, it is meant that the value of the next sample is not easily predicted from the values of the last N samples or the time of the sample. By simple it is meant that the value of the sample is easily predictable from the values of the last N samples and/or the time of the sample.
The sum of multiple simple waves is equivalent to the complex wave. The discrete FFT and its inverse simply translate a limited amount of data from one side of this equivalence to the other, between the complex waveform and the sum of simple waves. The discrete FFT can be used to translate a series of samples representing amplitude over time (the complex wave, representing a digital audio recording) into the same number of samples representing total spectral energy in a given range of frequencies (the simple wave components) at a particular instant of time. This instant is the time in the middle of the original amplitude/time samples. The inverse discrete FFT translates the data in the other direction, producing the complex waveform, from its simpler parts.
Each 128 sample window will be used as an input to the discrete FFT, resulting in 128 bins representing each of 128 frequency bands, ranging from 0 Hz to 22 Khz (the Nyquist frequency, or ½ the sampling rate).
Information can be encoded into the audio signal in the frequency domain or in the time domain. In the latter case, no FFT or inverse FFT is necessary. However, encoding in the frequency domain is recommended, since its effects are scattered over the resultant time domain samples, and not easily predicted. In addition, frequency domain encoding makes it more likely that randomization will result in noticeable artifacts in the resultant signal, and therefore makes the stega-cipher more defensible against such attacks. It is in the frequency domain that additional information will be encoded into the audio signal for the purpose of this discussion. Each frequency band in a given time slice can potentially be used to store a small portion of some additional information to be added to the signal. Since these are discrete estimates, there is some room for error which will not significantly effect the perceived quality of the signal, reproduced after modification, by the inverse FFT operation. In effect, intentional changes, which cannot be distinguished from random variations, are introduced in the frequency domain, for the purpose of storing additional information in the sample stream. These changes are minimized so as not to adversely affect the perceived quality of the reproduced audio signal, after it has been encoded with additional information in the manner described below. In addition, the location of each of these changes is made virtually impossible to predict, an innovation which distinguishes this scheme from simple steganographic techniques.
The saved, unwatermarked data-reduced signal (signal 40) is subtracted from the original input signal 11, yielding a remainder signal 60 composed only of the data that was lost during the data-reduction. A second watermark is then applied using a desired watermarking protocol (process step 301) to remainder signal 60 to generate a watermarked remainder signal 70. Finally, the watermarked remainder 70 and the watermarked, data-reduced signal 50 are added to form an output signal 80, which is the final, full-bandwidth, output signal.
The two watermarking techniques (process steps 300 and 301) may be identical (i.e., be functionally the same), or they may be different.
To decode the signal, a specific watermark is targeted. Duplicating the data-reduction processes that created the watermark in some cases can be used to recover the signal that was watermarked. Depending upon the data-reduction method, it may or may not be necessary to duplicate the data-reduction process in order to read a watermark embedded in a remainder signal. Because of the data-reduction, the decoding search can occur much faster than it would in a full-bandwidth signal. Detection speed of the remainder watermark remains the same as if there were no other watermark present.
Additionally, the watermarking described in connection with this embodiment above may be done with a plurality of predetermined keys or key pairs associated with a single watermark “message bit,” code object, or text. Keys or key pairs may also be stored or archived in a central certification authority, such that there will be a verified and official version of a particular key or key pair whenever access to such key or key pair, or verification or identification of the legitimacy and authorization of the use of a particular data signal or file associated with that key, is required. The central certification authority could be, for instance, a secure computer server archive maintained by a copyright holder to store keys relating to copyrighted files watermarked using such keys.
Signal Processing in a Single Watermark System
Additionally, the watermarking described in connection with this embodiment above may be done with a plurality of predetermined keys or key pairs associated with a single watermark “message bit,” code object, or text.
In either process, an external key can be used to control the insertion location of either watermark. In a copy-control system, a key is not generally used, whereas in a forensic system, a key must be used. The key can also control the parameters of the data-reduction scheme. The dual scheme can allow a combination of copy-control and forensic watermarks in the same signal. A significant feature is that the copy-control watermark can be read and rewritten without affecting the forensic mark or compromising its security.
Signal Processing in a Multi-Scrambler System (a Plurality of Streams May be Scrambled)
The saved, unscrambled data-reduced signal (signal 45) is subtracted from the original input signal 12, yielding a remainder signal 65 composed only of the data that was lost during the data-reduction. A second scrambling technique is then applied (process step 501) to remainder signal 65 to generate a scrambled remainder signal 75. Finally, the scrambled remainder signal 75 and the scrambled data-reduced signal 55 are added to form an output signal 85, which is the final, full-bandwidth, output signal.
The two scrambling techniques (process steps 500 and 501) may be identical (i.e., be functionally the same), or they may be different.
Additionally the scrambling described in connection with this embodiment may be done with a plurality of predetermined keys or key pairs associated with a single scrambling operation containing only a “message bit,” code object, or text.
To decode the signal, unscrambling follows the exact pattern of the scrambling process except that the inverse of the scrambling transfer function is applied to each portion of the data, thus returning it to its pre-scrambled state.
Signal Processing in a Single Scrambling Operation
Additionally the scrambling described in connection with this embodiment may be done with a plurality of predetermined keys or key pairs associated with a single scrambling operation containing only a “message bit,” code object, or text.
Another embodiment may combine both watermarking and scrambling with data reduction. Speed, performance and computing power may influence the selection of which techniques are to be used. Decisions between data reduction schemes ultimately must be measured against the types of keys or key pairs to use, the way any pseudo random or random number generation is done (chaotic, quantum or other means), and the amount of scrambling or watermarking that is necessary given the needs of the system. It is quite possible that some derived systems would yield a fairly large decision tree, but the present invention offers many benefits to applications in security that are not disclosed in the art.
As a further illustrative example of an advantageous embodiment, the following briefly describes an implementation of the present invention using sample rate reduction as the chosen data reduction method for watermarking in connection with an audio data signal.
I. Encoding:
In connection with the above-described embodiment, alternative step IIB is illustrated because decoding the open watermark may have to occur on consumer electronic devices, and therefore, generally, fewer processing steps may be desirable in consumer electronic devices. The secure watermark is not as time-critical during the decode process, and can therefore be afforded more processing time. Note further that the original sample rate during the encode does not have to be the same as the original sample rate for decode. Any intervening sample rate conversion will be ignored, as long as it never drops below the same rate of the signal to which the watermark is applied (for example, 10 kHz for the secured watermark of the prior example, or 5 kHz for the open watermark of the prior example).
The embodiments described herein may advantageously be implemented in connection with a data signal recipient's personal computer system or workstation (comprising a computer processor such as an Intel Pentium processor, spreadsheet software such as Microsoft Excel, and implementing a communications module such as a common web browser such as Internet Explorer or Netscape), linked by a World Wide Web connection to a data signal or file provider utilizing similar standard computer hardware and software, but may also be implemented in connection with any output device having appropriate electronic memory and/or processing capacity to implement the techniques set forth herein (which could include, for instance, consumer electronics output devices other than microcomputers). Because the digital watermarking techniques and systems disclosed herein are substantially universal, however, they may be applied across a variety of computer hardware and software configurations, for use with a variety of transmitted data signals or files, over a variety of public or private networks (although the utility of the present invention for digital watermarking of audio or video files transmitted over public networks such as the interne is obvious). The network communication link between the data signal/file recipient and the signal/file provider may further be provided with some network-default level of encryption (perhaps a relatively weak level such as 56 bit encryption). Similarly, known computer programming techniques and languages (for instance, Visual Basic, JAVA, C++) may be adapted in a variety of fashions for use in either the data reduction steps discussed herein, the cryptographic/scrambling processes disclosed, the specific watermarking techniques applied, or any combination of the above, for customized data reduction and digital watermarking, and output of an output signal, in the fashion most amenable to a particular user's needs. The ability to adapt a wide range of data processing algorithms (including but not limited to algorithms for data reduction, encryption/decryption, and compression) to yield various desired data signal outputs, to apply customizable digital watermarking procedures, and to allow customizable and maximally-efficient forensic or copy control watermarks to popular and useful data transmission protocols, all across a broad range of computer system platforms (i.e., various hardware, software, computer language, and operating system combinations) provides the present invention with considerable versatility.
The present invention as implemented with such computer systems permits secured delivery of valuable data streams over a variety of networks. Specifically, the present invention provides great utility for the delivery (commercial or otherwise) of video, audio, or other such files on media or over a public network such as the internet in a fashion that impedes theft or unauthorized use or resale of such files. For instance, the methods of the present invention could be applied to all the digitized commercial music files of a music vendor (to impose, for instance, a copy control watermark thereupon). Subsequently, those watermarked music files may be delivered to end users. End user attempts to make unauthorized copies can thus be controlled. Alternatively, output devices may be programmed to detect watermarks embedded in files by use of the present invention, such that if the file does not contain an appropriate watermark, the output device will not execute or “play” the file.
It is important to note that the watermarks embedded using the present invention may be embedded at a wide variety of points along the distribution chain for the data signals. For instance, in an embodiment in which the present invention is used to watermark commercial music or video files downloaded by an individual end user from a central server over an internet connection through an internet service provider, the present invention could be used to impose a forensic watermark (uniquely identifying the customer and download transaction) at the central server (or at the server of the internet service provider). When a suspected unauthorized copy of the file in question was detected, the watermark therein could be sensed/decoded in order to identify the source of the unauthorized copy. As has been emphasized, the techniques of the present invention may be applied to a wide variety of data signals, whether stored multimedia or computer code files, streamed files transmitted in real time, or other files or data signals, and may be applied in context-sensitive fashion to maximize protection (and effective signal transmission and output) for a particular data stream. It is also an aspect of this invention that the novel techniques for watermarking using data reduction herein can be exploited at the end user point of the distribution chain for data signals; that is, using the unique watermark/key information associated with a file watermarked using the techniques described hereinabove, a file may be analyzed (whether by representatives of a file copyright owner, for instance, or by hardware, software, or other appropriate analyzer, such as an embedded firmware chip, etc. contained in or supplied to an end user output device). Once the data signal is analyzed at the end user point, information relative to the any watermark or key actually contained on the file at that point may be derived and analyzed to determine if the file has been properly distributed to the end user. If it has not, the output device may be programmed to deny output or to manipulate the data signal in a destructive way (or to take other appropriate legal or copyright control action as may be desired by the file owner). The present invention includes such uses of (and devices for) data reduction-derived watermark detection and output control.
Those of ordinary skill in the art will appreciate that the foregoing discussion of certain preferred embodiments is illustrative only, and does not limit the spirit and scope of the present invention, which are limited only by the claims set forth below.
This application is a continuation of U.S. patent application Ser. No. 11/519,467, filed Sep. 12, 2006 now U.S. Pat. No. 7,664,264 which is a divisional of U.S. patent application Ser. No. 09/594,719 (issued as U.S. Pat. No. 7,123,718), filed Jun. 16, 2000, which is a continuation-in-part of International Application No. PCT/US00/06522, filed Mar. 14, 2000, which PCT application claimed priority to U.S. Provisional Application No. 60/125,990, filed Mar. 24, 1999. The previously identified patents and/or patent applications are hereby incorporated by reference, in their entireties. This application also is related to the following applications: U.S. patent application Ser. No. 09/046,627 (issued as U.S. Pat. No. 6,598,162), filed Mar. 24, 1998, entitled “Method for Combining Transfer Function with Predetermined Key Creation”; U.S. patent application Ser. No. 09/053,628 (issued as U.S. Pat. No. 6,205,249), filed Apr. 2, 1998, entitled “Multiple Transform Utilization and Application for Secure Digital Watermarking”; U.S. Provisional Patent Application Ser. No. 60/169,274 (which corresponds to U.S. patent application Ser. No. 09/731,040 which issued as U.S. Pat. No. 7,159,116), filed Dec. 7, 1999, entitled “Systems, Methods and Devices for Trusted Transactions”; and U.S. Patent Application Ser. No. 60/147,134 (which corresponds to U.S. patent application Ser. No. 10/049,101 which issued as U.S. Pat. No. 7,475,246), filed Aug. 4, 1999, entitled, “A Secure Personal Content Server.” All of the patent applications previously identified in this paragraph are hereby incorporated by reference, in their entireties.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3947825 | Cassada | Mar 1976 | A |
| 3984624 | Waggener | Oct 1976 | A |
| 3986624 | Cates, Jr. et al. | Oct 1976 | A |
| 4038596 | Lee | Jul 1977 | A |
| 4200770 | Hellman et al. | Apr 1980 | A |
| 4218582 | Hellman et al. | Aug 1980 | A |
| 4339134 | Macheel | Jul 1982 | A |
| 4390898 | Bond et al. | Jun 1983 | A |
| 4405829 | Rivest et al. | Sep 1983 | A |
| 4424414 | Hellman et al. | Jan 1984 | A |
| 4528588 | Lofberg | Jul 1985 | A |
| 4672605 | Hustig et al. | Jun 1987 | A |
| 4748668 | Shamir et al. | May 1988 | A |
| 4789928 | Fujisaki | Dec 1988 | A |
| 4827508 | Shear | May 1989 | A |
| 4876617 | Best et al. | Oct 1989 | A |
| 4896275 | Jackson | Jan 1990 | A |
| 4908873 | Philibert et al. | Mar 1990 | A |
| 4939515 | Adelson | Jul 1990 | A |
| 4969204 | Jones et al. | Nov 1990 | A |
| 4972471 | Gross et al. | Nov 1990 | A |
| 4977594 | Shear | Dec 1990 | A |
| 4979210 | Nagata et al. | Dec 1990 | A |
| 4980782 | Ginkel | Dec 1990 | A |
| 5050213 | Shear | Sep 1991 | A |
| 5073925 | Nagata et al. | Dec 1991 | A |
| 5077665 | Silverman et al. | Dec 1991 | A |
| 5113437 | Best et al. | May 1992 | A |
| 5136581 | Muehrcke | Aug 1992 | A |
| 5136646 | Haber et al. | Aug 1992 | A |
| 5136647 | Haber et al. | Aug 1992 | A |
| 5142576 | Nadan | Aug 1992 | A |
| 5161210 | Druyvesteyn et al. | Nov 1992 | A |
| 5210820 | Kenyon | May 1993 | A |
| 5243423 | DeJean et al. | Sep 1993 | A |
| 5243515 | Lee | Sep 1993 | A |
| 5287407 | Holmes | Feb 1994 | A |
| 5319735 | Preuss et al. | Jun 1994 | A |
| 5341429 | Stringer et al. | Aug 1994 | A |
| 5341477 | Pitkin et al. | Aug 1994 | A |
| 5363448 | Koopman et al. | Nov 1994 | A |
| 5365586 | Indeck et al. | Nov 1994 | A |
| 5369707 | Follendore, III | Nov 1994 | A |
| 5379345 | Greenberg | Jan 1995 | A |
| 5394324 | Clearwater | Feb 1995 | A |
| 5398285 | Borgelt et al. | Mar 1995 | A |
| 5406627 | Thompson et al. | Apr 1995 | A |
| 5408505 | Indeck et al. | Apr 1995 | A |
| 5410598 | Shear | Apr 1995 | A |
| 5412718 | Narasimhalv et al. | May 1995 | A |
| 5418713 | Allen | May 1995 | A |
| 5428606 | Moskowitz | Jun 1995 | A |
| 5450490 | Jensen et al. | Sep 1995 | A |
| 5469536 | Blank | Nov 1995 | A |
| 5471533 | Wang et al. | Nov 1995 | A |
| 5478990 | Montanari et al. | Dec 1995 | A |
| 5479210 | Cawley et al. | Dec 1995 | A |
| 5487168 | Geiner et al. | Jan 1996 | A |
| 5493677 | Balogh et al. | Feb 1996 | A |
| 5497419 | Hill | Mar 1996 | A |
| 5506795 | Yamakawa | Apr 1996 | A |
| 5513126 | Harkins et al. | Apr 1996 | A |
| 5513261 | Maher | Apr 1996 | A |
| 5530739 | Okada | Jun 1996 | A |
| 5530751 | Morris | Jun 1996 | A |
| 5530759 | Braudaway et al. | Jun 1996 | A |
| 5539735 | Moskowitz | Jul 1996 | A |
| 5548579 | Lebrun et al. | Aug 1996 | A |
| 5568570 | Rabbani | Oct 1996 | A |
| 5579124 | Aijala et al. | Nov 1996 | A |
| 5581703 | Baugher et al. | Dec 1996 | A |
| 5583488 | Sala et al. | Dec 1996 | A |
| 5598470 | Cooper et al. | Jan 1997 | A |
| 5606609 | Houser et al. | Feb 1997 | A |
| 5613004 | Cooperman et al. | Mar 1997 | A |
| 5617119 | Briggs et al. | Apr 1997 | A |
| 5625690 | Michel et al. | Apr 1997 | A |
| 5629980 | Stefik et al. | May 1997 | A |
| 5633932 | Davis et al. | May 1997 | A |
| 5634040 | Her et al. | May 1997 | A |
| 5636276 | Brugger | Jun 1997 | A |
| 5636292 | Rhoads | Jun 1997 | A |
| 5640569 | Miller et al. | Jun 1997 | A |
| 5646997 | Barton | Jul 1997 | A |
| 5657461 | Harkins et al. | Aug 1997 | A |
| 5659726 | Sandford, II et al. | Aug 1997 | A |
| 5664018 | Leighton | Sep 1997 | A |
| 5673316 | Auerbach et al. | Sep 1997 | A |
| 5677952 | Blakely et al. | Oct 1997 | A |
| 5680462 | Miller et al. | Oct 1997 | A |
| 5687236 | Moskowitz et al. | Nov 1997 | A |
| 5689587 | Bender et al. | Nov 1997 | A |
| 5696828 | Koopman, Jr. | Dec 1997 | A |
| 5719937 | Warren et al. | Feb 1998 | A |
| 5721788 | Powell et al. | Feb 1998 | A |
| 5734752 | Knox | Mar 1998 | A |
| 5737416 | Cooper et al. | Apr 1998 | A |
| 5737733 | Eller | Apr 1998 | A |
| 5740244 | Indeck et al. | Apr 1998 | A |
| 5745569 | Moskowitz et al. | Apr 1998 | A |
| 5748783 | Rhoads | May 1998 | A |
| 5751811 | Magnotti et al. | May 1998 | A |
| 5754697 | Fu et al. | May 1998 | A |
| 5757923 | Koopman, Jr. | May 1998 | A |
| 5765152 | Erickson | Jun 1998 | A |
| 5768396 | Sone | Jun 1998 | A |
| 5774452 | Wolosewicz | Jun 1998 | A |
| 5790677 | Fox et al. | Aug 1998 | A |
| 5799083 | Brothers et al. | Aug 1998 | A |
| 5809139 | Girod et al. | Sep 1998 | A |
| 5809160 | Powell et al. | Sep 1998 | A |
| 5818818 | Soumiya | Oct 1998 | A |
| 5822432 | Moskowitz et al. | Oct 1998 | A |
| 5828325 | Wolosewicz et al. | Oct 1998 | A |
| 5832119 | Rhoads | Nov 1998 | A |
| 5842213 | Odom | Nov 1998 | A |
| 5848155 | Cox | Dec 1998 | A |
| 5850481 | Rhoads | Dec 1998 | A |
| 5859920 | Daly et al. | Jan 1999 | A |
| 5860099 | Milios et al. | Jan 1999 | A |
| 5862260 | Rhoads | Jan 1999 | A |
| 5870474 | Wasilewski et al. | Feb 1999 | A |
| 5884033 | Duvall et al. | Mar 1999 | A |
| 5889868 | Moskowitz et al. | Mar 1999 | A |
| 5893067 | Bender et al. | Apr 1999 | A |
| 5894521 | Conley | Apr 1999 | A |
| 5903721 | Sixtus | May 1999 | A |
| 5905800 | Moskowitz et al. | May 1999 | A |
| 5905975 | Ausubel | May 1999 | A |
| 5912972 | Barton | Jun 1999 | A |
| 5915027 | Cox et al. | Jun 1999 | A |
| 5917915 | Hirose | Jun 1999 | A |
| 5918223 | Blum | Jun 1999 | A |
| 5920900 | Poole et al. | Jul 1999 | A |
| 5923763 | Walker et al. | Jul 1999 | A |
| 5930369 | Cox et al. | Jul 1999 | A |
| 5930377 | Powell et al. | Jul 1999 | A |
| 5940134 | Wirtz | Aug 1999 | A |
| 5943422 | Van Wie et al. | Aug 1999 | A |
| 5949055 | Fleet | Sep 1999 | A |
| 5963909 | Warren et al. | Oct 1999 | A |
| 5973731 | Schwab | Oct 1999 | A |
| 5974141 | Saito | Oct 1999 | A |
| 5991426 | Cox et al. | Nov 1999 | A |
| 5999217 | Berners-Lee | Dec 1999 | A |
| 6009176 | Gennaro et al. | Dec 1999 | A |
| 6029126 | Malvar | Feb 2000 | A |
| 6041316 | Allen | Mar 2000 | A |
| 6044471 | Colvin | Mar 2000 | A |
| 6049838 | Miller et al. | Apr 2000 | A |
| 6051029 | Paterson et al. | Apr 2000 | A |
| 6061793 | Tewfik et al. | May 2000 | A |
| 6067622 | Moore | May 2000 | A |
| 6069914 | Cox | May 2000 | A |
| 6078664 | Moskowitz et al. | Jun 2000 | A |
| 6081251 | Sakai et al. | Jun 2000 | A |
| 6081587 | Reyes et al. | Jun 2000 | A |
| 6081597 | Hoffstein | Jun 2000 | A |
| 6088455 | Logan et al. | Jul 2000 | A |
| 6131162 | Yoshiura et al. | Oct 2000 | A |
| 6141753 | Zhao et al. | Oct 2000 | A |
| 6141754 | Choy | Oct 2000 | A |
| 6148333 | Guedalia | Nov 2000 | A |
| 6154571 | Cox et al. | Nov 2000 | A |
| 6192138 | Yamadaji | Feb 2001 | B1 |
| 6199058 | Wong et al. | Mar 2001 | B1 |
| 6205249 | Moskowitz | Mar 2001 | B1 |
| 6208745 | Florenio et al. | Mar 2001 | B1 |
| 6226618 | Downs | May 2001 | B1 |
| 6230268 | Miwa et al. | May 2001 | B1 |
| 6233347 | Chen et al. | May 2001 | B1 |
| 6233684 | Stefik et al. | May 2001 | B1 |
| 6240121 | Senoh | May 2001 | B1 |
| 6263313 | Milstad et al. | Jul 2001 | B1 |
| 6272634 | Tewfik et al. | Aug 2001 | B1 |
| 6275988 | Nagashima et al. | Aug 2001 | B1 |
| 6278780 | Shimada | Aug 2001 | B1 |
| 6278791 | Honsinger et al. | Aug 2001 | B1 |
| 6282300 | Bloom et al. | Aug 2001 | B1 |
| 6282650 | Davis | Aug 2001 | B1 |
| 6285775 | Wu et al. | Sep 2001 | B1 |
| 6301663 | Kato et al. | Oct 2001 | B1 |
| 6310962 | Chung et al. | Oct 2001 | B1 |
| 6330335 | Rhoads | Dec 2001 | B1 |
| 6330672 | Shur | Dec 2001 | B1 |
| 6345100 | Levine | Feb 2002 | B1 |
| 6351765 | Pietropaolo et al. | Feb 2002 | B1 |
| 6363483 | Keshav | Mar 2002 | B1 |
| 6373892 | Ichien et al. | Apr 2002 | B1 |
| 6373960 | Conover et al. | Apr 2002 | B1 |
| 6374036 | Ryan et al. | Apr 2002 | B1 |
| 6377625 | Kim | Apr 2002 | B1 |
| 6381618 | Jones et al. | Apr 2002 | B1 |
| 6381747 | Wonfor et al. | Apr 2002 | B1 |
| 6385324 | Koppen | May 2002 | B1 |
| 6385329 | Sharma et al. | May 2002 | B1 |
| 6385596 | Wiser | May 2002 | B1 |
| 6389538 | Gruse et al. | May 2002 | B1 |
| 6405203 | Collart | Jun 2002 | B1 |
| 6415041 | Oami et al. | Jul 2002 | B1 |
| 6418421 | Hurtado | Jul 2002 | B1 |
| 6425081 | Iwamura | Jul 2002 | B1 |
| 6430301 | Petrovic | Aug 2002 | B1 |
| 6430302 | Rhoads | Aug 2002 | B2 |
| 6442283 | Tewfik et al. | Aug 2002 | B1 |
| 6446211 | Colvin | Sep 2002 | B1 |
| 6453252 | Laroche | Sep 2002 | B1 |
| 6457058 | Ullum et al. | Sep 2002 | B1 |
| 6463468 | Buch et al. | Oct 2002 | B1 |
| 6484264 | Colvin | Nov 2002 | B1 |
| 6493457 | Quackenbush | Dec 2002 | B1 |
| 6502195 | Colvin | Dec 2002 | B1 |
| 6522767 | Moskowitz et al. | Feb 2003 | B1 |
| 6522769 | Rhoads et al. | Feb 2003 | B1 |
| 6523113 | Wehrenberg | Feb 2003 | B1 |
| 6530021 | Epstein et al. | Mar 2003 | B1 |
| 6532284 | Walker et al. | Mar 2003 | B2 |
| 6539475 | Cox et al. | Mar 2003 | B1 |
| 6557103 | Boncelet, Jr. et al. | Apr 2003 | B1 |
| 6584125 | Katto | Jun 2003 | B1 |
| 6587837 | Spagna et al. | Jul 2003 | B1 |
| 6590996 | Reed | Jul 2003 | B1 |
| 6598162 | Moskowitz | Jul 2003 | B1 |
| 6606393 | Xie et al. | Aug 2003 | B1 |
| 6611599 | Natarajan | Aug 2003 | B2 |
| 6647424 | Pearson et al. | Nov 2003 | B1 |
| 6658010 | Enns et al. | Dec 2003 | B1 |
| 6665489 | Collart | Dec 2003 | B2 |
| 6668246 | Yeung et al. | Dec 2003 | B1 |
| 6668325 | Collberg et al. | Dec 2003 | B1 |
| 6674858 | Kimura | Jan 2004 | B1 |
| 6687683 | Harada et al. | Feb 2004 | B1 |
| 6725372 | Lewis et al. | Apr 2004 | B1 |
| 6754822 | Zhao | Jun 2004 | B1 |
| 6775772 | Binding et al. | Aug 2004 | B1 |
| 6784354 | Lu et al. | Aug 2004 | B1 |
| 6785815 | Serret-Avila et al. | Aug 2004 | B1 |
| 6785825 | Colvin | Aug 2004 | B2 |
| 6792548 | Colvin | Sep 2004 | B2 |
| 6792549 | Colvin | Sep 2004 | B2 |
| 6795925 | Colvin | Sep 2004 | B2 |
| 6799277 | Colvin | Sep 2004 | B2 |
| 6813717 | Colvin | Nov 2004 | B2 |
| 6813718 | Colvin | Nov 2004 | B2 |
| 6823455 | Macy et al. | Nov 2004 | B1 |
| 6834308 | Ikezoye et al. | Dec 2004 | B1 |
| 6842862 | Chow et al. | Jan 2005 | B2 |
| 6853726 | Moskowitz et al. | Feb 2005 | B1 |
| 6857078 | Colvin | Feb 2005 | B2 |
| 6865747 | Mercier | Mar 2005 | B1 |
| 6931534 | Jandel et al. | Aug 2005 | B1 |
| 6957330 | Hughes | Oct 2005 | B1 |
| 6966002 | Torrubia-Saez | Nov 2005 | B1 |
| 6968337 | Wold | Nov 2005 | B2 |
| 6977894 | Achilles et al. | Dec 2005 | B1 |
| 6978370 | Kocher | Dec 2005 | B1 |
| 6986063 | Colvin | Jan 2006 | B2 |
| 6990453 | Wang | Jan 2006 | B2 |
| 7007166 | Moskowitz et al. | Feb 2006 | B1 |
| 7020285 | Kirovski et al. | Mar 2006 | B1 |
| 7035049 | Yamamoto | Apr 2006 | B2 |
| 7035409 | Moskowitz | Apr 2006 | B1 |
| 7043050 | Yuval | May 2006 | B2 |
| 7046808 | Metois et al. | May 2006 | B1 |
| 7050396 | Cohen et al. | May 2006 | B1 |
| 7051208 | Venkatesan et al. | May 2006 | B2 |
| 7058570 | Yu et al. | Jun 2006 | B1 |
| 7093295 | Saito | Aug 2006 | B1 |
| 7095874 | Moskowitz et al. | Aug 2006 | B2 |
| 7103184 | Jian | Sep 2006 | B2 |
| 7107451 | Moskowitz | Sep 2006 | B2 |
| 7123718 | Moskowitz et al. | Oct 2006 | B1 |
| 7127615 | Moskowitz | Oct 2006 | B2 |
| 7150003 | Naumovich et al. | Dec 2006 | B2 |
| 7152162 | Moskowitz et al. | Dec 2006 | B2 |
| 7159116 | Moskowitz | Jan 2007 | B2 |
| 7162642 | Schumann et al. | Jan 2007 | B2 |
| 7177429 | Moskowitz et al. | Feb 2007 | B2 |
| 7177430 | Kim | Feb 2007 | B2 |
| 7206649 | Kirovski et al. | Apr 2007 | B2 |
| 7231524 | Bums | Jun 2007 | B2 |
| 7233669 | Candelore | Jun 2007 | B2 |
| 7240210 | Michak et al. | Jul 2007 | B2 |
| 7266697 | Kirovski et al. | Sep 2007 | B2 |
| 7286451 | Wirtz | Oct 2007 | B2 |
| 7287275 | Moskowitz | Oct 2007 | B2 |
| 7289643 | Brunk et al. | Oct 2007 | B2 |
| 7343492 | Moskowitz et al. | Mar 2008 | B2 |
| 7346472 | Moskowitz et al. | Mar 2008 | B1 |
| 7362775 | Moskowitz | Apr 2008 | B1 |
| 7363278 | Schmelzer et al. | Apr 2008 | B2 |
| 7409073 | Moskowitz et al. | Aug 2008 | B2 |
| 7457962 | Moskowitz | Nov 2008 | B2 |
| 7460994 | Herre et al. | Dec 2008 | B2 |
| 7475246 | Moskowitz | Jan 2009 | B1 |
| 7530102 | Moskowitz | May 2009 | B2 |
| 7532725 | Moskowitz et al. | May 2009 | B2 |
| 7568100 | Moskowitz et al. | Jul 2009 | B1 |
| 7647502 | Moskowitz | Jan 2010 | B2 |
| 7647503 | Moskowitz | Jan 2010 | B2 |
| 7664263 | Moskowitz | Feb 2010 | B2 |
| 7743001 | Vermeulen | Jun 2010 | B1 |
| 7761712 | Moskowitz | Jul 2010 | B2 |
| 7779261 | Moskowitz | Aug 2010 | B2 |
| 20010010078 | Moskowitz | Jul 2001 | A1 |
| 20010029580 | Moskowitz | Oct 2001 | A1 |
| 20010043594 | Ogawa et al | Nov 2001 | A1 |
| 20020009208 | Alattar | Jan 2002 | A1 |
| 20020010684 | Moskowitz | Jan 2002 | A1 |
| 20020026343 | Duenke | Feb 2002 | A1 |
| 20020056041 | Moskowitz | May 2002 | A1 |
| 20020071556 | Moskowitz et al. | Jun 2002 | A1 |
| 20020073043 | Herman et al. | Jun 2002 | A1 |
| 20020097873 | Petrovic | Jul 2002 | A1 |
| 20020103883 | Haverstock et al. | Aug 2002 | A1 |
| 20020161741 | Wang et al. | Oct 2002 | A1 |
| 20030002862 | Rodriguez | Jan 2003 | A1 |
| 20030126445 | Wehrenberg | Jul 2003 | A1 |
| 20030133702 | Collart | Jul 2003 | A1 |
| 20030200439 | Moskowitz | Oct 2003 | A1 |
| 20030219143 | Moskowitz et al. | Nov 2003 | A1 |
| 20040028222 | Sewell et al. | Feb 2004 | A1 |
| 20040037449 | Davis et al. | Feb 2004 | A1 |
| 20040049695 | Choi et al. | Mar 2004 | A1 |
| 20040059918 | Xu | Mar 2004 | A1 |
| 20040083369 | Erlingsson et al. | Apr 2004 | A1 |
| 20040086119 | Moskowitz | May 2004 | A1 |
| 20040093521 | Hamadeh et al. | May 2004 | A1 |
| 20040117628 | Colvin | Jun 2004 | A1 |
| 20040117664 | Colvin | Jun 2004 | A1 |
| 20040125983 | Reed et al. | Jul 2004 | A1 |
| 20040128514 | Rhoads | Jul 2004 | A1 |
| 20040225894 | Colvin | Nov 2004 | A1 |
| 20040243540 | Moskowitz et al. | Dec 2004 | A1 |
| 20050135615 | Moskowitz et al. | Jun 2005 | A1 |
| 20050160271 | Brundage et al. | Jul 2005 | A9 |
| 20050177727 | Moskowitz et al. | Aug 2005 | A1 |
| 20050246554 | Batson | Nov 2005 | A1 |
| 20060005029 | Petrovic et al. | Jan 2006 | A1 |
| 20060013395 | Brundage et al. | Jan 2006 | A1 |
| 20060013451 | Haitsma | Jan 2006 | A1 |
| 20060041753 | Haitsma | Feb 2006 | A1 |
| 20060101269 | Moskowitz et al. | May 2006 | A1 |
| 20060140403 | Moskowitz | Jun 2006 | A1 |
| 20060251291 | Rhoads | Nov 2006 | A1 |
| 20060285722 | Moskowitz et al. | Dec 2006 | A1 |
| 20070011458 | Moskowitz | Jan 2007 | A1 |
| 20070028113 | Moskowitz | Feb 2007 | A1 |
| 20070064940 | Moskowitz et al. | Mar 2007 | A1 |
| 20070079131 | Moskowitz et al. | Apr 2007 | A1 |
| 20070083467 | Lindahl et al. | Apr 2007 | A1 |
| 20070110240 | Moskowitz et al. | May 2007 | A1 |
| 20070113094 | Moskowitz et al. | May 2007 | A1 |
| 20070127717 | Herre et al. | Jun 2007 | A1 |
| 20070226506 | Moskowitz | Sep 2007 | A1 |
| 20070253594 | Lu et al. | Nov 2007 | A1 |
| 20070294536 | Moskowitz et al. | Dec 2007 | A1 |
| 20070300072 | Moskowitz | Dec 2007 | A1 |
| 20070300073 | Moskowitz | Dec 2007 | A1 |
| 20080005571 | Moskowitz | Jan 2008 | A1 |
| 20080005572 | Moskowitz | Jan 2008 | A1 |
| 20080016365 | Moskowitz | Jan 2008 | A1 |
| 20080022113 | Moskowitz | Jan 2008 | A1 |
| 20080022114 | Moskowitz | Jan 2008 | A1 |
| 20080028222 | Moskowitz | Jan 2008 | A1 |
| 20080046742 | Moskowitz | Feb 2008 | A1 |
| 20080075277 | Moskowitz et al. | Mar 2008 | A1 |
| 20080109417 | Moskowitz | May 2008 | A1 |
| 20080133927 | Moskowitz et al. | Jun 2008 | A1 |
| 20080151934 | Moskowitz et al. | Jun 2008 | A1 |
| 20090037740 | Moskowitz | Feb 2009 | A1 |
| 20090089427 | Moskowitz et al. | Apr 2009 | A1 |
| 20090190754 | Moskowitz et al. | Jul 2009 | A1 |
| 20090210711 | Moskowitz | Aug 2009 | A1 |
| 20090220074 | Moskowitz et al. | Sep 2009 | A1 |
| 20100002904 | Moskowitz | Jan 2010 | A1 |
| 20100005308 | Moskowit | Jan 2010 | A1 |
| 20100064140 | Moskowitz | Mar 2010 | A1 |
| 20100077219 | Moskowitz | Mar 2010 | A1 |
| 20100077220 | Moskowitz | Mar 2010 | A1 |
| 20100098251 | Moskowitz | Apr 2010 | A1 |
| 20100106736 | Moskowitz | Apr 2010 | A1 |
| 20100153734 | Moskowitz | Jun 2010 | A1 |
| 20100182570 | Matsumoto et al. | Jul 2010 | A1 |
| 20100202607 | Moskowitz | Aug 2010 | A1 |
| 20100220861 | Moskowitz | Sep 2010 | A1 |
| Number | Date | Country |
|---|---|---|
| 0372601 | Jun 1990 | EP |
| 0372601 | Jun 1990 | EP |
| 0565947 | Oct 1993 | EP |
| 0565947 | Oct 1993 | EP |
| 0581317 | Feb 1994 | EP |
| 0581317 | Feb 1994 | EP |
| 0649261 | Apr 1995 | EP |
| 0651554 | May 1995 | EP |
| 0651554 | May 1995 | EP |
| 0872073 | Jul 1996 | EP |
| 1547337 | Mar 2006 | EP |
| 1354276 | Dec 2007 | EP |
| 1354276 | Dec 2007 | EP |
| 100523 | Sep 1998 | NL |
| 1005523 | Sep 1998 | NL |
| WO 9514289 | May 1995 | WO |
| WO 9514289 | May 1995 | WO |
| WO9701892 | Jun 1995 | WO |
| 9629795 | Sep 1996 | WO |
| WO 9629795 | Sep 1996 | WO |
| WO 9642151 | Dec 1996 | WO |
| WO9726733 | Jan 1997 | WO |
| 9724833 | Jul 1997 | WO |
| WO 9724833 | Jul 1997 | WO |
| WO9726732 | Jul 1997 | WO |
| WO 9744736 | Nov 1997 | WO |
| WO9802864 | Jan 1998 | WO |
| WO9837513 | Aug 1998 | WO |
| WO9837513 | Aug 1998 | WO |
| WO 9952271 | Oct 1999 | WO |
| WO 9962044 | Dec 1999 | WO |
| WO 9962044 | Dec 1999 | WO |
| WO 9963443 | Dec 1999 | WO |
| WO 0057643 | Sep 2000 | WO |
| WO0118628 | Mar 2001 | WO |
| WO0143026 | Jun 2001 | WO |
| WO0203385 | Jan 2002 | WO |
| WO02003385 | Oct 2002 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20100153734 A1 | Jun 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60125990 | Mar 1999 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 09594719 | Jun 2000 | US |
| Child | 11519467 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 11519467 | Sep 2006 | US |
| Child | 12655036 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US00/06522 | Mar 2000 | US |
| Child | 09594719 | US |
