Patent Application
20070242581
The present invention relates to methods and apparatus for copy protecting data carrying discs, such as compact discs (CD).
The CD is a read only storage medium. Originally, CDs were used for the storage of music and other audio data. However, formats, such as the CD-ROM format, have been developed facilitating the reliable storage of data for use by computers and other digital devices. The CD-ROM format has become very popular and has become the mass data storage medium for computer programs and other files.
CDs can store about 74 minutes of high quality stereo sound or about 650 megabytes of data or some combination of both. The value of a CD to the end user, because of the value of the data it stores, is typically much more than its physical cost. This has made it an attractive target for counterfeiters who can produce replicas of an original CD at a nominal cost and sell the replicas, at the market price for the data, to make large profits.
With distribution media such as audio tapes or video cassettes, the quality of counterfeit copies tends to be lower than that of the originals due to the corruption of the analogue signal in the copying process. There is no such degradation in the case of CDs, however, as all of the information is stored digitally. Counterfeiters can therefore produce counterfeit copies of CDs which are almost indistinguishable from original or bona fide CDs.
If it were possible to distinguish between an original or bona fide CD and a counterfeit, then the problems caused by counterfeiters could be substantially reduced. Law enforcement officials, for example, would have a means of identifying counterfeit CDs, and could more easily obtain search warrants and secure prosecutions. If a CD carries a program, that program could be used to check that it was loaded from a bona fide, original CD.
In addition, there is always a need for new copy protection techniques.
The present invention seeks to identify counterfeit CDs and other data carrying discs and to develop copy protection techniques.
According to a first aspect of the present invention there is provided a method of determining the provenance of a data carrying disc, the method comprising the steps of determining physical characteristics of selected data recorded on a first, test disc whose provenance is to be determined, and comparing the physical characteristics determined with the physical characteristics of the same data on a second, data carrying disc produced from a known source to determine whether the test disc has been produced from the known source, wherein the physical characteristics determined have arisen inherently out of the manufacturing process from which the first and second discs were produced.
The manufacturing process for a data carrying disc, such as a CD, begins with the production of a master, and in the physical manufacturing process of the master, variations are introduced which give the master distinctive physical characteristics and cause errors in the data. The master is used to form generations of discs, and the physical characteristics of the master are passed on, down the generations. A group of discs manufactured from the same source, therefore, consistently exhibit the physical characteristics of the master, and those physical characteristics can be used as a “fingerprint” for, or identification of, that source. Additionally and/or alternatively, a disc may have predetermined physical characteristics impressed thereon to provide copy protection signals which may be used either to signify that a disc is genuine or as a means to disable access to a copied, counterfeit disc.
This invention has relevance to all data carrying discs, such as CDs and DVDs (digital versatile discs) and other optical discs, where physical characteristics arising out of the physical manufacturing process, are consistently reproduced in all discs originating from the same source.
Thus, in a method of the invention, the physical characteristics determined arise from the physical manufacturing process and act as a fingerprint for a particular known source in the manufacturing process.
An embodiment of a method of the invention may be used to identify that a test is genuine when the physical characteristics determined from the test disc correlate with the physical characteristics of the same data from a known and genuine source. Similarly, the absence of correlation between the physical characteristics may be used to identify a counterfeit disc.
CDs carrying popular music or computer programs, for example, may originate from a number of masters from manufacturing plants in different countries. What is more, the or each master will not be used to directly form the CDs. Each master is used in the production of a number of stampers which are used to produce the discs sold in the marketplace. It will, therefore, be appreciated that although the fingerprint of one master will be apparent in its progeny, each stage of the process will also introduce fingerprints from the sources used in those stages.
In an embodiment, the method further comprises the steps of extracting selected data from a test disc and determining the physical characteristics thereof, and at substantially the same time extracting the selected data from a disc from a known source and determining the physical characteristics thereof, whereby the comparing step is performed in real time.
For example, each of a test disc, whose provenance is to be determined, and a genuine disc from a known source are read in a respective disc reader and the determined information output from each reader is fed, for example, to processing means to enable the comparison to be undertaken.
Additionally and/or alternatively, the method may further comprise the steps of extracting selected data from a disc from a known source and determining the physical characteristics thereof, and recording the physical characteristics determined as a fingerprint of the known source.
There are different ways in which a method of the invention as defined above may be implemented. For example, measurements may be taken to determine the physical characteristics and to thereby enable the comparison. This would be an implementation in hardware.
Alternatively, data extracted from a disc can be processed to reveal information as to its location on the disc, for example, whereby physical characteristics thereof are determined. This would be an implementation in software.
Of course, an implementation which combines the two approaches would also be possible.
The present invention also extends to apparatus for determining the provenance of a data carrying disc by a method as defined above.
According to a further aspect of the present invention there is provided a method of copy protecting a data carrying disc, the method comprising:
recording selected data onto a data carrying disc, and
predetermining the physical characteristics of said selected data on the data carrying disc.
The physical characteristics of said disc which are predetermined may be chosen as appropriate. In an embodiment, the physical characteristics predetermined are the locations of the selected data on the disc.
For example, the physical characteristics may be the total length of the data carrying spiral track, or the radial positioning of specified sections of data. In one embodiment, the method further comprises recording said selected data at predetermined angular orientations on the disc. The angular orientations may be absolute or relative.
In a presently preferred embodiment, the selected data is arranged to lie substantially on a single radius of the disc.
The selected data whose physical characteristics are predetermined may be dedicated data sections. For example, the dedicated data sections may be introduced during manufacture along a radius of a master to provide copy protection for the discs made therefrom.
In an embodiment, the selected data whose physical characteristics are predetermined are selected frames of the data recorded on the disc.
Preferably, the selected frames of data are at evenly spaced radial positions on a data carrying disc from a known source.
The present invention also extends to a method of accessing data on a copy protected data carrying disc, where a data carrying disc has been copy protected by recording selected data thereon such that the recorded selected data has predetermined physical characteristics, the accessing method comprising:
extracting the selected data from said copy protected disc and determining the physical characteristics thereof; and
enabling access to the data on said disc if the physical characteristics determined from the disc are substantially the same as said predetermined physical characteristics.
In an embodiment, said predetermined physical characteristics are recorded as a fingerprint.
Preferably, a method of accessing data on a copy protected data carrying disc further comprises:
rotating said disc in a disc reader to extract the selected data therefrom;
determining a location in time of said selected data which has been extracted from said disc; and
measuring the angular orientation on said disc of said selected data.
For example, data may be recorded onto said copy protected disc according to CD standards with timing information recorded in the Q-subchannel, and wherein timing information from Q-subchannel blocks is utilised both in determining the time location, and in measuring the angular orientation of said selected data.
In an embodiment, the disc reader has a motor for rotating the disc, and an encoder associated with the motor, and wherein the angular orientation is measured using said encoder.
According to a further aspect of the present invention there is provided a theoretical model of a data carrying disc for use in copy protecting data carrying discs, the model describing the information to be carried on the disc and the intended locations of the information on the disc in a manner enabling the relative locations of any two pieces of the information to be calculated.
Preferably, the information and its intended locations are described in a manner enabling the relative angular orientations of any two pieces of the information to be calculated.
The present invention also extends to a fingerprint for use in a method of determining the provenance of a data carrying disc, the fingerprint identifying selected data recorded on a data carrying disc and identifying physical characteristics of said selected data.
In an embodiment, the identified physical characteristics of said selected data are the locations of the selected data on said disc.
For example, identified physical characteristics of said selected data are the angular orientations of the selected data on the disc.
A method of copy protecting a data carrying disc as defined above may further comprise writing the selected data to the disc at a bit rate controlled in dependence upon the predetermined physical characteristics of the selected data whereby the selected data is written at predetermined locations on the disc.
Preferably the method further comprises:
rotating the disc during the writing of data thereto;
writing the selected data to the disc as it is rotated; and
controlling the speed of rotation of the disc in dependence upon the predetermined physical characteristics of the selected data whereby the selected data is written at predetermined locations on the disc.
The present invention also extends to apparatus for producing a copy protected data carrying disc, said apparatus comprising:
a writing apparatus for writing data to a disc,
a drive for rotating the disc as it is written, and
control means for controlling the writing apparatus and the drive such that selected data is written to the disc to have predetermined characteristics.
Embodiments of the present invention will hereinafter be described, by way of example, with reference to the accompanying drawings, in which:
As is made clear above, the present invention can be used for any data carrying discs where physical characteristics of the discs arising out of the physical manufacturing process will be consistent for all discs originating from the same source. However, for simplicity the present invention is described herein with specific reference to CDs.
As is well known, a CD originates from a glass master which is cut by a laser whilst the disc is rotated on a mandrel. The laser beam intensity is modulated by the data to be recorded. Each master is used to make nickel copies which are used to make stampers. Commercial CDs are manufactured using these stampers. Every master is physically unique, and its physical characteristics result from imperfections in the glass substrate or photoresist coating, and arise out of variations, for example, in the speed of rotation of the mandrel and from the movement of the laser. These physical characteristics produce a fingerprint which is inherent to the master. Similarly, as the generations of copies to produce the stamper and then the CD discs are made directly and then indirectly from the master, additional unique fingerprints will be superimposed on each generation copy. These fingerprints are physical characteristics which are passed on to successive generations of copies. Of course, each generation, because of the manufacturing process used to produce it, has its own set of physical characteristics or its own fingerprint. Thus, every CD has superimposed fingerprints, one of which can be attributed to each of its ancestors.
With methods and apparatus of the invention, physical characteristics of a master or of a disc from a known source, such as a known stamper, are determined and are used as a fingerprint of that master or stamper. This fingerprint can be compared with the physical characteristics of a disc under test to determine whether or not the test disc has the same master or stamper as an ancester.
To further explain the specific embodiments of the present invention it is useful to look briefly at the data format of a CD.
Data is recorded on CDs according to the International Standard ISO/IEC 10149. The data is present on the CD as a sequence of pits of varying length in the reflective material of a CD. These pits lie on a tight spiral which runs continuously from near the centre of the reflective portion to near the outside of the reflective portion of a CD. A CD reader uses a laser to scan along the spiral and it detects the edges of the pits by measuring the reflectivity of the disc as it scans. The presence of a pit causes less of the light to be reflected back.
The pits represent binary bits of information. A data byte comprises 8 binary bits and a large number of these data bytes are grouped together in sectors along the spiral track.
The laser produces a beam of light which is reflected from the disc for its intensity to be measured with a photodetector. The photodetector produces an analogue signal which can be identified with the reflectivity of the disc along the spiral. The signal is amplified and converted to produce a stream of zero bits with a set bit for every edge detected. This is called the EFM (eight to fourteen modulation) stream and is the first digital signal from the disc.
The EFM stream is divided into blocks separated by a 24 bit synchronisation pattern. The synchronisation pattern is followed by three “merging” bits and then 33 lots of 14 bit words each followed by three “merging” bits. The 14 bit words are passed through a EFM demodulator (for example, a look-up table) which translates the words into 8 bit bytes. Each EFM block separated by the 24 bit synchronisation pattern therefore translates into a “frame” of 33 bytes. One byte of every frame is used for the “subcode” and the remaining 32 bytes are passed to C1 and C2 decoders which apply error correction.
Error corrected data comes out of the decoders in 24 byte blocks. These blocks are assembled sequentially, 98 at a time, into 2352 byte sectors. These 2352 bytes encode audio data but computer data has another layer of error correction on top, leaving 2048 bytes of user data. The subcode bytes are assembled vertically 98 at a time into subcode blocks. The first two subcode bytes are synchronisation bytes and the rest of the bytes are divided into the P, Q, R, S, T, U, V and W subchannels. The P-subchannel consists of the high order bit of the 96 non-synchronisation subcode bytes. The Q-subchannel block is made from the second highest order bit of the same bytes and so on.
The first four bytes of the Q-subchannel block are the “control” field, the second four make up the “ADR” field. 72 DATA-Q bits then follow, the interpretation of which depends on the value of the ADR field. There then follows a 16 bit CRC on the three previous fields. The CRC detects errors but does not correct them.
There are very nearly the same number of Q-subchannel blocks as there are sectors on a CD. It may appear from the description that Q-subchannel blocks correspond to sectors on a one to one basis; but this is not quite true as the C1 and C2 decoders delay some bytes for the purposes of coping with burst errors.
Normally the Q-subchannel blocks are used by the drive of a CD reader to navigate around the CD. When the “ADR” field contains “0001” the DATA-Q bits encode the position of the block in the track from the start of the disc. Each Q-subchannel block therefore has a unique address which increases monotonically.
The Q-subchannel block can be used to record other information such as the UPC/EAN which is effectively a bar code or an ISRC as defined in DIN-31-621. These Q-subchannel blocks do not contain much position data and hence are not so useful for navigating around the drive. They are placed at regular but fairly large intervals in the Q-subchannel so as not to interfere with the CD drive's navigation requirements.
Generally, during the manufacture of a master the track pitch and linear velocity are very closely controlled at constant values by servo systems, although no mastering system will provide zero deviation from the controlled values. Thus, two masters containing identical digital data will have physical differences in the total length of the spiral of information pits, the angular orientation of separate data areas relative to one another, and the radial positioning of the data. There will also be variations in the positions, both radial and angular, of sectors in which data is grouped, and in the positions of any markers on the CD.
An example of the differences in the physical characteristics is shown in
An example of an embodiment of the invention is illustrated in FIGS. 3 to 5. In this embodiment, it is the relative angular orientations of data on a CD which is the physical characteristic determined.
It would be possible to utilise an incremental shaft encoder 8 arranged to generate absolute index pulses. However, in the illustrated embodiment, and for simplicity, the shaft encoder 8 is arranged to generate incremental pulses and, reliance is placed on error-free storage of the cumulative count in a counter 25 (
The main data channel comprises digital information which, for CDs containing digital audio data, is output at a rate of 176,400 8-bit bytes every second. As described above, these bytes are grouped into sectors of 2,352 bytes, output at a rate of 75 frames per second. Embedded in parallel with the main data channel is the subcode data channel which contains timing and other information. This is grouped into the subcode blocks. The 96 non-synchronisation subcode bytes are output in parallel at a rate of one bit of each of the 8 subcode bytes simultaneously as an 8-bit byte for every 24 bytes of main data channel data. Consequently the 8 subcode bytes are output completely at a rate of 75 times per second, with the main timing information located in the second, Q-subchannel block. This timing information includes an indication of the absolute location of the main data channel in the format of minutes: seconds: frames, thereby locating the main data channel to a resolution of 1/75th of a second, or 2,352 bytes. This time count, which starts from 00:00:00 at the start of the data area on the CD, is commonly referred to as ‘absolute time’ or ‘A-time’, and is used by CD players and CD-ROM drives to navigate the disc during playback.
The substantially conventional data extraction circuit is, as is shown in
The use of the additional components 20 and 21 enables any commercially available CD player or CD-ROM drive from which the HF signal can be extracted to be modified for use in a method of the invention. This is preferred because it is not always possible to extract the Q-subchannel signal, the Q-subchannel sync signal and the Q-subchannel clock signal from commercially available CD players or CD-ROM drives. Even where it is possible to extract these signals, the timing delays of the digital data stream in different manufacturers' EFM decoder devices 19 are not necessarily identical.
The Q-subchannel signal 23 is a 96-bit serial bit stream which is clocked into a 96-bit binary shift register 27 by the Q-subchannel clock signal 24. The output of the 96-bit shift register 27 is fed to a 96 bit latch 32 such that the value of the Q-subchannel bit stream is latched by the Q-subchannel sync signal 22. In this manner the output values of both of the 32-bit latch 29 and of the 96-bit shift register 32 are latched synchronously with the Q-subchannel sync signal 22. It will be appreciated that the output of the latch 29 will be a count representative of angular position, whereas the output of the shift register 32 is the value of the Q-subchannel.
The values of the count from the latch 29, of the Q-subchannel, and of the Q-subchannel sync signal 22 are fed to a computer to enable the relative angular orientation of main channel data areas on the CD identified by the Q-subchannel to be determined. One method for making this determination involves:
1. Start the compact disc playing.
2. Reset the 32-bit counter 25 using the reset signal 26.
3. Monitor the Q-subchannel sync signal 22 to observe a state transition change from logical one to logical zero, which indicates that a fresh 96-bit Q-subchannel signal 23 has been completely serially-shifted into the 96-bit shift register 27.
4. Read the latched output values of the latch 29 and the latch 32.
5. To continue the process repeat from step 3.
It is possible to produce a fingerprint of a CD by using the technique described above to determine, for example, the relative angular orientations of 10 specific frames, each of duration 1/75th of a second, and the frames being at evenly spaced radii on the CD. The procedure for this can be performed automatically by a computer controlled CD player or CD-ROM drive and is listed hereafter in a form suitable for realisation in software:
1. Read the Table of Contents from the CD in order to determine the length of the data area. This is a standard procedure performed by all CD players and CD-ROM drives prior to accessing the main channel data on the CD. From this information calculate the absolute time values of 10 frame positions on the CD (1 frame is a main channel data segment of duration 1/75th of a second).
2. Start the CD playing, then reset the 32-bit counter 25.
3. Play the CD at a point five or more frames earlier than the first frame position calculated at stage 1.
4. Record the Q-subchannel value and the count representative of the relative angular position for all frames from the point five or more frames earlier than the first frame position calculated at step 1 to a point five or more frames later than the first frame position calculated at step 1. From the resulting recorded sequence of Q-subchannel values and count values of the relative angular positions, the location of the frame position of the first frame position calculated at step 1 can be determined either directly or by interpolation. The use of an interpolation technique enables CDs to be read where there are data-errors corrupting the Q-subchannel data, as is commonly the case.
5. Repeat steps 3 and 4 for the remaining nine frame positions calculated at step 1.
6. Construct a fingerprint for the CD by normalising the ten recorded relative angular positions to the first and calculating the resulting angular orientations. The fingerprint then takes the form illustrated in
Thus, the position of specific sections of the data stream of a master or of a known genuine disc may be measured to provide an authentication fingerprint. That authentication fingerprint may be stored, for example, in a suitable memory or storage device and/or may be recorded on paper or in other hard copy format. The measurement may be carried out using apparatus as described above. Then, at any time in the future, if it is desired to authenticate a disc which is believed to have been produced from that master, apparatus as described above may again be used to obtain the fingerprint of the corresponding sections of data stream on the test disc. The authentication fingerprint and the resulting fingerprint from the test disc are then compared to determine whether there is close correlation.
If the fingerprint data is stored electronically, such correlation may be carried out by suitable data processing apparatus. If the fingerprints are reduced to paper or other hard copy form, human input will be required. The latter may be advantageous when, for example, fingerprints of masters are stored at one location (for example, kept with the original manufacturer), whilst the fingerprints of discs under test replicas are obtained elsewhere, (for example, by police or customs authorities). The comparison of the fingerprints enables the test disc to be authenticated.
In the embodiment discussed above, the physical characteristics of the discs are determined by physical measurement of the discs. However, it is alternatively possible to determine physical characteristics by calculation, and/or by processing data obtained from the discs and/or by other software methods.
The position of sectors or markers on a CD is calculated by measuring the time taken to read the sector, immediately after reading a reference sector or other datum such as the Table of Contents (TOC). That is, the drive of a conventional CD reader is forced to seek from a particular reference point to a sector of interest. The total time taken to read the sector is taken as representative of the position of the sector header.
After processing, timing data derived from consecutive sectors or markers describes a characteristic saw tooth pattern as shown diagrammatically in
As discussed above, sectors that lie on any given radius will always be the same for CDs generated from the same master as shown, for example, in
In a first method to extract data representative of the interrelationship between sectors, a clock of the processor 63 is synchronised with data stored at the table of contents (TOC) on an original CD or master CD in reader 65. The disc under test is then placed in reader 60 and data representing sector position is extracted.
The time intervals between sectors or markers of the test disc are recorded. The timing information is compared with corresponding time intervals from stored information from a known source. If the disc under test is bona fide there will be no (or only minor) differences. However, if significant discrepancies between the two sets of timing information occurs then it is determined that the test disc is not derived from the known source and that they do not have common ancestors. So as to take account of extremely small discrepancies arising from thermal noise or drift discrepancies in timing, the information may be normalised. This depends on the extent of variation in measurement which arises due to Gaussian noise. A tolerance band is adopted. Discrepancies which fall outside the tolerance band are deemed to have been derived from a disc from a different source. Neural networks (not shown) may be taught to recognise such discrepancies.
In a second method, the apparatus in
In a third method, data about the inter-sector or inter-marker or marker timing for particular CDs, is supplied on a data carrier and is input independently into the processor 63 where it is compared with data derived from a test disc.
In any of the methods described above, it is possible simply to check that the disc under test has all of the same physical characteristics as the known source. Preferably, however, data from the test disc and corresponding data from the known source are correlated to calculate their degree of similarity. This may be done by cross referencing the two sets of data, identifying common features, and then making a calculation using statistical methods of the probability of the common features occurring purely by chance. A calculation of the probability that the discs are from different manufacturings can be made. A certain probability is taken as the threshold above which the hypothesis will be accepted.
The statistical method used can be chosen as required. Currently Bayesian statistical methods are preferred.
The invention has been described above with reference to methods of determining the provenance of a CD. However, and as noted above, it is also possible to use the described techniques to copy protect a CD. In this case, specific data is selected and this selected data is then written to the CD such that it has a predetermined physical characteristic. For example, selected sectors may be given a particular relative angular orientation. This selected data can then act as a signature for the disc.
In this respect, and as is well known, in a signature method software loaded onto the disc requires the disc reader to ascertain that the disc carries the required signature before access to the data on the disc is enabled. In this case, the software would include information identifying the selected data and its physical characteristics, such as its location, and access to the data on the disc would then be enabled if the disc were found to be carrying the selected data at the specified locations.
In order to copy protect a disc using this technique it is necessary to create a theoretical model of the disc to be created, or, at least, of the selected data and its location or other physical characteristics. This model, for example, could look like the fingerprint shown in
The apparatus shown in
As is also shown in
The data to be written to the disc 80, in the data storage medium 86, includes not only the information and control data to be written to the disc but also identifies from amongst all of the data, selected data which is to be given predetermined physical characteristics, for example, is to be put at predetermined locations on the disc. The apparatus shown in
An advantage of using the present technique to add copy protection signals to a disc is that it is very difficult for a counterfeiter to realise that the disc is copy protected. In this respect, the copy protection added mimics the normal physical properties of a disc and it is therefore not obvious which physical characteristics are being checked by the authentication.
It is important, of course, that the copy protection signature or fingerprint is not duplicated when a copy of the original disc is made. However, and as is indicated by consideration of
It will be appreciated that variations in and modifications of the present invention may be made within the scope of this application as defined by the appended claims
This application is a continuation-in-part of U.S. patent application Ser. No. 09/582,827 filed Dec. 22, 1998.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10382821 | Mar 2003 | US |
| Child | 11818267 | Jun 2007 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 09582827 | Jun 2000 | US |
| Child | 10382821 | Mar 2003 | US |
