Method for copy protecting a record carrier with a pattern of logical errors

ES 2 247 261 T3

DESCRIPTION

A method of copy-protecting a log medium with an error logic pattern.

This invention relates to a method of copy-protecting a record carrier, a copy-protected record carrier, and a method of detecting access control information.

A system for copy-protecting a record carrier, a copy-protected record carrier and an arrangement for reading are known from EP-0545472 (document D1 of the list of related documents). The known record carrier comprises a pre-arranged guide track, a so-called pre-groove. In the track determined by the previous groove, the information that is written in a previously defined way is represented by optically readable patterns that are formed by the variation of a first physical parameter, such as the height of the scanned surface. The pregroove has variations in a second physical parameter, such as a deflection in a transverse direction, also referred to as wobble. The oscillation in the previous groove is modulated in FM and that modulation represents access control information that is related to the information, such as a decoding code to retrieve the stored information as encoded information. The known device comprises reading means for reading the patterns and retrieval means for recovering the access control information. The known device and the record carrier form a system for the reproduction of the controlled information. For this purpose, the device comprises means for reproducing the information depending on the access control information. If the information is copied onto a record medium that can be written to, the information from that copy will not be reproduced, because only the patterns are written during the writing process and the copy itself does not contain access control information. some. A known system problem is that the reading means must be capable of recovering the access control information by detecting variations in the second physical parameter.

It should be noted that document WO 95/03655 (document D3) describes a CD PROM encryption system, in which the information on a CD-ROM is encrypted by means of a key, the key of which is programmed into the CD-ROM. after their manufacture, thereby damaging selected sectors to make them unreadable by the usual reading systems. Selected sectors are physically damaged by a high power laser. Record carriers are individually empowered to have a specific key for a certain user or group of users.

An object of the invention is to provide a system for copying record carriers, which is not based on variations in physical parameters, while counteracting the making of usable copies on writable record carriers. The invention is defined in the independent claims. The dependent claims define advantageous embodiments.

In the following, other preferred advantageous embodiments of the copy-protected record carrier, the retrieval arrangement and the methods according to the invention are described.

In EP-A-0 533 204 an information recording / reproducing apparatus is described, which has a protection function to protect the information recorded on an information recording medium. The information on the information recording medium is protected against access by executing a data conversion process after normal error encoding, such that the error correction capabilities are exceeded. In embodiments, the data conversion process is based on a secret key, which must be known to the reader in order to access the data. Knowledge of that secret key, which is transferred separately from the information recording medium, enables the reader to perform a reverse data conversion when the data is read. Then the normal error correction takes place. Without the knowledge of that secret key, the reader cannot read the data at all, as many uncorrectable errors are detected, thereby exceeding the error correction capabilities of the reader.

These and other aspects of the invention will become apparent from, and will be further clarified with reference to, the embodiments described by way of example in the description that follows, and with reference to the accompanying drawings, in which:

Figure 1 represents a copy-protected record carrier;

Figure 2 represents a logical map of the record area of ​​a copy-protected record carrier;

Figure 3 represents an error correction unit;

Figure 4 represents a bit error pattern;

Figure 5 depicts an arrangement for retrieving information from a copy-protected record carrier;

Figure 6 depicts a schematic diagram for copy-protecting a record carrier.

A disk-shaped copy-protected record carrier 1 is schematically depicted in Figure 1. The record carrier comprises a track 9 for storing information, the track of which is arranged in a helical pattern of windings around a central hole 10. The windings may also be arranged concentrically, rather than helically. The record carrier 1 is of an optically readable type, in which a transparent substrate is covered by a recording layer and by a protective layer, such as the well-known Compact Disc (CD). Information on the record carrier is represented by optically readable mark patterns. For example, the position and / or length of the marks then represent a binary information signal. Markings can be made with presses, as is customary for read-only CDs, such as a CD-ROM in which patterned holes and plateaus between holes represent the information. The invention can be used for any type of recording medium on which information is recorded according to predetermined error correction rules, such as high-density optical disc DVD (Digital Versatile Disc), optical tape or tape. magnetic for digital video. Track 9 comprises the marks and

ES 2 247 261 T3 can be scanned by a read head to read the stored information. The marks represent a sequence of bits according to a channel code, such as the EFM for CD (Eight to Fourteen Modulation). The bit sequence represents information according to predetermined error correction and formatting rules, such as the CIRC (Reed-Solomon Cross Interleaved Code) for CD. According to the formatting rules, the record medium can be subdivided into addressable sectors, such as on CD-ROM. CD-ROM is described in ISO 10149, the CD-ROM specification.

The information on the copy-protected record carrier 1 comprises main information and access control information to control access to the main information, so as to prevent access to the main information copied in an illegal copy, the copy of which does not understand all necessary access control information. According to the invention, the copy-protected record carrier 1 is provided with logic errors 2, the logic errors of which constitute an error pattern representing at least some of the access control information. The error pattern has at least one logic error at a predetermined position, but preferably a logic error pattern at a relatively large number of predetermined positions. A different error pattern can be generated for each new title to be distributed on a copy-protected record medium. In one embodiment of the error pattern the error positions should be interspersed with positions without error or with essential information. Furthermore, the error pattern may comprise a number of isolated error positions between error-free positions, but also some consecutive error positions. Preferably, the error position pattern is a pseudo-random pattern comprising approximately 50% error positions and 50% error-free positions, the pseudo-random pattern being generated from a source value, by means of a predetermined algorithm. . The error pattern has to be verified by an access control procedure, the procedure of which is inextricably embedded in the procedure for using the main information. The check should include at least one error position and preferably also at least one error-free position in the vicinity of an error position. This prevents a malicious party from easily mimicking or mimicking the error pattern by physically damaging a few positions. A logic error is constituted by a series of bit errors in the bit sequence, the number of bit errors of which is uncorrectable by said error correction rules. Bit errors cannot be copied using a normal recording device, since such device will accept the information to be recorded without error correction bit. The recorder will process this information in accordance with its built-in default formatting and error correction rules to generate a new sequence of bits, including inherently correct next-generation and error correction bits. Therefore, this new bit sequence will not understand any errors, and is not accessible to change bits in such normal register devices. The new sequence of bits can be recorded on a writable record carrier, but that copy will not understand the error pattern. It is to be noted that the logical errors have to be applied to the bit sequence after the error modification step, before writing the physical pattern of marks, so that it is uncorrectable by the error decoding step after reading the errors. brands. Application errors at a higher system level prior to error coding, for example by intentionally changing EDCs (error detection codes) in a sector or in a sector header on a CD-ROM, can be easily imitated or camouflaged by a malicious party, because the higher level formatting process is usually carried out by means of software, and is therefore accessible by manipulation. A working, albeit illegal, copy comprising the highest level errors can be made with normal recording devices and (adapted) software on the connected computer system, for example, a bit copy program available to make copies of Audio CDs.

Since burst errors will occur due to dirt or scratches on the surface of a recording medium, error correction rules, for example on a CD, are specially designed to correct burst errors by applying an interleaved before storing and deinterleaving after reading. The bit errors that constitute such burst errors will be mixed with a much larger number of other bits in the bit sequence by the deinterleaving rules that are part of the error correction and formatting rules. A number of consecutive bit errors sufficient to cause the uncorrectable errors must be greater than the longest correctable burst error. The error correction rules are described in relation to Figure 3. In a preferred embodiment, only selected bits of the sequence of bits are errored, the bits of which are selected to be accumulated by deinterleaving, the deinterleaving of which is part of the error correction and formatting rules. This results in a position on the record carrier exhibiting a high concentration of errors, while contiguous positions exhibit only a few or no errors. An example of bit errors has been described in Figure 4. Usually, the error correction rules, and especially the (de) interleaving rules, will operate on symbols, for example on octets (groups of 8 bits), while the error correction process is applied to error words from a series of accumulated symbols by deinterleaving. The error symbols are selected to accumulate during deinterleaving to an uncorrectable number in one or a few error words.

An effective way of applying bit errors is to invert each bit from a selected symbol into the original error-free bit sequence, the selected symbol of which is to be provided with a bit error. Alternatively, bit errors can be applied to symbols, when said symbols are translated into the physical mark pattern, for example using a controllable EFM encoder. For such symbols, the EFM encoder could be controlled to change some of the physical marks to be different from the originally intended marks based on the sequence of

ES 2 247 261 T3 bits without errors. Preferably, the resulting physical marks comply with the limitations specified for physical marks, as this ensures reliable operation of the reading and decoding processes.

Figure 2 shows a logical map of the record area of ​​a copy-protected record carrier. The registration area is subdivided into addressable sectors from the top at address 00 to address MAX. The first area 21 may be a leading area guide area or pre-gap, such as the 2 second silence area on a CD. The second area is a system area 22, which comprises system information about the contents of the disc, such as the PVD (Primary Volume Descriptor) on a CDROM. The remaining part of the disk is available for user data, such as main and directory information files. The user data area can be subdivided into several areas, the subdivision of which is free and is not limited to the map shown in Figure 2. In this map, the remaining area comprises a third area 23 that comprises user files 28, a fourth area 24 that does not comprise user data, and a fifth area 25 that comprises new user data. According to the invention, the fourth area 24 comprises a fill area 26. The fill area 26 has error sectors 11 indicated by x, and error-free sectors 12, the error sectors of which comprise the logic errors and constitute the pattern of mistake. The padding area 26 may comprise a large number of sectors, for example 20 Mbytes (Mega octets), and may cover substantially the entire record area not covered by the main information. This has the advantage that when the presence of the error pattern is to be checked, a small number, or only one, of the error sectors can be selected each time from the large number available in the fill area. A malicious person trying to eavesdrop on the check will not see a recurring test of a specific sector or a few sectors. rather, mainly different sectors of a large range of addresses will be read to detect the presence of logical errors. This will effectively prevent the malicious party from designing simple interception means to spoof the output of a sector read. Another advantage can be achieved if the writable disk containing the illegal copy has a smaller data capacity than the pressed disk. For example, in most CD-ROMs, a large part of their capacity is not used, but can be completely filled by the fill area, without increasing the manufacturing cost, whereas a recordable CD or a Rewritable CDs have less capacity than a fully filled CDROM. In that case, not all information (user information and fill area) can be transferred to the illegal copy. In one embodiment, additional access control information is comprised in system area 22 or some other area that is not directly accessible on a normal reading device, such as the leaderboard, exit queue, or an area. of pre-parting space 21. That additional access control information could be a license code indicating the party making use of the system for copy-protected record media, and / or it could be indicative of the error pattern, for example to be used as a source in an algorithm for generating the error pattern.

Figure 3 shows an error correction unit used in the CD system, called CIRC (ReedSolomon Cross Interleaving Code). A detailed description of the CIRC error correction rules can be seen in document GB 2076569 (PHQ 80009), document D2, the decoder with Figure 7 being described therein. In Figure 3 the input (on the left) is a block of 32 bytes stored consecutively on the record carrier, indicated by the column of bytes numbered 0 to 31, comprising 12 data bytes 0-11, 4 bytes of error correction in C2 12-15, versus 12 bytes of information 16-27 and 4 bytes of error correction in C1. The odd-place bytes are delayed in one cycle by a first delay unit 31, and the resulting 32-byte error words are error corrected by the C1 unit 32. The C1 unit can correct the one-byte error and detect errors octets 2 and 3, while errors in octets 4-32 are detected with a very low failure rate. If unit C1 detects an uncorrectable error, it will mark all bytes that are unreliable. The output of unit C1 is delayed by a second delay unit 33, which delays byte 0 by 27 x 4 = 108 blocks, byte 1 by 26 x 4 = 104 blocks, etc. The output of the second delay unit 33 constitutes a second error word, which is error corrected by the unit C2. Drive C2 usually corrects up to 2 errors, but can correct up to 4 bytes per erase, if drive C1 marks all detected errors. The output of unit C2 is decoded by decoding unit 35. The functions described are the complement of the inverse functions in the encoder, all functions being well known from the CD system and having been described in detail in document D2. According to the invention, a logic error results from bit errors that are uncorrectable, so bit errors must be present in the input blocks accumulating up to at least 2, but preferably 3 or more errors in the words of drive input error C1. Also at least 3, but preferably at least 5 errors should be present in at least a second error word at the input of drive C2. For 5 errors in a second error word, this requires at least 17 output blocks with consecutive error marks from C1, having errors in a group of 5 consecutively numbered information bytes. For example, errors in octet 0 in block 0 will accumulate in the input of C2 with errors in octet 2 in block 8 and with errors in octet 4 in block 16. To create a group of logical errors within a specified data unit, such as a sector in the CD system, preferably more bit errors should be applied than the previous minimum of 5 errors in 17 consecutive blocks. One more layer of error correction, such as that used on CD-ROM for error correction within a sector, can correct some uncorrectable logic errors, according to the above error correction rules. Therefore, a greater number of logical errors should be included. A realization with which a safe margin is achieved without the risk of spreading errors over a large area

ES 2 247 261 T3 has errors in all the first or second twelve consecutively numbered information bytes. Said errors will be spread due to the second delay unit over 12 x 4 = 48 blocks and an additional block due to the first delay unit, and having 24 consecutive blocks with errors, over 48 + 1 + 24 = 73 blocks. As a block comprises 24 bytes of information, this affects 73 x 24 bytes = 1752 bytes, which is well within a sector of the CD-ROM format (2352 bytes) as long as the error blocks are located within of said sector. The sector comprises 98 blocks, so a maximum of 49 consecutive blocks can have errors without affecting neighboring sectors. It should be noted that applying errors to all the octets in the blocks, which would be the case if the entire disk area were provided with errors, for example by physically damaging the area, is not a solution to create logical errors. . Such errors will be spread over at least 28 x 4 + 1 = 113 blocks, and with a minimum of 17 consecutive error blocks over 130 blocks. The number of octets affected is 130 x 24 = 3232, which is considerably more than one sector.

When errors that are uncorrectable are detected in drive C1 and / or C2, the bytes will not be changed, but will be marked as errors. Bit errors in the information bits result in bit errors being propagated to the output of the error correction unit, while the error correction bits are used in the error correction unit and will not be visible at the exit. Therefore, in a preferred embodiment the bit errors are present in the information bits, and not in the error correction bits.

Furthermore, bit errors should not be spread by deinterleaving to contiguous sectors that are to remain error free. Therefore, in a preferred embodiment the part of the bit sequence corresponding to error-free sectors contiguous to error sectors does not substantially comprise bit errors. Although some bit errors can be corrected in contiguous sectors, there is a higher risk of uncorrectable errors if additional errors, for example caused by dirt, are combined with such bit errors. In that case, an errored sector can be mistakenly classified as an errored sector.

Although bit errors accumulate in a few selected error words, and therefore directly affect only the selected positions, a large number of symbols (= octets) will be marked by the C1 unit as unreliable (in fact, all symbols in all c1 words that have 2 or more bytes of error). The error unit C2 will first calculate a syndrome to detect any possible errors, which syndrome indicates whether errors are present. Additional calculations will indicate the number of errors and possibly indicate which symbols need to be corrected. For correction several approaches can be used, for example errors in only 1 or 2 symbols can be corrected directly, and to correct 2 to 4 symbols with error the marks coming from the previous C1 unit can be used, to indicate which symbol is used. is to be replaced (erasure correction). Usually the C2 drive will not use any flags from C1 if it detects 0 or 1 error. Therefore, the large number of symbols marked, but not changed, will not be observed or classified as errors by the C2 unit. Preferably, for a sector format, said symbols with the C1 mark should be as much as possible within the selected error sector, since additional errors caused by dirt, etc., can cause uncorrectable C2 errors, in combination with the symbols marked in C1. For a detailed description of the error processing rules, reference is made to D2. In a different embodiment, where double layer C2 / C1 error coding is used, errors are introduced during coding, after the C2 coding step but before the C1 coding step. Therefore, during decoding, no errors are detected in the decoder of C1, and no marking of symbols occurs in C1. However, in the decoding step of C2, the errors appear and are uncorrectable. All errors can be easily controlled to be within a sector, since interleaving and deinterleaving take place after the C2 encoding and before the C2 decoding step. Alternatively, a combination of C2 and C1 errors can be used.

The invention can be applied in systems where different error correction rules are used, such as on DVD. A bit error matching pattern can be found to counteract interleaving according to the above description. In other applications a more elaborate approach to error correction might be a repeated application of the error correction rules, first interleaving the output of the first error correction process as in the encoder, and secondly deinterleaving and re-applying the error correction rules. As some errors can be corrected in the first process, such a second error correction process could correct other errors. To prevent logic errors from being corrected with such an approach, preferably the bit rate and positioning of the bit errors are such that they are uncorrectable in each error correction layer.

In Figure 4 a bit error pattern has been represented for the CIRC error correction rules as described in Figure 3. Bit errors are designed to be accumulated in the error words in C1, as well as in the error words in C2, in both cases up to 5 errors. The error pattern may be shifted to other information octets (0-11, 16-27), but they should cover only information octets and not error correction octets. In Figure 4 the errors in the octets have been indicated by the letters a, b, c, d, e, while the octets without error have not been marked. To affect 20 consecutive words of C1, 21 errored blocks have been provided, the odd octets starting one block before and the even blocks stopping one block later, to compensate for the first delay unit 31. Due to the delays in the second delay unit 33, the errors marked with the same letter will accumulate in the words in C2, so that 4 consecutive blocks of C2 will have 5 errors "e", the next 4 blocks of C2 5 "d" errors, up to the last 4 blocks of C3, with "a" errors. No error will be compensated because in each one

ES 2 247 261 T3 of the error words in C1 or C2 will accumulate either 0 or 5 errors. This error scheme can easily be extended for more errors in more consecutive error words, as required, or for other interleaving rules. It can also be applied a few times within a sector, to avoid any subsequent burst error correction process correcting the logic errors. As mentioned above in Figure 3, the C1 checker has a 100% probability of detecting words with 2 and 3 errors, while 4-32 errors could occasionally be falsely corrected. Therefore, a preferred embodiment where only C1 errors are used, has 3 errors accumulating in the C1 words. From Figure 4 it is possible to deduce an effective error pattern that has said 3 C1 errors, just by applying the errors a, b, c and omitting the errors d and e. Alternatively, instead of having 4 consecutive blocks of C1 with errors, only the first of each quadruple can have errors, resulting in an error in the block of C2. The error pattern that starts at octet 0, as indicated in Figure 4, results in that almost all of C1 is marked, but unchanged octets precede the error octets, the first octet being 27 of block 2 that precedes the logic errors by 108 blocks. After the logic errors, only a few marked octets will follow, but not changed, that is, the last one being octet 0 of block 21 with a 16-block delay. Preferably, by synchronizing the affected blocks with the sectors, the marked but unchanged octets should be positioned within the error sector, for example, the error pattern of Figure 4 should be applied at the end of the error sector. Correspondingly, when errors are applied to the highest numbered octets (eg 23-27), most of the marked but unchanged blocks will lag behind the logic errors.

An arrangement for retrieving information from a copy-protected record carrier 1 and processing the information is depicted in Figure 5. The arrangement comprises a reading unit for reading the sequence of bits from the record carrier 1. The reading unit comprises a reading head 41 for scanning the track and generating a reading signal corresponding to the physical marks on the record carrier, and a translation unit 42 for translating the reading signal into the sequence of bits, for example, an EFM decoder for decoding on a CD system. The sequence of bits is coupled to an error correction unit 43 to retrieve the information and correct possible errors, for example, the CIRC corrector in a CD system. The retrieved information is coupled to access control means 47 to control access to the information. The access control information is available for further processing at the output 48 of the access control means 47. During reading, the read head 41 is positioned on the track by a servo unit 44 of the conventional type, while the record carrier is rotated by a motor unit 45. The reading of the information is controlled through a controller 46, whose controller controls the motor unit 45, the servo unit 44 and the error correction unit 43, and is arranged to receive read commands, for example through an interface with the access control means 47 . The access control means 47 can be realized in circuits incorporated in a reading device that also comprises the above reading means. This has the advantage that the information will not be delivered at the output 48 if the access control information is not present on the record carrier. The access control means may alternatively be realized on a computer connected through an interface to a normal reading device, such as a CD-ROM drive mechanism. In the computer, the access control means can be incorporated in an interface box, or they can be carried out by means of software running on a central processing unit. The software for effecting the access control can be supplied to a user on the copy-protected record medium. This is advantageous in that no dedicated hardware is required, and because the user has all the necessary means to access the copy-protected information incorporated into the record carrier.

Access control according to the invention will be carried out as follows. First, the access control means will acquire the access control information indicative of the error pattern. This access control information can be a pattern stored on the record medium, for example the license code described in Figure 2, or an access code supplied over a network, for example the Internet, or on paper. In one embodiment, the error pattern can be generated using a source value and a previously defined algorithm, the source value being stored on the copy-protected record carrier. Second, it must be checked for the presence of logical errors, to ensure that the disk is a copy-protected original disk, and not an illegal copy. The access control means will select one or more error positions in the record carrier, or error sector (s) if the record carrier is formatted in addressable sectors, the error conditions of which shall have a logical error according to the error pattern. Third, the presence of an error is verified by reading the position of the selected error. Since the reading device will generate an error message on the interface in case a sector to be read comprises uncorrectable errors, the presence of the logic errors can be effectively detected. As a copy will not understand logical errors, the copy will be rejected and access to the information will be prevented. Preferably, a few error positions are selected at random from a large number of error positions available on the record carrier, for example, in a padding area dedicated to understanding the error pattern having error sectors and blank sectors. error. Alternatively, the error sectors can be intermixed with valid sectors that comprise the information for the normal user. In order to get a quick response for access control to take place, preferably only one error sector is selected and read. In practice, reading an error sector that has uncorrectable errors could only cause a delay of a few seconds, the delay of which is caused by the device.

ES 2 247 261 T3 read when trying to read the sector a few times. Such repeated tests are normal practice to improve the chances of successful information retrieval in the event of dirt or scratches. The error pattern can also be verified by selecting at least one position without error, but not all positions without error, whose positions without error should not have a logical error according to the error pattern, and verifying the absence of a error by reading the selected error-free position. Since the reading of a sector without error will be very fast, for example 0.2 seconds, preferably a greater number of sectors without error are selected and read, for example from 10 to 40. Said sectors without error are preferably selected to through a random selection process from all available sectors without error. In one embodiment the content of the sectors can be verified without error, for example by including a check value in such sector to be deduced from the sector number and the license code by means of a predetermined encryption algorithm. In another embodiment, error-free sectors are selected and checked before and after the selected error sector. This has the advantage that, for example, 21 to 81 sectors selected apparently randomly are read, of which only one has to understand an error. This has the advantage that a malicious party will have great difficulty trying to mimic or mimic that verification process. Since the error pattern according to the invention is designed to accumulate all the bit errors in the error sectors, without affecting the contiguous sectors, preferably some error-free sectors contiguous to the error sectors should be selected. This will likely expose the corrupted illegal copies in certain areas to mimic the error pattern. Access control for a computer program can be effected as follows. The information carrier contains the computer program, while some essential data is included in the access control information. The computer program may itself be encrypted, while a short startup program takes the place of the program and controls access to the main program. The access control information can be, for example, a decryption key, a serial number, or an access code, or possibly a small part of the program code (a subroutine, object or module). By requiring these essential data, the program can only function well if both the information and the access control information are available. The startup program can read access control information from a hidden location on the record carrier, such as the license code included in area 22 of the system (see Figure 2) and / or other access data files. . You can then check for the presence of the error pattern, as described above.

A schematic diagram for copy-protecting a record carrier is shown in Figure 6. The following text ensures that the invention is applied under the control of a licensor. In the first step 61 the editor creates a new title 71 of the software. The publisher does not need to make any changes to the software in order to use the invention. The system can first be fully developed and tested. In a second step 62 the publisher encrypts the main executable file (for example, for a DOS or Windows system) to create a set of encrypted files 72. Using a utility provided by the licensor the publisher encrypts using a license number as the key cryptography. The licensor provides unique license numbers for each product produced. As an example, if the software title to be protected has a program named, for example, foo.exe, the encryption process encrypts it, changes its name to foo.icd, and includes a new program, provided by the licensor with renamed to foo.exe. This program is then responsible for carrying out the security checks, verifying the license number and launching the encrypted program (ffo.icd). In a third step 63, the publisher creates an image file 73, such as an ISO 9660 image, from the set of encrypted files 72. Using a CD Authorization package, the publisher creates an image file 73 of that title. of complete software on a hard disk system. In a fourth step 64, the image file 73 is modified to create the entire content 74 of the copy-protected record carrier, which has a logical map as described in Figure

two. A second utility provided by the licensor modifies the image file 73 by adding a license structure including the license number to area 22 of the system. Image file 73 is further modified to fill the size of the image to include the fill area. The total length is preferably more than 74 minutes (333,000 sectors). The license structure contains a reference to the starting and ending sectors of the fill area. In the filling area, the sectors will be good or bad, and the distribution of the sectors will be determined by a pseudo-random process, which can be originated from the license number. In a fifth step 65 the image file is processed to create a master disk 75, for example by a "Mastering House". Image file 74 is sent to the Mastering House on magnetic tape or other suitable medium. Using modified Laser Beam Recorder software in accordance with the invention, a master disk 75 is produced. The Laser Beam Recorder software makes use of the license structure to determine which sectors of the filled area are marked as bad, and which ones are good. The ratio of sectors marked to good should be set, for example, at approximately 50%. The errors are applied according to the error patterns described above in Figure 4. In a sixth step 66 copy-protected record media 76, such as CD-ROMs, are produced by duplicating the master disk 75. In this step, the error pattern is transferred to each disk.

Although the invention has been explained by an embodiment using the CD-ROM as an example that follows the CIRC error correction rules, it will be clear that other recording media, magnetic or optical tape, etc., can be used in the invention. whether such record carriers comprise information protected by previously defined error protection rules. For example, the high-density DVD disc also uses an error correction process. Although the invention has been described with reference to

ES 2 247 261 T3 cia to preferred embodiments thereof, it is to be understood that these are not limiting examples. Thus, various modifications may be apparent to those skilled in the art, without exceeding the scope of the invention, as defined in the Claims. For example, applying error patterns to error-protected data transmitted over a network, such as the Internet, can provide access control in accordance with the invention. Furthermore, the invention resides in each and every new feature or feature combination.

List of related documents (D1) EP-0545472 (PHN 13922)

Closed information system with physical copy protection (D2) GB 2076589 (PHQ 80009)

The CIRC for error detection and correction (D3) WO 95/03655

CD PROM cryptography system.