[go: up one dir, main page]

WO2004066295A1 - Protection de contenu numerique - Google Patents

Protection de contenu numerique Download PDF

Info

Publication number
WO2004066295A1
WO2004066295A1 PCT/GB2004/000269 GB2004000269W WO2004066295A1 WO 2004066295 A1 WO2004066295 A1 WO 2004066295A1 GB 2004000269 W GB2004000269 W GB 2004000269W WO 2004066295 A1 WO2004066295 A1 WO 2004066295A1
Authority
WO
WIPO (PCT)
Prior art keywords
noise
frequency
spoiling
digital signal
signal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/GB2004/000269
Other languages
English (en)
Inventor
Winston D. Keech
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
DarkNoise Technologies Ltd
Original Assignee
DarkNoise Technologies Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from GB0301403A external-priority patent/GB0301403D0/en
Priority claimed from GB0313807A external-priority patent/GB0313807D0/en
Priority claimed from US10/460,851 external-priority patent/US20040252615A1/en
Application filed by DarkNoise Technologies Ltd filed Critical DarkNoise Technologies Ltd
Publication of WO2004066295A1 publication Critical patent/WO2004066295A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/00086—Circuits for prevention of unauthorised reproduction or copying, e.g. piracy
    • G11B20/00731—Circuits for prevention of unauthorised reproduction or copying, e.g. piracy involving a digital rights management system for enforcing a usage restriction
    • G11B20/00818—Circuits for prevention of unauthorised reproduction or copying, e.g. piracy involving a digital rights management system for enforcing a usage restriction wherein the usage restriction limits the signal quality, e.g. by low-pass filtering of audio signals or by reducing the resolution of video signals
    • G11B20/00826—Circuits for prevention of unauthorised reproduction or copying, e.g. piracy involving a digital rights management system for enforcing a usage restriction wherein the usage restriction limits the signal quality, e.g. by low-pass filtering of audio signals or by reducing the resolution of video signals wherein a spoiler signal is added to degrade the signal quality
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B20/00—Signal processing not specific to the method of recording or reproducing; Circuits therefor
    • G11B20/00086—Circuits for prevention of unauthorised reproduction or copying, e.g. piracy

Definitions

  • the present invention relates to digital content protection for digital content such as, for example, content stored on CDs, DVDs, digital tapes, or in digital data files and the like.
  • a protected digital medium or protected digital content when converted into an analogue format e.g. an analogue audio output, is generally unprotected and may easily be copied by analogue-to-analogue or analogue-to-digital means such as a microphone and a PC soundcard.
  • this data (generally audio data) is then prone to being recorded or captured by an analogue device or an analogue-to-digital device such as, for example, a tape recorder or PC soundcard or the like. It would be advantageous to prevent, or at least reduce or dissuade, such copying to safeguard revenue streams for owners of recorded audio or other copyright material. This also applies to any other medium such as DNDs and the like.
  • a first aspect of the present invention provides a method of creating a digital signal representing an audio signal that is resistant to copying, the method comprising: receiving a source digital signal having a predetermined sampling frequency; generating a spoiling noise signal having a centre frequency below half said predetermined sampling frequency, said spoiling noise signal being imperceptible to a human with normal hearing if reproduced faithfully and having a frequency and amplitude such that when transformed with a non-linear transformation a human- perceptible noise is created; and generating an output digital signal that is resistant to copying by combining said source digital signal and said spoiling noise signal.
  • the non-linearities of the recording system are exploited to move energy at predetermined frequencies to create spoiling noise.
  • the spoiling noise is arranged to fall within the audible frequency range at a level that is perceivable by humans.
  • the present invention provides spoiling noise components that are transferred by any non-linearities of a playback or recording device used for re-recording, especially any analogue stage in that process, to create unacceptable noise.
  • the non- linearities are preferably characteristics of components of the device, e.g. amplifiers or transducers, but effects with other noise signals generated by the device may also be exploited.
  • noise components that interact with commonly used tape bias frequencies may be employed.
  • a particularly advantageous form of noise signal comprises an ultrasonic or non-audible carrier frequency and modulated by a modulation envelope forming substantially symmetric packets.
  • the packets have a size and shape determined so that a component of the reproducing or rerecording device breaks the symmetry of the packet and creates noise in the audible region.
  • the component may be an automatic level or gain control or a feature of a digital coding scheme that aims to reduce the data volume.
  • This bias noise may also be encoded in such a way as to introduce packets of ultrasound and/or near-audible sound. That is, ultrasound tones (or near audible tones) in packets of calculated shape and duration may be placed amongst the bias and ultrasound/near audible balanced spike noise.
  • the packets are of correct geometry, then no or very low or practically imperceptible audible noise will be added due to their presence.
  • the packet durations may represent a frequency that is easily audible (e.g. 2000 Hz).
  • the ultrasound (or near audible) tones produce no or very little audible component during playback
  • their modulation may affect the automatic level control of a recording device therefore modulating the recording level of the copy at an audible frequency.
  • automatic level controls are generally characterised by a fast attack, slow relax profile. Therefore, a detected ultrasound/near audible sound packet of appropriate duration (e.g.
  • the packet repetition rate may also contribute to this process.
  • suitable packet geometries may be devised (e.g. a rampdown, rampup. rampdown profile). In this way, the rate of level change may be controlled so as to reduce any possible audible spike noise and the rise/fall rate ratio may be controlled for the purpose of optimising noise when psychoacoustic compression is encountered, e.g.
  • MP3 M-PEG layer 3 compression
  • the packet duration may be devised so that it corresponds to a frequency component within the important or preferred psycho-acoustic range (e.g. 2 kHz) for the codec, or beyond that region should it be necessary to include or exclude the noise packet for noise component or data encoding purposes.
  • the ramp rate ratio that is the ramp rate divided by the packet width
  • the packet width is preferably in the range of 100 samples to 10,000 samples (at CD rate) with a 1 to 5 ramp rate ratio preferred at the lower end of that range.
  • the ramp period is preferably of the order of 1 millisecond, which is within the Haan window of ear basillar membrane reflex period, typically 5 milliseconds.
  • the packet length after the end of ramp up period should be equal to the Haan window as this is the period that will be discarded by the codec producing the greatest asymmetry in the coded packet.
  • the packet length should be longer. Multiple packets with different lengths, geometries and spacings may be used.
  • Preferred embodiments provide a method further comprising the step of encoding data within or using at least some of the noise packets.
  • the encoded data can be used for a number of purposes such as, for example, identifying the original of the base band signal.
  • embodiments provide a method in which the data provides an indication of at least one of information associated with the digital content, information associated with a publisher of the digital content, information associated with a purchaser of the digital content.
  • embodiments provide a method in which the data provides an indication of at least the composer of a digital content, a track name associated with the digital content.
  • Such noise packets may be introduced in a prescribed manner so as to represent Pulse Interval Coded information.
  • the presence (or absence) of a packet among the other encoded noise may be detected by either recognition of its known parameter range (frequency of noise contained, ramp rate, profile, duration, level limits etc.) or the change of encoded noise parameters (absence/presence of frequency/frequencies noise level, sound level etc.) such as may be routinely performed by realtime frequency analysis using a frequency /spectrum analyser or Fourier-based program performing that function.
  • the spacing between each subsequent packet would represent a value of 1 to 16 (or 0 to 15) so as to represent one half of one byte of information.
  • two subsequent packets would represent one byte of information which could in turn represent one ASCII or Unicode character.
  • several of these characters could make a repeated character string such as serial number, purchaser's name etc. In this way, a short excerpt could be analysed so as to retrieve the encoded information/character string even if the packets have been degraded (or removed) by compression codes.
  • noise may be introduced by carefully prescribed modulation geometry of various noise components so that any noise reproduced in the relevant playback system is below the threshold of human hearing or so that noise of any frequency that is added in any other way may be reproduced by the playback system (before copying) at a level that is below the threshold of human hearing.
  • ultrasonic noise is defined as noise having a frequency just beyond an upper threshold of the human range of hearing. In general, this threshold is taken to be 20 kHz, although it may be defined as 25kHz, 30kHz, 35kHz or even 40kHz. Each or any one of these thresholds may be taken as the human threshold of hearing when determining the threshold of human hearing in the context of the present invention.
  • Near audible noises may be defined in a similar manner, although in some instances near-audible noise may have a slightly lower frequency that may just be heard by some human listeners with particularly sensitive high-frequency hearing. Particularly preferred is near-audible noise that falls within a frequency range perceptible to a person with normal hearing, but at a sensitivity that makes its perception practically inaudible.
  • the spoiling noise signal includes one or more components with centre frequencies in the range of from 17kHz to half the sampling frequency. Where there are multiple components, the spacing should preferably be in a range of from 1kHz to 5kHz.
  • the centre frequency determines the frequency of the spoiling noise due to intermodulation effects, such as overtone alaising effects at frequencies equal to the difference between the frequency of the spoiling noise component and half the sampling rate. There is also the possibility of beat frequencies between the spoiling noise components.
  • a particularly preferred embodiment has at least one component with a centre frequency in the range from 17 to 19kHz, at least one component in the range from 19 to 21kHz and at least one component in the range from 21kHz to half the sampling rate.
  • ultrasound and/or near audible sine wave noise bands of a predetermined level may be encoded to provide biasing or beating frequencies for specific recording devices or formats.
  • 17.2 kHz noise bands will provide a suitable bias/overtone for tape recorders using metal, chromium or
  • central frequencies may be moved or rotated through as much as 400 Hz so as to modulate the beat produced, increasing the perceived harshness of the produced beat noise and increasing its complexity within the copy, making it more difficult to remove.
  • the invention is applied to a digital file but exploits effects in the analogue domain (whether actually carried out in the analogue domain or normal in digital processing).
  • the invention can be employed at any stage of the distribution of digital media - by physical media, by transfer of digital files, by streaming audio, by broadcast, live performances, or in public showings of recorded media.
  • the noise signals can be added at the time of creation of the source signal, e.g. during editing, mastering or at the point of distribution.
  • the noise bias band and packet encoding may be generated by a program which may be given its encoding parameters in the form of a key.
  • This key may be interpreted to provide the bias band frequencies, frequency rotation range, frequency rotation rate, bias noise start point within the output file, noise packet width, noise packet rampup/down rates, packet encoding start point and interval range, plus an interval sequence representing data or a function, as well as other appropriate information.
  • a second program may reproduce the noise encoding so as to remove it when necessary.
  • a digital audio file of the conjugate noise may be produced by either program for digital mixing and therefore cancellation of the encoded noise.
  • Such keys or anti-noise files may be issued as part of a copyright control system.
  • a programmed sequence or random sequence realtime encoding program may be used to mix such a noise file at a concert, auditorium, cinema or such public address system so as to prevent quality copying of concert or cinema, etc. performances (for example by way of portable DAT recorders or video cameras) so as to help protect copyright material.
  • Re-sampling, re-recording and copying errors may increase as copies are made, increasing the noise audible within each subsequent copy.
  • noise levels may be low or inaudible at first copy to minidisk recorder, but may be noticeably audible on second copy.
  • a particular advantage of this technique is that the increase of errors may be tailored so as to follow first copy, but not subsequent copies. In this way, a person may be allowed to make a personal copy without charge or at nominal charge, with the copyright owner knowing that the person will not be able to make high-quality subsequent copies for further, unauthorised distribution.
  • the encoding scheme may anticipate the effect of the compression codec and compensate for this while encoding the inaudible noise, so that the correct encoding is achieved once the compression codec has been applied. In this way, the compressed file will be free of audible noise when replayed.
  • the inaudible noise encoder may also be arranged so as to perform a similar encoding process on a compressed but otherwise unmodified version of the audio file.
  • the CD-ROM data file may contain the audio file in an enciypted or key-enabled, self extracting file format.
  • the inaudible noise-encoded file may either be delivered by the self-extraction program/process in its normal state as a file that may be interpreted and played as an audio file or as a stream (or series of data blocks comprising a stream) to an appropriate multimedia platform, e.g. Windows media player.
  • the file or stream may be extracted by a program or process that may provide the audio file in its original form by cancelling or removing the inaudible noise by either generating the cancellation components from a provided key or by digital mixing of an anti-noise file that corresponds to the particular audio file.
  • the noise-encoding key may be the same as the self-extraction encryption key, thereby allowing the user the benefit of having to provide only one such key to access his/her audio file.
  • the delivery method is known and fixed, it is also possible to allow the selective removal of inaudible noise components or to disable their encoding beforehand if so desired so as to optimise the level (and minimise quantization errors) for a subset of available noise components.
  • the additional level contribution of the inaudible noise may be configured to represent the maximum available range of values when added, a calculated factor of the maximum available range or an additional value proportional to the original audio level. In this way it may be increasingly difficult for the noise to be removed by sampling and comparison of quiet areas of the audio.
  • the level/ratio of inaudible noise added may also be so configured as to minimise any unwanted playback artefacts due to inadequacies of the playback platform so as to keep them at an acceptable or inaudible level.
  • Certain components of the inaudible noise may also be increased or decreased so as to minimise any unwanted artefact noise on one or more playback platforms. Further, unwanted noise produced as artefacts due to limitations of accuracy on one or more preferred playback formats/platforms may be assessed or sampled so that the encoding process may compensate for or cancel such unwanted noise. This may maintain or improve the preferred playback quality, as compared to the uncompensated version, as required.
  • Figure 1 shows a graph depicting the variation of human sensitivity to sound with both level and frequency
  • Figure 2 illustrates schematically a first principle of embodiments of the present invention, that is, translation of energy from one frequency outside of the range of human hearing to a frequency within the range of human hearing;
  • Figure 3 shows a second principle used by embodiments of the present invention, that is, constructive interference or intermodulation products that combine to produce a signal within the range of human hearing that is audible;
  • Figure 4 shows a further principle upon which embodiments of the present invention can rely, that is, translation of energy at frequencies within human hearing (but imperceptible to the human ear) to a frequency within human hearing to form a perceivable frequency component;
  • Figure 5 shows, schematically, a process for protecting digital content according to an embodiment of the present invention
  • Figure 6 shows a number of frequency spectrums for a number of such digitally encoded spoiling components
  • Figure 7 shows a number of output spectra of the audio output by a Realistic minisette having recorded thereon the signal corresponding to the input spectra shown in Figure 6.
  • Figure 8 shows a further principle or aspect of embodiments of the present invention for creating spoiling frequency components within the range of human hearing.
  • FIG. 9 shows an expanded portion of the noise packet described above with reference to Figure 8.
  • Figure 10 illustrates the effect produced by a noise packet due to an automatic level control response characteristic of an output device
  • Figure 11 illustrates the effect of such packets in conjunction, for example, with the signals represented by the fifth spectrum shown in Figure 6;
  • Figure 12 depicts a number of output spectra, that is, post re-sampling output spectra that correspond to 17.2kHz, 18.6kHz and 20.7kHz noise components shown in the first to third spectra of Figure 9;
  • Figure 13 illustrates a frequency spectrum of a signal output by a recording device following sampling of a base band signal that contains spoiling noise at 17.2kHz, 18.6kHz and 20.7kHz for a sampling rate of 22050 Hz;
  • Figure 14 depicts a further output spectrum produced from an audio signal output by an HP Pavilion N5461 laptop computer from a base band audio signal comprising 3 spoiling frequencies 17.2kHz, 18.6kHz and 20.7kHz with each of the centre frequencies having been rotated by 0.2kHz at a frequency of 2.2kHz; and
  • Figure 15 shows a third frequency spectrum 1800 for an audio signal output by the HP pavilion computer following re-sampling, at a frequency of 22050Hz, of an audio base band signal containing spoiling frequency components at 17.2kHz, 18.6kHz and 20.7kHz each of the frequencies of which were rotated by ⁇ 0.2kHz at a frequency of 2.2kHz together with the above described noise packets illustrated in and described with reference to figure 10.
  • FIG. 1 there is shown a graph 100 that illustrates the human sensitivity to sound across a predetermined frequency range and for a range of loudness.
  • the predetermined frequency range is, for the purposes of illustration, 0.02kHz to 20kHz.
  • the loudness measured in terms of decibels, is illustrated as varying from OdB to 120dB. It can be appreciated that at very low and very high frequencies, sound of any magnitude is inaudible to the human ear. There is a region 102 between these very low and high frequencies where sound is completely inaudible below an inaudible sound level threshold 104 for respective volumes.
  • a further region 106 where sound can be detected by the human ear but it is at or below the point or threshold 108 of perception. Sound that falls within a third region 110 is distinctly audible by the human ear. The audibility at approximately 2 to 2.5kHz represents a range of frequencies at which human hearing exhibits maximum sensitivity. Finally, there is a further region 112 at or in which pain may be experienced by a listener. The audible region 110 and the further region 112 are separated by a boundary or threshold 114 that also varies with frequency.
  • FIG 2 there is shown a first principle upon which embodiments of the present invention can be based.
  • the arrangement 200 of figure 2 shows a frequency spectrum 202 of an input signal.
  • the input signal is illustrated for the purpose of clarity only as comprising a single noise source 204 or spoiling frequency embedded within a base band signal (not shown).
  • a schematic representation of the frequency response 206 of a human ear It can be appreciated that the cut-off point 208 of the frequency response of the human ear will be substantially 20kHz.
  • a notional threshold 210 is also shown which corresponds to the threshold 108 as shown in Figure 1 at which sound becomes perceivable or audible. Again for the purpose of illustration only, this threshold 210 is shown as being substantially constant rather than varying with frequency as shown in Figure 1.
  • the signal having the frequency spectrum 202 shown in figure 2 is processed by a non-linear system 212 having a transfer function of H(w).
  • the non-linear system 212 represents, for example, an analogue system that is used to re-record the sound output from an audio system (not shown), which sound has been derived from a digital source.
  • the non-linear system 212 will impose non-linear effects upon the signal having the frequency spectrum 202 illustrated. It can be seen that the frequency component 204 of the spoiling noise is beyond the frequency range of human hearing.
  • the frequency spectrum 214 of the output signal comprises a frequency component 216 that is within the range of human hearing and above the perceivable threshold level 210. It can be seen that the energy of the original frequency component 204 has been translated from that frequency component 204 to the frequency component 216 so that it is perceivable by human hearing. It will be appreciated that a number of non-linear effects such as, for example, inter modulation products might give rise to such an energy transfer of energy from the first frequency component 204 to the second frequency component 216.
  • Figure 3 shows an arrangement 300 similar to that shown in figure 2 for creating spoiling noise 302 in the spectrum 304 of an output signal from a pair 306 and 308 of frequency components of the input spectrum 310 of an input signal (not shown). It can be seen from the frequency spectrum 310 of the input signal that the pair 306 and 308 of frequency components both lie below the perceivable threshold 210. Therefore, even though these frequency components are present in the audible output and are both within the audible frequency range, they do not interfere with the listeners listening pleasure.
  • the audible output is processed by a non-linear system 312, it can be appreciated that the energy formerly associated with the pair of frequency components 306 and 308 is transferred, by inter- modulation and constructive interference, it is, by non-linear effects, to a further frequency component 314 that is located both within the frequency spectrum of human hearing 206 and such that it has a magnitude that is greater than the audible or perceivable threshold 210.
  • the non-linear system 312 used to re-record the audible output creates a spoiling frequency component 314 that cannot be filtered without adversely effecting the reproduced base band signal. It can be appreciated that re-recording audio produced in accordance with embodiments of the present invention can be used to spoil or detract from a listener's listening pleasure.
  • FIG 4 there is shown an arrangement 400 for producing a spoiling frequency 402 from a pair 404 and 406 of spoiling frequencies contained within the output of a Hi-fi system (not shown) in response to playing digital media.
  • the spoiling frequencies 404 and 406 are both within the range of human hearing but they are also both below the perceivable threshold 210 and, as such, do not interfere with the listening pleasure of a listener.
  • the spoiling frequencies 404 and 406 have been selected to exploit the non-linearities of the non-linear system 408 to ensure that a relatively sizeable spoiling frequency 402 is produced, via, for example, inter-modulation products, due to the transfer of energy from the spoiling frequency components 404 and 406 to that further spoiling frequency 402. It can be appreciated that the spoiling frequency 402 has a magnitude that is greater than the perceivable human hearing perceivable threshold 210 and that it falls within the frequency spectrum 206 of human hearing.
  • a base band audio signal 802 is combined with a source of spoiling noise (not shown).
  • the spoiling noise has a frequency spectrum as shown by 804.
  • the audio base band signal and the spoiling noise are combined to produce digital content which, in the illustrated example, is shown as having been stored on a CD 806.
  • the digitised data is processed by a Hi-fi system 808, it produces, via the loud speakers 810, audio having an output spectrum 812 that comprises a faithful reproduction 814 of the original audio 802 together with a spoiling frequency component 816 that is derived from the spoiling noise shown in the spoiling noise spectrum 804. It can be appreciated that the spoiling noise or frequency component 816 falls outside of the frequency response of the human ear 818.
  • the spoiling noise frequency component 816 has at least one of its frequency and magnitude selected to create, within the frequency response of the human ear 818, an undesirable artefact or frequency component 822.
  • the tape recorder was a "Realistic minisette - 20" tape recorder which used an ordinary ferric oxide tape and the recording relied upon the built in electret condenser microphone.
  • a blank digital audio sound file was created as the encoded source or digital content.
  • the blank digital audio sound file only contained the spoiling components or sources of spoiling frequency components to allow the effect of each component to be assessed in the output spectrum of any signals derived from the illegitimate copy created using the realistic minisette.
  • Figure 6 shows a number of frequency spectrums 900 for a number of such digitally encoded spoiling components.
  • the first spectrum 902 comprises a frequency component 904 having a centre frequency of 17.2kHz and a magnitude of about -10 dB to -12 dB. In preferred embodiments, the magnitudes are at least -20 dB and greater.
  • a second spectrum 906 is shown for a signal having a frequency component 908 that is centred on 18.6kHz. The magnitude of this frequency component 908 is about -10 dB to -12 dB. In preferred embodiments, the magnitude is at least -20 dB and greater.
  • a third frequency spectrum 910 is shown as having a frequency component 912 centred on 20.7kHz and having a magnitude of -10 dB to -12 dB. In preferred embodiments, the magnitude is at least -20 dB and, preferably, greater.
  • a fourth frequency spectrum 914 is also illustrated.
  • the fourth frequency spectrum 914 comprises 3 frequency components 916 to 920 that are centred on 17.2kHz, 18.6kHz and 20.7kHz.
  • the fourth frequency spectrum 914 is a combination of the first three spectra 902, 906 and 910.
  • sixth frequency spectrum 922 that represents a signal having 3 frequency components 924 to 928 on 17.2kHz, 18.6kHz and 20.7kHz respectively.
  • the centre frequency of each frequency component is arranged to oscillate about the centre frequencies of 17.2kHz, 18.6kHz and 20.7kHz by between ⁇ 0.1 kHz and 0.4 kHz.
  • the oscillation is performed at a frequency of about 1kHz to 5kHz e.g. 2.2 kHz.
  • Figure 7 shows a number of output spectra 1000 of the audio output by the realistic minisette having recorded the signals corresponding to the input spectra shown in Figure 6.
  • the spectra shown in Figure 6 are input spectra from the perspective of the device used to re-record or re-process the signals corresponding to those spectra.
  • the spectra shown in Figure 6 are output spectra when viewed from the perspective of the legitimate reproduction device.
  • the first output spectra 1002 is the spectrum produced having recorded the first input spectrum 902 with its frequency component of 17.2kHz. It can be appreciated that much of the energy of the 17.2kHz frequency component 904 has been redistributed to relatively low frequency and low magnitude frequency components.
  • the magnitude of the spoiling components are between -45 dB and -55 dB for frequencies of approximately 50 Hz to 3.1 kHz together with a -45 dB peak at 4.6 kHz.
  • the second output spectrum 1006 corresponds to the audio signal produced in response to the tape recorder having recorded the second input spectrum 906 with its 18.6kHz frequency component 908. It can be appreciated that the spectrum 1006 comprises a frequency components having magnitudes of -45 dB to -55 dB for frequencies of 60 Hz to 2 kHz together with a relatively strong -32 dB component at approximately 3.5 kHz. It can be appreciated that a third output spectrum 1010, which corresponds to the third input spectrum 910, has a number of frequency components 1012 that are distributed over frequency range of 40 Hz to 3.5 kHz with magmtudes of between -45 dB and -55 dB together with a relatively strong -36 dB peak at approximately 1.5 kHz. It will be appreciated that this spoiling noise or these spoiling frequency components 1012 fall squarely within the range of maximum human hearing sensitivity and have a relatively large magnitude, especially as compared to the components generated by the 17.2kHz signal.
  • a fourth output spectrum 1014 is shown.
  • the fourth output spectrum 1014 is derived from the signal having the fourth spectrum 914 as shown in figure 6. It can be seen that the three frequency components 916 to 920 of that fourth spectrum 914 have produced a significant number of intermodulation products that are distributed over a frequency range of 5.4 kHz to 14.3 kHz. A number of the more significant inter modulation products 1016 are distributed over a frequency range of about 1 Hz to 5.4kHz.. It can be appreciated that the main frequency components 1018 to 1026 will represent significant spoiling noise that will adversely effect the listening pleasure of any illegitimate recording of an audio base band signal having spoiling frequency components 916 to 920.
  • a fifth frequency spectrum 1028 corresponding to the fifth input spectrum 922 is also shown in figure 7.
  • the fifth spectrum 1028 has a significant number of inter-modulation products 1030 that fall within the frequency response of the human hearing.
  • the higher frequency inter-modulation products 1032 have a much smaller magnitude as compared to corresponding inter-modulation products 1034 of the fourth output spectrum 1014. It can be seen that much of the energy associated with the high frequency inter-modulation products 1032 has been re-focussed to lower frequencies, which has resulted in much greater average magmtudes of the frequency components in these lower frequencies.
  • FIG 8 there is shown a further principle or aspect of embodiments of the present invention for creating spoiling frequency components within the range of human hearing.
  • Figure 8 shows a blank section of a digital base band signal 1102 that has been provided with noise in accordance with embodiments of the present invention.
  • the main body 1104 of the signal comprises sine wave noise band encoding which, as described above, takes the form of 3 sine wave components having respective frequencies of 17.2kHz, 18.6kHz and 20.7kHz together with an additional 22.05kHz noise component.
  • the sine wave component level is ramped down or reduced to zero as indicated by the decreasing portions 1106 of the signal 1102 prior to the introduction of a noise packet 1108 that has a very specific shape and, in preferred embodiments, comprises solely 22.05kHz noise.
  • the noise packets 1108 are regularly distributed in time throughout the base band signal 1102. After each noise packet 1108, the sine wave component level is ramped up or increased as indicated by signal portions 1110.
  • Figure 9 shows an expanded portion 1202 of the noise packet 1108 described above with reference to figure 8.
  • figures 8 and 9 show the base band signal 1102 as not containing any recorded audio information. This is in the interest of clarity and to show clearly how the noise signals are encoded. Recorded audio information is simply superimposed on and/or mixed into the noise signal shown in figures 8 and 9.
  • Figure 10 illustrates the effect produced by a noise packet due to an automatic level control response characteristic of, for example, the Realistic minisette 20 tape recorder.
  • the noise band or noise packet 1108 has a substantially symmetrical geometrical form or envelope, ramping up from 0 to a plateau 1302 and then ramping back down to 0.
  • the solid lines 1304 show the ALC level output (a modulation) of a typical ALC circuit, giving a fast attack response to a peak leading to a fast level reduction 1306, followed by a slow decay after the peak, which then leads to a gradual level increase 1308. It will be seen that, in contrast to the symmetrical geometry of the noise packet 1108 in the base band signal 1102, the ALC response is geometrically asymmetrical.
  • the symmetrical noise packets 1108 have sufficiently slow rates of change such that they do not produce any significant perceivable audio sound when reproduced from the original digital content.
  • the modulated sound packet output introduces a significant amount of noise at an audio frequency.
  • This modulation has been found to approximate to a continuous audio output level reduction if the noise packets 1108 are sufficiently frequently distributed throughout the base band signal 1102. It will be appreciated by those skilled in the art that such a significant reduction in the average audio output level due to the packets 1108 will interfere with any listening pleasure associated with playing an illegitimate copy of digital content having such a signal 1102 embedded therein or associated therewith.
  • Figure 11 shows the effect of such packets 1108 in conjunction, for example, with the signals represented by the fifth spectrum 922 shown in figure 6. It can be seen that the response, is very similar to the output spectrum 1028 with the exception of a slightly raised magnitude at about 40Hz.
  • Figure 11 shows a number of spectra 1400 for an embodiment that uses both the noise packets 1108 together with the signal having the rotating or oscillating frequency components 924 to 928 shown in the fifth spectrum 922.
  • the output spectrum produced by such an embodiment is shown the third spectrum 1402 of the figure 11.
  • the other two spectra correspond to the fourth and fifth spectra 1014 and 1028 respectively of figure 7 and are included for comparison purposes. It can be appreciated from the third output spectrum 1402 that there is a region 1404 having a significantly increased average power level.
  • Figure 12 shows a number of output spectra 1500, that is post re-sampling output spectra 1502, 1504 and 1506, that correspond to the 17.2kHz, 18.6kHz and 20.7kHz noise components shown in the first to third 902, 906 and 910 spectra of figure 6.
  • each frequency component 904, 908 and 912 clearly manifests itself as one or more sub-harmonic tones.
  • the magnitude of the sub-harmonic tones progressively increases with frequency but shows a marked roll off in amplitude at 20.7kHz.
  • the tones produced are wholly unacceptable and manifest themselves as noise that is clearly heard as sub-tones.
  • FIG 13 there is shown a frequency spectrum of a signal output by a recording device following sampling of a base band signal that contains spoiling noise at 17.2kHz, 18.6kHz and 20.7kHz for a sampling rate of 22050 Hz.
  • the 17.2kHz, 18.6kHz and 20.7kHz signal components can be clearly seen at reference numerals 1602, 1604 and 1606 respectively. It can be seen that due to at least one of inter modulation between these frequency components 1602 to 1606 and due to aliasing effects a significant number of harmonics 1608, for example, are produced well within the audio range of human hearing.
  • FIG 14 there is shown a further output spectrum 1700 produced from an audio signal output by the HP Pavilion N5461 laptop computer from a base band audio signal comprising 3 spoiling frequencies 1702, 1704 and 1706 that correspond to frequencies 17.2kHz, 18.6kHz and 20.7kHz respectively with each of the centre frequencies having been rotated by 0.2kHz at a frequency of 2.2kHz.
  • 3 spoiling frequencies 1702, 1704 and 1706 that correspond to frequencies 17.2kHz, 18.6kHz and 20.7kHz respectively with each of the centre frequencies having been rotated by 0.2kHz at a frequency of 2.2kHz.
  • FIG 15 there is shown a third frequency spectrum 1800 for an audio signal output by the HP pavilion computer following re-sampling, at a frequency of 22050Hz, of an audio base band signal containing spoiling frequency components at 17.2kHz, 18.6kHz and 20.7kHz each or this centre frequencies of which were rotated by ⁇ 0.2kHz at a frequency of 2.2kHz together with the above described noise packets illustrated in and described with reference to figure 10.
  • the 17.2kHz, 18.6kHz and 20.7kHz components 1802, 1804 and 1806 respectively have relatively large magnitudes.
  • the spoiling noise created and described with reference to the above embodiments results from at least one of the following processes 1. intermodulation products being generated from the spoiling frequency components 17.2, 18.6 and 20.7kHz signals; 2. aliasing due to an inadequate sampling frequency, that is, due to sampling at frequencies below the Nyquist frequency;
  • the frequency rotation that is, the oscillations of the centre frequency of one or more of the spoiling noise components at 17.2kHz, 18.6kHz and 20.7kHz is oscillated by ⁇ 0.1kHz and ⁇ 0.4kHz and is preferably oscillated by 0.2kHz.
  • the frequency of oscillation is preferably between 0.01 to 0.05 x the sampling frequency which, for the CD format, where the sampling frequency is 22050Hz, gives a frequency range of 0.22kHz to 11kHz.
  • the oscillation was 2.2kHz, that is, the frequency used was the sampling frequency (22050Hz) x 0.1.
  • embodiments described above have used substantially symmetrical noise packet, embodiments are not limited to such an arrangement. Embodiments can be realised that use substantially asymmetrical noise packets.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Signal Processing For Digital Recording And Reproducing (AREA)

Abstract

L'invention se rapporte à la protection de contenu numérique. Dans certains modes de réalisation, cette invention concerne un CD comportant des données qui correspondent à un contenu audio traditionnel ou des données desquelles un contenu audio peut être dérivé au moyen d'un système cible spécifique, ainsi que des données qui correspondent à un composant d'altération à partir duquel un bruit d'altération est créé lorsque les données sont traitées par un système différent du système cible.
PCT/GB2004/000269 2003-01-22 2004-01-22 Protection de contenu numerique Ceased WO2004066295A1 (fr)

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
GB0301403A GB0301403D0 (en) 2003-01-22 2003-01-22 Improvements relating to the prevention of copying of digital media
GB0301403.2 2003-01-22
US44257503P 2003-01-24 2003-01-24
US60/442,575 2003-01-24
GB0313807A GB0313807D0 (en) 2003-06-13 2003-06-13 Digital content protection
US10/460,851 2003-06-13
GB0313807.0 2003-06-13
US10/460,851 US20040252615A1 (en) 2003-06-13 2003-06-13 Digital content protection

Publications (1)

Publication Number Publication Date
WO2004066295A1 true WO2004066295A1 (fr) 2004-08-05

Family

ID=32777105

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2004/000269 Ceased WO2004066295A1 (fr) 2003-01-22 2004-01-22 Protection de contenu numerique

Country Status (1)

Country Link
WO (1) WO2004066295A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8019201B2 (en) 2002-06-28 2011-09-13 Dcs Copy Protection Limited Method and apparatus for providing a copy-protected video signal
US8032006B2 (en) 2003-06-05 2011-10-04 Dcs Copy Protection Limited Digital processing disruption systems
US8160423B2 (en) 2004-10-13 2012-04-17 Dcs Copy Protection Limited Audio copy protection system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4086634A (en) * 1975-07-30 1978-04-25 Cook Laboratories, Inc. Method and apparatus for preparing recorded program material to prevent unauthorized duplication by magnetic tape recording
JPS56124104A (en) * 1980-02-29 1981-09-29 Teac Co Secret sound recording and reproducing device
GB2164481A (en) * 1984-07-23 1986-03-19 Keen Paul David Audio and video recording technique
GB2199689A (en) * 1987-01-05 1988-07-13 Thomas Claude Henry Keen Recording of an information signal
US5559460A (en) * 1994-12-22 1996-09-24 International Business Machines Corporation Peak detection circuit for suppressing magnetoresistive thermal asperity transients in a data channel
US5907656A (en) * 1994-03-29 1999-05-25 Sony Corporation Apparatus and method for reproducing video signals with varying-magnitude AGC signals
WO1999057723A2 (fr) * 1998-05-04 1999-11-11 Spiro J. Pandelidis High Tech Applications Procedes et dispositifs antipiratage pour signaux d'informations numeriques

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4086634A (en) * 1975-07-30 1978-04-25 Cook Laboratories, Inc. Method and apparatus for preparing recorded program material to prevent unauthorized duplication by magnetic tape recording
JPS56124104A (en) * 1980-02-29 1981-09-29 Teac Co Secret sound recording and reproducing device
GB2164481A (en) * 1984-07-23 1986-03-19 Keen Paul David Audio and video recording technique
GB2199689A (en) * 1987-01-05 1988-07-13 Thomas Claude Henry Keen Recording of an information signal
US5907656A (en) * 1994-03-29 1999-05-25 Sony Corporation Apparatus and method for reproducing video signals with varying-magnitude AGC signals
US5559460A (en) * 1994-12-22 1996-09-24 International Business Machines Corporation Peak detection circuit for suppressing magnetoresistive thermal asperity transients in a data channel
WO1999057723A2 (fr) * 1998-05-04 1999-11-11 Spiro J. Pandelidis High Tech Applications Procedes et dispositifs antipiratage pour signaux d'informations numeriques

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
PATENT ABSTRACTS OF JAPAN vol. 0052, no. 02 (P - 095) 22 December 1981 (1981-12-22) *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8019201B2 (en) 2002-06-28 2011-09-13 Dcs Copy Protection Limited Method and apparatus for providing a copy-protected video signal
US9264657B2 (en) 2002-06-28 2016-02-16 Dcs Copy Protection Limited Method and apparatus for providing a copy-protected video signal
US8032006B2 (en) 2003-06-05 2011-10-04 Dcs Copy Protection Limited Digital processing disruption systems
US8837909B2 (en) 2003-06-05 2014-09-16 Dcs Copy Protection Ltd. Digital processing disruption systems
US8160423B2 (en) 2004-10-13 2012-04-17 Dcs Copy Protection Limited Audio copy protection system
US8639092B2 (en) 2004-10-13 2014-01-28 Dcs Copy Protection Limited Audio copy protection system

Similar Documents

Publication Publication Date Title
Steinebach et al. StirMark benchmark: audio watermarking attacks
US4644422A (en) Anti-copy system
US6175627B1 (en) Apparatus and method for embedding and extracting information in analog signals using distributed signal features
EP0822550B1 (fr) Appareil intégrant des données de droits d'auteur
JP2000163870A (ja) 音声情報制御装置および方法
US20020009000A1 (en) Adding imperceptible noise to audio and other types of signals to cause significant degradation when compressed and decompressed
WO1997037448A2 (fr) Procede et appareil pour coder et decoder des donnees supplementaires sous forme de signaux analogiques
JP2002513982A (ja) デジタル情報信号用複製防止方法およびデバイス
JP2005502920A (ja) Ｄｓｄ信号のための堅牢なウォーターマーク
Kefauver et al. Fundamentals of digital audio
WO2004066295A1 (fr) Protection de contenu numerique
JP4515643B2 (ja) 磁気テープレコーダでのコピーを防止するためのシステム
US20040252615A1 (en) Digital content protection
JP2008516374A (ja) 音声コピー防止装置
AU2003250397B2 (en) Access controlled optical disc and method therefor
JP2008516374A5 (fr)
US20030015085A1 (en) Musical-file-processing apparatus, musical-file-processing method and musical-file-processing method program
JP4311541B2 (ja) オーディオ信号圧縮装置
EP1640985B1 (fr) Filigrane audio
US8195317B2 (en) Data reproduction apparatus and data reproduction method
WO2001088915A1 (fr) Addition d'un bruit imperceptible a des signaux audio et a d'autres types de signaux visant a provoquer une degradation significative de ces signaux lorsqu'ils sont comprimes et decomprimes
US20070192091A1 (en) Protection against sound piracy by microphones
JP3331871B2 (ja) ノイズ低減方法、ノイズ低減装置及び記録媒体
JP2000182320A (ja) 圧縮符号化防止方法
JPH11219172A (ja) 楽音波形データの識別情報埋込み方法、作成方法および記録媒体

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): BW GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase