Patent Grant
7177430
This invention relates generally to a digital data manipulation. More specifically, the invention relates to a technique for progressive degradation of a digital audio signal in reproduction.
The combination of computer technology and digital audio recording technology has made it possible to reproduce or create copies of audio recordings with the same or almost the same quality as the original recordings on media such as CD (compact disc) or DAT (digital audio tape). The ability to recreate digital audio recordings, for example, music, without losing its original quality, can be a threat to copyright owners of the music and can seriously affect revenue generation for the music industry.
Thus, the Audio Home Recording Act (AHRA) of 1992 (Title 17, Chapter 10, of the United States Code) was enacted to prevent unrestrained and uncontrolled reproduction of music recordings published by the music industry. The AHRA includes provisions to place certain built-in limitations in digital music that are designed to prevent unauthorized reproduction of music recordings. For example, in order to avoid violation of the AHRA, an audio device must comply with the serial copy management system (SCMS), which requires the configuration that uses a single protection bit. Under the SCMS, the single bit protection is reset in the original master copy, and is set in all other second or higher generation copy versions in order to prevent unauthorized reproductions off the second or higher generation copy versions.
The existing techniques that implement the SCMS often creates ineffective and inefficient solutions to prevent unauthorized recordings. In particular, it is overly simple and inflexible due to the single bit protection scheme. The use of a single bit protection is also often vulnerable to attempts to circumvent it, and is often deciphered relatively easily because modern computers are powerful and well-equipped to break the code and decipher single bit protection schemes. Once a single bit protection scheme is known, it is relatively easy to neutralize the whole protection scheme and make reproductions of the original music recordings at will.
Although there are more sophisticated encryption schemes to prevent unauthorized reproduction such as cryptographic watermarking and private copy protection scheme, they are also vulnerable to circumvention and may be deciphered by using personal computers whose processing power continues to escalate with time. In addition, cryptographic encryption schemes can be prohibitively expensive if the encryption is strong enough to prevent all unauthorized decryption attempts.
The AHRA exempts analog recordings because analog recordings will inherently degrade each successive reproductions or copying. However, the analog exemption does not apply to popular audio recording media, such as CD, because they use a digital format for recordings, not the analog format.
In view of the foregoing, it is highly desirable to provide a flexible, cost-efficient copy protection mechanism in compliance with the Audio Home Recording Act. It is also desirable to provide a mechanism that would be applicable to popular digital recording media, such as CD and DAT, that would prevent unauthorized reproduction while being strong enough to prevent circumvention or deciphering.
The present invention provides a system and method for introducing white noise into a digital audio data stream so that there is progressive quality degradation after each reproduction of the audio signals in a fashion analogous to analog audio reproduction. The invention may include a white noise generator, a microprocessor, and a digital entroping unit coupled to the white noise generator. In one aspect of the invention, the digital entroping unit is embedded in the microprocessor as part of its firmware program.
In a preferred embodiment, the invention uses a hardware based random number generator as the white noise generator. Typically, an incoming audio signal comprises a plurality of digital data sampled at different time points, and the random number generator generates a random random number comprising sixteen (16) bits for each audio data sample. A masking or scaling data is applied to each random number generated in order to select a predetermined number of bits from the random number. The predetermined number of bits (the strength of the white noise) is then added to the original audio date to create degraded audio data. The predetermined number of bits is determined based on the level of desired degradation of the audio data. In particular, various desired degradation levels (of hisses and pops) can be achieved by using different predetermined algorithms for modifying the masking or scaling data. Because of the randomness of the white noise being added, it is nearly impossible to recover the original audio signal from the degraded audio output.
In one aspect of the invention, the predetermined number of bits of the random number is selected so that there is approximately 3 dB degradation of the input digital audio signal as a result of digital entroping. At 3 dB output noise, there are noticeable differences or degradations from the original audio signal. Repeated reproductions or copying of the audio data with 3 dB degradation will result in cumulative degradation in the quality of the audio data as the digital entroping process repeatedly injects additional white noise into the next generation copies. Eventually, the audio data will be degraded to such an extent that after multiple generation of reproductions, it becomes unrecognizable or unenjoyable to the listener. Thus, an audio quality degradation similar to analog reproduction can be achieved for digital audio data.
In another aspect of the invention, the randomness of the digital entroping process is in timing of adding a noise number to the input audio data. The invention determines whether a noise number is to be added to the input audio data or not. The noise number is then added to certain audio data randomly selected based upon an algorithm using random numbers.
In accordance with the invention, multiple attempts to reproduce or copy the multi-generation audio signal will result in further degradation in the quality of the audio signal as the digital entroping process is applied multiple times, thereby increasing the randomness or entropy of the audio data. After multiple reproductions the audio data will be so degraded that it becomes unrecognizable or unacceptable to the listener. An audio quality degradation similar to analog reproduction can thus be achieved.
For a better understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings, in which:
Like reference numerals refer to corresponding parts throughout the drawings.
The invention is particularly applicable to on-line and it is in this context that the invention will be described. It will be appreciated, however, that the digital entroping unit in accordance with the invention has greater utility, such as to other types of digital audio applications. To understand the digital entroping in accordance with the invention, the basic process of the digital entroping and its operations will be described.
Overview—Digital Entroping
The invention introduces white noise into a digital audio signal in order to degrade the audio signal. White noise is a random noise whose noise spectral level (noise-power density) is uniform over a wide frequency range. Once the white noise is added to the input audio signal, it is nearly impossible to reverse the process and recover the original input audio signal because of the randomness of the white noise. The degree of degradation can be varied by changing the amount of the white noise that is added to the input audio signal. Because the ideal white noise, which requires an average power of infinity, does not exist in real life, the invention approximates the ideal white noise by using a random number (RND) generator. In a preferred embodiment, the invention may use a hardware based random number (RND) generator. For example, the random number generator may be constructed using linear feedback shift registers (LFSRs) on a PLL/VCO (phase locked loop/voltage controlled oscillator) based clock generator. In a preferred embodiment, the initial condition of the LFSRs are randomly determined by the analog electro-mechanical-conditions of the PLL/VCO circuits that are determined by the manufacturing and environmental parameters. A digital entroping system of the invention will now be described.
Digital Entroping System
In a preferred embodiment, the random number generator 101 is implemented as described in greater detail in co-pending U.S. patent application No. 09/847,982 entitled, “Random Number Generation Method and System,” filed May 2, 2001, which is incorporated herein by reference. It will be appreciated by one skilled in the art that other suitable methods may be used to implement the random number generator 101. In yet another embodiment of the invention, software routines included in the firmware of the microprocessor 103 may be used as the random number generator 101. Also it will be apparent to one skilled in the art that any suitable psuedorandom number generator may be used in conjunction with the invention instead of a random number generator.
In one embodiment of the invention, the random number generator 101 generates sixteen (16) bit random numbers. However, it will be apparent to one skilled in the art that other sizes and structures may be used for the random number generator 101 so long as the generated random numbers do not exhibit periodicities, or a predictable algorithm.
In operation, an incoming audio signal 107 is received by the microprocessor 103. The microprocessor generates a random number and provides the random number and the input audio signal to the digital entroping unit 105 for digital entroping. After processing by the digital entroping unit 105, the microprocessor 103 produces a degraded audio output 109.
In more detail, the incoming digital audio signal 107 is typically a modulated signal. Especially for music sounds, the original audio signal is analog in nature, and it must be converted into a digital signal using well-known modulation techniques for digital processing.
The pulse code modulated (PCM) audio signal may be further encoded for compression using a codec. Codecs are devices used to encode and decode (or compress and decompress) various types of data—particularly those that would otherwise use up inordinate amounts of storage, such as audio (sound) and video files. Although a PCM technique is described in connection with a preferred embodiment, it will be appreciated by one skilled in the art that other modulation techniques can be used to digitize the analog sound signal. For example delta modulation (DM) can also be used to convert the analog sound signal into digital signal. A digital entroping process for input audio signal 107 will now be described in more detail.
Digital Entropying Process
Referring to
where >> operator denotes a double right shift bit operation, & denotes a bitwise AND operation, and && denotes a logical AND operation.
Referring to Table 1, in one embodiment of the invention, one (1) bit LSB (least significant bit) white noise is added to the digital audio input 107 by applying the mask data 0000000000000001 to each data of the digital audio signal such as D1–D9. Alternatively, two (2) bit LSB white noise may be added by using the mask data 0000000000000011. Similarly, one (1) bit and two (2) bit MSB (most significant bit) white noises can be added by using 1000000000000000 and 1100000000000000, respectively. Although data in Table 1 uses sixteen (16) bit masking data, it will be apparent to one skilled in the art that the size of the making data can be varied without departing from the scope of the invention. Thus, the level of white noise can be controlled using masking data as illustrated in Table 1. It will be appreciated by one skilled in the art that other levels and polarity of white noise can be achieved by using other suitable masking data and formats than those illustrated in Table 1.
Typically, the target degradation level is approximately 3 dB (decibel) and the appropriate masking data is selected to achieve the 3 dB degradation. The dB or decibel is a ratio of output power to input power expressed in logarithmic terms. At 3 dB, human ears can detect noticeable distortions or degradations from the original audio signal. Other levels of degradation may also be achieved by using different levels of white noise. For example, if the compliance with the AHRA requires a higher level of degradation than 3 dB, then three (3) bit or greater LSB white noise can be used as masking data in order to introduce a higher level of degradation into the input digital audio data.
In a preferred embodiment of the invention, step 303 is executed by the digital entroping unit 105. For this purpose, the digital entroping unit 105 may comprise means for selecting a predetermined number of bits of a random number generated and means for adding the predetermined number of bits of the random number to digital audio data D1–D7. Means for selecting a predetermined bits of a random number may further comprise means for formatting and applying masking/scaling data to the random number. In a preferred embodiment of the invention, means for selecting a predetermined number of bits of a random number and means for adding the predetermined number of bits are implemented by software. However, it will be apparent to one skilled in the art that means for selecting a predetermined number of bits of a random number and means for adding the predetermined number of bits may be implemented by hardware. For example, means for adding a predetermined number of bits of a random number may be implemented by an adder circuit. Also, means for applying masking data to the random number may be implemented using hardware circuitry such as a shift register and a multiplier in accordance with an alternate embodiment of the invention.
Referring back to
In step 309, the result of the addition in step 307 is produced as output audio data. In step 311, it is determined whether all input audio data have been processed. If so, the digital entroping process completes. Otherwise, the process returns to step 301 to continue the digital entroping process. The digital entroping process completes when all digital data such as D1–D9 are processed. The output signal 109 comprises a plurality of digital data processed by the digital entroping process. At completion of the process, the randomness or the entropy of the output audio signal 109 increases and the quality of the output audio signal 109 decreases accordingly from its original quality depending on the level of the white noise introduced.
In step 505, it is determined whether the adjusted noise number is to be added to the input audio data or not. In one embodiment of the invention, this decision may be made based on a random number. In one embodiment of the invention, a separate random number may be generated for this purpose. For example, an LSB or MSB of a random number may be used to make the determination of whether to add the adjusted number to the audio data. Alternatively, more complex algorithms for processing the random number may be used to make the determination as to which audio data should be added the noise number. In comparison to
If it is determined in step 505 that the adjusted noise number is to be added to the input audio data, digital audio data such as D1 is obtained by the digital entroping unit 105 in step 507. Otherwise, the process continues to step 506. The sequence of executing steps 505 and 507 may be varied without departing from the scope of the invention. For example, the steps 505 and 507 may be executed simultaneously or in any order. If step 507 precedes step 505 in an alternate embodiment of the invention, step 509 may follow step 505.
In step 509, the adjusted noise number is added to the input audio data. In step 511, the result of the addition in step 509 is produced as output audio data. In step 513, it is determined whether all input audio data have been processed. If so, the digital entroping process completes. Otherwise, the process returns to step 506 to continue the digital entroping process. At completion of the process, the randomness or the entropy of the output audio signal 109 increases and the quality of the output audio signal 109 decreases accordingly from its original quality depending on the level of the white noise introduced.
When the routine in
In principle, a degradation of an input audio data depends on the particular random number generated for the input audio data, and whether there will be a degradation or not is a random function. Because of this randomness, it is nearly impossible to determine which input data of the incoming audio signal 107 is degraded and which is not, and an effort to recover the original incoming audio signal 107 based on the degraded audio output 109 will thus fail. Even if a potential hacker is able to duplicate the exact architecture of the particular random number generator being used, it is still nearly impossible to recover the original incoming audio signal 107 from the degraded audio output 109 because the same random number generator produces a different number sequence each time it is activated and its outputs cannot be duplicated or predicted accurately unless the successive of the random number generator are also known.
In accordance with the invention, multiple attempts to reproduce or copy the multi-generation audio signal will result in further degradation in the quality of the audio signal as the digital entroping process in
The foregoing description, for the purposes of explanation, used specific nomenclature to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art that the specific details are not required in order to practice the invention. In other instances, well known circuits and devices are shown in block diagram form in order to avoid unnecessary distraction from the underlying invention. The foregoing descriptions of preferred embodiments of the present invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed. It will be appreciated by those skilled in he art that changes in this embodiment may be made without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5083224 | Hoogendoorn et al. | Jan 1992 | A |
| 5497154 | Komamura | Mar 1996 | A |
| 6145081 | Winograd et al. | Nov 2000 | A |
| 6209094 | Levine et al. | Mar 2001 | B1 |
| 6219634 | Levine | Apr 2001 | B1 |
| 6288747 | Avalos et al. | Sep 2001 | B1 |
| 6425098 | Sinquin et al. | Jul 2002 | B1 |
| 20020027994 | Katayama et al. | Mar 2002 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20030081780 A1 | May 2003 | US |
