Patent Grant
8520843
1. Field of the Invention
The present invention relates to encrypting discrete signals, such as to encrypting voice information, and to decrypting accordingly.
2. Description of Prior Art
In the use, transmission, administration and archiving of audio material it is often desirable to protect the respective contents from unauthorized access. In particular in the field of voice recording there is a necessity to prevent unauthorized playback or clandestine interception during the transmission. At the same time, however, the data format used is to remain valid so that the appliances used for playback do not transition to error conditions even in the event of unauthorized access. This applies particularly to compressing data formats such as data formats in accordance with standards MPEG2 Layer 3 and MPEG2/4 AAC (AAC=Advanced Audio Coding).
In audio applications there is the added aspect that the encrypted signals must not do any damage to the interception equipment in the event of intercepting without decryption. The encrypted signals should therefore be encrypted such that they do not create any crackling or rustling or other extreme dynamics discontinuity when played back without being decrypted. Whereas when encrypting music data it is often sufficient to limit the quality of unauthorized playback to a large extent, it is requested in particular, for voice contents, that in the event of unauthorized use, the playback quality of the data encrypted should no longer allow the voice information, which may be, e.g., interviews, reports etc., to be intelligible.
Patent application WO 99/51279 entitled “Vorrichtung und Verfahren zum Erzeugen eines verschlüsselten Audio-und/oder Videostroms” (apparatus and method for creating an encrypted audio and/or video stream) whose applicant is also Fraunhofer-Gesellschaft, describes a method of scrambling encoded audio data based on permuting lines in a frequency range. This method allows making music signals largely unrecognizable. With voice contents, however, the exact spectral composition of the signal is of little importance for its intelligibility, so that the content of the spoken words and/or the voice information remains intelligible even though the voice of a speaker is alienated to a large extent.
It is the object of the present invention to provide a method and an apparatus for encrypting a discrete signal, and a method and an apparatus for decrypting accordingly, so that the encryption is as safe as possible, on the one hand, and does not give rise to errors in the event of unauthorized processing and is compatible with previous codings, on the other hand.
In accordance with a first aspect, the invention provides a method for encrypting a discrete signal consisting of successive samples, the method including the following steps: subdividing the successive samples into successive time blocks; coding the successive time blocks into encoded data blocks having a predetermined order; and altering the predetermined order of the encoded data blocks in accordance with a predetermined interchange specification.
In accordance with a second aspect, the invention provides an apparatus for encrypting a discrete signal consisting of successive samples, having: means for subdividing the successive samples into successive time blocks; means for coding the successive time blocks into encoded data blocks having a predetermined order; and means for altering the predetermined order of the encoded data blocks in accordance with a predetermined interchange specification.
In accordance with a third aspect, the invention provides a method for decrypting an encrypted signal having a plurality of encoded data blocks in an order and corresponding, in an encrypted form, to a discrete signal consisting of successive samples, the method including the following steps: altering the order of the encoded data blocks in accordance with a predetermined interchange specification; decoding the encoded data blocks in the altered order into successive time blocks having a predetermined order; forming the successive samples from the successive time blocks.
In accordance with a fourth aspect, the invention provides an apparatus for decrypting an encrypted signal having a plurality of encoded data blocks in an order and corresponding, in an encrypted form, to a discrete signal consisting of successive samples, having: means for altering the order of the encoded data blocks in accordance with a predetermined interchange specification; means for decoding the encoded data blocks in the altered order into successive time blocks having a predetermined order; means for forming the successive samples from the successive time blocks.
The present invention is based on the findings that a very high level of security of the encryption may be achieved by introducing temporal discontinuity, and that the occurrence of errors in unauthorized processing of signals encoded in such a manner maybe prevented, and the compatibility with standard codings may be ensured by performing the alteration of the chronological order after coding the discrete signal, i.e. with regard to encoded data blocks into which an encoder encodes the discrete signal. In this manner it is prevented, on the one hand, that a decoder receiving the encrypted signal enters into undefined states since in the encryption the temporal discontinuity is created in units of encoded data blocks. On the other hand, it is prevented that in the interaction with any coding process desired, such as a compressing coding process, the underlying temporal assumptions, such as the temporal and spectral masking, remain valid in the event of psycho-acoustic audio processes and that the inventive encryption is thus compatible with such codings, and that the implementation of the inventive encryption is simplified.
With an encryption in accordance with the present invention, the successive samples of a discrete signal are subdivided into successive time blocks which are then coded into encoded data blocks having a predetermined order. Subsequently, the predetermined order of the encoded data blocks is altered in accordance with a predetermined interchange specification.
With performing decryptions in accordance with the present invention, the order of the encoded data blocks of an encrypted signal which corresponds, in an encrypted form, to a discrete signal consisting of successive samples, is altered in accordance with a predetermined interchange specification and/or an inverse interchange specification whereupon the encoded data blocks are decoded, in an altered order, into successive time blocks having a predetermined order. Thereby the successive samples of the discrete signal are created from the successive time blocks.
In accordance with an embodiment of the present invention, the alteration of the predetermined order of the encoded data blocks is achieved, in the encryption, by permuting a predetermined number of successive data blocks of the encoded data blocks, a permutation vector being created as the interchange specification to this end. The permutation may be performed with regard to successive groups of encoded data blocks having the same size and/or length. A different permutation vector may be created and used for each permutation group. The creation of the permutation vectors occurs in a predetermined manner in the decoding, a correct decryption being ensured by creating and using, in the decryption, appropriate inverse permutation vectors for re-permuting the groups of encoded data blocks.
Preferred embodiments of the present invention will be explained in more detail below with reference to the accompanying Fig., wherein
Before explaining the present invention in more detail below with reference to
The psycho-acoustic encoder 10 includes means 14 for dividing the successive discrete samples making up the time signal into time blocks, and means 16 for coding the time blocks into encoded data blocks.
Means 12 for altering the order include means 18 for producing a permutation vector, writing means 20, a first latch 22, a second latch 24 and read-out means 26. An input of writing means 20 is connected to an output of the psycho-acoustic encoder 10 and/or means 16 for coding, whereas two outputs of same are connected to inputs of the first and second latches 22 and 24, respectively. An output of means 18 for producing a permutation vector is connected to an input of read-out means 26 so as to output a permutation vector to same, the read-out means comprising to further inputs connected to the outputs of latches 22 and 24. Readout means 26 are connected, at an output, to an output buffer 28 in order to output encoded data blocks in an encrypted form to same.
After having described above the structure of the encryption device of
The time signal is a discrete audio signal consisting of successive samples. The psycho-acoustic encoder 10 is based, for example, on an AAC standard coding process. Means 14 subdivide the successive samples in time blocks, for example, having a number of successive samples, the number equaling a power of 2. For handling aliasing effects, provisions may be made for a subdivision in mutually overlapping time blocks, so that each sample is assigned to two time blocks as is the case, for example in AAC coding.
Means 16 for coding the time blocks into encoded data blocks receive the time blocks from means 14 in a chronological order and then encode same. A time block may be encoded either individually, or in an isolated manner, on a time-block by time-block basis, or as a function of previous and subsequent time blocks in order to allow for psycho-acoustic models, such as temporal and spectral masking, for example. Means 16 for coding the time blocks outputs the encoded data blocks to writing means 20 in a predetermined order depending on the coding process. The data blocks may all have the same length or may have different lengths, such as, for example, in the case where the data blocks have a structure in conformity with MPEG2/4 AAC.
Writing means 20 receive the encoded data blocks and write the encoded data blocks into a current one of latches 22 and 24 one after the other, the latches cooperating to act as an alternating buffer, as will be described below. The size of latches 22 and 24 is sufficient for storing N encoded data blocks, N being an integer larger than 1 (N>1). Writing means 20 describe the current one of latches 22 and 24 in the order in which the encoded data blocks are transmitted from means 16 until there are N encoded data blocks in the current one of latches 22 and 24. If the current one of latches 22 and 24 is full, i.e. comprises N stored encoded data blocks, read-out means 26 read out latch 22 or 24 having just been filled, whereas writing means 20 write the encoded data blocks from means 16 to the other one of the two latches 22 or 24 in the order of their reception.
Read-out means 26 read latch 22 or 24, whichever was the last one to be fully written to, in a different order than used for writing to same. Specifically, read-out means 26 read the respective latch 22 or 24 in a permuted order specified by a permutation vector of size N which is created and delivered by means 18 for producing a permutation vector as will be described below. By means of the permuted readout, the order of the N encoded data blocks is altered in accordance with an interchange specification established by the permutation vector. The encoded data blocks read out in the permutated order combine to form a permutation group of encoded data blocks output by read-out means 26 to the output buffer 28 connected to a computer interface (not shown), for example.
Means 18 create the N-sized permutation vector anew for each permutation group, the N-sized permutation vector establishing the interchange specification, on the basis of which the encoded data blocks of a permutation group are permuted. The creation of a permutation vector is based on N pseudorandom numbers created by the pseudorandom number generator 30. For creating each permutation vector of the length N, the pseudorandom number generator 30 successively generates N pseudorandom numbers and outputs same to the sorter 34, the counter 32 incrementing an counter value and outputting same to the re-sorter 36 in the output of each pseudorandom number, the counter 32 starting with a value of 0 in order to output a value of 1 with the first pseudorandom number. In this manner, the pseudorandom numbers output by the pseudorandom number generator 30 are numbered in parallel with their generation and/or are provided with indexes in the order of their generation. The pseudorandom numbers generated by the pseudorandom number generator 30 combine to form a random number vector, or a random number array, of N pseudorandom numbers, whereas the numbers generated by counter 32 form an index vector, or an index array, consisting of ascending numbers of 1 to N. The sorter 34 receives the random number vector and sorts same using a suitable sorting method, for example in an ascending order. Sorter 34 is coupled to re-sorter 36 to allow the re-sorter 36 to re-sort the index vector received from counter 32 in parallel with sorting the random number vector. The re-sorted, or permuted, index array generated by re-sorter 36 represents the interchange specification for the N encoded data blocks which are next to be read by the read-out means, and will be output as a permutation vector to read-out means 26 by re-sorter 36, the read-out means using same, as has been described above, for defining the read-out order with regard to the respective latch 22 or 24.
Once read-out means 26 have read the N encoded data blocks from the one latch 22 or 24 and once, at the same time, writing means have filled the other latch with the next-in-line N encoded data blocks from encoder 10, writing means 20 and read-out means 26 change over to the other latch 22 or 24, respectively, the read-out process being performed with regard to the new encoded data blocks written to the alternating buffer, which data blocks are subsequently output to the output buffer in a permuted order. On the whole, an encrypted signal of encoded data blocks in a permuted order is yielded at the input and output of the output buffer, the signal preventing, in the event of unauthorized processing without decryption and in the case of voice, the voice information from being intelligible, as will be described in more detail with reference to
A decryption device in accordance with an embodiment of the present invention will be explained below with reference to
The device of
Means 38 comprise an arrangement similar to that of means 12 of the encryption device of
Decoder 40 includes means 52 for decoding the encoded data blocks output by read-out means 48 as well as means 54 downstream of means 52, for forming the successive samples, means 54 outputting the time signal to a digital-to-analog converter (not shown) or the like, for example.
After having described above the structure of the decryption device of
Writing means 42 receive the encoded data blocks present in an encrypted form, and output same, in the order in which they have been transmitted, to a current one of latches 44 and 46, which co-operate as an alternating buffer as in the encryption device of
Means 50 create the inverse permutation vectors per read-out operation by means of a same arrangement of means, for example, as is shown for means 18 in
The encoded N data blocks read out by read-out means 48 in a permuted order are fed to means 52 for decoding the encoded data blocks, the latter now being present in the predetermined order necessary for decoding the encoded data blocks in accordance with the coding process underlying the decoder 44, in order to obtain a correct time signal.
Once read-out means 48 have read out the respective latch, and once writing means 42 have completely filled the other latch, the read-out means read out the latch that has just been filled by writing means 42, while writing means 42 write to the latch that has just been read out by read-out means 48.
Means 52 decode the encoded data blocks and output time blocks in a predetermined order. Means 54 receive the time blocks and form the successive samples from same, of which samples the time signal consists, and output same to an analog-to-digital converter (not shown), for example.
After embodiments of encryption and/or decryption devices have been described above, an explicit embodiment will be described below with reference to
Samples of the time and/or audio signals, time blocks and/or data blocks are represented by means of rectangles in
64 shows a sequence of time blocks 66 as are created by means 14 of
68 shows a sequence of encoded data blocks A-N present in the predetermined order, as are output by means 16 of
72 shows a state such as results for the successive encoded data blocks 70 during the encryption with the encryption device of
74 represents the state obtained after five further encoded data blocks. The 5 further encoded data blocks F-J have been written to latch 2, while the encoded data blocks stored in latch 1 have been read out into the output buffer. For reading out the encoded data blocks stored in latch 1, the permutation vector as is indicated at 76, i.e. (4,3,5,2,1), has been used. In other words, permutation vector 76 assigns each encoded data block in latch 1 a number between 1 and 5 and/or N indicating the read-out order and/or the position at which that particular encoded data block is to be written to the output buffer, so that the encoded data blocks A-E are present in the order EDBAC in the output buffer.
78 represents the state obtained after 5 more encoded data blocks. As may be seen, the 5 subsequent encoded data blocks K-O have again been written to latch 1, while in the meantime latch 2 has been read out, by means of a permutation vector 80 (5,1,3,2,4), to the output buffer, where the encoded data blocks are yielded in the order GIHJE.
82 represents the flow and/or the sequence of encoded data blocks in an encrypted form, as are input into and/or output from output buffer 28. As may be seen, the encoded data blocks have been scrambled as compared to the predetermined order in which they are usually output due to the coding underlying the encoder 10, which is why, in the event that the audio data are carriers of voice information, this voice information is unintelligible in the event of decoding without decryption. Nevertheless it is prevented, in decoding without decrypting, that the decoder gets into invalid states, since the temporal discontinuity is defined in units of encoded data blocks.
If the coding underlying the psycho-acoustic decoder is in conformity with the AAC standard, for example, no crackling will occur at the block boundaries if the signal encrypted is decoded by a standard decoder, but rather is the temporal discontinuity expressed as an occurrence of aliasing portions due to the interchanged frames and/or data blocks, since the data blocks are retransformed into the time domain by means of the inverse modified discrete cosine transform (IMDCT), and since there is no more aliasing elimination at the overlap areas of the transformation windows.
If signal 82 is decrypted by a decoder and/or a decryption device in accordance with
At 84,
88 depicts the state such as is obtained after five more encoded data blocks FGHIJ. As may be seen, the next five encoded data blocks have been written to latch 2, while the encoded data blocks EDBAC are read out from latch 1 by means of an inverse permutation vector 90 in order to be transmitted to decoder 40 in the order ABCDE, the inverse permutation vector resulting from permutation vector 76 of
92 depicts the state such as results after reading out five more encoded data blocks from the flow of encoded data blocks 84. As can be seen, latch 1 has again been filled with the subsequent encoded data blocks K-O, while the encoded data blocks GHIJF have been read out in latch 2 and have been output to the decoder in a permuted order and/or inversely permuted order FGHIJ. The re-permutation is based on the inverse permutation vector 94 resulting from permutation vector 80 of
96 finally depicts the flow of successive encoded data blocks as is fed to the decoder. As can be seen, the order in which the encoded data blocks have been output from the encoder of the encryption device, i.e. ABCDEFGHIJKLMN . . . , is restituted, so that decoding may be performed according to standards.
The description given above with reference to
With regard to the above-described embodiments, it shall be pointed out in particular that the unintelligibility of the voice of the encrypted signal may be improved by the psycho-acoustic encoder 10 and/or means performing, between the encoder and the means, a frequency domain scrambling in accordance with the patent application WO 99/51279, mentioned in the introduction of the description, in order to alter the order.
After the present invention has been described above with reference to specific embodiments, it shall be pointed out that the present invention may be implemented both in hardware, such as in a form of an ASIC, an integrated circuit or the like, as well in software, such as in a software that may be run on a PC. In addition it shall be pointed out that, although the present invention has been described above with regard to the encryption of audio data and/or voice signals, the present invention may be generally applied to all fields where discrete signals are used and where, under certain circumstances, an coding of same is performed, such as in image and video processing or in data transmission in general. Accordingly, the coding preceding the creation of the temporal discontinuity in the encryption is not limited to psycho-acoustic coding. A JPEG coding with image or video data is also possible, for example. The present invention may generally be implemented with any coding process subdividing successive discrete samples into time blocks and coding same into encoded data blocks, or frames, or directly coding time blocks which already exist.
It shall additionally be pointed out that the exact implementation of the means for producing a permutation vector and of the means for producing the order of the encoded data blocks may vary, particularly, for example, with regard to the length of the interchange group N or the number and size of the latches used.
In addition, the means for producing a permutation vector may be implemented differently than described above. For example, the permutation vector could be the same for all interchange groups, in which case the inverse permutation vector would also be specified. It shall generally be pointed out that it is possible to depart from the principle of the permutation of successive interchange groups, which principle has been used in the previous embodiments, and that the variation in the order may also be carried out in other ways, such as by altering the order with regard to all encoded data blocks, in which case a latching of all encoded data blocks would be required to occur before altering the order in the encryption, and storing of all of the encoded data blocks would be required to occur before altering the order in the decryption.
While this invention has been described in terms of several preferred embodiments, there are alterations, permutations, and equivalents which fall within the scope of this invention. It should also be noted that there are many alternative ways of implementing the methods and compositions of the present invention. It is therefore intended that the following appended claims be interpreted as including all such alterations, permutations, and equivalents as fall within the true spirit and scope of the present invention.
| Number | Date | Country | Kind |
|---|---|---|---|
| 101 38 650 | Aug 2001 | DE | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP02/08661 | 8/2/2002 | WO | 00 | 2/5/2004 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO03/015328 | 2/20/2003 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 3773977 | Guanella | Nov 1973 | A |
| 3970790 | Guanella | Jul 1976 | A |
| 3978288 | Bruckner et al. | Aug 1976 | A |
| 4232194 | Adams | Nov 1980 | A |
| 4278840 | Morgan et al. | Jul 1981 | A |
| 4393276 | Steele | Jul 1983 | A |
| 4443660 | DeLong | Apr 1984 | A |
| 4600941 | Sakamoto et al. | Jul 1986 | A |
| 4612414 | Juang | Sep 1986 | A |
| 4747137 | Matsunaga | May 1988 | A |
| 4773092 | Huang | Sep 1988 | A |
| 5095525 | Almgren et al. | Mar 1992 | A |
| 5303302 | Burrows | Apr 1994 | A |
| 5339108 | Coleman et al. | Aug 1994 | A |
| 5436940 | Nguyen | Jul 1995 | A |
| 5717819 | Emeott et al. | Feb 1998 | A |
| 5799088 | Raike | Aug 1998 | A |
| 5825425 | Kazui et al. | Oct 1998 | A |
| 5991308 | Fuhrmann et al. | Nov 1999 | A |
| 6081784 | Tsutsui | Jun 2000 | A |
| 6084966 | Maebara et al. | Jul 2000 | A |
| 6134631 | Jennings, III | Oct 2000 | A |
| 6163576 | Lempel | Dec 2000 | A |
| 6226608 | Fielder et al. | May 2001 | B1 |
| 6266418 | Carter et al. | Jul 2001 | B1 |
| 6278783 | Kocher et al. | Aug 2001 | B1 |
| 6301268 | Laroia et al. | Oct 2001 | B1 |
| 6307940 | Yamamoto et al. | Oct 2001 | B1 |
| 6356545 | Vargo et al. | Mar 2002 | B1 |
| 6430222 | Okada | Aug 2002 | B1 |
| 6445797 | McGough | Sep 2002 | B1 |
| 6507672 | Watkins et al. | Jan 2003 | B1 |
| 6512758 | Sato et al. | Jan 2003 | B1 |
| 6642885 | Lobo | Nov 2003 | B2 |
| 6643729 | Sasaki et al. | Nov 2003 | B2 |
| 6650659 | Hamada et al. | Nov 2003 | B1 |
| 6810273 | Mattila et al. | Oct 2004 | B1 |
| 6963860 | Tsutsui et al. | Nov 2005 | B1 |
| 6985722 | Snelgrove et al. | Jan 2006 | B1 |
| 7016493 | Henson et al. | Mar 2006 | B2 |
| 7047196 | Calderone et al. | May 2006 | B2 |
| 7047222 | Bush | May 2006 | B1 |
| 7050495 | Mihara et al. | May 2006 | B2 |
| 7171246 | Mattila et al. | Jan 2007 | B2 |
| 7289951 | Ojanperaa | Oct 2007 | B1 |
| 7391714 | Blasco Claret et al. | Jun 2008 | B2 |
| 7675972 | Laksono et al. | Mar 2010 | B1 |
| 20010009604 | Ando et al. | Jul 2001 | A1 |
| 20010009605 | Ando et al. | Jul 2001 | A1 |
| 20010010755 | Ando et al. | Aug 2001 | A1 |
| 20010012443 | Ando et al. | Aug 2001 | A1 |
| 20010014201 | Ando et al. | Aug 2001 | A1 |
| 20010053220 | Kocher et al. | Dec 2001 | A1 |
| 20020002675 | Bush | Jan 2002 | A1 |
| 20020009000 | Goldberg et al. | Jan 2002 | A1 |
| 20020037103 | Hong et al. | Mar 2002 | A1 |
| 20020076049 | Boykin et al. | Jun 2002 | A1 |
| 20020122484 | Mihara et al. | Sep 2002 | A1 |
| 20020173967 | Law et al. | Nov 2002 | A1 |
| 20030095498 | Sato et al. | May 2003 | A1 |
| 20040030364 | Bange et al. | Feb 2004 | A1 |
| 20050027520 | Mattila et al. | Feb 2005 | A1 |
| 20060045364 | Mihara et al. | Mar 2006 | A1 |
| 20060142821 | Bange et al. | Jun 2006 | A1 |
| 20070211786 | Shattil | Sep 2007 | A1 |
| Number | Date | Country |
|---|---|---|
| 558993 | Mar 1973 | CH |
| 633 703 | Jan 1995 | EP |
| 920 209 | Jun 1999 | EP |
| 1458698 | Dec 1976 | GB |
| WO 9951279 | Oct 1999 | WO |
| WO 0051279 | Aug 2000 | WO |
| Entry |
|---|
| Franaszek, P. A.; Digital Speech Scrambler; IBM. |
| Goldburg, B. et al.; A Secure Analog Speech Scrambler Using the Discrete Cosine Transform. |
| Franaszek, P. Digital Speech Scrambler. IBM Technical Disclosure Bulletin. vol. 23. No. 1. Jun. 1980. |
| Number | Date | Country | |
|---|---|---|---|
| 20040196971 A1 | Oct 2004 | US |
