FI105736B - Method, arrangement and system for encryption of information sent as a packet-shaped digital signal - Google Patents

Description

105736

The present invention relates to a method, apparatus and system for encrypting information transmitted as a packet digital signal.

Digital signal transmission in television broadcasting is becoming more common and is already partly responsible for analog signal transmission. The advantage of digital signal transfer over analog transmission is e.g. the fact that signals from several different sources can be transmitted using the same transmission channel. This can be implemented e.g. such that the signals are divided into packets to be transmitted to the transport channel by some method known per se. The packets formed by the various sources are transmitted interleaved in time, so that preferably the packet has a header field which indicates which source the packet belongs to and a data field containing the actual information to be transmitted ._________

Before analog signals can be transmitted in packet form, the analog signals at the transmitting end must first be converted to digital, whereby the generated digital-20 signal can be arranged in packets for transmission to the transmission channel. At the receiving end, the analog signals converted to digital signals are converted back to analog.

Digital signal encryption differs from analog signal encryption. A variety of algorithms known per se can be used to encrypt the digital signal, and usually only the encryption keys known to the sender and the recipient (s). The encryption algorithm and / or key may be changed during transmission. It is also possible to use combinations of two keys, one of which is preferably a public key, which may be generally known, and the other is a secret key, respectively. In most applications, using the same encryption algorithm and one or more encryption keys is sufficient, so that the encryption key is only known to the sender and the recipient.

Upon receipt of an encrypted packet digital signal, the encryption algorithm used for encryption and the encryption key of those receiving devices having the right to decrypt are known. The ability of a recipient who does not know the decryption algorithm and / or key to benefit from the signal without decryption depends, e.g. 2 105736 rhythms used for encryption, the characteristics of a packet-shaped digital signal, and the depth of encryption, that is, how many packets and which packets are encrypted.

One important application for packet-based digital signal transmission is the upcoming 5 digital television system, where multiple programs are simultaneously transmitted on a single transmission channel. In this case, packets containing video, audio and data information of different programs are carried alternately on the transmission channel. An example is MPEG-2 (Motion Picture Experts Group), which defines a generic standard for information compression and video and audio compression that allows the transmission of a television image with less bits than if the television image were merely digitized directly into bits. A single video data packet usually contains image information of several macroblocks containing 16x16 pixels. Figures 1a and Ib show the packet structure of a digital signal according to the MPEG-2 standard. Figure 1a illustrates the top-level packet structure of the signal to be transmitted at the transmission-15 level, whereby the packet structure enables simultaneous transmission of several different programs on the same transport channel by splitting the signal over time into Transport Stream Packets (TSPs), each containing header information (HEADER) DF, Data Field). In Figure 1a, the data fields DF1 and DF3 contain video information related to program # 1, DF5 contains audio information related to program 20 # 1, data field DF2 contains video information related to another program # 2, and DF4 contains data information related to another program # 2. Figure Ib shows one packet comprising a header, i.e. HEADER, and a data field DF consisting of an Adaptation Field and a Payload (PAYLOAD). The adapter field may not be available in all packets sent.

, * 25 One packet contains 188 bytes. The MPEG-2 standard is further described in ISO / IEC 13818.

Source coding according to the MPEG-2 algorithm is based on the use of discrete cosine transform (DCT), motion compensation and variable coding. Some of the images contained in the sig-30 signal are encoded using only in-image information (intracoded or I-images). For the coding of other images, the information contained in the coded images may be utilized. Only motion vectors and prediction error are transmitted in encoded form to these images. The images are decoded in the receiver from the introdelected images by motion vectors and the result is corrected with a prediction error. 35 Images predicted over time from I-images are called P-images. P-pictures can also be predicted from the previous P-picture. The images between successive I and P images are so-called. B images and are predicted from the nearest previous and closest next P or I image.

3, 105736

Compressive coding, such as coding according to the MPEG-2 algorithm, makes the signal more sensitive to transmission errors. Therefore, receivers generally include "Error Concealment" features, i.e., the effect of errors detected in signal transmission is concealed. Errors can be detected either during channel-5 encoding or within the decoder.

The use of variable length codewords in encoding causes the receiver not to decode even the following bit stream that has been correctly transmitted after the error because the receiver lacks information about where the screening codeword starts and how many codewords have been lost due to the error. Therefore, the bitstream contains certain reserved codewords (Start Codes) to which the receiver can synchronize after a transmission error. For example, in the MPEG-2 standard, the smallest entity to which a receiver can synchronize after a transmission error is "Slice". "Slice" means a portion of an image consisting of a plurality of adjacent macroblocks. "Slice" can start at the beginning of an arbitrary macroblock and continue up to the right edge of the image. If the screen has 720 pixels horizontally, the maximum length of the "slice" is 45 macroblocks. Thus, after a transmission error, the receiver can decode the image information correctly from the beginning of the next image slice forward. In this case, the transmission error is shown in the image as a horizontal strip at least as high as the slice-20, and the corresponding image information is lost.

The information contained in the erroneous video packets may be attempted to be corrected at the receiver by replacing the particular video. a portion of the image based on the previous error-decoded image, possibly utilizing motion vectors.

The information contained in the erroneous audio packets can be corrected by muting the receiver during the erroneous information.

30 Full view of the information transmitted on a pay-TV system is only possible for viewers who pay separately for the reception of the program in question. This is preferably accomplished by encrypting the signal before transmission, and paying viewers having an encryption decoder attached to their receiver. In many cases, the decryption device requires different keys 35, which are sent against the view fee only to the respective decryption devices. Pay-TV operators aim at an encryption method that produces good quality picture and sound to the pay-viewer's receiver, in other words encryption must not impair the quality of the program being transmitted. Operators may also want to allow encrypted transmission to be viewed without decryption, but the picture and / or sound quality is poor. In this case, the quality of the picture and sound should preferably be such that the viewer will have an idea of the particular subject. operator programming, but still of poor quality to make long-term monitoring of the program meaningless. In this case, encrypted transmission provides a means of promoting the Operator's Programming to viewers who are on the same network but have not paid for the program. receiving the program.

The program is encrypted at the transmitting end by a device that performs signal hall 0 compilation using a valid encryption key and encryption algorithm. The decryption is performed at the receiver or before the receiver with a special decryption device or module. Decryption can only be successful if the receiver has the information needed to decrypt it, such as a decryption key and algorithm.

15

Most current encryption devices used in pay-TV applications are based on encryption of an analog video signal. These devices will not be able to be applied to future digital systems as such and will not take advantage of the special features of digital transmission. The encryption methods designed for digital ice systems, on the other hand, mix the signal so deeply that it cannot be understood without decryption.

Some prior encryption methods and arrangements are known from patent applications FI941315 and FI 941316.

25

It is an object of the present invention to provide a method and apparatus for encrypting a packet-shaped digital signal such that encryption is applied to whole packets and preferably only a portion of the packets to be transmitted are encrypted. The depth of encryption, i.e., the intelligibility of the signal, can be adjusted by changing the ratio between the unencrypted and the encrypted packets and selecting the packets to be encrypted appropriately.

[The present invention provides a method by which encryption can simply be accomplished, but it is practically impossible to decrypt without authorization. The invention also provides an apparatus for carrying out the method of the invention.

More particularly, the invention relates to a method for encrypting information transmitted as a packet digital signal comprising at least a data field comprising a header field and a payload, wherein method 510736 selects an encrypted packet according to a predetermined selection strategy, encrypts the encrypted data into at least characterized in that, as part of said selection strategy, the bit content of the data field of at least one packet is examined and said at least one packet is encrypted when the bit content of its data field corresponds to a predetermined selection criterion.

The device for encrypting a digital signal transmitted in packet form according to the invention comprises means for selecting encrypted packets from packets to be transmitted, means for marking the encryption information in the packet, and means for encrypting the packet to be encrypted.

15

It is characterized in that it comprises: means for identifying packets, means for storing comparison information, means for performing a comparison between a data field contained in the packets and said reference information, and means for making an encryption decision based on the result of said comparison.

The system according to the invention is characterized in that it comprises in at least one device the features of the device according to the invention described above.

25

The present invention utilizes the packet form of the transmitted signal. In this case, the encryption is easy to implement and the depth of encryption can be adjusted between fully encrypted and unencrypted. An advantage of the device according to the invention is e.g. the fact that the contents of the data field of the 30 packets being investigated, and possibly the data fields of one or more other packets, may also advantageously be taken into account when deciding the encryption decision for each packet.

* «

The encryption may be selectively implemented for one or more programs simultaneously transmitted in the transmission channel, whereby different encryption keys may be used for different signal components of the program. Encryption can also be implemented by targeting only certain signal components of the program, such as audio and / or video signals, whereby any other signal components of the program are not encrypted.

6, 105736

The device of the invention also enables the quality of the encrypted signal to be controlled by adjusting the encryption depth. If more packets are encrypted, the quality of the received signal without decryption will deteriorate. The choice of encrypted packets can be done in different ways. In order to control the characteristics of the received 5 signal without decryption, the ability of the receiver to conceal the error packets (Error Concealment) must be considered when selecting packets to be encrypted. For MPEG-2 television broadcasting, the most efficient encryption is achieved by encrypting packets belonging to I-images, whereby the receiver cannot fully compensate for the error caused by the encrypted I-images without decryption. On the other hand, a receiver with good error concealment capabilities may be able to eliminate virtually all P and B image encryption errors, so the pay-TV system operator may not have the benefit of encrypting these image types.

In the following, various embodiments of the invention and details of the invention will be described with reference to the accompanying drawings, in which Fig. 1a illustrates the transmission of a digital signal consisting of signal components of two programs according to the MPEG-2 standard; a preferred embodiment, Figure 3 illustrates the principle operation of a code comparison portion of an embodiment of the invention of Figure 2, Figure 4 illustrates a decision algorithm element of a preferred embodiment of the invention. 25 is a block diagram of principal operations, and FIG. 5 illustrates an arrangement of an encryption device according to the invention in a digital signal transmission system.

Figure 2 shows a preferred embodiment of the encryption device 1 according to the invention. The encrypted packet digital signal is passed through an input line IN. · To the pack synchronization element 2 and to the standard delay element 3. Typically, the packet begins with some synchronization data. The packet synchronization means 2 identifies the packet origins from the incoming packet digital signal. This can be done, for example, by the packet synchronization member 2 looking for a certain bit pattern, i.e. synchronization data, which is repeated at regular intervals in the incoming signal. When the synchronization element 2 has found such synchronization data repeatedly enough in the signal, it concludes that it has found the origin of the packet, i.e. it has synchronized to the incoming signal 7 105736. A separate input signal may also be used to indicate the origin of the packet, whereby the packet synchronization element 2 is not required.

Once the start of the packet has been found, the packet is directed to a packet identification element 4, whereby based on the identified information in the packet header field, such as the packet identifier, it is examined whether that packet belongs to a program or component to be encrypted. a reference memory to the packet sensing element to store! a packet identifier to be searched and a shift register for storing the packet identifier of the received 10 packets. The packet identifier is compared with the information stored in the reference memory. When such a packet belonging to the encrypted signal is found, information is transmitted to the decision algorithm 5 via the identification line 6 and this possibly encrypted packet is transmitted to the header field eraser 7.

The function of the header field deletion element 7 is to remove the header field HEADER and the possible adaptation field Adaptation Field from the detected packets and to transmit the payload of the packet to the PAYLOAD comparator 8. Since the length of the packet and its header field HEADER are The -20 member 7 can be implemented e.g. Information on the existence of a possible adaptation field is in the header field and the length of the adaptation field at the beginning of the adaptation field, whereby the payload is correspondingly shorter. Hereby, only the payload PAYLOAD of the packet is transmitted from the header field eraser 7 to the comparator 8.

25

In the embodiment of Figure 2, the comparing member 8 is formed from a plurality of separate comparators 8a, 8b, 8c, the function of which is explained in Figure 3. Each comparator 8a, 8b, 8c is preconfigured with at least one comparator register 9a, 9b is compared to the 30 data received through the reference line 10. The bit pattern to be searched can also belong to two or more data fields DF1, DF2, whereby the comparison has to be made on the basis of two or more packets 9. ____________ tea. The reference registers 9a, 9b may preferably be formed from the shift registers. In most typical applications, 32-bit data comparison is sufficient, and the transfer registers are preferably 32-bit. The data to be compared is directed to register-11 of data-35. The comparator 8a, 8b, 8c generates a comparison result 12 which is passed to the decision algorithm 5 via the result lines 13a, 13b, 13c. In the example of Fig. 3, the first reference element XOR, based on the first reference information MASK and the incoming data field DF of the first reference register 9a, generates a first intermediate result 14 per bit 8 105736, which is transmitted to the second reference element AND. In this example, the MASKI of the first reference data is XOXX 1XXX 1X01 XXI1 11X1 XXI1, the content of the second reference information MASK2 is 0100 1000 1011 0011 1101 0011 and the incoming data is XOXX 0ΧΧΧ 1X01 5 XX11 11X1 XXI1. Formation of the first intermediate result 14 is preferably performed by an absolute or function, whereby the bits to be compared are the same (both 1 or 0), the result is 0, otherwise 1. The X marks mean that the bit in question always 0 because the corresponding bit in the second reference MASK2 is 0.

10

The second comparator AND compares bit by bit the first comparator 14 from the first comparator XOR and the second comparator MASK2 in the second comparator register 9b, which compares to give a second intermediate result RES. Comparison is performed by the and function, whereby each of the comparator bits one also obtains the result of the comparison 1, otherwise 0. Thus, only the bits having equivalent bits in the other reference register 9b are relevant for the comparison in the AND comparator and the whole comparison. From the second intermediate result RES, output data of the comparator 8a, 8b, 8c is formed, i.e. the result of the comparison 12 for deriving to the decision algorithm body 5.

20

Comparators 8a, 8b, 8c, therefore, look for certain preset bit patterns in the incoming data. In the comparators 8a, 8b, 8c of the example of Figure 3, finding the correct code is evident from the fact that in the second intermediate result RES all bits are zero. In the computed example, no code was found because the fifth bit on the left was not the same as the first 25 data in the mask. The comparators 8a, 8b, 8c report the found bit pattern to the decision algorithm means 5 via the result lines 13a, 13b, 13c.

The comparator of Fig. 3 has a first control bus 38 provided in the first reference register, which can be used e.g. the contents of the first reference register. setting and changing as needed, and a second control bus 39 for setting the contents of the second reference register, respectively. The control paths 38, 39 are not required in applications where the contents of the reference registers do not need to be changed, whereby the reference registers can be set advantageously at the time of manufacture of the device.

35 The comparators 8a, 8b, 8c of Figure 2 have their own control paths 38a, 39a; 38b, 39b; 38c, 39c. The control buses can be operated as is known per se, e.g. by means of a separate logic or a microcontroller.

105736 The decision algorithm 5 executes the identification data received from the packet identification body 4 along line 6 and the comparison results 12 from the comparators 8a, 8b, 8c along the result lines 13a, 13b, 13c and a predetermined algorithm for packet encryption decision SP. In addition to the information received from comparators 8a, 8b, 8c, the decision algorithm-5 body 5 may also use the content of the data following the bit pattern found to make the encryption decision SP. Comparison result 12 and data field DF are irrelevant to encryption if the packet identifier is incorrect.

Fig. 4 shows a decision-algorithm element 5 of a preferred embodiment of the invention applied to the encryption of a television signal according to the MPEG-2 standard. The encryption decision SP may be influenced by four different encryption factors: random number generator 15, number of packets, image type, and slice location. The packet encryption decision SP can be formed by a combination of the four different encryption factors, whereby the control inputs VI, V2, V3, V4 select which encryption factors are taken into account in the packet encryption decision SP. In this description, a packet encryption decision SP is described as a logical zero (0) if the packet is to be encrypted and as a logical one (1) if the packet is not encrypted. Similarly, for the various factors used to form the packet encryption decision SP, a logical zero means that the encryption condition is fulfilled for that encryption factor, and the logical 20 does not implement the encryption condition. In a preferred embodiment of the device, a logic zero can mean a lower voltage (about OV) and a logic one a correspondingly higher voltage (about 5V).

The random number generator 15 as such generates random numbers in any of the "25 given intervals, e.g. integers from 0 to 9. is formed by a random number greater than or equal to the first lower bound A1 and less than or equal to the first upper bound Y1.

The packet identification data coming from the detection line 6 to the decision algorithm 5 is passed to a packet counter 17 which counts the number of packets included in the encrypted signal component. The packet counter 17 is accompanied by a demultiplexer 18 for selecting a reset condition of the packet counter 17. In the embodiment of Figure 4, the packet counter 17 is reset based on the comparison result of either the first comparator 8a or the second comparator 8b, depending on the state of the reset selection input VN. After the reset, the counting of the packet counter 17 starts from the beginning, whereby the application can advantageously count the incoming packets starting from e.g. the beginning of the picture. The packet counter 17 transmits the second encryption information P2 on the arrival of packets belonging to the encrypted signal component to the decision combining block 16. The second lower limit A2 and the second upper limit Y2 can be set on the packet counter, wherein the packet counter 17 transmits the second encryption data P2 5 preferably only the number is greater than or equal to the second lower bound A2 and less than or equal to the second upper bound Y2.

In the embodiment of Fig. 4, the second comparator 8b is arranged to search for the beginning 10 of the image. For a MPEG-2 video signal, the image's initial identifier is a 32-bit binary number 00000100H. The notation H denotes a hexadecimal number, that is, a 16-digit integer sequence. When the second comparator 8b detects the origin of the image, the information is passed to the image type identification block 19, which also receives the two bytes following the initial identifier containing the image type information. The image type identification block 15 19 transmits information about the image type (I, P, or B image) to the image type request block 20 which compares the image type to the image type selection inputs VI, VP, VB. If that image type is selected to be encrypted, the image type request block 20 forms a third encryption information P3 in decision combining block 16.

In the embodiment of Figure 4, the first comparator 8a is arranged to search for the starting points of the picture slices. In a video signal according to the MPEG-2 standard, the slice header begins with 000001H, followed by 8-bit vertical position information XX. The position XX of the slice of the image is between OlH and afH. When the first comparator 8a detects the origin of the image slice, information is transmitted. 25 to location detection block 21, which also receives location information XX. The location recognition block 21 transmits information about the location of the image slice to the location requirement block 22, which compares the location of the image slice to the given location lower boundary A3 and the upper position boundary Y3. If the location of said image slice is between the lower position A3 of said position and the upper position Y3 of the position, the position request block 22 forms four to 30 encryption data P4 into the decision merge block 16.

Decision combining block 16 examines the states of control inputs VI-V4 and generates an encryption decision SP based on the encryption information PI-P4 selected to be effective on encryption decision SP (0). Preferably, the decisions are combined with an AND function, 35 whereby the encryption decision of the packet is true (0) only when all the encryption information PI - P4 affecting the encryption decision SP is true (0). This can be represented in Boolean algebra by the formula (1). In the formula, the result of the addition (+) is zero (0) only if both of the additions are zero, otherwise the result is one (1). The result of multiplication (·) l 11 105736 is only one when both multipliers are one, otherwise the result is zero. The strikethrough denotes negation, that is, zero is converted to number one and number one to zero. Then, for example, P1 is zero when P1 is one.

5 (1) SP = (V '\ + P]) (V2 + T * 2) - (V3 + PZ) (V4 + P4)

In Formula (1), only four encryption factors affecting the encryption decision are used, but in some embodiments of the invention it may be necessary to use more factors that influence the encryption decision. A more general form of Formula (1) is represented by Formula (2), where P is a multiplication.

(2) SP = n (Vi + Pi), where i = 1 ... n and n = number of factors affecting the encryption decision 15 The decision algorithm 5 can be implemented entirely on a hardware basis, e.g. by using programmable logic circuits having the necessary logical connection in the decision, but especially with respect to the depth of encryption or changes in the algorithm, it is more flexible to use hardware with a microcontroller as one component, whereby the executing portion of the decryption decision may be implemented at least in part in the microcontroller program. The microcontroller is preferably one which has a sufficient number of access lines for transmitting data, e.g. a packet detecting means 4, a header field deleting member 7 and a comparing member 8 for a microcontroller and a microcontroller for an encryption field setting member 23.

* · 25 decision algorithm element 5 obtains packet encryption information, which is transmitted to encryption field field setting element 23, body. Preferably, the encryption data field has two distinct states, e.g., logical zero and one, wherein the first state represents an unencrypted packet and the second state represents an encrypted packet. Preferably, the encryption field is in the packet header field HEADER, which is preferably transmitted unencrypted, whereby the received packet header field HEADER 35 examines whether encryption information is set in it. If the encryption information is set, the packet is decrypted, otherwise the packet is forwarded as such to the receiver for further processing, such as demultiplexing video, audio and data signals.

The standard delay member 3 delays the packet from the IN line so that the decision algorithm 5in 5 can take the necessary steps to make the packet encryption decision. The standard delay member 3 can advantageously be implemented with a FIFO (First In, First Out) memory, the size of which is adjusted to suit.

The encryption of a packet with a specific packet identifier is described above. If the h-10 is allowed to process several signals at the same time, the devices shown in Fig. 2 can be connected sequentially. Each device makes encryption decisions for packets with a given packet identifier.

Multiple packet identifiers with different packet identifiers may also be encrypted in parallel, each packet identifier being encrypted requiring its own packet identification member 4 and header field deletion member 7. Comparison member 8 may not be required separately for each signal. In this case, one packet synchronization element 2, one constant delay element 3, one encryption field setting element 23 and one encryption block 24. The decision algorithm may support encryption of more than one signal component.

20

Figure 5 shows an example of how an encryption decision executing block according to the invention may be located in a television system. Encryption device 1 and blocks 25-30 are located at the transmitter end and blocks 32-37 are located at the receiving end. In blocks 25, 26 and 27, separate source coding for the video, audio and data signals is performed and digital packets are formed from the 25 coded signals. Block 28 multiplexes audio, video, and data signals associated with one program, and block 29 multiplexes multiple programs into a single bit stream. Thus, the entities formed by blocks 25-28 may be multiple in parallel, i.e., one entity per program whose transmissions are combined for transmission on the same transport channel 31 in the oh-30 multiplexer 29. The encryption device 1 performs the encryption decision according to the invention and any encryption for selected packets. The encryption decision is recorded in the HEADER header field of the packet and the actual encryption is performed in the encryption block 24 (Figure 2) of the encryption device 1 in the data field DF of the packets whose HEADER header field is labeled to be encrypted. The data fields DF of packets 35 that are not marked for encryption pass through encryption device 1 unchanged. In block 30, the received digital signal is subjected to channel coding and modulation and transmitted to the actual transmission channel represented by block 31, which may be e.g. a television cable network, satellite connection, or antenna connection between the transmitter and the receiver. In block 32, the signal received from the transport channel is demodulated and decoded by the channel. In block 33, decryption is performed. In block 34, i.e., the demultiplexer, from the programs included in the received bitstream, the audio, video and da-5 level signals of the desired program are selected for their own processing branches. In blocks 35-37, the data packets are decoded, i.e., the video, audio and data bit streams are formed from the packets, and the video, audio and data signals are separately decoded.

The encryption algorithm of the device according to the invention can be implemented either fixed or in such a way that the algorithm can be changed as needed, whereby the change of the encryption algorithm is transmitted in the bit stream,

The packet header field HEADER and the adaptation field Adaptation Field are typically transmitted unencrypted, whereby the encryption is applied to the payload of the data field 15 or part thereof.

It will be understood that the present invention is not limited to the above embodiments, but that the invention may be modified within the scope of the appended claims.

> ......

Claims (16)

A method for encrypting information transmitted as a packet digital signal comprising at least a header field (HEADER) and a data field (DF) comprising a payload (PAYLOAD), the method comprising selecting an encrypted packet according to a predetermined selection strategy, encrypting the encrypted data, part of the packet to be encrypted, characterized in that, as part of said selection strategy, the bit content of at least one data field (DF) of the packet 10 is examined and said at least one packet is encrypted when the bit content of its data field meets a predetermined selection criterion. Method according to claim 1, characterized in that said encryption information is marked in the header field (HEADER) of the packet. Method according to claim 2, characterized in that the payload (PAYLOAD) is encrypted. A method according to claim 2 or 3, characterized by selecting 20 packets to be encrypted by examining the data fields (DF) of at least two packets, wherein at least one of said at least two packets is encrypted when a predetermined search condition is fulfilled. Method according to claim 1, characterized in that the selection strategy of the encryption key is further based on random selection and / or selection according to a predetermined sequence. 6. A method according to claim 1, characterized in that the packet-shaped digital signal consists of packets of one or more signal sources which are transmitted interleaved. < A device (1) for encrypting a digital signal to be transmitted in a packet form, comprising means (2, 4, 5, 7, 8) for selecting encrypted packets from packets to be transmitted, means (23) for encrypting information in a packet, and means (24) means for encrypting the marked packet, characterized by: 105736 means (2, 4, 8) for identifying packets, means (9a, 9b) for storing reference information (MASK, MASK2), means (XOR, AND) for performing a comparison of the data contained in the packets. a field (DF) and said comparison information (MASK1, MASK2), and 5 means (5) for making an encryption decision based on the result of said comparison (12). The device for encrypting a packet-shaped digital signal according to claim 7, wherein the packets comprise at least a data field (DF) further comprising a header field (HEADER) and a payload (PAYLOAD), characterized in that the means (23) for 23) for marking the encryption information in the packet header field (HEADER), and the means (24) for encrypting the marked packet to be encrypted further comprises means (24) for encrypting the payload of the packet (PAYLOAD). Device (1) according to Claim 7, characterized in that the means (2, 4, 5, 7, 8) for selecting the packets to be encrypted further comprise means (5) for making the encryption decision randomly or according to a predetermined sequence. 20 Device according to one of Claims 7 to 9, characterized in that the packet-shaped digital signal is a compressed video signal. Device according to one of Claims 7 to 9, characterized in that the packet-shaped digital signal is an audio signal. Device according to one of Claims 7 to 9, characterized in that the packet-shaped digital signal is a data signal. Device according to one of Claims 7 to 12, characterized in that the packet-shaped digital signal consists of packets of one or more signal sources. Device according to Claim 7, characterized in that the means (2, 4, 7, 8) 35 for identifying the packets comprise means (8a) for recognizing the initial identifiers of the image and / or the image type. 16 105736 14 105736 Device according to Claim 7, characterized in that the means (2, 4, 7, 8) for identifying the packets comprises means (8b) for identifying the initial identifiers of the parts of the image. 5 A system for encrypting information transmitted as a packet-shaped digital signal, comprising: means (25, 26, 27) for encoding video, audio and data signals and generating packets, means (2, 4, 5, 7, 8) for selecting packets to be encrypted the packets to be transmitted, the means (23) for encrypting the encrypted information in the packet, the means (24) for encrypting the marked-to-encrypted packet, combining the packet-converted signals associated with one program of the multiplexer (28), the multiplexer (29) (1) encrypting the signal, changing the bitstream of the channel encoder (30) for transmission to the transmission channel (31), 20. means (32) for decoding the channel, means (33) for decrypting, means (34) for extracting program-related packet-shaped signals, and , 36, 37) packet unpacking for decoding signals, characterized in that the means (1) for encrypting the signal further comprises means (2, 4, 8) for identifying packets, means (9a, 9b) for comparing information (MASK, MASK2). means (XOR, AND) for performing a comparison between the data directory (DF) of the packets and said comparison information (MASK, MASK2), and * means (5) for making an encryption decision based on the result of said comparison (12). 105 '/ 36 17