JPH02285738A - Digital video scramble device - Google Patents

Abstract

PURPOSE:To reduce the reproduced noise level of a receiver not subscribed yet and to apply cryptographic processing with excellent confidentiality by applying the scrambling to a compression digital signal subjected to quasi- momentary compression with a PN code series as a conventional system, converting a range code into a modified range code according to a prescribed rule and sending the result. CONSTITUTION:When a range code is N (3-bit), the kind of code K capable of expression is 2N=8. When the kind used actually by a quasi-momentary compression means is L(=5) among K kinds, unused kinds M=K-L(=3) are obtained. Normally the relation of L>M exists. In this case, the range code of L(=5) kinds is converted into L kinds (=5) of modified range code included in the sum of a 1st group (M=3) comprising M(=3) kinds of unused range code and L-M(=2) range codes arranged in the range code representing a block with a small maximum amplitude in the original K kinds of range codes. Through the constitution above, reproduced noise is sufficiently low with the range code only corresponding to the small amplitude block used for a nonsubscribing receiver and since the code differs from the substantial range code, the confidentiality is sufficient, as well.

Description

[Detailed Description of the Invention] Field of Industrial Application The present invention relates to a digital audio scrambling device for wired broadcasting, etc. It is related to the conventional technology, satellite broadcasting currently being implemented in Japan, and In high-definition television satellite broadcasting, quasi-instantaneous compression PCM method or quasi-instantaneous compression differential PCM method is adopted as the audio transmission method.As is well known, in these methods, linear quantized digital audio signal or The differentially quantized digital audio signal is divided into blocks by appropriate periods (approximately 1 m SeC), and a range code is determined according to the maximum value of the digital audio signal included in each block. A compressed digital audio signal is generated by extracting only the bits at a predetermined position from a certain number of bits of a digital audio signal, and this compressed digital signal and the range code are output together. ζ When digital audio is encrypted (scrambled) by adding a PN (pseudo-random code) sequence, even within the shell, the addition of PN codes protects against secrecy when played back by a receiving device of a non-subscriber. It can be said that this method is excellent as a method of scrambling digital audio because the PN code generation method is high and the probability that the PN code generation method will be decoded is low. However, when this digital audio with the PN code added is played back by a receiving device of a non-subscriber, there is a high probability that noise with the maximum sound pressure will be played back, which will only cause psychological discomfort to the viewer. In extreme cases, there is a risk of damaging playback equipment such as speakers.To prevent this, for example, an identification code to identify whether or not it is a paid broadcast is placed in the transmitted signal, so that the receiving equipment of non-subscribers can One possible method is to stop audio playback when this is detected.However, many receiving devices already on the market do not have such measures, and it is difficult to promote subscriptions to paid TV programs. From this point of view, there is a strong desire to scramble and broadcast in a state that can be heard to some extent, so the above playback stop method cannot necessarily be said to be appropriate. The output compressed digital audio signal is added with PN code, but the range code is sent out as is without addition (Arashi, 1986).
Inquiry No. 17 "Technical Conditions for Paid Broadcasting in Television Broadcasting by Broadcasting Satellite" to the Satellite Broadcasting Department of the Telecommunications Technology Council of the Government Ministry, Report [Attachment]). Generally, when an audio (music) signal is observed for a long period of time, the probability of occurrence is high in the small amplitude region, and therefore the range code also tends to occur in the range that indicates the small amplitude period.As a result, the range code is also scrambled. The average level of noise in the reproduced output of the non-subscribing receiver will be lower than in the case of
If the range codes are scrambled, all ranges will occur with approximately the same probability). Note that this effect can be similarly obtained when the digital audio signal before quasi-instantaneous compression is differential PCM. Figure 2 shows the configuration of a conventional digital audio scrambling device. The range code and the compressed digital audio signal are extracted by the quasi-instantaneous compression means 2. Of these, the compressed digital audio signal is converted into a PN code from the PN code sequence generation means 4 controlled by the encryption key.
This scrambled compressed digital audio signal and the range code are combined and sent out by the synthesizing means 5. Note that the procedure on the receiving side is the reverse of this diagram, so the explanation is omitted. However, on the receiving side, only the subscriber can manually create the encryption key.Problems to be solved by the inventionThe above conventional example certainly reduces noise compared to the case where the range code is also scrambled. It cannot be said that this is a sufficient countermeasure - the first reason is that the amount of noise reduction is insufficient.a The range code is transmitted as is.The maximum value within each block during playback is equal to the original value. Power Length Equal to Audio Signal When comparing noise with the same maximum amplitude and a general music signal, the average energy of the noise is greater, so the scrambled signal sounds louder to the auditory sense. The second reason is that the secrecy for non-subscribers' receiving equipment is lowered, but if the range code is reproduced correctly, the strength information of the original sound can be reproduced compared to when the range code is also scrambled. Confidentiality decreases a little
For example, there may be cases where it is possible to distinguish whether the original sound is a human conversation or a musical tone. In view of the above points, the present invention reduces the noise level reproduced by a non-subscriber receiving device and provides confidentiality. It is an object of the present invention to provide a digital audio scrambling device which is excellent in the above-described problems. The range code may be scrambled using a method such as , etc., and then the range code is converted into a modified range code according to a predetermined rule before being sent out.The rules for converting a range code are as follows. Then, the number of types of range codes that can be expressed with N bits is 2N.
If the number of range codes actually used in the quasi-instantaneous compression means is equal to the number of range codes of these types, the number M of unused range codes is M=KL. Usually, more than half of K is used (if less than half is used, the range code should only require (N-1) bits), so L>M. In this case, the original L The types of range codes are arranged in the order of the first group consisting of M types of unused range codes, and the range code that originally indicates the block with the smallest maximum amplitude among the types of range codes ( LM) types of modified range codes included in the sum group with a second group consisting of L-M) types of range codes.

Operation First, in a receiving device having the right to descramble the scrambled audio, that is, a receiving device of a subscriber in pay television, the PN code sequence k for the compressed digital audio signal is transmitted.
Since the range code conversion rules are also known, the original speech can be completely restored. On the other hand, in a receiving device that does not have the right to cancel, that is, a receiving device of a non-subscriber, the following phenomenon occurs. First, the compressed digital audio signal has P
Since encryption is performed using the N code sequence, the content is almost unknown.Next, the range code that affects the average sound volume is the range code power that is not originally used.\\The sound volume is small. Even if a range code indicating
In the latter case, the compressed digital audio signal is expanded according to the range code value.Needless to say, the compressed digital audio signal is expanded only within a relatively small amplitude range. In this case, the most common action is to interpolate the previous value of the range code. In other words, when an unused range code is received, the range code indicating the small amplitude block that was received at the closest previous point is repeated and used.This causes the compressed digital audio to be decompressed. Since only the range code corresponding to the small amplitude block is used in the unsubscribed receiving device, the reproduced sound (substantially noise) also has a sufficiently low average volume. Furthermore, since the range code used in this way is different from the original range code, it is extremely advantageous in terms of confidentiality.Example The following is an example of the case where the present invention is applied to satellite broadcasting in Japan. I will explain this using drawings. Figure 3 shows the r14 system used in digital audio transmission for satellite broadcasting in Japan.
This is a bit arrangement diagram showing the relationship between the range code and the compressed digital audio signal in ``/10 quasi-instantaneous compression''. A total of five ranges are available, up to a range where up to 4 bits are omitted, and in each case, the audio is extracted as compressed digital audio. Therefore, the number of types of range codes that can be expressed is 21 = 8, but the number of types of range codes actually used is 5 as shown in Figure 3.
Therefore, the number of types of unused range codes M is M=
8-5 = 3, that is, "range oJ ~" range 4"
Five types are used, and three types from "Range 5" to "Range 7" are not used. Here, FIG. 1 is a block diagram showing the configuration of the digital audio scrambling device according to the present invention. Although the difference from the conventional example shown in FIG. 2 is that the code conversion means 6 is installed, the conversion rules are fixed or variable depending on the code solution. is the original range code “range 0”.
'' to ``Range 4'' Some of the modified range code candidates after conversion extracted corresponding to these 3 types of unused range codes ``Range 5'' to ``Range 7'' The other is two types (L-M-=
5-3=2) range codes “Range 4” and “Range 3”
This situation is shown in Figure 4. Result The original range code "Range 0" to "Range 4" is converted by the conversion means 6 to the modified range code "Range 3" to "Range 7", but in this case, the number of types of conversion combinations is 5! =1
Since there are 20 possible combinations, one of them can be specified using the encryption key as described above.The most general implementation method of the range code conversion means 6 is to use a ROM in which the conversion rules are written in advance. FIG. 5 shows an example of how the ROM is used when using the above 120 combinations. If the conversion is fixed, it can be realized using only a few logic gate circuits, as in the example shown in FIG. When received by a non-subscribing receiving device, the only valid range codes are "Range 3" and "Range 4", and all other reception range codes "Range 5" to "Range 7" are not used. The error detection circuit of the range code performs interpolation processing on the previous value, resulting in either "range 3" or "range 3."
Even if Range 4 is used repeatedly, as a result, the reproduced noise will be limited to the two ranges with the smallest amplitude among the five ranges, so it will not be reproduced at a sufficiently low volume. It also has the effect of improving confidentiality by masking most of the original voice strength information, which only changes between two types with adjacent ranges. Note that when transmitting an error correction code added to a range code, first convert the range code (X,
An error correction code can be added to the extracted modified range code. Code “range 3” ~
Even if a 4-bit correction code is added to each of "Range 7", the digital voice scrambling device according to the present invention has more than the effect of the invention. It has the advantage of being able to reduce the amount of data and providing excellent secrecy. Furthermore, the present invention requires only converting a range code of at most a few bits, and if this conversion is fixed, it can be configured with several logic gate circuits, and even if the conversion is variable, it can be configured with a small ROM.
Since it can be achieved with just one person, the circuit load inside the device is extremely light.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a digital audio scrambling device according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of a conventional digital audio scrambling device. FIG. 3 is a block diagram showing the configuration of a conventional digital audio scrambling device. Figure 4 is an explanation of an example of range code conversion. Figure 5 is a block diagram showing the configuration of a range code conversion means using a ROM. Figure 6 is a range code conversion means using a logic gate circuit. This is a block diagram showing the configuration of 4. 1. A/D conversion ratio 2. Quasi-instantaneous compression half-depression 4. PNN code generation view 5. Synthetic hair ratio 6
・・・・Name of range code conversion agent Patent attorney Shigetaka Awano and 1 other person

Claims (4)

[Claims] (1) Divide the digital audio signal series into blocks by dividing them into appropriate periods, determine a range code according to the maximum value of the digital audio signal included in each block, and use this range code to determine the range of the digital audio signal in the corresponding block. A compressed digital audio signal is generated by extracting only bits at a predetermined position with a constant number of bits, and the compressed digital audio signal is It includes an encryption means for encrypting a signal using a pseudo-random code sequence, and a range code conversion means for converting the range code to generate a modified range code, and the conversion of the range code is performed by converting the N bits constituting the range code. (N is a natural number) K types (K
=2^N), the quasi-instantaneous compression means uses L types (L is a natural number, L<K), and the number of types of unused ranges M (M is a natural number, M=K-L) is If it is smaller than L, the original L types of range codes are divided into a first group consisting of M types of unused range codes and a range code indicating a block with a small maximum amplitude among the original K types of range codes. A digital audio scrambling device characterized in that conversion is performed into a total of L types of modified range codes included in a sum group with a second group consisting of (LM) types of range codes arranged in order. (2) The combination of conversion from the original L types of range codes to L types of modified range codes is determined using the encryption key or another encryption key. The digital audio scrambling device according to scope 1. (3) The digital audio scrambling device according to claim 1, wherein the digital audio signal sequence is obtained by linearly quantizing an analog audio signal. (4) The digital audio scrambling device according to claim 1, wherein the digital audio signal sequence is obtained by differentially quantizing analog audio between adjacent sample values.