JPH0630483B2 - Confidential device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an improvement of a secret talk device, particularly a spectrum inversion rotation type secret talk device.

(Prior Art) Conventionally, various secret communication methods and secret communication devices have been proposed as a method for maintaining confidentiality of communication contents. One of them is to divide or rearrange a frequency spectrum of a signal to be transmitted into a required number of sections. Some are transmitted as an inverted rotation spectrum.

This spectrum reversal rotation method is superior in demodulation sound quality and easy to realize a circuit compared to others, but the simple reversal rotation method has a drawback that the secret strength is slightly low and a band filter that is difficult to manufacture in a device that realizes this. There was a problem that it required.

For this reason, for example, Japanese Patent Application No. Sho 59148 (148)
Alternatively, as shown in Japanese Patent Application No. 59-44141 (Secret method and apparatus), the inversion frequency position or rotation combination of the spectrum can be arbitrarily changed, and a band-pass filter can be replaced by a low-pass filter which is relatively easy to manufacture. By doing so, a method has been proposed in which confidentiality is enhanced and cost is reduced.

In order to facilitate the understanding of the present invention, the conventional method will be described below.

FIGS. 2 (a) and 2 (b) are a block diagram showing a conventional reversing rotary secret communication device and a signal spectrum diagram for explaining the operation thereof.

To briefly explain the device configuration shown in FIG. 2 (a),
As shown in (a) of FIG. 2 (b), the audio signal 1 having the maximum frequency limited to _m is branched into two, and one of them is directly connected to one input terminal of the adder 2.
Further, the other one is input to the other input end of the adder 2 through the frequency conversion circuit 3 which outputs only the sum of the two with the local frequency as m and is output to the output end of the adder.
As shown in (b) of (b), a spectrum is obtained by arranging two of the audio signals 1 adjacent to each other with a frequency shift.

Further, the signal is branched into _two, and a low pass filter LPF ₁ is inserted between the _two multipliers MULT ₁ and MULT ₂ for the first time.
Through the second series circuit route of the second series circuit route in which the second low-pass filter LPF ₂ is inserted between the two multipliers MULT ₃ and MULT ₄ in the same manner.
To the multipliers belonging to the first series circuit route via 90 ° phase shifters PS ₁ and PS ₂ and directly to the multipliers of the second series circuit route, the oscillators OSC ₁ and OSC ₂ outputs. Is applied as a local frequency to each of the pre-multipliers and the post-multipliers, so that the output of the second adder 4 has an inversely rotated voice spectrum as shown in FIG. 2 (b) (c). It is configured to obtain.

With this structure, it is possible to eliminate the need for a band-pass filter which is difficult to manufacture. However, a code generator is added to either one of the local oscillators OSC ₁ or OSC ₂ to change the local frequency according to a predetermined rule. Then, if the data is transmitted via a desired communication line, the confidentiality can be further enhanced. When receiving this, the original voice signal can be restored by controlling the local oscillator and demodulating in synchronization with the coding rule.

However, such spectrum inversion rotation processing can be performed on an analog signal as it is, but since the circuit element used for this is generally large and complicated, the
Usually, digital processing is performed after digital conversion (AD conversion), and by doing so, the apparatus itself is simplified.

However, since the above-mentioned conventional secret-talking device requires a minimum of 10 wave elements, 5 multipliers and 2 adder / subtractors, a total of 10 arithmetic elements, various problems occur when digitizing them.

That is, since the bit length for displaying data is generally finite in digital arithmetic, the shorter the bit length is, the more rounding error (also referred to as truncation error) occurs, and it acts as if noise is mixed. This problem is remarkable in fixed-point arithmetic, so it is desirable to use as few arithmetic elements as possible.

(Object of the Invention) The present invention has been made in order to solve the problems of the conventional spectrum inversion rotation type secret talking device as described above,
It is an object of the present invention to provide a secret-talking device suitable for performing digital processing by reducing the number of calculation elements while making use of the features of the conventionally-proposed secret-talking device.

(Summary of the Invention) To this end, in the present invention, a voice spectrum signal subjected to a required band limitation is divided into two, one of which is configured to extract a low frequency component of a mixed output with a local frequency of m. 1 frequency conversion circuit and local frequency _m
It is input to one input end of the adder circuit via the first route consisting of the + _k mixing circuit, and the other of the two branched speech spectrum components is set to have a local frequency of _k and a high frequency band is selected from the mixed components. A second frequency conversion circuit including a high-frequency wave filter that extracts only a component and a second mixing circuit that has a local frequency of m
The signal is input to the other input terminal of the adder via the route of, and only the component of m or less is extracted from the output of the adder to be the secret output.

Still further, by controlling the signal relating to k of the local frequency by a coded signal which changes irregularly in time, the confidentiality is further enhanced.

(Example) Hereinafter, the present invention will be described in detail based on illustrated examples.

FIG. 1 (a) is a block diagram showing an embodiment of the present invention.

In the same figure, 5 is a voice input terminal, and in that part (b)
As in (a), input the audio signal whose maximum frequency is limited to m, and branch it into two, and use one as the local frequency.
Is input to the other input terminal of the multiplier 7 that receives the output of the oscillator 6 that outputs the second oscillator through the low-pass wave filter 8 that sets the cutoff frequency to m. The output m + k (0 <k <m) of 9 is input to the second multiplier 10 having one input, and its mixed output is added by the adder 11
Connect to one input end. The other of the divided audio signals is the third oscillator 12 which outputs the local frequency k.
A third multiplier 13 having one input, a high-pass filter 14 having a cutoff frequency m, and a fourth multiplier 16 having an output of a fourth oscillator 15 outputting a local frequency m as one input. Is connected to the other input terminal of the adder 11 via the second route consisting of and the output of the adder 11 is input to the low-pass filter 17 having a cutoff frequency of m, and the desired confidential signal is output at the output terminal. To obtain

Of the voice signal applied to the input terminal 5 of the confidential device configured as described above, which has been band-limited so that the maximum frequency is m as shown in FIG. 1 (b) (a), is passed through the first route. The spectrum of what reaches the adder 11 is first mixed with the local frequency m in the first frequency conversion circuit consisting of the multiplier 7, the oscillator 6 and the low pass wave filter 8, and then only the component of m or less is extracted. Therefore, as shown in Fig. 1 (b) (b), the spectrum becomes symmetrical with the frequency zero axis as the center, and it is further shifted to the position of frequency k + m in the next stage mixing circuit. The spectrum signal as shown reaches the adder 11.

On the other hand, the sound signal passing through the second route is similarly mixed with the local frequency k in the third multiplier 13, and then only the components of m or more are extracted by the high frequency wave filter 14, so that the spectrum thereof is the same as that in FIG. (D), and is mixed with m by the fourth multiplier 16 to form a spectrum as shown in FIG.

Therefore, the signal passing through the two routes is added to the output of the adder 11, but the low-pass filter 1 having m as a cutoff frequency is added.
Since m or more is cut by 7, the output end has a frequency of m or less and k as shown in FIG.
In this case, it is possible to obtain the voice spectrum signal which is rotated in the reverse direction.

As is clear from the above description, a similar inverted rotation spectrum signal can be obtained with a configuration having one multiplier and one adder in total, which is smaller than that of the conventional secret-talking device, so that a similar inverted rotation spectrum signal can be obtained. It is possible to reduce the rounding error and improve the calculation speed.

Also in the present invention, as in the conventional device, it is also possible to concurrently use the operation of increasing the confidential talk intensity by changing k among the local frequencies irregularly in time. In that case, it may be necessary to change the local frequency in synchronization with the irregular change.

Furthermore, in the above-described embodiment, the transmission side device is described as an example, but a reception demodulation device can be realized by the same configuration,
In this case, the second local frequency may be mk.

Therefore, if the same device is also used as a transmitter / receiver, the third device
As shown in the figure, a controller 18 for switching the second local oscillation frequency to m ± k is added, and when transmitting, m +
When k is received, m-k may be configured to be a local frequency.

Further, as indicated by the dotted line in the figure, the controller generates a coded signal that randomly changes k of the local frequency with time, and the second oscillator 9 and the third oscillator are generated by the coded signal. If it is configured to control 12 k, the confidentiality can be further enhanced as described above.

In implementing the present invention, it is not necessary to limit to the above-mentioned example, and other configurations, for example, the same oscillator frequency is shared by the same local frequency, or m of the fourth oscillator 15 is used.
It is possible to freely set, for example, changing the band limiting frequency of the audio signal by changing the frequency to another frequency. Further, it will be apparent that the present invention can be an analog signal processing device.

(Effects of the Invention) As described above, according to the present invention, the same processing as in the prior art is used to reduce the number of arithmetic processing elements, so that noise due to rounding error is reduced, the arithmetic speed is increased, and the apparatus is reduced. It is extremely effective in providing a confidential device that reduces the cost of.

[Brief description of drawings]

1 (a) and 1 (b) are a block diagram showing an embodiment of the present invention and a spectrum diagram for explaining the operation thereof, and FIG.
(a) and (b) are a block diagram showing a conventional secret-talking device and a spectrum diagram for explaining the operation thereof, and FIG. 3 is a block diagram showing a modified example of the present invention. 5 ... Audio signal input terminal, 6, 9, 12 and 15 ... Oscillator, 7, 10, 13 and 16 ... Multiplier, 8 and 17 ...
… Low-pass filter, 14 …… High-pass filter, 11 …… Adder,
18 ... Controller.

Claims (4)

[Claims] 1. A first frequency conversion circuit that divides an input signal into two parts, one of which has a local frequency of _m and extracts a component of _m or less, and a local frequency of _m + _k ( _m > _k ).
Is input to the first route in which the mixing circuit is connected in series, and the other is a second frequency conversion circuit for extracting a frequency component of _m or more with the local frequency being _k and the local frequency is _m ′ ( _m ≧ _m ′). ) Is input to a second route connected in series with each other, the outputs of the two routes are added, and then only a component of _m or less is extracted from the added signal to obtain a desired signal. A secret-talking device characterized by being configured as described above. 2. The secret communication device according to claim 1, wherein _k of the local frequencies is irregularly or regularly changed in a range of 0 < _k < _m . 3. A first frequency conversion circuit which divides an input signal into two parts, one of which has a local frequency of _m and extracts a component of _m or less, and a local frequency of _m − _k ( _m > _k ).
Is input to the first route in which the mixing circuit is connected in series, and the other is a second frequency conversion circuit for extracting a frequency component of _m or more with the local frequency being _k and the local frequency is _m ′ ( _m ≧ _m ′). ) Is input to a second route connected in series, the outputs of the two routes are added together, and then only the component of _m or less is extracted from the added signal. , A secret-talking device characterized by demodulating a signal subjected to secret-talk processing. 4. A first frequency conversion circuit that divides an input signal into two parts, one of which has a local frequency of _m and extracts a component of _m or less, and a local frequency of _m − _k or _m + _k.
A second frequency conversion circuit for inputting to a first route in which a mixing circuit capable of switching to ( _m > _k ) is connected in series, and for extracting the frequency component of _m or more with the other having a local frequency of _k and the local frequency. _M ′ ( _m ≧
_m ') is input to a second route connected in series with a second mixing circuit, and the outputs of the two routes are added together, and then only the components of _m or less are extracted from the added signal. As a result, the confidential communication device can be switched between the confidential transmission circuit and the demodulation circuit for the signal subjected to the confidential processing.