JPH0349221B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a privacy device for preventing a third party from understanding the contents of a conversation in a communication device that transmits voice by emitting radio waves.

Conventionally, this type of device divides the audio spectrum into, for example, five equal subbands, and swaps each subband and inverts the spectrum. There are two methods: changing the audio signal, or encoding the audio signal and converting the code to create a confidential message. Among these methods, the method of spectrum inversion replacement has the disadvantage that the sense of rhythm of the voice is transmitted as is, and confidentiality is low, while the method of changing the replacement on a time basis and the method of encoding the audio transmit the synchronization signal on the transmitter side. This system had the disadvantage that it was necessary to synchronize the audio waveform with the receiving side, and if the audio waveform was replaced in a certain time unit, synchronization noise would occur due to waveform discontinuity.

The present invention aims to solve the drawbacks of such prior art, and without using a synchronization signal,
It is characterized by erasing the rhythmic feeling of the secret voice,
The object of the present invention is to provide a confidential communication device that can be used even on communication lines with low quality or strong fading. Embodiments of the present invention shown in the drawings will be explained in detail below.

In FIG. 1, 1 is a switcher for switching between microphone input and confidential voice input, 2 is an amplifier, 3 is a mixer circuit that converts the microphone input or confidential voice input into an intermediate frequency, 4 is an intermediate frequency bandpass filter, 5
is a mixer circuit that returns the intermediate frequency to baseband,
6 is a delay element; 7 is a switch that switches between a delayed signal passing through the delay element 6 and undelayed audio; 8 is a variable gain amplifier; 9 is a low-pass filter that removes harmonic components; 10 1 is a power amplifier; 11 is a switch for switching between secret message output and secret message reproduction output; 1;
Reference numeral 2 is a local oscillator having eight outputs for converting baseband signals into intermediate frequencies, and reference numeral 13 is a local oscillator for returning each intermediate frequency signal to baseband, which oscillates 16 frequencies and outputs 8 of them.

14 is a delay element controller that controls the delay time of each signal returned to the baseband. 15 is a variable gain amplifier controller.

Next, to operate this, first switch the switch 1 to the microphone input side and apply the baseband audio signal to the amplifier 2. The output of the amplifier 2 is divided into eight equal parts according to the frequency band by a mixer circuit 3 and an intermediate frequency bandpass filter 4. This frequency relationship assumes that the audio input is from 300Hz to 2700Hz, and for each 300Hz band there are 300
_S1 from Hz to 600Hz, from 600Hz to 900Hz
_S2 , similarly from 2400Hz to 2700Hz, _S8 , intermediate frequency is _IF , output of local oscillator 12 is _1L1 to _1L8
Then, _IF = _SN + _1LN ...(1) _1LN = _IF -300×N...(2) (Here, N can be expressed as an integer from 1 to 8, the number of band divisions.

For example, if we assume that _IF = 8000~8300Hz,
The output _1L8 of the local oscillator 12 for the baseband signal _S8 from 2400Hz to 2700Hz is obtained by substituting the lower frequency of _IF 8000Hz into equation (2), _1L8 = 8000 - 300 x 8 = 5600 [Hz]. Also, equation (1) is (8000-8300) = (2400-2700) + 5600 [Hz].

Next, the intermediate frequency of the intermediate frequency hand-pass filter 4 is returned to baseband audio using the local oscillator 13 and the mixer circuit 5, and the frequency of the local oscillator 13 at this time performs frequency swapping and spectrum inversion. This frequency relationship is shown below.

Here, the signal frequency-converted and returned to the baseband is converted from ' _S1 to ' _S8 for each 300Hz band from the low range, and the local oscillator output is converted from ' _S8 from _2L1 to convert ' _S1 . _{If we take 2L8 to _} Here, N is an integer from 1 to 8 used in formulas (1) and (2), and M is an integer from 1 to 8 as well.

When exchanging frequencies, for N=1, M should be an integer other than 1 up to 8, and similarly for N=2, M should be an integer other than 2 from 1 to 8, and the frequency in this case The permutation combinations are permutations of integers 1 to 8, so 8! =40320
be.

Further, to perform frequency inversion, it is sufficient to select + from ± in equation (4), and if - is selected, frequency inversion is not performed. Depending on whether or not the divided bands are inverted, there are 2 ⁸ = 256 combinations, and in this method there are a total of 8! ×
There are 2 ⁸ = 10321920 permutations. The method up to this point is known as the normal spectrum inversion replacement method.

Next, for each mixer output of the mixer circuit 5, the delay element 6, the switch 7, and the delay element controller 14 are used to determine the delay time of the audio signal of each subband, and the variable gain amplifier controller 15 is used. do,
The amplitude of each subband is determined, the signals separated into each band are combined by the output of the variable gain amplifier 8, the spurious components are removed by the low-pass filter 9, and the signals are amplified by the power amplifier 10. Then, the switch 11 outputs the private message output to the transmission line.

Next, we will discuss how to reproduce the audio with the secret message back to the original audio.

When the switch 1 is switched to the confidential speech input side and the confidential speech audio signal is applied to the amplifier 2, the output of the amplifier 2 is divided into eight equal parts according to the frequency band by the mixer circuit 3 and the intermediate frequency bandpass filter 4. The frequency relationship here is 300Hz to 600Hz as in the case of telling the secret story mentioned above.
The band goes to IF1, the 600Hz to 900Hz component goes to IF2,
Similarly, 2400Hz to 2700Hz is converted to IF8.

Next, the local oscillation frequency of the local oscillator 13 is selected using the local oscillator 13 and the mixer circuit 5 so that the intermediate frequency of the intermediate frequency bandpass filter 4 becomes the spectrum of the original audio signal. This reproduction method is also already known.

Next, for each mixer output of the mixer circuit 5, the delay element 6, the switch 7, and the delay element controller 14 are used, and each subband is delay element controller 1.
Determine the control output of 4. At this point, the audio signals of each band return to their original spectra, and also return to the original audio signals that match in time. These signals are returned to their original amplitude by a variable gain amplifier 8, combined with the amplifier output of the variable gain amplifier 8, spurious components are removed by a low-pass filter 9, and the power amplifier 10 amplifies the signal, and a switch 11 outputs the private conversation reproduction output to the transmission line.

FIGS. 2A to 2F show the spectrum and time delay relationship for the secret message operation and the secret message playback operation in the above-mentioned embodiment of the secret message device according to the present invention.
Figure A represents the microphone input audio as a spectrum at time T=to. Here, B1 to B8 are the names of bands to be divided into bands. B1 is 300Hz to 600Hz band, B2 is 600Hz to 900Hz band, similarly B8 is 2400
The band ranges from Hz to 2700Hz. Figure 2B shows the spectrum obtained by performing frequency inversion and swapping on this spectrum.
Shown below. Here, B1 to B8 with dashes indicate spectrum inversion, for example, the original audio is in the B5 band 1500Hz to 1800Hz frequency band.
This shows that the frequency has been swapped between 300Hz and 600Hz. Figure 2C is the spectrum shown in Figure 2B that has been delayed.
The 1800Hz component is delayed and the amplitude of each spectrum is set and sent. FIG. 2D shows the received confidential speech with the same spectral time relationship as FIG. 2C. Next, when the frequency regeneration operation is performed, the spectrum returns to the original audio signal as shown in Figure 2E, but the time is B2,
B3 and B5 bands are behind. Figure 2 F shows the result in which the spectrum remains as it is, but the delay playback operation and amplitude playback operation are performed to delay the B1, B4, B6, B7, and B8 bands, and all bands are delayed from the microphone input by a delay time _t3 . In this case, the original microphone input audio can be played back.

Since the present invention is such a system, by selectively performing a time delay operation on each band, which is a feature of the present invention system, it is possible to combine the hidden secrets of the usual frequency inversion and replacement system, in this case by dividing the frequency into 8. 2 ⁸
×8! = 256 times more combinations than 10321920, or 2 ⁸ × 2 ⁸ × 8! ≒2.6×10 ⁹ is obtained, and the periodicity of the voice is disturbed even for an eavesdropper who does not have a secret speech device, so the periodicity of the speech is disturbed compared to the conventional frequency inversion and replacement method, which retains a sense of the rhythm of the speech. This significantly improves performance. In addition, the present invention does not generate click noise that occurs when waveforms are discontinuous due to signal separation and synthesis, as in the conventional time-switching method, so the naturalness of the reproduced audio is maintained. be done.

Therefore, in shortwave communications, it is difficult to synchronize or synchronize, such as long-distance communications with poor S/N ratios or lines with strong fading. Even in such cases, by using the present invention, it is possible to perform highly confidential communication.

As explained above, according to the present invention, it is possible to change the secret speech over time and disturb the rhythm of the speech without using a synchronization signal. It is possible to carry out highly confidential communication without providing any means of secrecy, and it has remarkable effects such as being able to be used even on lines with low S/N ratios and lines with strong fading. .

[Brief explanation of drawings]

FIG. 1 is a block circuit diagram showing one embodiment of the present invention, and FIGS. 2A to 2F are graphs showing the spectrum and time delay relationship of the audio signal in this embodiment. 1...Switcher, 2...Amplifier, 3...Mixer circuit, 4...Intermediate frequency band pass filter, 5...
Mixer circuit, 6...Delay element, 7...Switcher, 8
...Variable gain amplifier, 9...Low pass filter,
10...Power amplifier, 11...Switching device, 12...
Local oscillator, 13... Local oscillator, 14... Delay element controller, 15... Variable gain amplifier controller.

Claims (1)

[Claims] 1. A first mixer of the same number as the number of divisions into which the frequency band of the audio signal is divided into equal frequency divisions into which the audio signal or the confidential audio signal is input, and the audio signal or the confidential audio signal for each of the frequency bands that are divided into the frequency equal parts. a first local oscillator that generates the same number of local oscillation frequencies as the number of divisions for converting into the same intermediate frequency band and supplies them to each of the mixers; The same number of intermediate frequency bandpass filters as the number of divisions to be passed, and second mixers of the same number as the number of divisions, each receiving a signal of the same intermediate frequency band that has passed through each of the intermediate frequency bandpass filters; converting the signal in the same intermediate frequency band into an audio frequency signal different from the original audio spectrum when inputting the audio signal, and into a signal with the original audio spectrum when inputting the confidential audio signal; a second local oscillator that generates the same number of local oscillation frequencies as the number of divisions and supplies them to each of the second mixers; and a second local oscillator that temporally delays the signal output from each of the second mixers. the same number of delay circuits as the number of divisions, the same number of switchers as the number of divisions for switching between signals that have passed through the delay circuits and signals that have not passed through the delay circuits, and a delay element controller that controls each of the switchers; variable gain amplifiers having the same number as the number of divisions, each inputting a signal from each of the switching devices, and a variable gain amplifier controlling the gain of each of the variable gain amplifiers to respectively control the amplitude of the signal from each of the switching devices. a gain amplifier controller; and a low-pass filter that synthesizes the same number of output signals as the division number from the variable gain amplifier, removes unnecessary waves such as harmonics, and outputs the secret speech signal or the reproduced speech signal. What is claimed is: 1. A confidential communication device, characterized in that it is configured to improve the confidentiality of voice by changing the delay, amplitude, and frequency of signals for each frequency band of voice.