Field of the invention

The invention is related to the field of generating artificial interference for masking electromagnetic verbal information leak channels, and can be used for provision of verbal information protection in digital radio communication lines under conditions of the necessity to comply with electronic apparatus compatibility requirements, preset energy characteristics of the radio transmission system, and probability of erroneous reception.

Technology intelligence methods are based on the fact that information processing, storage and transmission processes always give rise to generation of stray dispersion fields of various physical nature, or to induction of currents and voltages in non-informational physical circuits. Said circumstances cause increasing the requirements imposed on performance of means and methods that provide information protection. Therefore, development of methods for practical protection of verbal information against leaking through various technical channels is a crucial task.

Background of the invention

The problem of microphonics effect occurrence is widely known, in particular, it is known that an electromagnetic technical verbal information leak channel (sometimes called a feedthrough channel) may be formed due to the presence of random electro-acoustic transducers, so called random microphones, in individual components of the technical system, which is described, e.g., in the educational guidance “Information protection by technical means” by Y.F. Katorin, A.V. Razumovsky, A.I. Spivak / Ed. by Y.F. Katorin - St. Petersburg: NIU ITMO, 2012, pp.174-178, in the educational guidance“Information protection against leaking through technical channels” by V.K. Zhelezniak - St.Petersburg: GUAP, 2006, pp.23, 44, 63, and in the application manual «Capturing and Measuring Microphonics and Phase-Hits with the RTSA» / Application Note 37W-18577-1 - Tectronix, 2005, pp. 1-19.

Disclosure of the invention

A leak channel generation mechanism may be as follows. In the course of a verbal message transmission, acoustic pressure acts not only on the headset microphone but also, for instance, on reference generator components in the radio transmitting device, which results in undesirable narrow-band modulation of the generator output frequency by the law of acoustic pressure variation. Further, as a rule, in the frequency conversion process, the undesirable modulation products penetrate the main transmission channel of the radio device, they are amplified, and, after radiation into the air, the possibility of unauthorized interception of the verbal message arises. One of approaches to ensuring protection of verbal information against leaking through an electromagnetic channel is decreasing the level of acoustic pressure on electronic components susceptible to the microphonics effect by means of damping and sound insulation. For instance, such approaches include the methods disclosed in the patents [1, 2]. The principal drawback of those methods is the fact that they are directed just to decreasing the level of the undesirable modulation products and are not The principal drawback of those methods is the fact that they are directed just to decreasing the level of the undesirable modulation products and are not able to eliminate them completely. In addition, application of sound-insulating materials in radio transmitting devices has a detrimental effect on their performance, since it leads to a change in temperature operating conditions of key radio components.

Also known is a wide class of passive transmitter protection methods based on the use of screening and filtration of electromagnetic pickups and interferences. For instance, the prior art methods disclosed in the patents [3, 4] may be assigned to that class. Said methods suffer from the same drawbacks as the sound insulation methods. In addition, the above-mentioned methods are not efficient in elimination of pickups penetrating the main transmission channel.

Another line in information protection by technical means is generation of artificial interference in electromagnetic information leak channels by masking signal generators having a power higher than the power of electromagnetic information leak signals. This decreases significantly the possibility of information detection and reception. The prior art includes the methods [5-7] for protection of radio communication lines by generation of masking interference in a broad frequency band, the interference power exceeding that of electromagnetic signals of stray electromagnetic radiation and pickups.

The drawback of said methods is creation of unintentional interference for a wide variety of radio-electronic devices located in a close vicinity of the masking radiation generator. The prior art method most similar in its technical essence to the proposed one is a method for masking an electromagnetic verbal information leak channel in digital radio communication lines described in the article [8], which is taken as a pertinent prior art.

The pertinent prior art method is as follows.

A noise (masking) signal in the required frequency band is formed, the signal bandpass filtering is carried out, the signal is pre-amplified, modulated by a low-frequency noise signal, amplified, radiated into the air and excites the antenna. In this process, summation of powers of the masking signal and the undesirable modulation products takes place in the air.

The method is implemented as follows. A noise signal comes to the input of a bandpass filter from a low-power noise generator that generates a noise signal in the required frequency band. A noise signal in the preset frequency band appears at the output of the bandpass filter. Then the noise signal arrives at a preamplifier where it is amplified and fed to the first input of the modulator. The preamplifier has also an input for receiving a signal from a noise output power control device; this signal defines power of the signal at the preamplifier output. The second input of the modulator receives a signal from a noise contamination bandwidth control device. From the modulator output, the noise signal formed by means of the modulation process, arrives at a power amplifier. Here, it is amplified up to the required level and transmitted to an antenna. The second input of the modulator receives a low-frequency noise signal from the noise contamination bandwidth control unit, e.g., a noise-like signal formed by some automatic machine. Masking of undesirable modulation products by a verbal signal is provided by addition, in the air, of powers of the masking signal and the signal radiated by the radio transmitting station. The prior art method does not provide any service information exchange with every radio station, therefore, the noise generator cannot receive from them any information on turning on a useful signal radiation, frequency and bandwidth of the radiated signal. This, in turn, will lead to deterioration of the radio station receiver sensitivity, since the noise generator is radiating noise at signal reception moments. Also, the device implementing the prior art method will create interference for radio-electronic apparatus that use the adjacent channels, since it is necessary to mask with noise the entire range of operating frequencies if the radio station changes frequency channels during operation, or to mask the widest possible frequency band if the radio station uses different signal bands for transmission.

The method works efficiently in case of the absence of strict electromagnetic compatibility requirements for radio electronic means, and prevents or decreases significantly the electromagnetic leak of information, including verbal information.

The drawback of the prior art method consists in the fact that it implies operation of an outside noise radiator to an individual antenna(s). Difference of actual detection patterns of antennas of the radio station and the noise radiator, which is caused, among other factors, by location of the antennas and influence of nearby objects, does not allow provision of proper-quality masking over all azimuthal directions, which constitutes potential vulnerability.

The objective of the proposed invention is development of an efficient method for masking an electromagnetic verbal information leak channel in digital radio communication lines that does not violate electromagnetic compatibility requirements for radio-electronic apparatus, and that deteriorates power characteristics of the radio system insignificantly. For attaining the above objective, the method for masking electromagnetic verbal information leak channels in digital radio communication lines comprises generation of a masking noise signal in the required frequency range, filtration of the signal, addition of the masking signals and undesirable modulation products, wherein, according to the invention, addition of the masking noise signal is performed inside the radio station, without radiation into the air or to an antenna external in respect of the transmitting radio station; during filtration, a masking noise signal band coinciding with the bandwidth of the radiated useful signal is isolated; in scaling, the masking noise signal level is set by the criterion of not increasing the probability of the received signal error.

The method according to the invention consists in the following.

The noise signal source generates a noise signal having the required spectral characteristics. A low-pass filter restricts the masking signal frequency band to the band of the useful signal being transmitted. The masking signal level is scaled so that its addition with the transmitted useful signal will not increase the probability of the received signal error. The formed masking signal is added to the transmitted useful signal inside the transmitting radio station, the signal/noise ratio for undesired modulation products being decreased significantly. Signal amplification and radiation into the air by antenna excitation is performed when undesired products of modulation by the verbal signal are already masked with noise.

In other words, the problem of providing adaptive masking of electromagnetic verbal signal leak channel is solved by using several operations: finding a simple and cheap way of forming a masking noise signal in the band of the radiated useful signal, and restricting the masking signal level to a level that does not lead to a significant increase in probability of the received signal error.

The first solution consists in software implementation of the processes of the masking signal (noise) generation, filtration, level setting and superimposing. In this case, the masking noise generator is located directly inside the radio transmitting device and has an access to all required information on radiation activation, on the radiated signal frequency and bandwidth. Thus, it becomes possible to superimpose the masking noise at the stage of modulation by the useful signal already. It should be noted that introduction of a masking noise generator in digital radio stations including a signal processor or PLIC as a digital processing unit comes down to simple and cheap upgrading of configuration files.

The second solution consists in calculating a maximum possible power of the masking noise that, when added to the useful signal band, practically does not increase the probability of the useful signal reception error; however, for undesirable modulation products, the signal/noise ratio changes significantly, and reception of such modulation products becomes difficult or impossible.

Many digital software-defined radio transmitting devices where speech masking means can be used include integrated circuits of a signal processor or a field programmable logic device that serve as a basis, in particular, for a functional signal forming and digital processing unit. The simplest implementation of the method implies, for instance, software modification of implementing the signal forming and digital processing unit. Brief description of the drawings

A structural diagram of the device that implements the proposed method is shown in Fig. 1, where the following is denoted:

7 - a signal forming and digital processing unit including: 1 - a noise signal digital source module, 2 - a digital low-pass filter module, 3 -a digital signal scaling module, 4.1 - 4.2 - a digital summing module, 5 -an interference-resistant coding module, 6 - a digital modulator module.

The signal forming and digital processing unit 7 includes the following components connected in series: an interference-resistant coding module 5 and a digital modulator module 6, two outputs of the latter being connected to respective inputs of the first 4.1 and second 4.2 digital summing modules. Connected in series are a digital noise signal source module 1, a digital low- pass filter module 2 and a digital signal scaling module 3 having its output connected to respective inputs of the first 4.1 and the second 4.2 digital summing modules, outputs of which are the first and second outputs of the unit 7, the input of the unit 7 being the input of an interference-resistant coding module 5.

Fig. 2 illustrates the method of masking the electromagnetic channel of verbal information leak in the frequency range.

An embodiment of the invention

An input verbal signal, digitized, encoded by a vocoder and ciphered in a scrambler (not shown in Fig. 1) comes into the signal forming and digital processing unit 7, is encoded in the interference-resistant coding module 5 and arrives at the input of the digital modulator 6. At this stage, the signal may be free of undesirable modulation products; nevertheless, the digital noise signal source module 1 forms a masking noise signal that is filtered in the low-pass filter module 2 in order to restrict the noise signal band to a width not exceeding the width of the useful signal formed by the digital modulator 6. The digital signal scaling module 3 sets the level of the restricted-band masking noise signal by way of multiplying signal samples by a value calculated based on information about the level of undesirable products of modulation in feedthrough of the verbal signal and on requirements to magnitude of the erroneous reception probability. The masking signal thus formed is summed in the summing modules 4.1 - 4.2 with output signals of the digital modulator 6 and arrives at the outputs of the unit 7. Thus, preliminary noise contamination is provided in the signal band in which an electromagnetic verbal information leak channel could be formed at subsequent signal conversion stages.

Industrial applicability The proposed method can be easily implemented by means of software algorithms that are well known and widely used in the industry.

The digital noise signal source module 1, low-pass filter module 2, digital signal scaling module 3 and summing module 4 are built in the traditional way. The digital noise signal module 1 may be made by means of a linear- feedback shift register, e.g., in accordance with the algorithm and the source text from the book“Applied cryptography. Protocols, algorithms, source texts in C language”. Shnayer B. - Triumph, 2013. (section 16.2, p. 279).

The low-pass filter module 2 may be designed and built, e.g., in accordance with the algorithm from the book“Digital signal processing. Practical guide for engineers and researchers” Smith S. - Dodeca-XXI, 2012. (p. 334-336).

The digital signal scaling module 3 is implemented by way of multiplying by a constant, which is a basic operation like summing in the module 4, and does not require any special algorithm for implementation.

The method according to the invention has been introduced in a radio transmitting device by way of software implementation of the proposed algorithm. The test results have shown that masking is ensured for undesirable modulation products generated as a result of feedthrough of a verbal signal , and no interference occurs in adjacent frequency channels.

The cost of modifications and additional means turned out to be incommensurably lower than that of prior art solutions of the problem, and electromagnetic compatibility was significantly higher.

Below, we present a proof of efficiency of the method according to the invention.

In a particular case, let us consider the influence of masking noise addition to the signal band on deterioration of reception quality by an example of a coherent receiver of binary orthogonal frequency- shift-keying (FSK) signals. As is known, error probability in reception of such signal by a

coherent demodulator is described by the formula [9, p. 245]

where Q(x) is Gauss error integral function defined as

E _b is energy per binary information symbol; in this case, it coincides with energy of one transmitted symbol;

N ₀ -spectral noise density at the receiver input. In transmission of FSK signals, the receiver bandwidth W in Hertzs is usually equal to symbols transmission speed ; we can consider that TW = 1

[9, p. 254]. Then noise power N can be found as N = N ₀ · W , then

Power of the received signal S, respectively, is · W , then Let us express the ratio through signal and noise powers in the

received signal band at the receiver output.

With the set ratio of signal and noise powers at the input, we obtain the expression for error probability with the set ratio of signal and noise powers in the reception band:

Assume that the information transmission channel provides a maximum error of 5 % [10], then, according to [9, p. 1060], power ratio is 1.65, or SNR= 10 1g(1.65) -2.175 dB.

After adding a masking interference to the signal band, the noise power at the receiver input will increase by the masking signal power. Let us take the signal power as 1. Then the noise power is N = 1/1.65 = 0.60606. Let the masking interference have a power level 100 times lower than the signal, P _msk = 0.01. The masking interference and the noise are independent processes. Therefore, their sum will have power + N = 0.60706 , which will result in lowering the SNR value to a level of 1/ 0.60706 =1.647 (2.168 dB).

SNR decrease due to addition of the masking interference in the case under consideration will amount to 0.007 dB, which practically will not the received signal error probability. However, the signal/noise ratio for undesirable modulation products will change more noticeably. For instance, if, as a result of feed through, the parasite signal level at the transmitter output is -40 dBs (Fig. 2a) relative to the useful signal, then, after adding the masking interference of -20 dBs level (Fig. 2b), the feedthrough signal at the transmitter output will be lower by 20 dB than the level of the masking interference added, and reception of such signal will be very difficult or impossible.

Therefore, the method according to the invention provides concurrent masking of the electromagnetic verbal information leak channel and fulfilment of requirements for electromagnetic compatibility of radio- electronic means in digital radio communication lines with preset energy characteristics of the transmitting radio system and erroneous reception probability. Thus, the proposed changes in the method for masking the electromagnetic verbal information leak channel allow obtaining an undoubtedly significant technical result by simple means.

The attained technical result is concurrent masking of the electromagnetic verbal information leak channel and fulfilment of requirements for electromagnetic compatibility of radio-electronic means in digital radio communication lines with preset energy characteristics of the transmitting radio system and erroneous reception probability.

Sources of information 1. Inventor’s Certificate No.34635. A device for elimination of the microphone effect in superheterodyne receivers. Binshtok V.V., Grachev E.T. 28.02.1933.

2. US 20100246147. Microphonics suppression in high-speed communications systems. Y. Qin, F.S. Matsumoto, E. Tiongson. 16.11.2006.

3. US 4831211. EMI/RFI sealed microphonics isolation apparatus and methods. L. D. McPherson, E. Saenz. 08. 06.1988.

4. US 6239359. Circuit board RF shielding. F.L. Peter, F.W. Verdi.

11.05.1999. 5. Patent No. 2277758. A method for masking of electromagnetic verbal signals leak channels in sound amplifying apparatus. - Zhelezniak V.K. IPC H04K 3/00, H03B 29/00 - 8 pp. - Publ. 10.06.2006, Bull. No.16.

6. Patent No. 2360365. A radio masking device. -Ivanov V.P. IPC H04K 3/00. - 13 pp. - Publ. 27.06.2009, Bull. No.18. Patent No. 2493594. A method for protection of information being processed by computer means by way of noise contamination of stray electromagnetic radiation and pickups, an information protection device for implementation of the method. - Lepekha Y.P. IPC G06F 21/00, H04K 3/00. - 10 pp.— Publ. 27.06.2013, Bull. No.18. Zemlianukhin P.A. A multi-channel adaptive noise generator for masking stray electromagnetic radiations and pickups // Izvestiya Yu.FU. - Pp.82-93. Skliar B. Digital communication. Theoretical fundamentals and practical application. Edition 2, revised : Translated from English. - M.

: Williams Publishing House, 2003. - 1104 pp. Makoviy V.A. Comparative assessment of economical costs of communication suppression in the tactical link /V.A. Makoviy, V.I. Nikolayev, V.N. Povetko // Theory and technology of radio communication. - 2016. - No.3. - Pp.88-93. Grayver E. Implementing software defined radio. - Springer. 2013th Edition. - 2014. p. 270.