RU2652791C1 - Method of the low-visible radar stations probing radio signals recognition - Google Patents

Abstract

FIELD: radio engineering and communications.

SUBSTANCE: invention relates to radio engineering, namely to methods of the low-visible radar stations probing radio signals type and modulation parameters recognition. It can be used for recognition of the low-visible radar stations radio signals with a complex time-frequency structure. Said result is achieved due to: setting the reference radio signals, forming the reference radio signals time-frequency portrait by performing of the Choi-Williams frequency-time conversion operation over a plurality of the reference radio signals digital samples, forming the received radio signal to be recognized time-frequency portrait by performing a Choy-Williams frequency-time conversion operation over the plurality of the received radio signal digital samples; with the received to be recognized radio signal time-frequency portrait, performing an additional noise suppression operation using a discrete S-conversion (Stockwell conversion), comparing the to be recognized radio signal time-frequency portrait with the reference radio signals time-frequency portrait; deciding whether the to be recognized radio signal belongs to one of the reference radio signals.

EFFECT: increase in the radio signals recognition probability.

1 cl, 5 dwg

Description

The invention relates to pattern recognition, and in particular to methods for recognizing radio signals, in particular to methods for recognizing the type and modulation parameters of probing radio signals of stealth radars. The method can be used in technical means that perform the task of identifying radio signals of stealth radars with a complicated time-frequency structure.

A known method of signal recognition [Omelchenko V. Recognition of signals by the power spectrum in the optimal Karunen-Loeva basis. - Proceedings of the universities MB and MTR of the USSR. Ser. Radio electronics, 1980, No. 12, pp. 11-18], in which the energy spectrum of the signal is calculated, then the Karunen-Loev transform is performed on it and based on the obtained characteristics, the signals are first selected for useful and interfering, and then, in the case of useful signal, carry out its comparison with the existing reference classes and assignment to one of them.

The disadvantage of this method is the relatively low probability of correct recognition ¹ (The probability of correct recognition is the relative frequency of making the right decision when assigning the received radio signal to one of the reference classes. The correct recognition event is the opposite (additional) to the error recognition event (P _dec = 1-R _osh ) - see J. Tu, R. Gonzalez. Principles of pattern recognition. Transl. from English. - Moscow: Mir, 1978 // P. 142-152) of signals having similar spectra, which is due to weak contrast the density of the formed recognition features.

There is also known a method for recognizing radio signals based on the singular decomposition of the pseudo-frequency-time-distribution (pseudo-CWR) Wigner - Wille [Marinovic N., Eichmann G. An expansion of Wigner distribution and its applications. - Proc. IEEE ICASSP-85, 1985, pp. 1021-1024], in which the energy distribution matrices of the reference radio signals are first formed on the basis of the Wigner-Ville pseudo-FWM, their spectral decomposition is performed, the parameters of the reference radio signals are formed, then the recognized radio signal is received, sampled and quantized, the energy distribution matrix of the received radio signal is generated, they identify the signs of the received radio signal, compare them with the parameters of the reference radio signals and identify the received radio signal from the results of the comparison.

The disadvantage of this method is the relatively low probability of correct recognition of radio signals with a complex time-frequency structure, as well as radio signals when exposed to noise and interference, which is due to the peculiarities of the Wigner-Ville [Cohen L., time-frequency distributions used for recognition of pseudo-CVMs. Review // TIIER, 1989, v. 77, No. 10. S. 72-121]. The probability of correct recognition is reduced due to the appearance of interference background and false power peaks in the Wigner-Ville pseudo-FWM, which distort the real picture of the signal energy distribution in frequency-time coordinates.

The closest analogue in technical essence to the claimed one is the method of recognition of radio signals according to RF patent No. 2356064 C2, dated 05/20/2009. In the closest analogue, reference radio signals are preliminarily set; performing a frame wavelet transform operation on a plurality of discrete and quantized samples of reference radio signals; generate energy distribution matrices of the reference radio signals after excluding the insignificant wavelet coefficients from the sequences of normalized and ranked wavelet coefficients; form feature vectors of reference radio signals by line-by-line concatenation of elements of energy distribution matrices; performing a frame wavelet transform operation on a plurality of discrete and quantized samples of a received recognizable radio signal; form a matrix of the energy distribution of the recognized radio signal after excluding the insignificant wavelet coefficients from the sequences of normalized and ranked wavelet coefficients; form a feature vector of a recognizable radio signal by line-by-line concatenation of elements of an energy distribution matrix; comparing the feature vectors of the recognizable radio signal with the feature vectors of the reference radio signals; make decisions on whether the recognized radio signal belongs to one of the reference radio signals.

The disadvantage of the prototype method is the relatively low probability of correct recognition of radio signals with a low signal to noise ratio (from minus 4 to 4 dB).

The purpose of the claimed technical solution is to develop a method for recognizing radio signals, which increases the likelihood of recognition by eliminating the noise component in the generated time-frequency porter.

The claimed technical solution expands the arsenal of funds for a similar purpose.

This goal is achieved by the fact that in the known method of recognizing radio signals, instead of the operation of the frame wavelet transform [Dyakonov V. Wavelets. From theory to practice. - M .: SOLON-R, - 2002, 448 S. // S. 106] to obtain the time-frequency portrait, the Choi-Williams transform [G.J. Upperman, T.L.O. Upperman, D. Fouts, and P. Pace, '' Efficient timefrequency and bi-frequency signal processing on a reconfigurable computer, '' IEEE Asilomar Conf., Pp. 176-180, 2008] followed by noise reduction using the discrete S-transform (Stockwell transform) [Y. Wang and J. Orchard, '' The Discrete Orthonormal Stockwell Transform for Image Restoration, '' in 16th IEEE International Conference on Image Processing, 2009, pp. 2761-2764; H. Huang, F. Sun, '' Medical-Image Denoising and Compressing Using Discrete Orthonormal Stransform, '' in 2nd International Conference on Electrical, computer Engineering and Electronics (ICECEE 2015), 2015, vol. 291, no. Icecee, pp. 291-296].

To identify the signals, a criterion for the difference (similarity) of the obtained time-frequency portrait with a standard is used. As a metric, a two-dimensional correlation coefficient is used:

[Theoretical foundations of the time-frequency analysis of short-term signals: Monograph / S.N. Agievich, S.V. Janitors, A.M. Kudryavtsev. - SPb .: YOU, 2010, p. 240 // p. 140].

A recognized radio signal is considered incident to the reference radio signal if the two-dimensional correlation coefficient of their time-frequency portraits is maximized.

Thanks to a new set of essential features, the claimed method provides an increase in the probability of recognition of radio signals with a low signal-to-noise ratio (from minus 4 to 4 dB) due to the fact that the discrete S-transform performs the noise reduction operation better than the frame wavelet transform [S. Saoud, M. Ben Naser New Speech Enhancement based on Discrete Orthonormal Stockwell Transform // International Journal of Advanced Computer Science and Applications, Vol. 7, No. 10, 2016, p. 196; Y. Wang, '' Efficient Stockwell transform with applications to image processing '', PhD thesis, University of Waterloo, Ontario Canada, 2011]. This increases the likelihood of recognition.

The claimed method is illustrated by drawings, which show:

FIG. 1 is a generalized structural diagram of a recognition process;

FIG. 2 is an example of a frequency-time portrait of a reference radio signal obtained based on a Choi-Williams transform;

FIG. 3 is an example of a frequency-time portrait of a recognized radio signal prior to performing a noise reduction operation;

FIG. 4 is an example of a time-frequency portrait of a recognizable radio signal after performing a noise reduction operation using S-conversion;

FIG. 5 is a generalized graph of the dependence of the probability of correct recognition on the signal-to-noise ratio (SNR) for the prototype method and the proposed method.

распознаваемого радиосигнала

и его последующая очистка от шума, в результате чего формируется частотно-временной портрет

; сравнение частотно-временного портрета распознаваемого радиосигнала с частотно-временными портретами эталонных радиосигналов; принятие решения о принадлежности распознаваемого радиосигнала к одному из N эталонных радиосигналов.In general, the recognition process (Fig. 1) includes the following procedures: the formation of time-frequency portraits {P ₁ ... P _N } for the set {S ₁ (t) ... S _N (t)} of reference radio signals; formation of a frequency-time portrait

recognizable radio signal

and its subsequent cleaning from noise, as a result of which a time-frequency portrait is formed

; comparing the time-frequency portrait of a recognizable radio signal with the time-frequency portraits of reference radio signals; making a decision on whether the recognized radio signal belongs to one of the N reference radio signals.

The claimed method allows due to better noise reduction to increase the probability of recognition of radio signals and thereby resolve the contradiction due to the need to increase the probability of recognition of radio signals at low signal-to-noise ratios (from minus 4 to 4 dB) with an increasing volume of energetically secretive radio signals.

The implementation of the claimed method is explained as follows.

Pre-set n reference radio signals, the number and types of which cover the possible number and types of real radio signals to be recognized in accordance with the task for their recognition. Then, a set of operations is performed to form a time-frequency portrait of each n-th reference radio signal, where n = 1, ..., N. For this, each reference radio signal, presented in digital form, is subjected to the time-frequency Choy-Williams transformation, which is described in [Gj Upperman, T.L.O. Upperman, D. Fouts, and P. Pace, '' Efficient timefrequency and bi-frequency signal processing on a reconfigurable computer, '' IEEE Asilomar Conf., Pp. 176-180, 2008]. In FIG. 2 shows an example of a time-frequency portrait of a reference radio signal. Over the time-frequency portrait of the received recognizable signal, an additional noise reduction operation is performed using the discrete S-conversion described in [S. Saoud, M. Ben Naser New Speech Enhancement based on Discrete Orthonormal Stockwell Transform // International Journal of Advanced Computer Science and Applications, Vol. 7, No. 10, 2016, p. 196; Y. Wang, '' Efficient Stockwell transform with applications to image processing '', PhD thesis, University of Waterloo, Ontario Canada, 2011]. The threshold processing for noise reduction is performed in accordance with the rules set forth in [Smolentsev N.K. Fundamentals of the theory of wavelets. Wavelets in MATLAB. - M.: DMK Press, 2005, p. 304 // p. 218]. The specific threshold value depends on the type of signal, the required recognition probability, and the time taken for recognition. In FIG. Figure 3 shows an example of a time-frequency portrait of a recognized radio signal before performing a noise reduction operation at a signal-to-noise ratio of 0 dB.

The introduction of an additional operation of noise reduction using S-conversion leads to an increase in the probability of recognition of a radio signal at a low signal to noise ratio (from minus 4 to 4 dB). In FIG. 4 shows an example of a time-frequency portrait of a recognizable radio signal after performing a noise reduction operation at a signal-to-noise ratio of 0 dB.

A comparative assessment of the probability of correct recognition of R _dec using the claimed method and the prototype method was carried out using mathematical modeling on a computer in MATLAB. As recognizable signals, signal models of subtle radar stations with frequency and phase modulation were used [Voinov, D.S. Signals of stealth radar / D.S. Voinov [et al.]; Cherepovets: CHVVIURE, 2015 - 13 pp., Ill. - Dep. at TsVNI of the Russian Federation Defense Ministry on October 14, 2015, No. B8677]: multi-frequency signal according to the Costas code (frequency arrays (100 500 400 600 200 300) MHz, (500 400 600 200 300 100) MHz (200 400 800 500 1000 900 700 300 600 100) MHz), phase-manipulated (according to Barker codes (7, 11, 13), P1, P2, P3, P4, T1, T2, T3, T4, Frank).

The radio signals are normalized to the average power level. The parameters of the standards were formed for 100 samples of each signal. To assess the probability of correct recognition of P _dec , the Monte Carlo method was used for 100 samples for each signal for various ratios of the signal power P _c and the noise power R _w in the range from minus 4 to 20 dB (Fig. 5). The results of comparative calculations showed that the probability of correct recognition of radio signals by the claimed method is higher than the prototype method (Fig. 5) in the region of low (from minus 4 h - 4 dB) values of R _s / R _w in 1.4-3.1 times . Which indicates the possibility of achieving a technical result when using the claimed method is an increase in the probability of recognition of radio signals.

Based on the above description, technical means can be manufactured from known components using known technological equipment in the electronics industry that performs the task of recognizing radio signals. Thus, the claimed method meets the criteria of the invention of "applicability".

Claims (1)

The method of recognition of radio signals, which consists in the fact that the reference radio signals are pre-set, their time-frequency portraits are formed, after which a recognized radio signal is received, its time-frequency portrait is formed; performing a noise canceling operation on a frequency-time portrait of a recognizable radio signal; then, the time-frequency portrait of the recognized radio signal is compared with the time-frequency portrait of the reference radio signals; decide on whether the recognized radio signal belongs to one of the reference radio signals, characterized in that for the formation of time-frequency portraits, the Choi-Williams transform is performed on digital samples of the reference and recognizable radio signals, and the noise reduction operation on the time-frequency portrait of the recognized radio signal is performed using S- transformations.

Images