RU2546330C1 - Method for polarisation-sensitive radio monitoring of mobile objects - Google Patents

Abstract

FIELD: radio engineering, communication.

SUBSTANCE: method is carried out using novel operations for useful radio signal feedback adaptive processing, which increase sensitivity and the dynamic range when forming horizontal and vertical polarisation components of a two-component complex frequency-time image of radio signals scattered by objects in the analysed region of Doppler frequencies and time delays.

EFFECT: high probability of detecting remote and weakly scattering objects.

3 dwg

Description

The invention relates to radio engineering and can be used in monitoring systems of air, land and sea space using direct and scattered by mobile objects radio signals emitted by many uncontrolled and controlled transmitters of electronic systems for various purposes.

The technology of secretive detection and tracking of objects using natural radio illumination of targets created at a variety of frequencies by radio emissions from transmitters for various purposes: broadcast (VHF FM broadcasting, UHF digital television), information (communication) and measuring (control, navigation), have not yet received sufficient distribution, despite the fact that it can significantly increase the efficiency of detection, spatial localization and identification of a wide class of moving objects.

The received radio signal, as a rule, includes powerful direct radio signals and the signal components of the selected target radio illumination transmitter scattered from the earth's infrastructure. In addition, it contains time-delayed and shifted by the frequency of the Doppler shift signals scattered by objects, as well as signals of other uncontrolled sources operating at a frequency that matches the frequency of reception. For effective detection and accurate spatial localization of a wide class of objects (cars, ships, planes and unmanned aerial vehicles, helicopters, rockets, descent vehicles), high-quality selection of weak radio signals scattered from objects against the background of a powerful direct signal of the selected radio backlight transmitter, as well as against the background of signals other unwanted sources. In most typical situations, the noise level is 40-60 dB higher than the level of scattered signals.

Covert radar systems include a direct signal reception channel for the backlight transmitter and a reconnaissance channel.

Traditionally, in stealth radar systems, partial suppression of interference in the form of a direct signal from the backlight transmitter is achieved by minimizing side lobes, generating zero in the antenna radiation pattern, or adaptive spatial filtering of useful signals in the reconnaissance channel.

Additional suppression of the direct signal can be achieved by using an antenna with a polarization orthogonal to the polarization of the radio signal of the backlight transmitter in the reconnaissance channel.

However, the best characteristics of stealth radar systems can be achieved by using two reconnaissance channels with orthogonal polarizations. This is due to the fact that the signal scattered by the target, as a rule, has random polarization. As a result, incoherent summation of images in the coordinates “time delay (range) - Doppler frequency (speed)”, formed using radio signals of two orthogonal polarizations, provides an increase in the average signal-to-noise ratio compared to using a single fixed polarization. In addition, this increases the stability of the detection procedure to channel and inter-channel interference, as a rule, having a different polarization from the useful signal.

A known method of polarization-sensitive radio monitoring of moving objects [1], including the reception of radio signals scattered by moving objects of unknown polarization by a low-base antenna array consisting of orthogonally located antennas with combined phase centers, the formation of an ensemble of radio signals depending on the time and antenna number, synchronous conversion of the ensemble radio signals received into digital signals, the conversion of digital signals of pairs of opposite antennas into complex quadrature compositions -governing the dipole and quadrupole output signals, detect moving objects scattered radio signals and their arrival directions of signals of the quadrature components of the dipole and quadrupole output signals.

This method provides increased detection stability and spatial localization to polarization errors. However, this method belongs to the class of low-base direction finding methods, which is a fundamental limitation on the path to achieving the potential accuracy of spatial localization of moving objects.

A known method of polarization-sensitive radio monitoring of moving objects [2], free from these shortcomings and adopted as a prototype. According to this method:

use direct and scattered by mobile objects radio signals emitted by broadband transmitters of various electronic systems;

receive a lattice of N antennas components of horizontal and vertical polarization of the vector multipath electromagnetic field of direct and scattered radio signals;

synchronously transform the ensemble of radio signals received by the antennas into digital signals;

и рассеянные $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

сигналы для выбранных азимутально-угломестных направлений приема, где h и v - индексы компонент горизонтальной и вертикальной поляризации, которые совместно со значением азимутально-угломестного направления приема запоминают;digital signals are converted into two-component direct $$\tilde{s}=\left[\begin{array}{c}{\tilde{s}}^h\\ {\tilde{s}}^v\end{array}\right]$$

and scattered $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

signals for the selected azimuthal elevation directions of reception, where h and v are the indices of the components of horizontal and vertical polarization, which together with the value of the azimuthal elevation direction of reception are stored;

и $${\Vert {\tilde{s}}^v\Vert }^2$$

прямого сигнала;calculate and compare component energy $${\Vert {\tilde{s}}^h\Vert }^2$$

and $${\Vert {\tilde{s}}^v\Vert }^2$$

direct signal;

;choose the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

;

в матричный сигнал комплексной фазирующей функции A, включающий гипотетические сигналы, рассеиваемые каждым потенциальным объектом;convert the direct signal component with maximum energy $${\tilde{s}}_{max}^{h,v}$$

in the matrix signal of the complex phasing function A, including hypothetical signals scattered by each potential object;

for each selected azimuthal elevation direction, the scattered signal s is converted into the signals of the components of the complex time-frequency image h ^h = (A ^H A) ^-1 A ^H s ^h and h ^v = (A ^H A) ^-1 A ^H s ^v , where A ^H is a Hermitian conjugate matrix with A;

, где $$h_z^h$$

и $$h_z^v$$

- z-е элементы сигналов компонент h^h и h^v, определяют число рассеянных радиосигналов, по параметрам которых - значениям временной задержки, доплеровского сдвига и азимутально-угломестного направления приема - выполняют обнаружение и пространственную локализацию подвижных объектов.after which, according to local maxima of the sum of the squares of the modules of the elements of the components of the complex time-frequency image $${\left|h_z^h\right|}^2+{\left|h_z^v\right|}^2$$

where $$h_z^h$$

and $$h_z^v$$

- the z-th elements of the signal components h ^h and h ^v , determine the number of scattered radio signals, the parameters of which - the values of time delay, Doppler shift and azimuth-elevation direction of reception - perform detection and spatial localization of moving objects.

The prototype method implements a sufficiently effective detection and spatial localization of moving objects in the conditions of unknown polarization of the signals scattered by objects.

However, the prototype method, when generating the signals of the horizontal and vertical polarization components of a complex time-frequency image, uses operations based on the formation of a classical two-dimensional cross-correlation function, which, in addition to the main lobe, contains high side lobes masking the signals of distant and weakly scattering objects.

Thus, the disadvantage of the prototype method is the low probability of radio monitoring of distant and poorly scattering objects.

The technical result of the invention is to increase the likelihood of detection and proper spatial localization of distant and poorly scattering objects.

An increase in the probability of detection and proper spatial localization is achieved through the use of new adaptive processing operations with feedback on a useful radio signal, which provide increased sensitivity and dynamic range when forming components of the horizontal and vertical polarization of a two-component complex time-frequency image of radio signals scattered by objects in the analyzed region of Doppler frequencies and time delays.

и рассеянные $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

сигналы для выбранных азимутально-угломестных направлений приема, где h и v - индексы компонент горизонтальной и вертикальной поляризации, которые совместно со значением азимутально-угломестного направления приема запоминают, вычисляют и сравнивают энергию компонент $${\Vert {\tilde{s}}^h\Vert }^2$$

и $${\Vert {\tilde{s}}^v\Vert }^2$$

прямого сигнала, выбирают компоненту прямого сигнала с максимальной энергией $${\tilde{s}}_{max}^{h,v}$$

, преобразуют компоненту прямого сигнала с максимальной энергией $${\tilde{s}}_{max}^{h,v}$$

в матричный сигнал комплексной фазирующей функции A, включающий гипотетические сигналы, рассеиваемые каждым потенциальным объектом, согласно изобретению из матричного сигнала комплексной фазирующей функции A формируют блочный матричный сигнал фазирующей функции $$\tilde{A}=\left[\begin{array}{cc}A& 0\\ 0& A\end{array}\right]$$

, блочный матричный сигнал фазирующей функции $$\tilde{A}$$

запоминают, для каждого выбранного азимутально-угломестного направления приема преобразуют рассеянный сигнал s в двухкомпонентный сигнал комплексного частотно-временного изображения $$h^{\left(0\right)}={\left({\tilde{A}}^H\tilde{A}\right)}^{-1}{\tilde{A}}^{H}s$$

, где $$h^{\left(0\right)}=\left[\begin{array}{c}{h}^{\left(0\right)h}\\ h^{\left(0\right)v}\end{array}\right]$$

, $${\tilde{A}}^H$$

- матрица, эрмитово сопряженная с $$\tilde{A}$$

, двухкомпонентный сигнал h⁽⁰⁾ запоминают и используют в качестве начального приближения, а также итерационно формируют зависящий от предыдущего решения вспомогательный матричный сигнал $$\Lambda \left(h^{\left(k-1\right)}\right)=diag\left\{{\left({\left|h_z^{{}^{\left(k-1\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k-1\right),v}}\right|}^2\right)}^{-1/2}/2\right\}$$

, $$h_z^{{}^{\left(k-1\right),h}}$$

и $$h_z^{{}^{\left(k-1\right),v}}$$

и - z-е элементы компонент h^(k-1),h и h^(k-1),v сигнала h^(k-1), k=1, 2, … - номер итерации, блочный вспомогательный матричный сигнал $$\tilde{\Lambda }\left(h^{\left(k-1\right)}\right)=\left[\begin{array}{cc}\Lambda \left(h^{\left(k-1\right)}\right)& 0\\ 0& \Lambda \left(h^{\left(k-1\right)}\right)\end{array}\right]$$

и двухкомпонентный сигнал очередного приближения комплексного частотно-временного изображения $$h^{\left(k\right)}={\left[{\tilde{A}}^H\tilde{A}+\lambda \tilde{\Lambda }\left(h^{\left(k-1\right)}\right)\right]}^{-1}{\tilde{A}}^{H}s$$

, где λ - множитель Лагранжа, до тех пор, пока энергия разности текущего и запомненного предыдущего частотно-временных изображений не достигнет заданного малого значения $$\left[{\Vert h^{\left(k\right),h}-h^{\left(k-1\right),h}\Vert }^2+{\Vert h^{\left(k\right),v}-h^{\left(k-1\right),v}\Vert }^2\right]\le \delta$$

, после чего по локальным максимумам суммы квадратов модулей элементов компонент текущего частотно-временного изображения $${\left|h_z^{{}^{\left(k\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k\right),v}}\right|}^2$$

определяют число рассеянных радиосигналов, по параметрам которых - значениям временной задержки, доплеровского сдвига и азимутально-угломестного направления приема - выполняют обнаружение и пространственную локализацию подвижных объектов.The technical result is achieved by the fact that in the method of polarization-sensitive radio monitoring of moving objects, which consists in the use of direct and scattered by moving objects radio signals emitted by broadband transmitters of electronic systems for various purposes, the array of horizontal and vertical polarization components of a vector multi-beam electromagnetic field is received from N antennas direct and scattered radio signals, synchronously transform the ensemble of received radio signals into antennas digital signals, digital signals are converted into two-component direct $$\tilde{s}=\left[\begin{array}{c}{\tilde{s}}^h\\ {\tilde{s}}^v\end{array}\right]$$

and scattered $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

signals for the selected azimuthal elevation directions of reception, where h and v are the indices of the horizontal and vertical polarization components, which, together with the azimuthal elevation direction of the reception, memorize, calculate and compare the component energy $${\Vert {\tilde{s}}^h\Vert }^2$$

and $${\Vert {\tilde{s}}^v\Vert }^2$$

direct signal, select the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

transform the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

in the matrix signal of the complex phasing function A, including hypothetical signals scattered by each potential object, according to the invention, a block matrix signal of the phasing function is formed from the matrix signal of the complex phasing function A $$\tilde{A}=\left[\begin{array}{cc}A& 0\\ 0& A\end{array}\right]$$

, block matrix signal of the phasing function $$\tilde{A}$$

remember, for each selected azimuthal elevation direction, the scattered signal s is converted into a two-component signal of a complex time-frequency image $$h^{\left(0\right)}={\left({\tilde{A}}^H\tilde{A}\right)}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

where $$h^{\left(0\right)}=\left[\begin{array}{c}{h}^{\left(0\right)h}\\ h^{\left(0\right)v}\end{array}\right]$$

, $${\tilde{A}}^H$$

is a Hermitian conjugate matrix $$\tilde{A}$$

, the two-component signal h ^{(0) is} stored and used as an initial approximation, and an auxiliary matrix signal depending on the previous solution is iterated $$\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)=diag\left\{{\left({\left|h_z^{{}^{\left(k-\mathrm{one}\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k-\mathrm{one}\right),v}}\right|}^2\right)}^{-\mathrm{one}/2}/2\right\}$$

, $$h_z^{{}^{\left(k-\mathrm{one}\right),h}}$$

and $$h_z^{{}^{\left(k-\mathrm{one}\right),v}}$$

and - the zth elements of the components h ^{(k-1), h} and h ^{(k-1), v of the} signal h ^(k-1) , k = 1, 2, ... - iteration number, block auxiliary matrix signal $$\tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)=\left[\begin{array}{cc}\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)& 0\\ 0& \Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)\end{array}\right]$$

and a two-component signal of the next approximation of a complex time-frequency image $$h^{\left(k\right)}={\left[{\tilde{A}}^H\tilde{A}+\lambda \tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)\right]}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

, where λ is the Lagrange multiplier, until the energy of the difference between the current and remembered previous time-frequency images reaches a predetermined small value $$\left[{\Vert h^{\left(k\right),h}-h^{\left(k-\mathrm{one}\right),h}\Vert }^2+{\Vert h^{\left(k\right),v}-h^{\left(k-\mathrm{one}\right),v}\Vert }^2\right]\le \delta$$

, after which, according to local maxima of the sum of the squares of the modules of elements of the components of the current time-frequency image $${\left|h_z^{{}^{\left(k\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k\right),v}}\right|}^2$$

determine the number of scattered radio signals, the parameters of which - the values of time delay, Doppler shift and azimuth-elevation direction of reception - perform the detection and spatial localization of moving objects.

The operation of the method is illustrated by drawings:

figure 1 is a structural diagram of a device for polarization-sensitive radio monitoring of moving objects.

figure 2 - simulation results of the process of polarization-sensitive detection and time-frequency localization of moving objects by the proposed method.

figure 3 - simulation results of the process of polarization-sensitive detection and time-frequency localization of moving objects when using the prototype method.

To assess the comparative effectiveness of the proposed method, a simulation was performed on a PC.

When modeling, we used semi-natural data based on the sound signal of channel 49 of analogue television measured with a sampling frequency of 318785 Hz. The length of the sequence of the analyzed signal was 65536 samples.

The measured signal was used as a direct signal. He, with the addition of white noise with a level of minus 30 dB and reduced in amplitude, time-delayed and Doppler-shifted copies of the measured signal, was used as a reconnaissance signal. The scenario under consideration included a direct signal from the backlight transmitter and signals scattered by six objects. The first two objects are stationary, and the rest are mobile. The signal levels of stationary objects are 20 dB lower than the direct signal level, and the signal levels of other objects are on average lower by 60 dB.

The dimensions of the delay-Doppler shift grid were chosen equal to 101 × 101, the delay step was 3.1369 μs, and the Doppler shift was 4.8643 Hz.

Figure 2 and figure 3 presents the time-frequency image scattered by the objects of the radio signals generated by the proposed method and the prototype method, respectively.

From a comparison of these images it follows that the proposed method provides detection and frequency-time localization of signals from all six objects. At the same time, the prototype method provides detection of only merged signals from the first and second objects.

The device (figure 1), which implements the proposed method, contains a series-connected reception and preprocessing system 1 and computer system 2.

In turn, the reception and pretreatment system 1 includes an antenna array 1-1, a direct and scattered signal receiving path, including a frequency converter 1-2, ADC 1-3 and an adaptive spatial filtering device 1-4.

At the same time, system 2 has an output intended for connection to external systems.

System 1 is an analog-to-digital device and is designed for adaptive spatial filtering of useful direct and scattered radio signals by objects.

. Каждая антенна обеспечивает одновременный ненаправленный или направленный прием двух скалярных полей - ортогональных составляющих поляризованной волны в точке приема, и имеет два отдельных выхода для радиосигналов горизонтальной (h) и вертикальной (v) поляризаций.Antenna array 1-1 consists of N antennas with numbers $$n=\overline{\mathrm{one},N}$$

. Each antenna provides simultaneous non-directional or directional reception of two scalar fields - the orthogonal components of the polarized wave at the receiving point, and has two separate outputs for horizontal (h) and vertical (v) polarization radio signals.

The spatial configuration of the antenna array can be arbitrary: flat rectangular, flat annular or three-dimensional, in particular conformal.

The frequency converter 1-2 is a 2N-channel, made with a common local oscillator and with the bandwidth of each channel, variable in accordance with the width of the spectrum of the received radio signal. A common local oscillator provides multi-channel coherent signal reception.

ADC 1-3 is also 2N-channel and is synchronized by the signal of one reference oscillator (for simplicity, the reference oscillator is not shown in the diagram). If the resolution and speed of the ADC are sufficient for direct analog-to-digital conversion of the input signals, then frequency selective bandpass filters and amplifiers can be used instead of the frequency converter 1-2. In addition, the frequency converter 1-2 provides the connection of one of the antennas instead of all the antennas of the array for periodic calibration of the receiving channels using an external signal source. Calibration is possible using an internal generator, the output of which is also connected instead of all antennas for periodic calibration of channels (to simplify, the internal generator is not shown in the diagram).

The adaptive spatial filtering device 1-4 is a computing device.

Computing system 2 is intended for iterative formation of a two-component signal of a complex time-frequency image of radio signals scattered by objects in the analyzed region of Doppler frequencies and time delays, as well as for the detection and spatial localization of moving objects.

The device operates as follows.

In system 2, based on data from external systems, a set of transmitters emitting spread-spectrum radio signals is identified, selected, and periodically updated.

The parameters of the selected set of transmitters (number, carrier frequency, spectral width, shape, synchronization parameters and power of the emitted signal, coordinates or distance and angular position relative to the receiving point) are stored in system 2 and are also used to configure the converter 1-2. In order to simplify the control circuit of the converter are not shown.

The frequency converter 1-2 according to the signals of system 2 is tuned to a given frequency of reception.

и вертикальной $$x_n^v\left(t\right)$$

поляризаций поступает на входы преобразователя частоты 1-2.The vector multipath electromagnetic field of the direct and scattered radio signals received by each antenna with the number n of the array 1-1 is in the form of horizontal time-dependent radio signals $$x_n^h\left(t\right)$$

and vertical $$x_n^v\left(t\right)$$

polarization goes to the inputs of the frequency converter 1-2.

и $$x_n^v\left(t\right)$$

фильтруется по частоте и переносится на более низкую частоту.In the 1-2 frequency converter, each received radio signal $$x_n^h\left(t\right)$$

and $$x_n^v\left(t\right)$$

filtered by frequency and transferred to a lower frequency.

и $$x_n^v\left(t\right)$$

синхронно преобразуется с помощью АЦП 1-3 в цифровые сигналы $$x_1^h\left(i\right)$$

,…, $$x_N^h\left(i\right)$$

и $$x_1^v\left(i\right)$$

,…, $$x_N^v\left(i\right)$$

, где i - номер временного отсчета сигнала, которые поступают в устройство 1-4, где запоминаются.1-2 radio signal ensemble formed in the converter $$x_n^h\left(t\right)$$

and $$x_n^v\left(t\right)$$

synchronously converted using ADC 1-3 to digital signals $$x_{\mathrm{one}}^h\left(i\right)$$

, ..., $$x_N^h\left(i\right)$$

and $$x_{\mathrm{one}}^v\left(i\right)$$

, ..., $$x_N^v\left(i\right)$$

, where i is the number of time counts of the signal that enter the device 1-4, where they are stored.

и двухкомпонентные рассеянные сигналы $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

для выбранных азимутально-угломестных направлений приема, где h и v - индексы компонент горизонтальной и вертикальной поляризации.In device 1-4, digital signals are converted into a two-component direct signal $$\tilde{s}=\left[\begin{array}{c}{\tilde{s}}^h\\ {\tilde{s}}^v\end{array}\right]$$

and two-component scattered signals $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

for the selected azimuthal elevation directions of reception, where h and v are the indices of the components of horizontal and vertical polarization.

и двухкомпонентные рассеянные сигналы s для выбранных азимутально-угломестных направлений приема осуществляется известными способами адаптивной пространственной фильтрации [3].Converting digital signals to a two-component direct signal $$\tilde{s}$$

and two-component scattered signals s for the selected azimuthal elevation directions of reception is carried out by known methods of adaptive spatial filtering [3].

,…, $$x_N^h\left(i\right)$$

формируется сигнал пространственной корреляционной матрицы входных сигналов R. Сигнал корреляционной матрицы R преобразуется в сигнал оптимального весового вектора для формирования прямого w=R^-1η сигналов, где η - вектор наведения, определяемый азимутально-угломестным направлением приема прямого радиосигнала, длиной волны (частотой f_k) и геометрией решетки.In this case, for example, from digital signals of horizontal polarization $$x_{\mathrm{one}}^h\left(i\right)$$

, ..., $$x_N^h\left(i\right)$$

the signal of the spatial correlation matrix of the input signals R is generated. The signal of the correlation matrix R is converted to the optimal weight vector signal to generate a direct w = R ^-1 η signal, where η is the guidance vector determined by the azimuth-elevation direction of direct radio signal reception, wavelength (frequency f _k ) and lattice geometry.

,…, $$x_N^h\left(i\right)$$

объединяются в матричный цифровой сигнал X, преобразованием которого формируется сигнал s^h=w^HX, являющийся векторным сигналом компоненты горизонтальной поляризации прямого сигнала $${\tilde{s}}^h={\left\{s^h\left(1\right),\dots ,s^h\left(i\right),\dots ,s^h\left(I\right)\right\}}^T$$

, где $${\left\{\right\}}^T$$

- означает транспонирование, I - число временных отсчетов сигнала, принятого в выбранном азимутально-угломестном направлении.After that, digital signals $$x_{\mathrm{one}}^h\left(i\right)$$

, ..., $$x_N^h\left(i\right)$$

are combined into a matrix digital signal X, the conversion of which produces a signal s ^h = w ^H X, which is a vector signal of the horizontal polarization component of the direct signal $${\tilde{s}}^h={\left\{s^h\left(\mathrm{one}\right),...,s^h\left(i\right),...,s^h\left(I\right)\right\}}^T$$

where $${\left\{\right\}}^T$$

- means transposition, I is the number of time samples of the signal received in the selected azimuth-elevation direction.

, а также компонент s^h и s^v рассеянных сигналов для выбранных азимутально-угломестных направлений приема.Similarly, the formation of the component of the vertical polarization of the direct signal $${\tilde{s}}^v$$

, as well as the components s ^h and s ^{v of the} scattered signals for the selected azimuth-elevation directions of reception.

The physically described adaptive spatial filtering operations provide simultaneous directional reception from the given directions of two components of the useful direct signal of the selected backlight transmitter and two components of the useful scattered signal, while simultaneously suppressing a wide class of interference coming from other directions.

The obtained two-component signals, together with the value of the azimuthal elevation direction of reception, are received in computer system 2, where they are stored.

In computing system 2, the following actions are performed:

и $${\Vert {\tilde{s}}^v\Vert }^2$$

прямого сигнала;- component energy is calculated and compared $${\Vert {\tilde{s}}^h\Vert }^2$$

and $${\Vert {\tilde{s}}^v\Vert }^2$$

direct signal;

;- selects the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

;

в матричный сигнал комплексной фазирующей функции A, включающий гипотетические сигналы, рассеиваемые каждым потенциальным объектом.- the component of the direct signal with maximum energy is converted $${\tilde{s}}_{max}^{h,v}$$

into the matrix signal of the complex phasing function A, which includes hypothetical signals scattered by each potential object.

в матричный сигнал A осуществляется по следующей формуле: $$A=\left[D{\tilde{s}}_{max,0}^{h,v},\dots ,D{\tilde{s}}_{max,q}^{h,v},\dots ,D{\tilde{s}}_{max,Q-1}^{h,v}\right]$$

, где $${\tilde{s}}_{max,q}^{h,v}={\left[{\tilde{s}}_{max,\left(1-q\right)}^{h,v},\dots ,{\tilde{s}}_{max,\left(I-q\right)}^{h,v}\right]}^T$$

- векторы размером I×1, являющиеся сдвинутыми по времени на qT_s версиями прямого сигнала, $${\tilde{s}}_{max}^{h,v}$$

,…, q=0,…,Q-1, Q - число временных задержек прямого сигнала, T_s - период выборки сигнала;Transformation of direct signal components with maximum energy $${\tilde{s}}_{max}^{h,v}$$

in the matrix signal A is carried out according to the following formula: $$A=\left[D{\tilde{s}}_{max,0}^{h,v},...,D{\tilde{s}}_{max,q}^{h,v},...,D{\tilde{s}}_{max,Q-\mathrm{one}}^{h,v}\right]$$

where $${\tilde{s}}_{max,q}^{h,v}={\left[{\tilde{s}}_{max,\left(\mathrm{one}-q\right)}^{h,v},...,{\tilde{s}}_{max,\left(I-q\right)}^{h,v}\right]}^T$$

- vectors of size I × 1, which are time-shifted qT _s versions of the direct signal, $${\tilde{s}}_{max}^{h,v}$$

, ..., q = 0, ..., Q-1, Q is the number of time delays of the direct signal, T _s is the sampling period of the signal;

D = [D- _L , ..., D- _l , ..., D ₀ , ..., D _{+ l} , ..., D _{+ L} ], $$D_l=\left[\begin{array}{cccc}\mathrm{one}& 0& ...& 0\\ 0& e^{j2\pi l/\mathrm{<figure-callout\; id="0"\; label="I"\; filenames="00000092.png,00000093.png"\; state="\{\{state\}\}">I</figure-callout>}}& & 0\\ ...& ...& ...& ...\\ 0& 0& & e^{j2\pi l(I-\mathrm{one})/I}\end{array}\right]$$

- Doppler shift matrices, l = 0, ..., ± L, L - coordinate grid size by Doppler shift. The sizes of the matrices D _l and D are respectively equal to I × I and I × I (2L + 1).

, а размер этой матрицы I×Q(2L+1) определяется числом отсчетов в разведываемом сигнале (длительностью интервала наблюдения) и размерами координатной сетки по временному запаздыванию и доплеровскому сдвигу частоты;Thus, the columns of matrix A are time-delayed and frequency-shifted Doppler shift versions of the direct signal $${\tilde{s}}_{max}^{h,v}$$

, and the size of this matrix I × Q (2L + 1) is determined by the number of samples in the reconnaissance signal (the duration of the observation interval) and the size of the coordinate grid according to time delay and Doppler frequency shift;

- from the matrix signal of the complex phasing function A, a block matrix signal of the phasing function is formed $$\tilde{A}=\left[\begin{array}{cc}A& 0\\ 0& A\end{array}\right]$$

запоминается.- block matrix signal of the phasing function $$\tilde{A}$$

remembered.

After that, in computing system 2, for each selected azimuthal elevation direction of reception, the following actions are performed:

, где $$h^{\left(0\right)}=\left[\begin{array}{c}{h}^{\left(0\right)h}\\ h^{\left(0\right)v}\end{array}\right]$$

, $${\tilde{A}}^H$$

- матрица, эрмитово сопряженная с $$\tilde{A}$$

;- the scattered signal s is converted into a two-component signal of a complex time-frequency image $$h^{\left(0\right)}={\left({\tilde{A}}^H\tilde{A}\right)}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

where $$h^{\left(0\right)}=\left[\begin{array}{c}{h}^{\left(0\right)h}\\ h^{\left(0\right)v}\end{array}\right]$$

, $${\tilde{A}}^H$$

is a Hermitian conjugate matrix $$\tilde{A}$$

;

- the two-component signal h ^{(0) is} stored and used as an initial approximation;

, где $$h_z^{{}^{\left(k-1\right),h}}$$

и $$h_z^{{}^{\left(k-1\right),v}}$$

и - z-е элементы компонент h^(k-1),h и h^(k-1),v сигнала h^(k-1), k=1, 2, … - номер итерации, блочный вспомогательный матричный сигнал $$\tilde{\Lambda }\left(h^{\left(k-1\right)}\right)=\left[\begin{array}{cc}\Lambda \left(h^{\left(k-1\right)}\right)& 0\\ 0& \Lambda \left(h^{\left(k-1\right)}\right)\end{array}\right]$$

и двухкомпонентный сигнал очередного приближения комплексного частотно-временного изображения $$h^{\left(k\right)}={\left[{\tilde{A}}^H\tilde{A}+\lambda \tilde{\Lambda }\left(h^{\left(k-1\right)}\right)\right]}^{-1}{\tilde{A}}^{H}s$$

, где λ - множитель Лагранжа, до тех пор, пока энергия разности текущего и запомненного предыдущего частотно-временных изображений не достигнет заданного малого значения $$\left[{\Vert h^{\left(k\right),h}-h^{\left(k-1\right),h}\Vert }^2+{\Vert h^{\left(k\right),v}-h^{\left(k-1\right),v}\Vert }^2\right]\le \delta$$

;- an auxiliary matrix signal depending on the previous solution is iteratively formed $$\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)=diag\left\{{\left({\left|h_z^{{}^{\left(k-\mathrm{one}\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k-\mathrm{one}\right),v}}\right|}^2\right)}^{-\mathrm{one}/2}/2\right\}$$

where $$h_z^{{}^{\left(k-\mathrm{one}\right),h}}$$

and $$h_z^{{}^{\left(k-\mathrm{one}\right),v}}$$

and - the zth elements of the components h ^{(k-1), h} and h ^{(k-1), v of the} signal h ^(k-1) , k = 1, 2, ... - iteration number, block auxiliary matrix signal $$\tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)=\left[\begin{array}{cc}\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)& 0\\ 0& \Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)\end{array}\right]$$

and a two-component signal of the next approximation of a complex time-frequency image $$h^{\left(k\right)}={\left[{\tilde{A}}^H\tilde{A}+\lambda \tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)\right]}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

, where λ is the Lagrange multiplier, until the energy of the difference between the current and remembered previous time-frequency images reaches a predetermined small value $$\left[{\Vert h^{\left(k\right),h}-h^{\left(k-\mathrm{one}\right),h}\Vert }^2+{\Vert h^{\left(k\right),v}-h^{\left(k-\mathrm{one}\right),v}\Vert }^2\right]\le \delta$$

;

определяется число рассеянных радиосигналов, по параметрам которых -значениям временной задержки, доплеровского сдвига и азимутально-угломестного направления приема - выполняется обнаружение и пространственная локализация подвижных объектов.- by local maxima of the sum of the squares of the modules of the elements of the components of the current time-frequency image $${\left|h_z^{{}^{\left(k\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k\right),v}}\right|}^2$$

the number of scattered radio signals is determined, the parameters of which are the values of time delay, Doppler shift and azimuth-elevation direction of reception — the detection and spatial localization of moving objects is performed.

The following actions are performed:

- the values of the Doppler shift of each scattered signal are compared with the threshold, and when the threshold is exceeded, a decision is made to detect a moving object in the analyzed azimuth-elevation direction of reception.

The threshold is selected based on minimizing the probability of missing an object.

When determining the geographical coordinates of the detected moving object in computing system 2, the following actions are performed:

- from the value of the time delay of the signal τ, the apparent distance to the object is determined D = τc, where c is the speed of light;

- the spatial coordinates of the detected object are determined by the apparent range D and azimuth values α and elevation angle β of the reception of scattered signals, for example, in accordance with [4].

Moreover, for the pair “detection device - transmitter”, an ellipsoid of equal apparent distances corresponding to the geometrical place of points in space is built, the sum of the distances to which (from the transmitter to the object and from the object to the detection device) is equal to the found value of the apparent distance D. At the intersection of the ellipsoid and direction values (azimuth and elevation angle) of scattered signal reception are determined by the geographical coordinates of the detected object.

The results of the detection and spatial localization of airborne objects are displayed to increase information content.

From the above description it follows that a device that implements the proposed method provides an increase in the probability of detection and proper spatial localization of distant and weakly scattering objects due to the use of new adaptive processing operations with feedback on a useful radio signal.

Thus, through the use of new adaptive processing operations with feedback on a useful radio signal, which provide increased sensitivity and dynamic range when forming components of horizontal and vertical polarization of a two-component time-frequency image of radio signals scattered by objects in the analyzed area of Doppler frequencies (speeds) and time delays (ranges), it is possible to solve the problem with the achievement of the specified technical result.

Information sources

1. RU, patent, 2158002, cl. G01S 13/14, 2000

2. US patent 7304603 B2, cl. G01S 13/02, 2007

3. Ratynsky M.V. Adaptation and superresolution in antenna arrays. M .: Radio and communication. 2003 year

4. RU, patent, 2444754 C15, cl. G01S 13/02, 2012

Claims (1)

The method of polarization-sensitive radio monitoring of moving objects, which consists in using direct and scattered by mobile objects radio signals emitted by broadband transmitters of electronic systems for various purposes, using an array of N antennas, the horizontal and vertical polarization components of the vector multipath electromagnetic field of direct and scattered radio signals, synchronously transform the ensemble of radio signals received by the antennas into digital signals, I convert digital signals t in a two-component line $$\tilde{s}=\left[\begin{array}{c}{\tilde{s}}^h\\ {\tilde{s}}^v\end{array}\right]$$

and scattered $$s=\left[\begin{array}{c}{s}^h\\ s^v\end{array}\right]$$

signals for the selected azimuthal elevation directions of reception, where h and v are the indices of the components of horizontal and vertical polarization, which, together with the azimuthal elevation direction of reception, memorize, calculate and compare the component energy $${\Vert {\tilde{s}}^h\Vert }^2$$

and $${\Vert {\tilde{s}}^v\Vert }^2$$

direct signal, select the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

transform the component of the direct signal with maximum energy $${\tilde{s}}_{max}^{h,v}$$

into the matrix signal of the complex phasing function A, including hypothetical signals scattered by each potential object, characterized in that a block matrix signal of the phasing function is formed from the matrix signal of the complex phasing function A $$\tilde{A}=\left[\begin{array}{cc}A& 0\\ 0& A\end{array}\right]$$

, block matrix signal of the phasing function $$\tilde{A}$$

remember, for each selected azimuthal elevation direction, the scattered signal s is converted into a two-component signal of a complex time-frequency image $$h^{\left(0\right)}={\left({\tilde{A}}^H\tilde{A}\right)}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

where $$h^{\left(0\right)}=\left[\begin{array}{c}{h}^{\left(0\right)h}\\ h^{\left(0\right)v}\end{array}\right]$$

, $${\tilde{A}}^H$$

is a Hermitian conjugate matrix $$\tilde{A}$$

two-way signal $$h^{\left(0\right)}$$

remember and use as an initial approximation, and also iteratively form an auxiliary matrix signal depending on the previous solution $$\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)=diag\left\{{\left({\left|h_z^{{}^{\left(k-\mathrm{one}\right)},h}\right|}^2+{\left|h_z^{{}^{\left(k-\mathrm{one}\right)},v}\right|}^2\right)}^{-\mathrm{one}/2}/2\right\}$$

, $$h_z^{{}^{\left(k-\mathrm{one}\right),h}}$$

and $$h_z^{{}^{\left(k-\mathrm{one}\right),v}}$$

- ze elements of components $$h^{\left(k-\mathrm{one}\right),h}$$

and $$h^{\left(k-\mathrm{one}\right),v}$$

signal $$h^{\left(k-\mathrm{one}\right)}$$

, k = 1, 2, ... - iteration number, block auxiliary matrix signal $$\tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)=\left[\begin{array}{cc}\Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)& 0\\ 0& \Lambda \left(h^{\left(k-\mathrm{one}\right)}\right)\end{array}\right]$$

and a two-component signal of the next approximation of a complex time-frequency image $$h^{\left(k\right)}={\left[{\tilde{A}}^H\tilde{A}+\lambda \tilde{\Lambda }\left(h^{\left(k-\mathrm{one}\right)}\right)\right]}^{-\mathrm{one}}{\tilde{A}}^{H}s$$

, where λ is the Lagrange multiplier, until the energy of the difference between the current and remembered previous time-frequency images reaches a predetermined small value $$\left[{\Vert h^{\left(k\right),h}-h^{\left(k-\mathrm{one}\right),h}\Vert }^2+{\Vert h^{\left(k\right),v}-h^{\left(k-\mathrm{one}\right),v}\Vert }^2\right]\le \delta$$

, after which, according to the local maxima of the sum of the squares of the modules of the elements of the components of the current time-frequency image $${\left|h_z^{{}^{\left(k\right),h}}\right|}^2+{\left|h_z^{{}^{\left(k\right),v}}\right|}^2$$

determine the number of scattered radio signals, the parameters of which - the values of time delay, Doppler shift and azimuth-elevation direction of reception - perform the detection and spatial localization of moving objects.

Images