RU2766536C1 - Method of beam formation in aperture digital antenna array - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: invention relates to antenna engineering and specifically to methods of beam formation of receiving digital antenna arrays (AA) for communication and radar systems. Summary: in the method of generating a multibeam pattern, the received complex signal from each of the M AA modules is divided into two quadrature components which form M pairs of signals. Using ADC, quadrature components are digitized and time sequences are formed from them. For the formation of the beam pattern in each receiving module using the device for digital formation of the beam pattern in each receiving module and under the control of the device for generating coefficients for forming the beam pattern in the receiving module corresponding digital signals are mixed into corresponding quadrature pairs. Each pair of quadrature components is multiplied into N groups according to the number of beams and combined according to the attribute of belonging to the beam pattern of separate beams. To form a multi-beam antenna array pattern using a device for digital formation of a multi-beam pattern antenna array and under control of a device for generating coefficients for forming a multi-beam pattern antenna array corresponding digital signals are added to the corresponding quadrature pairs of each group according to the number of beams. Each quadrature component of each channel, obtained after digitization, is additionally refined in accordance with the expressions:

,

where I₀, Q₀ are specified in-phase and out-of-phase quadrature components at the current moment of time;

is digitized in-phase and out-of-phase quadrature components of time sequence at corresponding moments of time; L is the sliding window duration, which is a multiple of the signal carrier period and the sampling period; 2l is a moment in time within the sliding window duration.

EFFECT: lowering the level of side lobes of the multibeam beam pattern (BP).

1 cl, 1 tbl, 5 dwg

Description

The invention relates to antenna technology, and in particular to methods of beamforming receiving digital antenna arrays (DA) for communication and radar systems.

Known antenna post of a radar station with a receiving antenna array (AR) containing M receiving antenna modules, the outputs of which are connected to the inputs of the device for converting analog signals into digital form (signal digitization) of the receiving antenna array, a digital device for generating coefficients for generating amplitude-phase distribution in opening of the receiving antenna array for each of the scanning beams - radiation patterns (DN) connected to the device for digital formation of K scanning beams ([1], RU 2395140 C2, IPC G01S 13/42; 27.01.2010).

The disadvantages of the method implemented by this device are typical for all antenna arrays with direct conversion of signals in spatial channels into digital form. Ensuring the coherence of the signals at the output of each of the channels is possible only with a very small jitter of the reference signal of the analog-to-digital converter (ADC) (fractions of picoseconds), which for high-speed digital signals is almost more difficult than the wiring of microwave oscillations. At the same time, the power consumed by the ADC increases significantly with increasing sampling frequency (for example, for 2-2.5 GHz, the power consumption is about 1 W ([2], p. 37). The amount of digital data, even when using subsampling, is excessive and requires decimation (decimation).This leads to irrational use of the speed properties of the ADC.In this regard, the direct conversion of the analog microwave signal into digital form is impractical.In addition, the process of converting analog signals into digital form is associated with the generation of sampling and quantization errors, which, in the end, As a result, in combination with natural noise at the ADC input and jitter errors, in the process of forming the CAR pattern, they lead to an additional increase in the level of side lobes.

There is a method according to which signals are received by means of an antenna array consisting of M antenna elements. The weight coefficients are calculated for each receiving channel in the azimuth and elevation planes, according to the number of generated beams (1,...k,...,K), where K is the number of generated beams. Produce synchronous sampling and quantization of each signal from the output of each of the antenna elements. The sequence of samples from the output of each ADC is converted into a sequence of quadrature samples, filtering and decimation of the frequency of the digital sequence by K times are performed. For each receiving channel, a sequence of weighted readings is formed by multiplying each quadrature reading by weight coefficients, separately for the azimuth and elevation planes. The resulting sequence of readings of K beams of the radiation pattern is formed by summing the weighted readings for each k-th beam related to the same sampling times ([3], RU 2495447 C2, IPC G01S 3/80; 20.05.2013).

In this case, the disadvantages are completely similar to the disadvantages of the previous method.

The closest in technical essence (prototype) of the proposed method is a method of diagram formation, implemented by a digital receiving AR for a radar station, in which an AR containing M receiving antenna modules is used to form a multibeam DN. The received complex signal from each module is divided into two quadrature components forming M pairs of signals. By synchronous time sampling and level quantization, the quadrature components are digitized and time sequences are formed from them. Then, to form a pattern in each receiving module, using a device for digital pattern formation in the receiver module and under the control of a device for generating weight coefficients to form a pattern in the receiver module, corresponding digital signals are mixed to the corresponding quadrature pairs, each pair of which is then multiplied by K groups by the number rays and unite on the basis of belonging to the pattern of individual rays. And to form a multipath AP AP using a device for digital formation of a multipath AP AP and under the control of a device for generating weight coefficients for generating a multipath AP AP, corresponding digital signals are mixed again to the corresponding quadrature pairs of each group according to the number of beams. All quadrature signals of each beam are summed in pairs, forming K pairs of output quadrature signals according to the number of generated beams ([4], RU 2584458 C1, G01Q 3/00; 05/20/2016).

The disadvantage of the prototype is the increased level of side lobes of the multipath DN in comparison with the level corresponding to the absence of noise due to sampling and quantization errors acting in conjunction with natural noise and jitter noise at the input of the ADC.

The technical problem to be solved by the present invention is the increased level of side lobes of the multipath DN in comparison with the level corresponding to the absence of noise due to sampling and quantization errors acting in combination with natural noise and jitter noise at the input of the ADC.

To solve this technical problem, a method for beamforming in a receiving digital array is proposed, consisting in the fact that an array containing M receiving antenna modules is used to form a multipath RP. The received complex signal from each module is divided into two quadrature components forming M pairs of signals. By synchronous time sampling and level quantization, the quadrature components are digitized with the help of an ADC and time sequences are formed from them. Then, to form a pattern in each receiving module, using a digital pattern generator in the receiver module and under the control of a coefficient generator for generating pattern in the receiver module, corresponding digital signals are mixed to the corresponding quadrature pairs, each pair of which is then multiplied into K groups according to the number of beams and combined on the basis of belonging to the pattern of individual beams. To form a multipath AP AP using a device for digital formation of a multipath AP AP and under the control of a device for generating coefficients for generating a multipath AP AP, corresponding digital signals are mixed again with the corresponding quadrature pairs of each group according to the number of beams. All quadrature signals of each beam are summed in pairs, forming K pairs of output quadrature signals according to the number of generated beams.

According to the invention, the digitized quadrature signal components of each receiving module are further refined using digital filters in accordance with the expressions:

where I, Q - updated in-phase and out-of-phase quadrature components at the current moment of time;

- оцифрованные синфазная и несинфазная квадратурные составляющие временной последовательности на соответствующие моменты времени;

- digitized in-phase and out-of-phase quadrature components of the time sequence at the corresponding points in time;

2l - moment of time within the duration of the tracking window L.

Thus, the proposed method has the following distinguishing features in the sequence of its implementation from the prototype method shown in table 1.

From the presented comparison table of the implementation sequences of the prototype method and the proposed method, it can be seen that the following new operation has been introduced: the digitized quadrature component of the signal of each channel is further refined in accordance with the expressions:

where I, Q - updated in-phase and out-of-phase quadrature components at the current moment of time;

- оцифрованные синфазная и несинфазная квадратурные составляющие временной последовательности на соответствующие моменты времени;

- digitized in-phase and out-of-phase quadrature components of the time sequence at the corresponding points in time;

2l - moment of time within the duration of the tracking window L.

The introduction of one operation allows, in comparison with the prototype method, to ensure the achievement of a technical result consisting in lowering the level of side lobes of a multipath DN due to sampling and quantization errors acting in conjunction with natural noise and jitter noise at the input of the ADC by reducing their influence on diagramming process.

The analysis of technical solutions made it possible to establish that analogues, characterized by a set of features identical to all the features of the proposed technical solution, are absent in known sources from the prior art, which indicates that the proposed method complies with the condition of patentability "novelty".

The results of the search for known solutions in this and related fields of technology in order to identify features that match the distinguishing features from the prototype showed that they do not follow explicitly from the prior art. From the prior art, the influence of the transformations envisaged by the essential features on the achievement of the specified technical result has not been revealed either. Therefore, the claimed technical solution meets the condition of patentability "inventive step".

The invention is illustrated graphically with figures 1-5.

In FIG. 1 shows a diagram of a device that implements the proposed method.

In FIG. 2 shows a digital filter circuit.

In FIG. Figure 3 shows the spectra of the quadrature components of the signal.

In FIG. 4 shows estimates of the signal-to-noise ratio (SNR) in the implementation of the proposed method and in the implementation of the prototype.

Figure 5 shows the pattern of the AR, taking into account the influence of noise.

The method of diagram formation in the receiving digital AA includes the following operations:

- for the formation of multipath DN use AR containing M receiving antenna modules - 1;

- the received complex signal from each module is divided into two quadrature components, forming M pairs of signals - 2;

- by synchronous time sampling and level quantization, the quadrature components are digitized and time sequences are formed from them - 3;

- the digitized quadrature component of the signal of each channel is additionally refined in accordance with the expressions:

where I, Q - updated in-phase and out-of-phase quadrature components at the current moment of time;

- оцифрованные синфазная и несинфазная квадратурные составляющие временной последовательности на соответствующие моменты времени;

- digitized in-phase and out-of-phase quadrature components of the time sequence at the corresponding points in time;

2l - time point within the duration of the tracking window L - 4;

- then, to form a pattern in each receiving module, using a device for digital pattern formation in the receiver module and under the control of a device for generating coefficients for forming a pattern in the receiver module, corresponding digital signals are mixed to the corresponding quadrature pairs, each pair of which is then multiplied into K groups by the number rays and unite on the basis of belonging to the DN individual rays - 5;

- and for the formation of a multipath DN AR using a device for digital formation of a multipath DN AR and under the control of a device for generating weight coefficients for the formation of a multipath DN AR, the corresponding digital signals are again mixed into the corresponding quadrature pairs of each group according to the number of beams - 6;

- all quadrature signals of each beam are summed in pairs, forming K pairs of output quadrature signals according to the number of generated beams - 7.

The block diagram of the device that implements the beamforming method in the receiving digital AA is shown in Fig. 1 (possible implementation).

The device according to Fig. 1 contains:

- antenna array 1 of M receiving antenna modules, each of the modules, in turn, contains: a linear part - 1.1.1-1.1.M; frequency converter - 1.2.1-1.2.M;

- reference frequency synthesizer - 2;

- M synchronous analog-to-digital converters (ADC) - 3.1-3.M;

- M pairs of digital filters for the I and Q quadrature components in each pair - 4.1.I-4.M.I and 4.1.Q-4.M.Q, respectively;

- M devices for digital pattern formation in the receiving module - 5.1-5.M;

- device for generating coefficients for the formation of RP in the receiving module - 6;

- M×K devices for digital formation of multibeam AP AP - 7.m.k. (m=1,2,…,M;k=1,2,…,K);

- a device for generating coefficients for the formation of a multibeam AP AR - 8;

- K totalizers - 9.1-9.K.

At the same time, the AR 1 of the receiving antenna modules contains serially interconnected output and input, M linear parts 1.1.1-1.1.M and M frequency converters 1.2.1-1.2.M. Pair quadrature outputs of frequency converters 1.2.1-1.2.M are also connected in series at the output and input with the corresponding ADC 3.1-3.M, digital filters for quadrature components 4.1.I-4.M.I and 4.1.Q-4.MQ and devices for digital formation of pattern in the receiving module 5.1-5.M. The second inputs of the frequency converters 1.2.1-1.2.M and the second inputs of the ADC 3.1-3.M are connected to the corresponding outputs of the reference frequency synthesizer 2. The second inputs of the digital DN formation devices in the receiving module 5.1-5.M are connected to the corresponding outputs of the device generation of coefficients for the formation of RP in the receiving module 6. Paired outputs of each of the M devices for digital formation of RP in the receiving module 5.1-5.M are connected to the corresponding K inputs (according to the number K of the generated beams) of the digital formation devices for multibeam RP AP 7.mk (m =1,2,…,M; k=1,2,…,K), the paired outputs of which, grouped according to the attribute of belonging to one of the K beams, are connected to the corresponding K inputs of the adders 9.1-9.K. The second inputs of devices for digital formation of multipath DN AR 7.m,k connected to the output of the device for generating coefficients for the formation of multipath DN AR 8. Paired outputs K adders 9 are the outputs of the multipath receiving AR.

Let us consider the operation of the diagram-forming device in the receiving digital antenna array. AP 1 contains: M receiving antenna modules. The linear part of the receiving modules 1.1.1-1.1.M provides the necessary amplification of the radio signal, and the frequency converters 1.2.1-1.2.M - lowering the carrier frequency to an intermediate one and forming the quadrature components of the signal at an intermediate frequency. To synchronize the operation of the receiving modules, a reference frequency synthesizer 2 is used as a reference oscillator of the frequency converters. The quadrature components from each receiving module are fed to the inputs of the corresponding synchronous analog-to-digital converters (ADC) 3.1-3.M. In this case, the sampling rate for the ADC 3.1-3.M is also set using the reference frequency synthesizer 2. The intermediate frequency and sampling rate are chosen so as to ensure the most rational use of the speed properties of the ADC 3.1-3.M while minimizing power consumption. Any increase in power consumption is highly undesirable, since in this case, given the large number of AP 1 modules and limited dimensions, problems often arise with the removal of excess heat. If the spectrum of the useful signal is sufficiently narrowband, then the subsampling principle should be used to lower the sampling frequency, choosing a sampling frequency comparable to the intermediate frequency or even less than it. Using this principle, by means of synchronous time sampling and level quantization, the quadrature components are digitized using ADC 3.1-3.M and time sequences are formed from them, which, in order to reduce the influence of natural noise, jitter errors, sampling and quantization, are subjected to additional processing by the introduced, according to the invention, digital filters for quadrature components 4.1.I-4.MI and 4.1.Q-4.MQ Digital filters for quadrature components 4.1.I-4.M.I and 4.1.Q-4.M.Q are discrete optimal filters of the integrating type, consistent with the flow of digitized intermediate frequency oscillations. In this case, the noise contained in the digital signal is averaged. Due to this, the effect of reducing the influence of natural noise, jitter errors, sampling and quantization on the level of side lobes of a multipath DN is achieved. Quadrature components refined in this way are fed to the inputs of the device for digital formation of DN in each module 5.1-5.M, which, by analogy with the prototype, operate under the control of devices for generating weight coefficients for generating DN in each module 6. Device outputs 5.1-5.M by analogy with the prototype are connected in pairs with the inputs of the device for digital formation of multibeam DN of the antenna array 7.mk (m=1,2,…,M; k=1,2,…,K). The control inputs of the latter are connected to the device for generating weight coefficients for the formation of a multibeam pattern of the antenna array 8. The outputs of the devices 7.mk, grouped by means of adders 9.1-9.K according to belonging to one of the K beams of the multipath pattern, are the outputs for signals in the form of pairs quadrature components, which are further used for primary radar processing.

The execution of the blocks and nodes shown in Fig. 1, possibly using a modern element base of digital AA.

The proposed algorithm for the operation of digital filters is described by the expressions:

It can be implemented using specialized computing tools, or in hardware: in the form of a discrete digital filter, shown in Fig. 2.

Let us consider the rationale for the proposed method of diagram formation in the receiving digital antenna array. Digital AR is a set of digital spatial channels-modules, each of which has its own digital signal at the output. DN is formed by setting the required phase distribution along the AR opening. In this case, in the radio-frequency linear and converting parts of the receivers, errors may occur due to the spread of the phases of the active elements and phase shifters. To compensate for these phase errors, additional phase shifters are used, for example, in the form of segments of transmission lines, dielectric inserts, filters, etc. ([5], pp. 486, 487). However, with the transition to digital signal processing methods, during the operation of digital arrays, random (uncontrolled) phase errors may appear, caused by quantization, sampling, jitter noise and noise in the input analog signal of the ADC. These errors will lead to significant amplitude and phase errors in the spatial digital signal, which leads to in-phase distortion and, consequently, to an increased level of multipath sidelobes in comparison with the level corresponding to the absence of noise.

As is known from the statistical theory of antennas, aperture errors have a uniform distribution and are statistically independent. The total phase error of digital signal processing (DSP) is defined as follows ([5], pp. 10, 11):

where Δϕ _d is the phase error caused by the sampling process;

Δϕ _j - phase error introduced by jitter noise (jigger (English jitter - jitter) or phase jitter of a digital data signal);

Δϕ _k - phase error arising in the process of quantization;

Δϕ _N is the phase error caused by noise in the input analog signal.

The phase error of the sampling is found through the period (frequency) of the sampling and the period (frequency) of the signal:

where Δt _d - sampling period;

F _max - maximum frequency in the signal spectrum.

The jitter noise phase error can be expressed in a similar way:

The quantization phase error occurs due to the rounding of the amplitude of the quantized signal - the appearance of quantization noise. It can be found through the number n of bits of the ADC code:

The phase error caused by the presence of noise in the input analog signal is expressed similarly to the quantization phase error: through the width of the noise track on average over the period of the harmonic process:

where q is the signal-to-noise ratio.

Thus, the total phase error will depend on the sampling rate, the amount of jitter noise, the number of ADC code bits, and the ratio of signal power to noise power. If we take into account that the processes occurring in the processing device are similar to the processes in the beam-forming antenna switching circuit, then to assess the directional properties of a digital array, in this case, we can apply the relation known from the statistical theory of antennas using the phase errors found above:

where D is the gain of the antenna array in the presence of phase errors;

D ₀ - KND ideal aperture;

- величина фазовой ошибки ЦОС в апертуре решетки.

- the value of the phase error of the DSP in the array aperture.

As a rule, the conversion of signals into digital form is carried out at some intermediate frequency with a period T.

The signals at the ADC input of each i-th receiving module (hereinafter, the processing channel) can be represented as:

- квадратурные составляющие, содержащие фазовый шум, l-го временного отсчета

в i-ом канале (i=0…N);where

- quadrature components containing phase noise, l-th time reference

in the i-th channel (i=0…N);

- случайные фазы квадратурных составляющих, приведенные ко входу АЦП, в i-ом канале антенной решетки;

- random phases of the quadrature components, reduced to the input of the ADC, in the i-th channel of the antenna array;

- промежуточная частота;T - oscillation period:

- intermediate frequency;

- частота дискретизации;Δt - sampling period:

- sampling frequency;

A _i - complex amplitude of the signal in the channel, taking into account the position of the phase center of the antenna element of the i-th channel and the initial phase of the source signal.

The sampling frequency of the ADC is determined by the width of the spectrum of the received signal and the required rate of control of the amplitude-phase distribution. If the signal is sufficiently narrowband, then the principle of subsampling can be used, when the sampling rate is comparable to or even lower than the carrier frequency. This simplifies the technical implementation of the ADC when it is necessary to convert a high-frequency analog signal. In real digital arrays, a sampling frequency of the order of 100-200 MHz is usually sufficient ([5], p. 24). For example, for the company "Texas Instruments)) it is typical to use the following combination of frequencies: the input intermediate frequency of the analog signal is 75 MHz; sampling rate 100 MHz. Let's take this combination of frequencies as a reference in the simulation. Let us check the condition for the realizability of the subsampling principle. It looks like:

where 2Δƒ is the width of the spectrum of the received signal;

m is the subsampling order.

It is easy to verify that this condition for the adopted combination of frequencies is satisfied at m=1 and at 2Δƒ<50 MHz. For example, at 2Δƒ=40 MHz, the inequality becomes: 95 MHz<100 MHz<110 MHz.

и квадратурной

частям элемента пространственного поля. Первый номер последовательности соответствует текущему моменту времени (l=0). L - 1-й номер последовательности соответствует временному положению на L элементов в прошлое. Таким образом, процесс выборки данных соответствует принципу скользящего окна. С целью обеспечения равных условий осреднения длительность временной числовой последовательности L должна быть кратна периоду сигнальной несущей и периоду дискретизации.:

и

, где k_PR и k_D - целые. В частности для частот взятых в качестве примера эти коэффициенты могут быть взяты в наборах: k_PR=3, k_D=4; k_PR=6, k_D=8; k_PR=9, k_D=12 и т.д.For each i-th channel, we select two time numerical sequences of size L corresponding to the in-phase

and quadrature

parts of the element of the spatial field. The first sequence number corresponds to the current time (l=0). L - the 1st sequence number corresponds to the time position L elements in the past. Thus, the data sampling process follows the sliding window principle. In order to ensure equal averaging conditions, the duration of the time sequence L must be a multiple of the period of the signal carrier and the sampling period.:

And

, where k _PR and k _D are integers. In particular, for frequencies taken as an example, these coefficients can be taken in sets: k _PR =3, k _D =4; _kPR =6, _kD =8; k _PR =9, k _D =12, etc.

и квадратурной

компонент в каждом канале обрабатываются совершенно одинаково, путем спектрального разложения этих компонент на интервале L - F (I^*) и F(Q^*). В силу конечного числа элементов на интервале спектр будет носить линейчатый характер с ярко выраженными максимумами модулей спектральных составляющих, так как числовые последовательности

и

получены путем оцифровки периодического сигнала. Максимальные значения по модулю этих спектральных составляющих характеризуют энергию несущей сигнала. Эти спектральные составляющие являются комплексно сопряженными и расположенными в положительной и отрицательной частотных областях. Остальные обусловлены шумовой компонентой и могут быть исключены из рассмотрения. Сумма оставшихся является действительным числом и равна квадратурной компоненте на данный момент времени.The resulting two numeric sequences for in-phase

and quadrature

the components in each channel are treated in exactly the same way, by spectral decomposition of these components on the interval L - F (I ^* ) and F(Q ^* ). Due to the finite number of elements in the interval, the spectrum will have a line character with pronounced maxima of the moduli of the spectral components, since the numerical sequences

And

obtained by digitizing a periodic signal. The maximum modulo values of these spectral components characterize the signal carrier energy. These spectral components are complex conjugate and located in the positive and negative frequency regions. The rest are due to the noise component and can be excluded from consideration. The sum of the remaining ones is a real number and is equal to the quadrature component at this point in time.

With the chosen interval L, the sum of the spectral components characterizing the signal energy is described by the relations

determining the operation algorithm of the proposed digital filter.

As an example, in FIG. 3, a and 3, b shows the spectra of quadrature components from a harmonic signal with a carrier amplitude A=10 when choosing k _PR = 9 and k _D = 12, respectively.

. Таким образом, отфильтровывая спектральные составляющие под номерами №3 и №9, можно существенно уменьшить влияние шума на квадратурные компоненты.From the results in FIG. 3 it follows that the specified value of the amplitude A=10 and calculated through the amplitudes of the quadrature components coincides with the accuracy adopted during rounding. The error is quite small and is due to the influence of noise in a narrow range - in the vicinity of frequencies

. Thus, by filtering out the spectral components numbered #3 and #9, it is possible to significantly reduce the effect of noise on the quadrature components.

In an analytical expression for the in-phase component of an arbitrary channel, this action will look like:

where I is the updated value of the in-phase component at the filter output at the current time.

After opening the brackets and bringing similar expressions, expression (11) will be significantly simplified:

The expression for the quadrature component Q ₀ will be exactly the same:

Relations (12) and (13) define the digital filter operation algorithm.

As an example, estimates were obtained of the beneficial effect achieved by using the relations (12) and (13) in the process of diagram formation in comparison with the prototype.

When performing numerical studies, the following initial data were used:

- the amplitude of the harmonic signal at the intermediate frequency - 10;

- intermediate frequency, MHz - f _PR =75⋅10 ⁶ Hz;

;- sampling step when modeling a carrier at an intermediate frequency

;

- ADC sampling frequency - f _D =10 ⁸ Hz;

- spectrum width of the received signal Δƒ=20⋅10 ⁶ MHz;

- subsampling order m=1;

;- reduced sampling frequency and reduced intermediate frequency -

;

.- window length

.

In FIG. 4 shows estimates of the signal-to-noise ratio (SNR) in the implementation of the proposed method and in the implementation of the prototype. The abscissa shows the values of the phase noise deviation in the channels of the digital array Δϕ in radians, and the ordinate shows the SNR values at the outputs of the channels of the digital array. The symbols "□" correspond to the results of a numerical experiment when implementing the prototype method, and the symbols "x" - when implementing the proposed method. For convenience, according to the results of the experiment, approximating dependences were obtained, which are indicated by the dashed and solid curves, respectively.

. При выборе иного сочетания коэффициентов k_PR, k_D, частоту среза можно оценить по формуле

. В любом случае требуется, чтобы частота среза была меньше максимальной частоты в полосы информационного сигнала, т.е. выполнялось условие: ƒ_cp<Δƒ. Для случая, взятого в качестве примера, условие выполняется: 10 МГц<20 МГц.As follows from the above simulation results, the use of a digital filter at the ADC output increases the signal-to-noise ratio. For example, when choosing the coefficient k _PR =9, k _D =12, the signal-to-noise ratio increases two to three times. In this case, the digital filter is an integrating filter that narrows the receiver bandwidth with a cutoff frequency approximately equal to

. When choosing a different combination of coefficients k _PR , k _D , the cutoff frequency can be estimated by the formula

. In any case, it is required that the cutoff frequency be less than the maximum frequency in the information signal band, i.e. the following condition was fulfilled: ƒ _cp <Δƒ. For the case taken as an example, the condition is met: 10 MHz<20 MHz.

In FIG. 5, and the curve corresponds to the RP in the absence of noise ƒ ₀ (θ) for M=128-element equidistant linear digital antenna array with interelement spacing d=0.5λ (λ is the wavelength of the signal source). The amplitude distribution of the form

where θ ₀ is the direction of the maximum of the formed beam of the antenna array.

In accordance with the known data and numerical calculation, the achievable level of side lobes in this case corresponds to -40 dB.

To form a pattern taking into account the impact of noise, a point source of a harmonic signal of unit amplitude was placed at each observation point. To this signal was added a random component of the phase noise with deviation Δϕ=π10°/180°. At the same time, in each AA processing channel, the model of the received signal was described by the expression

where ξ _l,i ∈[-Δϕ/2, Δϕ/2] is the random noise component.

The generated RP in each observation direction θ at time l=0 was calculated by the formula

In FIG. 5, b shows the curve obtained by calculating the RP using the formula (16) taking into account the impact of noise and without processing with a digital filter (according to the prototype).

In FIG. 5, c shows the DN curve when calculated by the formula

those. taking into account the impact of noise and the results of processing by a digital filter (according to the claimed method).

It follows from a comparison of the results that the gain in the level of side lobes is 5...6 dB.

The conducted studies have shown that the gain depends on the size of the sliding window (the number of terms in expressions (10)). An increase in the size of the sliding window leads to an increase in the gain, however, in this case, the dynamic properties of the antenna array deteriorate. In this regard, the choice of window sizes is a compromise between the desire to solve the problem of approaching the level of side lobes corresponding to the absence of noise and the dynamic properties of the antenna array.

Thus, the above examples and justification of expressions (12) and (13) demonstrate the feasibility of the proposed method and the achievement of the formulated technical result - an increase in SNR in the CAR channels by 5...6 dB.

The above materials on the possible implementation of the method based on known blocks and devices confirm compliance with the criterion of "industrial applicability" of the proposed method.

List of sources used

1 Pat. No. 2395140, Russian Federation, IPC G01S 13/42. Antenna post of the radar station / L.R. Karev, G.A. Morozov, V.V. Samulevich, A.N. Sergeev, Ya.S. Khasin. - 2008130749, claim. 07/24/2008; publ. 27.01.2010, Bull. No. 3;

2 Malakhov, R.Yu. Digital onboard antenna array module. Thesis for the degree of candidate of technical sciences (supervisor Dobychina E.M.) / R.Yu. Malakhov. - M.: MAI, 2015. - 156 p.;

3 Pat. No. 2495447, Russian Federation, IPC G01S 3/80. The method of forming the radiation pattern / V.V. Zadorozhny, A.Yu. Larin, O.V. Ovodov. - 2011146408, Appl. 11/15/2011; publ. 05/20/2013 Bull. No. 14;

4 Pat. No. 2584458, Russian Federation, IPC G01Q 3/00. Digital scanning receiving antenna array for a radar station / G.A. Morozov, T.I. Sukhacheva, Ya.S. Khasin. - 2014141907, claim. 10/17/2014; publ. 05/20/2016 Bull. No. 14;

5 Microwave devices and antennas. Design of phased antenna arrays: Textbook for universities / Pod. ed. DI. Resurrection. - M.: Radio engineering, 2012. -744 p.;

6 Shmachilin, P.A. Characteristics of onboard digital AFAA microwave Specialty 05.12.07 - "Antennas, microwave devices and their technologies." Abstract of the dissertation for the degree of candidate of technical sciences / P.A. Shmachilin. - M.: MAI, 2011. - 20 p. (Head Doctor of Technical Sciences, Professor D.I. Voskresensky).

Claims (6)

A method for forming a multipath radiation pattern, which consists in the fact that the received complex signal from each of the M modules of the antenna array is divided into two quadrature components forming M pairs of signals, and by synchronous sampling in time and quantization in terms of level, the quadrature components are digitized and temporal components are formed from them sequence, then to form the beamforming in each receiving module, using the digital beamforming device in the receiving module and under the control of the device for generating beamforming coefficients in the receiving module, the corresponding digital signals are mixed to the corresponding quadrature pairs, each pair of which is then multiplied by N groups according to the number of beams and are combined on the basis of belonging to the radiation patterns of individual beams, and to form a multi-beam radiation pattern of the antenna array using a digital shaper the corresponding digital signals are again mixed to the corresponding quadrature pairs of each group according to the number of beams, characterized in that each quadrature component of each channel obtained after digitization is additionally refined in accordance with the expressions

where I ₀ , Q ₀ - updated in-phase and out-of-phase quadrature components at the current time;

- digitized in-phase and out-of-phase quadrature components of the time sequence at the corresponding points in time; L is the duration of the sliding window, which is a multiple of the signal carrier period and the sampling period;

- the moment of time within the duration of the sliding window.

Images