RU2141706C1 - Method and device for adaptive spatial filtering of signals - Google Patents

Abstract

FIELD: adaptive antenna arrays. SUBSTANCE: method involves reception of signals at N points spaced apart, where N 2, calculation of weight coefficients, generation of first and second reference signals whose frequencies are respectively chosen from equations ω₁<(ω₀-Δω) and ω₂>(ω₀Δω),, where ω₀ and Δω - are carrier frequency and spectrum width of useful signal received, separation of each of them into N equal-amplitude components, shifting of each i-th component, where i = 1, 2 ... N, through preset phase, summing up of i-th signals received and i-th components of first and second reference signals, and weighed addition of signals shaped. Directivity pattern main lobe shaped in the process covers angular spectrum within which arrival of useful signal is expected. EFFECT: improved noise immunity in signal reception. 2 cl, 4 dwg, 2 tbl

Description

 The proposed objects of the invention are united by a single inventive concept, relate to radio engineering and communication, namely to adaptive spatial filtering of signals implemented in adaptive antenna arrays, and can be used in radio communication systems operating in difficult jamming environments, for example, in radio communication systems with moving objects (ON), if necessary, to achieve a higher level of noise immunity of signal reception.

 Known methods of adaptive spatial filtering (ACE) of signals implemented in adaptive antenna arrays (AAR), for example, the method of ACE signals implemented in AAR according to the application of Japan N 57-112564 from 04/02/1992

This method involves receiving signals at N spatially separated points, where N ≥ 2, selecting from N received signals M _l signals, where 2≤M _l ≤N, l is the number of selected signals (l = N, N-1, N-2 , ..., 2), calculation of weighting coefficients (VK), weighted summation of M _l selected signals, estimation of the weighted summed (output) signal. (Evaluation in this case characterizes the quality of the received useful signal, by which we mean the signal / (interference + noise) (SINR) ratio at the output of the AAR). In the case when the quality of the output signal does not satisfy the required value, a decision is made on the choice of new M ₁ signals and so on. If necessary, the procedure for selecting new M ₁ signals is cyclically repeated. In this case, the calculation of VC is carried out according to the algorithm for minimizing the signal power (MMV) at the output of the AAR.

However, this analogue has disadvantages:
low noise immunity of signal reception, since VC calculation should be performed at those times when a useful signal is absent at the AAP input (P _s = 0) or a useful signal is present, but the condition P _p > P _s is fulfilled (here P _p , P _s - interference power and useful signal, respectively), and if at the input of the AAR P _p ≤ P _s , unintentional suppression of the useful signal during adaptation can occur [1];
inefficient use of the resource of antenna elements (AE), because based on the evaluation of the output signal, a decision is made on the choice of the number of AEs, and accordingly, the number of signals M _l , for the subsequent ACE and there may be situations when one or more AEs will not be used to receive signals .

 There is also a known method of ACE signals implemented in AAP according to the patent of the Russian Federation N 2090960 from 09.20.1997.

 This method involves receiving signals at N spatially separated points, where N≥2, calculation of N weighting factors, the choice of the i-th VK, where i-1, 2,. . .N, weighted summation of the received signals, quality assessment of the weighted summed (output) signal. In this case, if according to the evaluation results the quality of the output signal is lower than the required value, then a decision is made on the choice of new previously calculated VCs. If necessary, the procedure for selecting new VK is cyclically repeated. In this case, the calculation of VC is carried out according to the MMB algorithm.

However, this analogue has the disadvantage: - low noise immunity of signal reception, since VC calculation should be performed at those times when there is no useful signal at the AAP input (P _s = 0) or a useful signal is present, but the condition P _p > P _s is satisfied, and if at the input of the AAR P _p ≤ P _s , unintentional suppression of the useful signal during adaptation can occur [1].

 The closest in technical essence to the claimed method is the ACE signal method implemented in AAP according to US patent N 4.713.668 of 12/15/1987.

где W_i(t) - текущее значение i-го ВК, Δt - интервал временной дискретизации, V_yi - значение i-го управляющего сигнала, Y(t) - взвешенно суммированный (выходной) сигнал, X_i(t) - сигнал, принятый i-м АЭ, μ - коэффициент усиления (передачи), ^* - обозначение операции комплексного сопряжения.This prototype method of ACE signals provides for the reception of signals at N spatially separated points, the calculation of the VC and the weighted summation of the received signals. In this case, the calculation of the VC is carried out according to the algorithm for maximizing the signal / (noise + noise) ratio (MOSP) at the output of the AAR, according to the expression

where W _i (t) is the current value of the i-th VK, Δt is the time sampling interval, V _yi is the value of the i-th control signal, Y (t) is the weighted summed (output) signal, X _i (t) is the signal, adopted by the i-th AE, μ - gain (transmission), ^* - designation of the operation of complex pairing.

точно определяют направление прихода полезного сигнала, его частоту и учитывают характеристики собственно АР, то есть V_yi = V_ci, где V_ci - сигналы, принимаемые i-ми АЭ (сигналы, описывающие волновой фронт полезного сигнала). В дальнейшем для упрощения записи управляющие сигналы V_yi, а также сигналы V_ci, описывающие волновой фронт полезного сигнала, представим в виде N-мерных векторов

и

, здесь θ_y и θ_c - предполагаемое и реальное направление прихода сигнала, T - знак, обозначающий операцию транспонирования. Кроме того, для удобства записи и упрощения математических выкладок выражение (1) представим в векторной форме

где

- N-мерный вектор весовых коэффициентов (ВВК)

- вектор, составляющими которого являются сигналы с выходов АЭ

- взвешенно суммированный сигнал (выходной сигнал ААР).The application of this method is based on the assumption that the control signals V _yi ,

precisely determine the direction of arrival of the useful signal, its frequency and take into account the characteristics of the AP itself, that is, V _yi = V _ci , where V _ci are the signals received by the i-th AEs (signals describing the wavefront of the useful signal). Further, to simplify the recording, the control signals V _yi , as well as the signals V _ci describing the wavefront of the useful signal, are represented as N-dimensional vectors

and

, here θ _y and θ _c are the estimated and actual direction of arrival of the signal, T is the sign denoting the transpose operation. In addition, for the convenience of writing and simplifying mathematical calculations, expression (1) can be represented in vector form

Where

- N-dimensional vector of weights (VVK)

- vector, the components of which are signals from the outputs of AE

- weighted summed signal (AAR output signal).

Вместе с тем в реальных условиях направление прихода полезного сигнала θ_c известно, как правило, лишь с точностью до некоторого углового сектора θ_c∈[θ_y1÷θ_y2], где θ_y1,θ_y2 - левая и правая границы сектора, в пределах которого ожидается приход полезного сигнала. Данная неопределенность ведет в тому, что предполагаемое направление прихода полезного сигнала θ_y будет не равно истинному направлению прихода полезного θ_c(θ_y≠ θ_c) и вектор управляющих сигналов

не будет коллинеарен

(не совпадает с точностью до постоянного коэффициента). Вместе с тем, если

не коллинеарен

, то применение данного способа АПФ сигналов может привести к непреднамеренному подавлению полезного сигнала, причем глубина непреднамеренного подавления полезного сигнала прямо пропорционально зависит от входного отношения сигнал/шум и величины ошибки в априорных данных о направлении прихода полезного сигнала (углового расстояния между векторами

Известны устройства, реализующие АПФ сигналов, например, ААР по патенту РФ N 2099838, опублик. 20.12.1997 года, бюл. N 35.However, the prototype method has the disadvantage of low noise immunity of signal reception, since the basis for the applicability of this method is the coincidence of the control signals (control signals of the directivity characteristic of the AAR) V _yi , with signals V _ci received by the i-AE (signals describing the wavefront of the useful signal )

At the same time, under real conditions, the direction of arrival of the useful signal θ _{c is} known, as a rule, only up to a certain angular sector θ _c ∈ [θ _y1 ÷ θ _y2 ], where θ _y1 , θ _y2 are the left and right boundaries of the sector, within which is expected to receive a useful signal. This uncertainty leads to the fact that the expected direction of arrival of the useful signal θ _y will not equal the true direction of arrival of the useful θ _c (θ _y ≠ θ _c ) and the vector of control signals

will not be collinear

(does not coincide up to a constant coefficient). However, if

not collinear

, then the use of this method of ACE signals can lead to unintentional suppression of a useful signal, and the depth of unintentional suppression of a useful signal is directly proportional to the input signal-to-noise ratio and the magnitude of the error in a priori data on the direction of arrival of the useful signal (angular distance between vectors

Known devices that implement ACE signals, for example, AAP according to the patent of the Russian Federation N 2099838, published. 12/20/1997, bull. N 35.

 This device consists of N AEs connected through complex weight multipliers (HLCs) with the corresponding inputs of a common adder, N adaptive circuits and a signal power maximizing unit. The first inputs of N adaptive circuits are connected to the outputs of the corresponding AEs, the second inputs of N adaptive circuits are connected to the output of a common adder, and the first outputs of N adaptive circuits are connected to the corresponding inputs of the HLC. The first and second inputs of the power maximization unit are connected respectively to the first and second outputs of the adaptive circuits, and the outputs are connected to the corresponding third inputs of the adaptive circuits. The output of the total adder is the output of the AAP.

 However, the known device has the disadvantage of low noise immunity of signal reception, due to the fact that VC calculation should be carried out at those times when there is no useful signal at the AAP input, that is, it is applicable only for receiving signals that have a pause in the process of their transmission (for example, signals with pseudo-random tuning of the operating frequency). Calculation of VC in the presence of a useful signal at the input can lead to its unintentional suppression.

 AAR is also known by the application of Japan N 57-112564 from 04/02/1992

 This device consists of N antenna elements, N weighing circuits (complex weight multipliers), N switches, an adder, a signal processing unit, a control unit, and the i-th AE through the switches connected to the first inputs of the i-weighting circuits and the signal processing unit, The i-th outputs of the signal processing unit are connected to the second inputs of the i-weighting circuits, the outputs of which are connected to the i-inputs of the adder, the output of which, connected to the second input of the signal processing unit and the input of the control unit, is an AAP output, and i-th outputs of the control unit are connected to i-th switches.

However, this device has disadvantages:
low noise immunity of signal reception, since VC calculation should be performed at those times when there is no useful signal at the AAP input (P _s = 0) or a useful signal is present, but the condition P _p > P _s is satisfied, and if at the input AAP P _p ≤ P _s , unintentional suppression of the useful signal during adaptation can occur [1];
inefficient use of the AE resource, because based on the evaluation of the output signal, a decision is made on the choice of the number of AE, and accordingly, the number of signals M _l for the subsequent ACE, and situations are possible when one or more AEs will not be used to receive signals.

 Closest to the claimed device in its technical essence is the AAP described in US patent N 4.713.668 from 12/15/1987.

 The prototype device consists of N antenna elements, N complex conjugation units, N complex weight multipliers, N-input common adder, control signal input device, subtracter, signal evaluation unit, N adaptation circuits, and the output of the i-th AE, where i = 1, 2, 3, ... N, is connected to the input of the i-th complex conjugation unit and with the first input of the corresponding i-th HLC, the output of the i-th complex conjugation unit is connected to the signal input of the i-th adaptation loop, i-th the output of the control signal input device is connected to the control input ohm of the i-th adaptation loop, the output of which is connected to the second input of the i-th HLC, the output of the i-th HLC is connected to the i-th input of the N-input common adder, the output of the adder connected to the first input of the subtractor and the input of the signal estimator is AAP output, the output of the signal estimator is connected to the second input of the subtractor, the output of which is connected to the information inputs of the N-adaptation blocks. In this case, the i-th VK is calculated according to expression (1).

не будет коллинеарен

. Применение

приведет к формированию максимума диаграммы направленности (ДН) ААР, ориентированного в направлении (θ_y), а следовательно, возможно непреднамеренное подавление полезного сигнала, причем глубина подавления полезного сигнала прямо пропорционально зависит от входного отношения сигнал/шум и величины ошибки в априорных данных (углового расстояния между векторами

Целью заявленных объектов изобретения является разработка способа и устройства АПФ сигналов, обеспечивающих более высокую помехозащищенность приема сигналов в случаях, когда направление прихода полезного сигнала известно с точностью до некоторого углового сектора [θ_y1÷θ_y2].
Поставленная цель достигается тем, что в известном способе, заключающемся в приеме сигналов в N пространственно разнесенных точках, где N ≥ 2, расчете ВК и взвешенном суммировании принятых сигналов, дополнительно генерируют первый и второй опорные сигналы с различными частотами. Каждый из этих сигналов разделяют на N равноамплитудных составляющих. После этого каждую из составляющих сдвигают по фазе на предварительно заданную величину, затем i-й принятый сигнал, где i=1, 2, ... N, и i-е составляющие первого и второго опорных сигналов суммируют. При этом расчет i-х ВК осуществляют согласно выражению
W_i(t+Δt) = W_i(t)-μ[V_yi-Y(t)Z $\genfrac{}{}{0ex}{}{\text{*}}{\text{i}}$ (t)], (3)
где W_i(t) - текущее значение i-го ВК, W_i(t+Δt) - значение i-го ВК в момент времени (t+Δt), Δt - интервал временной дискретизации, V_yi - значение i-го управляющего сигнала, Y(t) - взвешенно суммированный сигнал, Z_i(t) - суммарный сигнал, полученный в результате сложения i-го принятого сигнала и i-х составляющих первого и второго опорных сигналов, μ - коэффициент усиления (передачи).However, the prototype device has the disadvantage of low noise immunity of signal reception, so in a real signal-noise environment, the expected direction of arrival of the useful signal θ _y can differ significantly from the real θ _c and, accordingly, the vector of control signals previously calculated on the basis of a priori information about the parameters of the useful signal

will not be collinear

. Application

will lead to the formation of a maximum of the directivity pattern (DF) of the AAP oriented in the direction (θ _y ), and therefore, unintentional suppression of the useful signal is possible, and the depth of the suppression of the useful signal is directly proportional to the input signal-to-noise ratio and the magnitude of the error in a priori data (angular distance between vectors

The aim of the claimed objects of the invention is to develop a method and device of ACE signals that provide higher noise immunity of signal reception in cases where the direction of arrival of the useful signal is known accurate to a certain angular sector [θ _y1 ÷ θ _y2 ].
This goal is achieved by the fact that in the known method, which consists in receiving signals at N spatially separated points, where N ≥ 2, calculating the VC and weighted summation of the received signals, the first and second reference signals with different frequencies are additionally generated. Each of these signals is divided into N equal-amplitude components. After that, each of the components is phase shifted by a predetermined value, then the i-th received signal, where i = 1, 2, ... N, and the i-th components of the first and second reference signals are summed. In this case, the calculation of i-x VC is carried out according to the expression
W _i (t + Δt) = W _i (t) -μ [V _yi -Y (t) Z $\genfrac{}{}{0ex}{}{\text{*}}{\text{i}}$ (t)], (3)
where W _i (t) is the current value of the i-th VK, W _i (t + Δt) is the value of the i-th VK at the time (t + Δt), Δt is the time sampling interval, V _yi is the value of the i-th control signal, Y (t) is the weighted summed signal, Z _i (t) is the total signal obtained by adding the i-th received signal and the i-components of the first and second reference signals, μ is the gain (transmission).

где

- вектор, составляющими которого являются суммарные сигналы, полученные от сложения i-го принятого сигнала и i-х составляющих первого и второго опорных сигналов

Поставленная цель в заявленном устройстве достигается тем, что в известную ААР, содержащую N АЭ, N блоков комплексного сопряжения, N комплексных весовых умножителей, N-входовый общий сумматор, устройство ввода управляющих сигналов, N контуров адаптации, причем вход i-го блока комплексного сопряжения соединен с первым входом соответствующего i-го КВУ, выход i-го блока комплексного сопряжения соединен с сигнальным входом i-го контура адаптации, а i-й выход устройства ввода управляющих сигналов соединен с управляющим входом i-го контура адаптации, выход которого соединен со вторым входом i-го КВУ, выход i-го КВУ соединен с i-м входом N-входового общего сумматора, выход которого является выходом ААР, дополнительно введены первый и второй генераторы опорных сигналов, первый и второй блоки фазовращателей, N трехвходовых сумматоров (в дальнейшем первый и второй генераторы опорных сигналов для упрощения записи будем называть первый и второй генераторы). Выход каждого i-го АЭ подключен к первому входу i-го трехвходового сумматора. Выход первого генератора подключен к входу первого блока фазовращателей, а выход второго генератора подключен к входу второго блока фазовращателей, i-й выход первого блока фазовращателей соединен со вторым входом i-го трехвходового сумматора, а i-й выход второго блока фазовращателей соединен с третьим входом i-го трехвходового сумматора. Выход i-го трехвходового сумматора соединен с первым входом i-го КВУ. Выход N-входового общего сумматора соединен с информационными входами N контуров адаптации. Кроме того, расчет ВК осуществляют согласно выражению (3).For the convenience of writing and simplifying mathematical calculations, expression (3) can be represented in vector form

Where

- a vector, the components of which are the total signals obtained from the addition of the i-th received signal and i-components of the first and second reference signals

The goal in the claimed device is achieved by the fact that in the well-known AAR containing N AE, N complex conjugation blocks, N complex weight multipliers, N-input common adder, input signal control device, N adaptation loops, and the input of the i-th complex conjugation block connected to the first input of the corresponding i-th HLC, the output of the i-th complex conjugation unit is connected to the signal input of the i-th adaptation loop, and the i-th output of the control signal input device is connected to the control input of the i-th adaptation loop the output of which is connected to the second input of the i-th HLC, the output of the i-th HLC is connected to the i-th input of the N-input common adder, the output of which is the output of the AAR, the first and second reference signal generators, the first and second blocks of phase shifters are additionally introduced, N three-input adders (hereinafter, the first and second generators of reference signals to simplify the recording, we will call the first and second generators). The output of each i-th AE is connected to the first input of the i-th three-input adder. The output of the first generator is connected to the input of the first block of phase shifters, and the output of the second generator is connected to the input of the second block of phase shifters, the i-th output of the first block of phase shifters is connected to the second input of the i-th three-input adder, and the i-th output of the second block of phase shifters is connected to the third input i-th three-input adder. The output of the i-th three-input adder is connected to the first input of the i-th HLC. The output of the N-input common adder is connected to the information inputs of N adaptation loops. In addition, the calculation of VC is carried out according to expression (3).

 The above set of distinctive essential features, due to the transformation of signals received by AEs at different points of space that is not essential for a useful signal, allows, under conditions when the direction of arrival of the useful signal is known to within a certain angular sector, to receive the useful signal while suppressing interference without significant deterioration OSPSH at the output of a device that implements the inventive method of ACE signals. This is achieved due to the formation of a wider main lobe of the radiation pattern (AH) of the AAR, overlapping the angular sector, within which a useful signal is expected to arrive.

 The analysis of the prior art made it possible to establish that analogues that are characterized by a combination of features that are identical to all the features of the claimed technical solutions are absent, which indicates compliance of the claimed inventions with the condition of patentability "novelty".

 Search results for known solutions in this and related fields of technology in order to identify features that match the distinctive features of the claimed objects from prototypes have shown that they do not follow explicitly from the prior art. The prior art also did not reveal the popularity of the impact provided by the essential features of the claimed inventions of the transformations on the achievement of these technical results. Therefore, the claimed invention meets the condition of patentability "inventive step".

 The claimed method and device that implements it are illustrated by drawings: FIG. 1 is a structural diagram of the AAR; FIG. 2 is a structural diagram of an adaptation circuit; FIG. 3 is a block diagram of a phase shifter unit; FIG. 4 is a graph illustrating the results of the analysis of the effectiveness of the claimed method of ACE signals and the device that implements it.

 The possibility of implementing the claimed method is explained as follows.

 Adaptive spatial filtering of signals in the general case refers to the weighted summation of signals received at N-spatially separated points (received at N-spatially separated antennas), which, under the conditions of a priori uncertainty about the noise environment, optimizes the weighted-summed signal according to a criterion that implies an improvement in quality further signal processing, for example, according to the ISMT criterion. In this case, the ACE algorithm is usually called the mathematical expression (rule) of calculating the VK [3].

L - число помех (L ≤ N-1) и изотропного шума ш(t), причем X_i(t) = s_i(t) + p_i(t) + ш(t), здесь s_i(t), p_i(t) - полезный сигнал, l-я помеха, принимаемая i-м АЭ. Для наглядности будем считать, что s(t), p_l(t), ш(t) взаимно некоррелированные, стационарные, эргодические случайные процессы. Будем считать также, что несущие частоты сигнала и помех тождественны, а источники сигнала и помех точечные по угловым размерам и находятся в дальней зоне, антенная решетка согласована по поляризации с электромагнитной волной принимаемых сигналов.The known prototype ACE method consists in receiving in N spatially spaced places (receiving in N spaced apart antennas) a useful signal s (t), interference

L is the number of interference (L ≤ N-1) and isotropic noise w (t), and X _i (t) = s _i (t) + p _i (t) + w (t), here s _i (t), p _i (t) is the useful signal, the l-th interference received by the i-th AE. For clarity, we assume that s (t), p _l (t), w (t) are mutually uncorrelated, stationary, ergodic random processes. We will also assume that the carrier frequencies of the signal and interference are identical, and the signal and interference sources are point-wise in angular size and are in the far zone, the antenna array is polarized with the electromagnetic wave of the received signals.

 In this case, the VVK, providing control of the amplitude-phase distribution of currents at the outputs of the corresponding AEs, is calculated according to expression (2).

и предварительно рассчитанного (на основе априорной информации о частоте, направлении прихода полезного сигнала и с учетом характеристик собственно АР) вектора управляющих сигналов -

используется выходной сигнал АР (сигнал обратной связи) - Y(t). За счет использования сигнала обратной связи обеспечивается отслеживание текущих изменений сигнально-помеховой обстановки (подавление помех с изменяющимися пространственно-временными параметрами).In this method, when calculating the IHC, in addition to the signals from the AE outputs,

and pre-calculated (based on a priori information about the frequency, direction of arrival of the useful signal and taking into account the characteristics of the actual AR) the vector of control signals -

the output signal AP (feedback signal) is used - Y (t). Through the use of a feedback signal, tracking of current changes in the signal-noise environment is provided (suppression of interference with changing spatio-temporal parameters).

где R_пш - корреляционная матрица (КМ) помех и шума; β - нормирующий множитель.In the case when the VVK is calculated in the absence of a useful signal, the VVK calculated according to expression (2), as t _ → ∞, converges to the vector determined by the expression

where R _psh - correlation matrix (CM) of interference and noise; β is the normalizing factor.

где R_xx = R_cc + R_пш, R_xx - КМ входных сигналов, R_сс - КМ полезного сигнала.If there is a useful signal at the AAR input (when calculating the IHC), the algorithm for calculating the IHC (2) as t _ → ∞ converges to the vector determined by the expression

where R _xx = R _cc + R _psh , R _xx is the CM of the input signals, R _ss is the CM of the useful signal.

может быть определен как

где j - мнимая единица, φ_ci - фазовый сдвиг сигнала, принимаемого i-м антенным элементом, к точке, принятой за фазовый центр АР, определяемая выражением

здесь d_i - расстояние от АЭ, выбранного за фазовый центр АР (как правило это первый АЭ) до i-го АЭ, λ - длина волны принимаемого сигнала (помехи).For a linear equidistant AR consisting of identical and non-interacting AEs, the signal wavefront vector

can be defined as

where j is the imaginary unit, φ _ci is the phase shift of the signal received by the i-th antenna element to the point taken as the phase center of the AR, defined by the expression

here d _i is the distance from the AE selected for the phase center of the AR (usually the first AE) to the i-th AE, λ is the wavelength of the received signal (interference).

где φ_yi - фазовый сдвиг сигнала, принимаемого i-м АЭ, определяемый согласно выражению (8) с учетом замены реального направления прихода полезного сигнала θ_c предполагаемым направлением прихода θ_y.
Как уже отмечалось, недостатком рассмотренного способа-прототипа АПФ сигналов является низкая помехозащищенность приема сигналов при отсутствии точной априорной информации о направлении прихода полезного сигнала.Similarly, we can determine the vector of control signals

where φ _yi is the phase shift of the signal received by the ith AE determined according to expression (8), taking into account the replacement of the real direction of arrival of the useful signal θ _{with the} estimated direction of arrival θ _y .
As already noted, the disadvantage of the considered prototype method of ACE signals is the low noise immunity of signal reception in the absence of accurate a priori information about the direction of arrival of the useful signal.

(неточностям априорных данных о направлении прихода полезного сигнала θ_y), устраняя таким образом недостаток способа-прототипа.Additional actions on the signals allow, under conditions when the direction of arrival of the useful signal is known to an accuracy of a certain angular sector, to form a wider main lobe of the NAM AAR, oriented in the direction of this angular sector and overlapping it, and thereby provide low sensitivity of the proposed method of ACE signals to inaccuracies in the task of the vector of control signals

(inaccuracies of a priori data on the direction of arrival of the useful signal θ _y ), thus eliminating the disadvantage of the prototype method.

 The possibility of implementing the proposed method is explained as follows.

The inventive method, as well as the prototype method, provides for the reception of a useful signal, interference and isotropic noise in N spatially spaced places (reception of N antennas spaced in space). Simultaneously with the reception of a useful signal, interference and isotropic noise, two reference signals F ₁ (t), F ₂ (t) are generated with different frequencies different from the frequency of the received signal,
F ₁ (t) = Aexp (jω ₁ t), (10)
F ₂ (t) = Aexp (jω ₂ t), (11)
where A is the amplitude of the reference signals, ω ₁ , ω ₂ are the carrier frequencies of the first and second reference signals. In this case, the frequencies ω ₁ , ω ₂ are selected according to the condition
ω ₁ <(ω ₀ -Δω), ω ₂ > (ω ₀ + Δω), (12)
where ω ₀ is the carrier frequency of the received useful signal, Δω is the width of the spectrum of the received useful signal. This condition is introduced so that the reference signals F ₁ (t), F ₂ (t) are subsequently filtered out in the input circuits of the preselector of the radio receiving device, with which it is supposed to use a device that implements the inventive method of ACE signals. It should also be noted that when choosing the frequencies ω ₁ , ω ₂ it is necessary to provide that the reference signals F ₁ (t), F ₂ (t) do not get into the radio receiver through the mirror and other side reception channels.

где A_1i, A_2i - амплитуда i-х составляющих первого и второго опорного сигналов (причем

Фазовые сдвиги Δφ_1i,Δφ_2i выбирают согласно условий

где S_i(θ_c-Δθ),S_i(θ_c+Δθ) - сигналы на выходе i-го АЭ, приходящие соответственно с направлений θ_c-Δθ,θ_c+Δθ, где Δθ - максимальная ошибка в задании направления прихода полезного сигнала, причем θ_c-Δθ = θ_y1, θ_c+Δθ = θ_y2.
Необходимо заметить, что сигналы S_i(θ_y1) и S_i(θ_y2) реально на входе ААР не присутствуют, но их параметры можно рассчитать, так как известно направление их прихода θ_y1,θ_y2 и характеристики собственно АР. Расчет S_i(θ_y1) и S_i(θ_y2), представленных в виде N-мерных векторов

аналогичен расчету

в соответствии с выражениями (7), (8) с учетом θ_y1,θ_y2.
Значения амплитуды A сигналов на выходе первого и второго генераторов целесообразно выбирать согласно рекомендациям, представленным в табл. 1, с учетом сигнально-помеховой обстановки.Next, each of these reference signals is divided into N equal-amplitude components, then each of the obtained components is shifted by a predetermined phase value Δφ _1i , Δφ _2i , i.e., signals are generated

where A _1i , A _2i is the amplitude of the i-components of the first and second reference signals (and

The phase shifts Δφ _1i , Δφ _2i are selected according to the conditions

where S _i (θ _c -Δθ), S _i (θ _c + Δθ) are the signals at the output of the i-th AE coming respectively from the directions θ _c -Δθ, θ _c + Δθ, where Δθ is the maximum error in setting the direction of arrival useful signal, and θ _c -Δθ = θ _y1 , θ _c + Δθ = θ _y2 .
It should be noted that the signals S _i (θ _y1 ) and S _i (θ _y2 ) are not really present at the AAP input, but their parameters can be calculated, since the direction of their arrival θ _y1 , θ _y2 and the characteristics of the AP itself are known. Calculation of S _i (θ _y1 ) and S _i (θ _y2 ) represented as N-dimensional vectors

similar to calculation

in accordance with expressions (7), (8) taking into account θ _y1 , θ _y2 .
The values of the amplitude A of the signals at the output of the first and second generators are advisable to choose according to the recommendations presented in table. 1, taking into account the signal-noise situation.

Here N is the number of AEs, P _c is the signal power at the output of AEs, σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ is the dispersion of thermal noise, Δθ _max is the maximum error in a priori data on the direction of arrival of the useful signal.

i-х составляющих опорных сигналов F_1i(t), F_2i(t), для наихудшего случая (в смысле наихудшей сигнально-помеховой ситуации для АПФ сигналов) мы можем утверждать, что в других случаях (наличие помехи на входе ААР (P_п ≠ 0)) данное значение амплитуды также обеспечит приемлемое значение выходного ОСПШ.The amplitude values (Table 1) A of the signals at the output of the first and second generators were obtained in the case of the “worst” signal-noise situation for the operation of the AAR (by the “worst” signal-noise situation we mean the situation when there is no interference at the input of the AAR (P _n = 0), and the error in setting the directions of arrival of the signal and the input signal-to-noise ratio take their maximum values). It should be noted that knowing the value of the amplitude A of the signals at the output of the first and second generators, and accordingly the amplitudes

of the i-component reference signals F _1i (t), F _2i (t), for the worst case (in the sense of the worst signal-noise situation for ACE signals), we can state that in other cases (there is interference at the input of the AAR (P _p ≠ 0)) this amplitude value will also provide an acceptable value of the output SINR.

The value of P _c at the AE output can be directly measured by a signal power meter if there is no interference at the AAP input (P _p = 0). In the case when there is interference at the input of the AAR, that is, P _p ≠ 0, the value of P _c and σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ can be calculated by a well-known method (see, for example, the book by V. A. Meshalkin, B. V. Sosunov, “Fundamentals of the energy calculation of radio channels,” Leningrad, Military Communications Academy, 1991, pp. 7–20), taking into account the known characteristics of a radio transmitting device, antenna-feeder devices of a radio receiving and transmitting device, and actual propagation conditions of radio waves on a radio link path. The maximum error in a priori data on the direction of arrival of the useful signal (Δθ _max ) is defined as Δθ _max = (θ _y2 -θ _y1 ) / 2. In addition, the values of P _c and σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ can be specified with some “margin” based on the typical operating conditions of radio lines.

In the case of AAP implementation on digital signal processing processors (in this case, CM R _xx can be normalized), it is advisable to choose the values of the amplitudes A of the signals at the output of the first and second generators according to the recommendations presented in Table. 2 (excluding useful signal power).

Для удобства записи и упрощения математических расчетов сигналы X_i(t), F_1i(t), F_2i(t) запишем в виде N-мерных векторов

Ввод сигналов F_1i(t) и F_2i(t) в каналы ААР позволит наряду с реально принятыми сигналами X_i(t) имитировать прием сигналов S_i(θ_y1) = F_1i(t), S_i(θ_y2) = F_2i(t) с направлений θ_y1 и θ_y2.
Очевидно, что с учетом

выражение для расчета ВВК примет вид (4) и процедура расчета ВВК (4) при t _→ ∞ будет сходиться к выражению

где R_ZZ = R_XX + R_FF1 + R_FF2, R_FF1, R_FF2 - КМ первого и второго опорных сигналов.By converting the first and second reference signal (dividing the first and second reference signal by N equal-amplitude components and the phase shift of each component), we obtain N components of the first reference signal (13) and N components of the second reference signal (14), and the ith components The first and second reference signals will coincide with the signals that would be received at the i-th space points (i-AE) S _i (θ _y1 ) and S _i (θ _y2 ), from the directions θ _y1 and θ _y2, respectively. The peculiarity of these signals is that they are not actually received by the AR from the indicated directions, but are artificially formed and entered into the AAR after the antenna elements. The input signals F _1i (t) and F _2i (t) in the AAP channels is carried out by summing the i-components of the first and second reference signal with i-signals received at i-points of space

For the convenience of recording and simplifying mathematical calculations, the signals X _i (t), F _1i (t), F _2i (t) can be written in the form of N-dimensional vectors

The input of the signals F _1i (t) and F _2i (t) into the AAP channels will, along with the actually received signals X _i (t), simulate the reception of signals S _i (θ _y1 ) = F _1i (t), S _i (θ _y2 ) = F _2i (t) from the directions θ _y1 and θ _y2 .
Obviously, given

the expression for calculating the IHC will take the form (4) and the procedure for calculating the IHC (4) as t _ → ∞ will converge to the expression

where R _ZZ = R _XX + R _FF1 + R _FF2 , R _FF1 , R _FF2 - CM of the first and second reference signals.

(здесь H - знак, обозначающий операцию эрмитового сопряжения), а следовательно,
R_ZZ = R_ПШ + B, (22)
где

определяется согласно выражению (7).Moreover, for the correlation matrices R _FF1 , R _FF2 , the peer-to-peer approximation is valid

(here H is the sign denoting the operation of Hermitian conjugation), and therefore
R _ZZ = R _PN + B, (22)
Where

is determined according to expression (7).

= 1 и проанализировав ее элементы, несложно убедиться, что данные элементы имеют вид (здесь для наглядности мы полагаем, что P_с = 1, а также A=1)

где

- элемент, стоящий на пересечении k-й строки (k=1, 2, ... N) и i-го столбца матрицы B, φ_ki= φ_ck-φ_ci; φ_ck, φ_ci - аргументы k-го и i-го элементов вектора

(выражение (7)); Δφ_ki= φ_i(Δθ)-φ_k(Δθ); φ_i(Δθ), φ_k(Δθ), фазовые сдвиги, определяемые согласно выражению (8) при замене θ_c на величину Δθ.
С учетом формулы Эйлера запишем

,
а следовательно, выражение (23) примет вид

где β_ki= 1+2cos(Δφ_ki).
Вследствие того, что элементы матрицы B имеют вид (24), мы можем воспользоваться результатами работы [4] и утверждать, что использование алгоритма расчета ВВК (4) или (21) для АПФ сигналов в случае, когда направление прихода полезного сигнала известно с точностью до некоторого углового сектора, приведет к повышению ОСПШ на выходе ААР. Следовательно, помехозащищенность приема сигналов при использовании данной ААР увеличится.By normalizing the matrix B, based on the condition

= 1 and after analyzing its elements, it is easy to verify that these elements have the form (here, for clarity, we assume that P _c = 1, as well as A = 1)

Where

is the element at the intersection of the kth row (k = 1, 2, ... N) and the ith column of the matrix B, φ _ki = φ _ck -φ _ci ; φ _ck , φ _ci - arguments of the kth and i-th elements of the vector

(expression (7)); Δφ _ki = φ _i (Δθ) -φ _k (Δθ); φ _i (Δθ), φ _k (Δθ), phase shifts determined according to expression (8) when replacing θ _c with Δθ.
Given the Euler formula, we write

,
therefore, expression (23) takes the form

where β _ki = 1 + 2cos (Δφ _ki ).
Due to the fact that the elements of matrix B have the form (24), we can use the results of [4] and state that the use of the algorithm for calculating the VVC (4) or (21) for ACE signals in the case when the direction of arrival of the useful signal is known with accuracy to a certain angular sector, will lead to an increase in the SINR at the output of the AAR. Consequently, the noise immunity of signal reception when using this AAP will increase.

 The results of simulation allow us to conclude that the claimed method of ACE signals is not very sensitive to errors in a priori data on the direction of arrival of a useful signal (a description of simulation is given on page 26).

A device implementing this method of ACE signals is shown in FIG. 1 and consists of N antenna elements 1 (1 ₁ , 1 ₂ , ... 1 _N ), N blocks of complex conjugation 2 (2 ₁ , 2 ₂ , ... 2 _N ), N complex weight multipliers 3 (3 ₁ , 3 ₂ ,... 3 _N ), N-input total adder 4, control signal input device 5, N adaptation circuits 6 (6 ₁ , 6 ₂ , ... 6 _N ), the first and second generators 7 (7 ₁ , 7 ₂ ), the first block of phase shifters 8, the second block of phase shifters 9, N three-input adders 10 (10 ₁ , 10 ₂ , ... 10 _N ). The output of each i-th antenna element 1, where i = 1, 2, 3, ... N, is connected to the first input of the i-th three-input adder 10. The output of the i-th three-input adder 10 is connected to the first input of the i-th complex weighted the multiplier 3 and through the i-th block of complex pairing 2 is connected to the signal input of the i-th adaptation circuit 6. The output of the i-th adaptation circuit 6 is connected to the second input of the i-th HLC 3. The output of the i-th HLC 3 is connected to the i-th the input of the N-input common adder 4. The output of the N-input common adder 4, connected to the information inputs of N adaptation circuits 6, is AAR exit. The output of the first generator 7 _{1 is} connected to the input of the first block of phase shifters 8, the i-th output of the first block of phase shifters 8 is connected to the second input of the i-th three-input adder 10. The output of the second generator 7 _{2 is} connected to the input of the second block of phase shifters 9, the i-th output of the second block phase shifters 9 is connected to the third input of the i-th three-input adder 10. The control input of the i-th adaptation circuit 6 is connected to the i-th output of the input device of the control signals 5.

 Each ith adaptation circuit 6 shown in FIG. 2, is intended to perform the iterative calculation of the VVK in accordance with expression (3) and consists of a series-connected amplifier 6.1, a multiplier 6.2, an integrator (integrating RC filter) 6.3, a subtracter 6.4. The input of the amplifier 6.1 is the information input of the adaptation circuit 6, and the second input of the multiplier 6.2 and the second input of the subtractor 6.4 is the signal and control input of the adaptation circuit 6. The output of the subtractor 6.4 is the output of the adaptation circuit 6.

 Antenna elements 1 are designed to receive electromagnetic waves. Typical antenna devices used in antenna arrays, for example, asymmetric vibrators, etc. can be used as AEs.

 Each i-th block of complex pairing 2 is designed to perform the operation of complex pairing of signals. These blocks can be made in the form of signal decomposition schemes into in-phase and quadrature components with the inclusion of an additional phase shifter at π / 2 in the quadrature branch. The circuit block complex conjugation is described in the patent of the Russian Federation N 2099838, published December 20, 1997, Bull. N 35.

 Complex weight multipliers 3 are designed to perform complex weighing operations, that is, changing the amplitude and phase of the signals to the desired value. The HLC scheme is described in RF patent N 2099838, published on December 20, 1997, Bull. 35.

 Each N input adder 4 and a three-input adder 10 are designed to perform the operation of summing signals. These adders (depending on the frequency range) can be made in the form of high-frequency transformers on coaxial or microstrip lines. The scheme and principle of operation of these adders are described in the patent of the Russian Federation N 2099838, published on 12.20.1997, Bull. N 35. In addition, adders 4, 10 can be implemented on bridge circuits for adding and distributing high-frequency oscillation powers, see, for example, the book by V.V. Zaentsev, V.M. Katushkina, S.E. London, ed. Z. I. Model. Devices for adding and distributing power of high-frequency oscillations. -M .: Soviet Radio, 1980, pp. 9-17.

 Information input device 5 is designed to enter information about the sector within which a useful signal is expected to arrive, and is an adjustable voltage source, and when implemented using a digital element base, an online storage device (see I.P. Stepanenko, Fundamentals of Microelectronics. M.: Soviet Radio, 1980, pp. 378-385).

Generators 7 ₁ , 7 ₂ are designed to generate two electrical oscillations of a given amplitude and frequency. As generators 7 ₁ , 7 ₂ can be used, for example, a frequency synthesizer based on a PLL with a variable frequency divider, see the book "Radio receivers" edited by prof. A.P. Zhukovsky M .: Higher School, 1989, pp. 107-109.

The first block of phase shifters 8 is designed to control phase shifts of signals. A diagram of this block of phase shifters is shown in FIG. 3 and consists of N phase shifters 8.1 (8.1 ₁ , 8.1 ₂ , ... 8.1 _N ), the inputs of which are connected together and are the input of the first block of phase shifters, and the output of the ith phase shifter is the i-th output of the first block of phase shifters. The scheme and principle of operation of phase shifters 8.1 are known and they can be performed on ferrite elements, see, for example, D.N. Voskresensky's book "Antennas and microwave devices. Designing phased antenna arrays." - M.: Radio and communications, 1981. , pp. 57-58, or on operational amplifiers, see, for example, the book by V. V. Nikitchenko “Functional Units of Adaptive Interference Compensators” Part 1. Leningrad, Military Academy of Communications, 1990, pp. 125-126.

 The second block of phase shifters 9 in composition, purpose and principle of operation is similar to the first block of phase shifters 8.

 Amplifier 6.1 is designed to amplify the signal, while the amplitude of the input signal increases by a factor of μ (μ is the gain of this amplifier), and the phase of the signal remains unchanged. As an amplifier, an amplifier stage with a bipolar transistor can be used, switched on according to the scheme with a common emitter with negative feedback (see the Handbook of electronic devices: in 2 volumes T.1 / Burin L.I., Vasiliev V.P. , Kaganov V.I. et al., Edited by D.P. Linde. - M .: Energy. P. 33. Fig. 1-30), or on analog integrated circuits, for example, IC 235 UR 7, IC 171 HC 1A and others, see, for example, the book of V.V. Nikitchenko "Functional Units of Adaptive Interference Compensators" Part 1. L .: Military Academy of Communications, 1990, pp. 57-64.

 The correlation mixer 6.2 (multiplier) is designed to form a cross-correlation function of the input signals. The scheme of the correlation mixer is described in RF patent N 2099838, published on December 20, 1997, Bull. N 35.

 The integrator 6.3 is designed to perform the operation of selecting a low-frequency component proportional to the amplitude and phase of the signal. An integrating RC filter can be used as an integrator, see the book Monzingo R.A., Miller T.U. Adaptive antenna arrays. Introduction to the theory. - M .: Radio and communications, 1986, p. 183. The integrator circuit is described in RF patent N 2099838, published on 12.20.1997, Bull. N 35.

 Subtractor 6.4 is designed to perform the operation of subtracting signals. The scheme of the subtractor is described in the patent of the Russian Federation N 2099838, published on December 20, 1997, Bull. N 35.

 In addition, the blocks that make up the AAP block diagram can be implemented in software, in particular, on a digital signal processor, for example, the TMS320C30 chip (see Digital signal processing processors: Reference / A.G. Ostapenko, S.I. Lavlinsky , A.B.Sushkov et al .; Edited by A.G. Ostapenko. - M.: Radio and Communications, 1994. - p. 88). The possibility of using digital signal processing processors to implement ACE algorithms for radio signals is substantiated in an article by Gorodilina VV, Glushankova EI "The control microprocessor controller for implementing algorithms for adaptive processing of radio signals in the antenna array," Radio Engineering, N 9, 1984, p. 72.

 The claimed device operates as follows.

i-х составляющих опорных сигналов F_1i(t), F_2i(t), целесообразно выбирать согласно рекомендациям, представленным в таблице 1 (таблице 2) в зависимости от сигнально-помеховой обстановки (исходя из максимально возможного значения входного отношения сигнал/шум и ошибки в задании направления прихода сигнала). В результате на выходе фазовращателей 8₁-8_N получим N составляющих первого опорного сигнала (13), а на выходе фазовращателей 9₁-9_N получим N составляющих второго опорного сигнала (14), причем i-е составляющие первого и второго опорного сигнала будут совпадать с сигналами S_i(θ_y1),S_i(θ_y2). Особенность этих сигналов состоит в том, что они реально не принимаются АР, а искусственно формируются и вводятся в каналы ААР после АЭ. Далее, i-е составляющие первого и второго опорного сигнала (сигналы (13), (14)) поступают соответственно на второй и третий входы i-го трехвходового сумматора, в котором происходит выполнение операции суммирования i-х составляющих первого и второго опорного сигнала с сигналами, принятыми i-ми антенными элементами. На выходе i-го трехвходового сумматора будет образован сигнал (17). Данные действия с сигналами позволят наряду с реально принятым полезным сигналом, помехами и шумами (X_i(t)) имитировать прием сигналов F_1i(t), F_2i(t).The useful signal s (t) and interference p (t) from the sources located at angles θ _c , θ _p relative to the normal to the axis of the AAR (in the general case θ _c ≠ θ _p ) and the isotropic noise w ( t). The useful signal and interference arrive at the i-th AE with different phase values φ _ci ≠ φ _pi .
The sum of the useful signal, interference and noise (X _i (t)) from the outputs of the i-x AEs are supplied to the first inputs of the corresponding three-input adders. Simultaneously with the reception of a useful signal, interference and noise in the generators 7 ₁ , 7 ₂ , two reference equal-amplitude signals F ₁ (t), F ₂ (t) (signals (10), (11)) are formed with different frequencies other than the frequency received useful signal. In this case, the frequencies of the first and second reference signals are selected according to condition (12). In phase shifters 8 ₁ -8 _N, each i-th component formed from the separation of the first reference signal is shifted by a predetermined phase value Δφ _li , and in phase shifters 9 ₁ -9 _N each i-th component formed from the separation of the second reference signal, phase Δφ _{2i is} shifted by a predetermined value, that is, signals (13), (14) are generated. The phase shifts Δφ _1i , Δφ _2i are selected according to conditions (15), (16). The values of the amplitude A of the signals at the output of the first and second generator, and, accordingly, the amplitudes

of the i-component reference signals F _1i (t), F _2i (t), it is advisable to choose according to the recommendations presented in table 1 (table 2) depending on the signal-noise situation (based on the maximum possible value of the input signal-to-noise ratio and errors in setting the direction of arrival of the signal). As a result, at the output of phase shifters 8 ₁ -8 _N we get N components of the first reference signal (13), and at the output of phase shifters 9 ₁ -9 _N we get N components of the second reference signal (14), and the ith components of the first and second reference signal will be coincide with the signals S _i (θ _y1 ), S _i (θ _y2 ). The peculiarity of these signals is that they are not actually received by the AR, but are artificially formed and introduced into the AAR channels after the AE. Further, the ith components of the first and second reference signal (signals (13), (14)) are respectively supplied to the second and third inputs of the i-th three-input adder, in which the operation of summing the i-components of the first and second reference signal signals received by the i-th antenna elements. At the output of the i-th three-input adder, a signal (17) will be generated. These actions with the signals will allow simulating the reception of signals F _1i (t), F _2i (t) along with a really received useful signal, interference and noise (X _i (t)).

(здесь "-" обозначает операцию усреднения по времени) и соответствующего управляющего сигнала V_yi, поступающего из устройства ввода управляющих сигналов.The signals Z _i (t) from the outputs of the i-th three-input adders are fed to the first inputs of the i-th complex weight multipliers and through the i-th blocks of complex coupling to the signal inputs of the i-th adaptation circuits. In complex weight multipliers, the signals Z _i (t) are multiplied by the corresponding VK W _i (t) coming from the outputs of the i-th adaptation circuits, which ensures a change in the amplitude and phase of the signals Z _i (t). From the outputs of the complex weight multipliers, the "weighted" signals are fed to the ith inputs of the N input total adder, where they are summed, forming the output signal AAP
Y (t) = (W ₁ (t) Z ₁ (t) + W ₂ (t) Z ₂ (t) + ... + W _N (t) Z _N (t)). (25)
The output signal Y (t) is additionally supplied to the information inputs of the adaptation circuits, where the operation of calculating the mutual correlation of the output signal Y (t) with the input signal Z _i (t) is performed. The calculation of VK W _i (t) is constantly performed in accordance with expression (3) in adaptive circuits, taking into account the value of the cross-correlation coefficient

(here, “-” denotes the time averaging operation) and the corresponding control signal V _yi coming from the control signal input device.

 A detailed comparison of the capabilities of the proposed method and device that implements this method of ACE signals, and the method and device prototype ACE signals, was carried out using the simulation method, which further confirms the possibility of their technical implementation.

By way of illustration in FIG. 4 shows the dependence of the SINR obtained by simulation P _c / (P _p + σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ ) at the output of the AAR from the angle of arrival of the useful signal θ _c .
In the simulation, the following assumptions about the characteristics of the AR and the signal-noise environment were used:
- AR linear equidistant N = 3, interelement distance d = λ / 2 (λ is the wavelength of the signal), AE isotropic and noninteracting;
- the carrier frequencies of the signal and interference are identical f _c = f _p = 300 MHz (λ = 1 m);
- angle of arrival of interference (relative to the normal to the line of location of the AE) θ _p = 10 ^o ;
- input signal-to-noise ratio 10lg (P _c / σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ ) = 10 dB;
- input noise / noise ratio 10lg (P _p / σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ ) = 20 dB.

It was also assumed that the characteristics of the AR are known exactly, and the direction of arrival of the signal is known accurate to the angular sector θ _y1 = 35 ^o , θ _y2 = 65 ^o .

 When conducting simulation, the frequency, amplitude, and phase of the ith components of the first and second reference oscillations were selected based on the above recommendations.

1. The frequencies of the first and second reference signals F ₁ (t), F ₂ (t) were selected according to condition 12 and amounted to f ₁ = 290 MHz ((λ = 1,034 m) and f ₂ = 310 MHz ((λ = 0.967 m )

2. The amplitude A of the signal at the output of the first and second reference oscillator was calculated according to the recommendations presented in table 1, respectively, from the conditions ((P _c / σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ = 10 dB, at P _c = 12 mW, σ $\genfrac{}{}{0ex}{}{\text{2}}{\text{w}}$ = 0.1 mW, Δθ _y = 15 ^o ) and amounted to A = 0.142 V.

и составили φ₁₁= φ₂₁ = 0^o, φ₁₂ = 99,85^o, φ₁₃ = 199,7^o, φ₂₂ = 168,66^o, φ₂₃ = 337,31^o.3. The phase shifts of the ith components of the first and second reference oscillations φ _1i , φ _{2i were} calculated according to expression (8) as

and amounted to φ ₁₁ = φ ₂₁ = 0 ^o , φ ₁₂ = 99.85 ^o , φ ₁₃ = 199.7 ^o , φ ₂₂ = 168.66 ^o , φ ₂₃ = 337.31 ^o .

The estimated direction of arrival of the useful signal θ _{y is} selected in the middle of the known angular sector θ _y1 = 35 ^o , θ _y2 = 65 ^o , θ _y = 50 ^o .

 The curves indicated in FIG. 4 digits 1, 2, obtained respectively for the cases: 1 - the claimed AAR, 2 - AAR prototype. For comparison, the same drawing shows a curve characterizing the value of the SINR that is potentially achievable for a given signal-noise situation (curve 3).

 The graphs show that within the specified angular sector of the arrival of the useful signal, the efficiency of the AAR that implements the proposed ACE method is significantly higher than that of the prototype AAR.

 This advantage of the claimed AAR will help to increase the noise immunity of radio communication lines (systems), characterized by the lack of accurate a priori information about the direction of arrival of the useful signal, in particular, communication systems with moving objects, and, ultimately, will contribute to the implementation of AAR in these systems.

 Thus, the proposed device fully implements the inventive method of ACE signals.

Literature
1. Marchuk L.A., Giniyatullin N.F., Kolinko A.V. Analysis of algorithms to minimize the power of the output signal in the AAR. // Radio engineering and electronics, 1998, N 1.

 2. Marchuk L.A., Kolinko A.V. Minimizing the signal-to-disturbing ratio in adaptive spatial filtering tasks. // Radio engineering, 1998, N 1, p. 48-54.

 3. Cheremisin O.P., Ratynsky M.V. et al. Effective projection algorithm for adaptive spatial filtering. // Radio Engineering and Electronics, 1994, Volume 39, Issue 2, pp. 259-263.

 4. Marchuk L.A. Robust ACE signal algorithms with inaccurate parameters. // Radio engineering, 1997, N 11, p. 3-7.

Claims (2)

1. The method of adaptive spatial filtering of signals, which consists in receiving signals at N spatially separated points, where N ≥ 2, calculating weight coefficients, weighted summation of the received signals, characterized in that the first and second reference signals with different frequencies are additionally generated, each of which is shared them by N equal-amplitude components, phase-shift each component by a predetermined phase value, after which the i-th received signal, where i = 1, 2, ... N, and the i-th components of the first and second supports s signals are summed, the i-th weighting factor is calculated by the formula
W _i (t + Δt) = W _i (t) -μ [V _yi -Y (t) Z $\genfrac{}{}{0ex}{}{\text{*}}{\text{i}}$ (t)],
where W _i (t) is the current value of the i-th weight coefficient;
V _yi is the value of the i-th control signal;
Δt is the time sampling interval;
Y (t) is the weighted summed signal;
Z _i (t) is the total signal from the addition of the i-th received signal and i-components of the first and second reference signals;
* - designation of the operation of complex pairing;
μ is the gain (transmission), and the frequencies of the first ω ₁ and second ω ₂ reference signals are selected from the condition ω ₁ <(ω ₀ -Δω), ω ₂ > (ω ₀ + Δω), where ω ₀ and Δω are the carrier frequencies, respectively and the spectrum width of the received wanted signal. 2. Adaptive antenna array containing N antenna elements, where N ≥ 2, N complex conjugation blocks, N complex weight multipliers, N-input common adder, control signal input device, N adaptation loops, and the input of the i-th complex conjugation block, where i = 1, 2, 3, ..., N, is connected to the first input of the corresponding i-th complex weight multiplier, the output of the i-th complex conjugation unit is connected to the signal input of the i-th adaptation loop, i-output of the input device control signals connected to control input i -th adaptation circuit, the output of which is connected to the second input of the i-th complex weight multiplier, the output of the i-th complex weight multiplier is connected to the i-th input of the N-input common adder, the output of which is the output of the adaptive antenna array, characterized in that introduced the first and second reference signal generators, the first and second blocks of phase shifters, N three-input adders, and the output of each i-th antenna element is connected to the first input of the i-th three-input adder, the output of the first reference generator signal is connected to the input of the first block of phase shifters, and the output of the second generator of the reference signal is connected to the input of the second block of phase shifters, the i-th output of the first block of phase shifters is connected to the second input of the i-th three-input adder, and the i-th output of the second block of phase shifters is connected to the third the input of the i-th three-input adder, the output of the i-th three-input adder is connected to the first input of the i-th complex weight multiplier, the output of the N-input total adder is connected to the information inputs of N adaptation loops, and the first ω ₁ and second ω ₂ reference signals are selected from the condition ω ₁ <(ω ₀ -Δω), ω ₂ > (ω ₀ + Δω), where ω ₀ and Δω are the carrier frequency and spectrum width of the received useful signal, respectively.

Images