RU2407026C1 - Location finding method of narrow-band radio signals of short-wave range - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: narrow-band radio signals of short-wave range, each of which lies in the appropriate k (k=1,…K) frequency subrange of the specified interval of K frequency subranges, are received with each radiator of linear antenna array with independent receiving radiators elevation plane during the observance interval; received signals are separated at the inlet of each receiving radiator by means of K narrow-band filters, correlation matrix is calculated as per the received signals in each k (k=1,…K) frequency subrange, each of k correlation matrixes are decomposed as per own signal and noise vectors, own directivity patterns of array are determined in each k frequency subrange as directivity patterns in mode of excitation of independent receiving radiators with the appropriate complex-conjugated proper vectors, and directions of input and complex amplitudes of received signals in k frequency subrange are determined by solving the systems of equations.

EFFECT: high accuracy of location finding of signals with completely unknown structure and parametres and determination of their complex amplitudes at reception of those signals via multi-beam radio routes with essentially various amplitudes and at low Signal, Interference, Noise ratio.

5 cl, 9 dwg

Description

The invention relates to radio engineering and can be used to improve the accuracy of direction finding of radio signals in the short-wave (HF) frequency range. A characteristic feature of the HF range is multipath, i.e. the ability to propagate radio waves along several mismatched paths.

A known method of broadband detection and direction-finding [1], in which digital signals characterizing the spectra of the received signals are extracted from the output signals of each element of the antenna array, and for each selected frequency in the reception band, using the phase of the signals, a direct calculation of the spatial Fourier series that discretely describes angular power spectrum at the selected frequency. After restoration of the angular spectrum at all frequencies, the bearing of any source emitting signals at any of the frequencies within the current reception band is determined. This method of the maximum possible amplitude-phase information uses only the phase of the signal, has a low speed in determining the azimuth bearing and does not allow to resolve multipath signals, especially when they are close to the angular position.

There is also a method [2] for determining the bearing and frequency of radiation sources based on a two-dimensional Fourier transform (in time and space) of signals received by an annular array of N elements at each time frequency. At fixed elevation guidance values, each spatial frequency component formed as a result of the conversion is multiplied q times by the Fourier transform of the guidance function, which depends on the frequency, elevation angle, antenna array parameters and 2πq / M azimuthal cyclic shift. Next, q spatial inverse Fourier transforms of the resulting set of corrected spatial frequency components are produced, the presence of the signal is determined and estimates of the azimuth and elevation angle of its source at each frequency are obtained from the squared modulus of the two-dimensional complex angular spectrum. The disadvantages of this method include the fact that with a small number of emitters in the antenna array, the accuracy of determining the bearing is not high enough. Also, it does not allow the separation of multipath signals.

Closest to the claimed invention is a method of direction finding sources of radio emission in a multipath environment, described in [3]. This object is achieved in that the method includes receiving signals on an N-element antenna array (AP), generating scanning signals in a given angular sector, calculating weight coefficients and then multiplying the weight coefficients with the corresponding scanning signals, storing the received signals for all discrete values of the angle scan. Then the angle value corresponding to the signal with the maximum amplitude is selected from them, the signals are re-received, their weight coefficients are calculated, and then they are multiplied with the corresponding scanning signals for each discrete scanning angle value during the transmission of the information packet and the guard interval, respectively. Next, the resulting signal calculated during the guard interval is subtracted from the resulting signal calculated in the transmission interval of the information packet, the difference signals for all discrete scan angle values are stored, from which the angle value corresponding to the signal with maximum amplitude is selected. However, the described method requires knowledge of the guard interval for each of the signals, a large amount of computation, and does not provide the required accuracy in determining the directions of arrival, especially for nearby signals, as well as in determining the amplitudes of these signals. In addition, the accuracy of direction finding signals in the above methods is significantly impaired with small ratios of the power of direction finding signals to the power of noise and interference.

The aim of the invention is to ensure high accuracy of direction finding of signals with completely unknown structure and parameters and to determine their complex amplitudes when these signals arrive along multipath radio paths with significantly different amplitudes and with a small signal / (interference + noise) ratio.

вычисляют по принятым сигналам в каждом k-ом (k=1,…,K) частотном поддиапазоне, раскладывают каждую из k-ых корреляционных матриц

по собственным сигнальным

и шумовым

(j=1,…,(N_x-1)) векторам, определяют собственные диаграммы направленности антенны

и

в каждом k-ом частотном поддиапазоне как диаграммы направленности в режиме возбуждения независимых приемных излучателей соответствующими комплексно-сопряженными собственными векторами:

и

где

, k - номер частотного поддиапазона, λ_k - средняя длина волны сигнала в k-ом частотном поддиапазоне, n - номер независимого приемного излучателя, d - расстояние между независимыми приемными излучателями, а направления прихода

и комплексные амплитуды

принимаемых сигналов в k-ом частотном поддиапазоне определяют путем решения систем уравнений:The problem is achieved in that in the method of direction finding narrowband radio signals of the HF range, which consists in the fact that the narrowband radio signals of the HF range propagating along multipath radio paths, each of which lies in the corresponding k-th (k = 1, ..., K) frequency sub-band of a given interval K frequency subbands, take each emitter of a multi-element antenna and calculate the correlation matrix of received signals, according to the invention as a multi-element antenna using a linear antenna p Brushes consisting of N _x independent reception emitters perform reception of narrowband radio signals falling in elevation plane for observation interval, the received signals are separated by K narrowband filters installed at the inlet of each independent receiver transducer and spanning a predetermined interval K frequency subbands correlation the matrix

computed from the received signals in each kth (k = 1, ..., K) frequency subband, lay out each of the kth correlation matrices

by own signal

and noise

(j = 1, ..., (N _x -1)) vectors, determine the antenna’s own radiation patterns

and

in each k-th frequency subband as radiation patterns in the excitation mode of independent receiving emitters by the corresponding complex conjugate eigenvectors:

and

Where

, k is the number of the frequency subband, λ _k is the average wavelength of the signal in the kth frequency subband, n is the number of the independent receiving emitter, d is the distance between the independent receiving emitters, and the directions of arrival

and complex amplitudes

the received signals in the kth frequency subband is determined by solving systems of equations:

, j=1,…,(N-1),

, j = 1, ..., (N-1),

m=1,…, M, l=1,…,M.

m = 1, ..., M, l = 1, ..., M.

.Where

.

совпадают, где k=k₁, k₂,…,k_pl; l=1,…,L, с последующим объединением p_l частотных поддиапазонов в полосу частот одного l-го широкополосного сигнала.The problem is also achieved by the fact that from the found solutions of systems of equations choose p _l solutions for which the directions of arrival of signals

coincide, where k = k ₁ , k ₂ , ..., k _pl ; l = 1, ..., L, followed by combining p _l frequency subbands into the frequency band of one l-th wideband signal.

, где k=k₁, k₂,…,k_pl, l=1,…,L, объединенных частотных поддиапазонов.The task is also achieved by the fact that the spectrum of a broadband signal is formed from spectral components

where k = k ₁ , k ₂ , ..., k _pl , l = 1, ..., L, of the combined frequency subbands.

сигналы от каждого излучателя суммируют с весовыми коэффициентами, пропорциональными комплексно-сопряженному сигнальному собственному вектору

для этого направления, умноженному на обратную корреляционную матрицу помех

.The task is also achieved by the fact that for each of the found directions of arrival

the signals from each emitter are summed with weight coefficients proportional to the complex conjugate signal eigenvector

for this direction, multiplied by the inverse correlation matrix of interference

.

в угломестной и

азимутальной плоскостях, в течение интервала наблюдения с помощью дополнительной линейной антенной решетки, состоящей из N_y независимых приемных излучателей, расположенной перпендикулярно оси основной линейной антенной решетки из N_x независимых приемных излучателей, определяют обобщенные направления прихода

и

сигналов из решения систем уравнений:The task is also achieved by the fact that additionally receive narrow-band radio signals falling from an arbitrary direction

in elevated and

azimuthal planes, during the observation interval using an additional linear antenna array consisting of N _y independent receiving emitters located perpendicular to the axis of the main linear antenna array of N _x independent receiving emitters, determine the generalized directions of arrival

and

signals from solving systems of equations:

, j=1,…,(N_x-1) и

, j=1,…,(N_y-1),

, j = 1, ..., (N _x -1) and

, j = 1, ..., (N _y -1),

и

- собственные шумовые диаграммы направленности для основной и дополнительной линейных антенных решеток соответственно, после чего восстанавливают истинные направления прихода сигналов

и

по соотношениям:Where

and

- intrinsic noise radiation patterns for the primary and secondary linear antenna arrays, respectively, after which they restore the true direction of arrival of the signals

and

by ratios:

The invention is illustrated by drawings.

Figure 1 schematically shows an antenna array of N emitters for one-dimensional direction finding; figure 2 a, b shows its own amplitude radiation pattern (LH) of an antenna consisting of two emitters, provided one incident on the antenna signal with a ratio of signal power to noise power P _c / R _W = 10 dB and 0 dB; on figa, b shows the own amplitude DN of the antenna, consisting of four emitters, provided one incident on the antenna signal with a ratio of signal power to noise power P _c / R _W = 10 dB and 0 dB; on figa, b shows the own amplitude DN of the antenna, consisting of two emitters, provided there are two signals incident on the antenna with the ratio of the total signal power to noise power P _c / R _w = 10 dB and 0 dB; on figa, b shows the own amplitude DN of the antenna, consisting of four emitters, provided two incident on the antenna signals with a ratio of the total signal power to noise power P _c / R _W = 10 dB and 0 dB; figure 6 shows a diagram of the antenna part of the HF direction finder for implementing the method; 7 schematically shows an antenna array for two-dimensional direction finding; Figs. 8 and 9 show tables with data about errors in determining the parameters of incoming signals.

блок 4 управления весовыми коэффициентами; блок 5 сумматоров.The antenna part of the HF direction finder (Fig.6) contains a linear antenna array 1 consisting of N (1, ... n, ... N) independent receiving emitters; N blocks 2, each of which consists of K narrow-band filters tuned respectively to frequencies (ω ₁ ... ω _k ... ω _K ), while the number 3 denotes the reference plane in which the complex signal amplitudes are taken

block 4 control weights; block 5 adders.

для каждого m-го направления. В процессе пеленгации должны быть определены направления

и амплитуды

приходящих сигналов в каждом k-ом частотном поддиапазоне.The proposed method is based on dividing a given frequency range using narrow-band analog filters, which have a large dynamic range, into K more narrow-band frequency subbands. The division into K frequency subbands is carried out in such a way that in each of the k-th frequency subbands (k = 1, ... K) at a given time there is only one signal incident on a multi-element antenna with a certain average harmonic frequency ω _k . Due to the multipath propagation, this signal can arrive at the antenna array (AP) of the HF direction-finder simultaneously from one or more directions with unknown complex amplitudes

for each m-th direction. In the direction finding process, directions should be identified

and amplitudes

incoming signals in each kth frequency subband.

и с гармонической частотой ω_k с направления

на входе n-го излучателя возникает напряжениеConsider the method of direction finding the signal on the example of one-dimensional direction finding in a fixed plane (φ = 0) for an antenna of N independent receiving emitters located at a distance d from each other along the axis ox (figure 1). When a plane electromagnetic wave with a complex amplitude is incident

and with a harmonic frequency ω _k from the direction

at the input of the nth emitter voltage

- набег фазы между антенными элементами, определяемый направлением на источник радиосигнала,

- начальная фаза сигнала частоты ω_k с m-го направления, а

- комплексная амплитуда напряжения на входе n-го излучателя на частоте ω_k.Where

- phase incursion between the antenna elements, determined by the direction of the radio source,

is the initial phase of the frequency signal ω _k from the mth direction, and

- the complex amplitude of the voltage at the input of the nth emitter at a frequency ω _k .

As a rule, along with harmonic signals, a noise signal with a power R _w in the reception band is incident on the emitters. Then the complex signal + noise amplitude averaged over the observation period T at the input of the emitters can be written as

определяется корреляционная матрица напряжений на входах излучателей

, которая в предположении некоррелированности принимаемых шумов в соседних излучателях как с полезным сигналом, так и между собой имеет вид:According to complex amplitudes and

determined by the correlation matrix of voltages at the inputs of the emitters

, which, under the assumption that the received noise is not correlated in neighboring emitters, both with a useful signal and with each other, has the form:

- корреляционная матрица сигналов в отсутствии шумов,

обозначает вектор-столбец,

обозначает вектор-строку); [E] - единичная матрица.Where

- correlation matrix of signals in the absence of noise,

denotes a column vector,

denotes a row vector); [E] is the identity matrix.

при Р_ш>0 является неособенной. По корреляционной матрице определяется N отличных от нуля собственных значений и N соответствующих им собственных векторов. Собственные значения сигнально-шумовой корреляционной матрицы имеют вид:Matrix

when P _w > 0 is nonsingular. The correlation matrix determines N non-zero eigenvalues and N corresponding eigenvectors. The eigenvalues of the signal-noise correlation matrix are:

- суммарная мощность сигнала, принимаемая всеми излучателями AP.Where

- total signal power received by all AP emitters.

и называемого сигнальным

, при любом количестве падающих волн и количестве излучателей совпадает с сигнальным вектором входных напряжений на клеммах излучателей. Остальные N-1 собственных векторов, соответствующих λ₂,…,λ_N, называются шумовыми и обозначаются

(j=1,…,(N-1)). С учетом указанных обозначений разложение сигнально-шумовой корреляционной матрицы по собственным векторам имеет вид:The structure of the eigenvector corresponding to the eigenvalue

and called a signal

, for any number of incident waves and the number of emitters, it coincides with the signal vector of input voltages at the terminals of the emitters. The remaining N-1 eigenvectors corresponding to λ ₂ , ..., λ _N are called noise and are denoted by

(j = 1, ..., (N-1)). Given these notations, the expansion of the signal-noise correlation matrix in eigenvectors has the form:

как ДН в режиме возбуждения излучателей Ap комплексно-сопряженным вектором

. Тот факт, что возбуждение излучателей является комплексно-сопряженным вектору

, при условии одного падающего сигнала обеспечивает формирование собственной сигнальной ДН AP с максимумом и собственных шумовых ДН с глубоким провалом (в идеале - нулевым значением) в направлении прихода сигнала (фиг.2) для двух излучателей и (фиг.3) - для четырех излучателей на частоте

при соотношении мощности сигнала к мощности шума Р_с/Р_ш=10 дБ и 0 дБ). Поэтому для одного сигнала (M=1) направление его прихода определяется по совпадающим нулям собственных шумовых ДН.We define the proper MD Fq ^k (θ) corresponding to the vector

as a DN in the excitation regime of emitters Ap by a complex conjugate vector

. The fact that the excitation of emitters is a complex conjugate vector

, under the condition of one incident signal, it provides the formation of its own signal beam AP with a maximum and its own noise beam with a deep dip (ideally, a zero value) in the direction of arrival of the signal (Fig. 2) for two radiators and (Fig. 3) for four radiators on frequency

with the ratio of signal power to noise power R _s / R _W = 10 dB and 0 dB). Therefore, for one signal (M = 1), the direction of its arrival is determined by the coincident zeros of the intrinsic noise patterns.

Own signal and noise DNs look like:

, где λ_k - длина волны сигнала в k-ом частотном поддиапазоне.Where

where λ _k is the wavelength of the signal in the kth frequency subband.

, принимаемого излучателями AP. Для M плоских волн, падающих с направлений θ_m на AP из излучателей, уровень полезного сигнала, принимаемого по j-й шумовой ДН, определяется выражением:The signal and noise intrinsic signals are used to determine the level of the useful signal

received by the emitters AP. For M plane waves incident from the directions θ _m on the AP from the emitters, the level of the useful signal received from the jth noise beam is determined by the expression:

при соотношении мощности сигналов к мощности шума P_c/Р_ш=10 дБ и 0 дБ), как для случая одного сигнала. Поэтому по минимумам собственных шумовых ДН нельзя определить направления двух и более сигналов. Для определения направлений θ_m нескольких (M>1) сигналов решается система функциональных уравнений:As the analysis shows, in the case of two or more signals arriving from different directions, the directions of the minimums of the noise DNs do not coincide with the directions of arrival of the signals (Fig. 4 for two radiators and Fig. 5 for four radiators at a frequency

when the ratio of signal power to noise power P _c / R _W = 10 dB and 0 dB), as for the case of a single signal. Therefore, it is impossible to determine the directions of two or more signals by the minima of their own noise DNs. To determine the directions θ _{m of} several (M> 1) signals, a system of functional equations is solved:

which follows from the fact that the level of the useful signal received from its own noise DNs is zero.

(m=1,…,M) приходящих сигналов при найденных из (9)

используется М дополнительных уравнений:To determine the absolute values of the amplitudes

(m = 1, ..., M) of incoming signals when found from (9)

M additional equations are used:

or, taking into account that the structure of the eigen signal vector coincides with the signal vector, the system of equations (10) takes the form:

The curves for the signal and noise DNs shown in Figs. 2, 3, 4, 5, very weakly depend on the ratio P _c / P _W , which indicates the possibility of applying the method of high-precision direction finding described below and at small ratios P _c / Р _w .

с помощью весовых коэффициентов

, вводимых устройствами управления амплитудой и фазой сигнала на выходе каждого n-го излучателя (фиг.6). С этой целью вектор весовых коэффициентов

выбирается из условияAfter determining from the systems of equations (9), (11) the direction of arrival and the amplitude of the m-th plane wave, the operation of maximizing the ratio C / (P + W) for this direction is performed

using weights

input devices for controlling the amplitude and phase of the signal at the output of each n-th emitter (Fig.6). To this end, the vector of weights

is selected from the condition

- корреляционная матрица помех, в которой учитываются как шумовые составляющие, так и сигнальные составляющие со всех направлений за исключением направления

;

- собственный сигнальный вектор для m-го направления прихода сигнала с частотой ω_k:Where

- a correlation matrix of interference, which takes into account both noise components and signal components from all directions except for the direction

;

- eigen signal vector for the m-th direction of arrival of the signal with a frequency ω _k :

After operation (12), the signal is sent for further processing.

To implement the above method of determining the directions of arrival of signals and their amplitudes, an HF direction finder is used (Fig. 6).

.The method is as follows. The signal arriving from an arbitrary mth direction enters the antenna array, consisting of N independent receiving emitters, each of which has a system of narrow-band filters. A harmonic signal is removed from the output of each kth filter of the nth emitter

.

, пришедших в каждом k-ом частотном поддиапазоне. Направления прихода сигналов находятся из условия равенства нулю уровня полезного сигнала, принимаемого по шумовым собственным диаграммам направленности, определяемым по собственным векторам корреляционной матрицы, соотношение (9). Для максимизации отношения сигнал/(помеха + шум) для направлений прихода сигналов, найденных на предыдущем этапе, производится суммирование сигналов с каждого n-го излучателя для каждого k-го поддиапазона с весовыми коэффициентами

, выбираемыми в соответствии с законом (12).After narrow-band filtering, relation (3) calculates the correlation matrix of complex signal amplitudes,

coming in every k-th frequency subband. The directions of arrival of the signals are found from the condition that the level of the useful signal, taken from the noise eigenvectors determined by the eigenvectors of the correlation matrix, be equal to zero, relation (9). To maximize the signal / (interference + noise) ratio for the arrival directions of the signals found in the previous step, the signals from each n-th emitter for each k-th subband with weight coefficients are added

selected in accordance with law (12).

, найденными из (11) в каждом из k-ых частотных поддиапазонов, входящих в его частотный интервал.The described method for determining the directions of arrival and amplitudes of harmonic signals can be used not only for harmonic signals, but also for broadband signals or signals with a frequency-varying frequency. In this case, the algorithm should be supplemented with the following operations: after all possible directions of arrival of signals in all frequency subbands have been determined, it is necessary to combine those frequency subbands in which the directions of arrival of the signals are the same in one frequency interval. As for the amplitude spectrum of a broadband signal, it can be found as a set of spectral components with complex amplitudes

found from (11) in each of the kth frequency subbands included in its frequency interval.

и азимутальных

углов прихода сигналов должна быть выполнена процедура определения обобщенных координат

и

в соответствии с алгоритмом нахождения направлений прихода сигналов на линейную антенную решетку из N излучателей для случая одномерной пеленгации.For two-dimensional direction finding, two orthogonally located linear antenna arrays are used (Fig. 7). Moreover, to find elevated

and azimuthal

angles of arrival of signals, the procedure for determining the generalized coordinates should be performed

and

in accordance with the algorithm for finding the directions of arrival of signals to a linear antenna array of N emitters for the case of one-dimensional direction finding.

и

, а затем в соответствии с алгоритмом для одномерной пеленгации определяются истинные углы прихода сигналов и их амплитуды.The difference between two-dimensional direction finding and one-dimensional direction is that first the generalized coordinates are determined

and

and then, in accordance with the algorithm for one-dimensional direction finding, the true angles of arrival of the signals and their amplitudes are determined.

и с гармонической частотой ω_k с направления

на входах n_x-го и n_y-го излучателей возникают комплексные амплитуды напряжений:When a plane electromagnetic wave with a complex amplitude is incident

and with a harmonic frequency ω _k from the direction

at the inputs of the n _x th and n _y th emitters, complex voltage amplitudes arise:

N_x и N_y - количество излучателей вдоль осей ox и oy соответственно, d_x и d_y - расстояние между излучателями в первой и второй линейных антенных решетках соответственно.Where

N _x and N _y are the number of emitters along the axes ox and oy, respectively, d _x and d _y are the distances between emitters in the first and second linear antenna arrays, respectively.

и

в соответствии с алгоритмом определения углов, справедливым для одномерной пеленгации.Relations (14) and (15) can be considered independently of each other and thus determine

and

in accordance with the algorithm for determining angles, valid for one-dimensional direction finding.

The generalized coordinates are related to the desired signal parameters by the relations

и

в пространстве.Using (16) and (17), from the found generalized coordinates, the true directions of signal arrival are restored

and

in space.

для двумерного пеленга до размерности двух независимых одномерных матриц соответствующих линейных AP. Данный способ существенно сокращает объем вычислений при определении углов прихода

и

.Thus, the use of an additional linear antenna array, orthogonal to the first, reduces the dimension of the signal-noise matrix

for a two-dimensional bearing to the dimension of two independent one-dimensional matrices of the corresponding linear APs. This method significantly reduces the amount of calculation when determining the angles of arrival

and

.

которые показывают, что ошибка определения угловых координат источников радиосигналов и их амплитуд зависит от точности решения систем уравнений (9), (11) и может быть сколь угодно малой. В частности, при задании шага решения уравнений (9), (11) в интервале 0.1°÷1° максимальная ошибка определения угловых координат сопоставима с величиной шага, а относительная точность определения амплитуд составляет несколько процентов, см. табл.1-6 (фиг.8, 9). В этих таблицах введены следующие обозначения: θ₀₁ и θ₀₂ - истинные направления прихода сигналов, θ₁ и θ₂ - направления, определяемые из решения системы (9), Δθ₁ и Δθ₂ - ошибки определения направлений прихода сигналов, A₀₁ и A₀₂ - истинные значения амплитуд сигналов, принимаемых AP с направлений θ₀₁ и θ₀₂, A₁ и A₂ - значения амплитуд сигналов, полученные в результате решения системы (11), ΔA₁ и ΔA₂ - величины относительной ошибки определения амплитуды первого и второго сигналов соответственно, P_c01 и P_c02 - мощности сигналов, принимаемых AP с направлений θ₀₁ и θ₀₂.As an illustrative example, the simulation results for one-dimensional direction finding of one and two harmonic narrow-band signals with a frequency

which show that the error in determining the angular coordinates of radio signal sources and their amplitudes depends on the accuracy of solving systems of equations (9), (11) and can be arbitrarily small. In particular, when setting the step for solving equations (9), (11) in the range 0.1 ° ÷ 1 °, the maximum error in determining the angular coordinates is comparable to the step size, and the relative accuracy in determining the amplitudes is several percent, see Table 1-6 (Fig. .8, 9). The following notation is introduced in these tables: θ ₀₁ and θ ₀₂ are the true directions of arrival of signals, θ ₁ and θ ₂ are directions determined from the solution of system (9), Δθ ₁ and Δθ ₂ are errors in determining the directions of arrival of signals, A ₀₁ and A ₀₂ - the true values of the amplitudes of the signals received by the AP from the directions θ ₀₁ and θ ₀₂ , A ₁ and A ₂ - the values of the amplitudes of the signals obtained as a result of solving the system (11), ΔA ₁ and ΔA ₂ - the values of the relative error in determining the amplitudes of the first and second signals, respectively, P _c01 and P _c02 are the power of the signals received by the AP from the directions θ ₀₁ and θ ₀₂ .

The simulation results in table 1-6 (Fig, 9) are obtained for different ratios P _c / P _W As follows from these results, the accuracy of direction finding of multipath signals remains very high even at P _c / P _w ≥ -10 dB.

Information sources

1. Patent US 4626859, CL G01S 5/04.

2. RF patent 2150122, cl. G01S 3/14.

3. RF patent 2141675, cl. G01S 3/00.

Claims (5)

1. The method of direction finding narrowband radio signals of the KB range, which consists in the fact that narrowband radio signals of the KB range, propagating along multipath radio paths, each of which lies in the corresponding k-th (k = 1, ..., K) frequency sub-band of a given interval K of frequency sub-bands, taking each emitter array antenna, and calculating the correlation matrix of the received signals, characterized in that the array antenna is used, consisting of N _x independent linear reception antenna array zluchateley, carried reception of narrowband radio signals falling in elevation plane for observation interval, the received signals are separated by K narrowband filters installed at the inlet of each independent receiver and transmitter predetermined interval overlapping K frequency subbands correlation matrix

calculated from the received signals in each k-th (k = 1, ..., K) frequency sub-band, lay out each of the k-th correlation matrices

by own signal

and noise

(j = 1, ..., (N _x -1)) vectors, determine the antenna’s own radiation patterns

and

in each kth frequency subband as radiation patterns in the mode of excitation of independent receiving emitters by the corresponding complex conjugate eigenvectors:

and

Where

, k is the number of the frequency subband, λ _k is the average wavelength of the signal in the kth frequency subband, n is the number of the independent receiving emitter, d is the distance between the independent receiving emitters, and the directions of arrival

and complex amplitudes

the received signals in the kth frequency subband are determined by solving systems of equations:

, j = 1, ... (N _x -1),

m = 1, ... M, l = 1, ..., M,
Where

,

. 2. The method according to claim 1, characterized in that from the found solutions of the systems of equations, p ₁ solutions are selected for which the directions of arrival of the signals

coincide, where k = k ₁ , k ₂ , ..., k _p1 ; l = 1, ..., L, with the subsequent integration of p ₁ frequency subbands into the frequency band of one l-th broadband signal. 3. The method according to claim 2, characterized in that the spectrum of the broadband signal is formed from spectral components

, where k = k ₁ , k ₂ , ..., k _p1 , l = 1, ..., L, are combined frequency subbands. 4. The method according to claim 1, characterized in that for each of the found directions of arrival

the signals from each emitter are summed with weight coefficients proportional to the complex conjugate signal eigenvector

for this direction, multiplied by the inverse correlation matrix of interference

. 5. The method according to claim 1, characterized in that it further receive narrowband radio signals falling from an arbitrary direction

in elevated and

the azimuthal planes, during the observation interval, using the additional linear antenna array consisting of N _y independent receiving emitters, located perpendicular to the axis of the main linear antenna array of N _x independent receiving emitters, determine the generalized directions of arrival

and

signals from solving systems of equations:

, j = 1, ..., (N _x -1) and

, j = 1, ..., (N _y -1),
Where

and

- own noise radiation patterns for the primary and secondary linear antenna arrays, respectively,
then restore the true direction of arrival of signals

and

by ratios:

,

.

Images