RU2365051C1 - Autocorrelation device for opening of wavelength-time structure of signals of digital communication systems - Google Patents

Abstract

FIELD: physics; radio.

SUBSTANCE: invention concerns radio measuring technique and can be used in digital communication systems and radio monitoring, in particular, in clocking and receiving devices of phase-shift keyed signals. The device contains two autocorrelators with quarter-phase processing, each of them includes two multipliers, a phase shifting device, a delay circuit, two switched integrators, two squaring devices, an adder and a device taking the square root, and also two threshold devices, two counters, a calculating device, the control circuit which forms commands on the basis of external target designation, a clock generator.

EFFECT: reliability increase at opening of a spectral spectrum of the phase-shift keyed signals (PKS) and antijamming ability increase at estimation of PKS time parametres.

1 dwg

Description

The invention relates to radio measurement technology and radio monitoring and can be used in digital communication systems and radio monitoring, in particular in devices for synchronizing and receiving phase-shifted signals (PMS).

Widespread in digital communication systems received multiple access with time division multiplexing (TDMA). Typically, digital TDMA communication systems are used to transmit groups of channels in one direction based on the use of a group digital signal, which is split into packets and frames. Each packet or frame contains the preamble necessary for synchronization, and an informative part. In the formation of the preamble, as a rule, processes with a discrete spectrum are used, and in the formation of the informative part, FMSs with a pseudo-random manipulating function having a continuous spectrum are used.

In radio monitoring of FMS with an unknown shape, spectral and correlation methods are used to open the spectral-temporal structure. When receiving high-speed FMS, autocorrelation methods provide, in comparison with spectral methods, higher accuracy in estimating the time structure with lesser hardware complexity and deserve special attention.

In modern systems with mdvr there are two ways to organize the digital stream:

1) for a fixed duration of frames using packets of variable duration;

2) for a fixed duration of packets, the repetition period of packets in a frame is changed.

A known autocorrelation meter of random FMS parameters [1], containing a series-connected multiplier, to the input of which a delay line output, a band-pass filter and a low-pass filter (integrator) are connected, while the output of the delay line tuning speed generator is connected respectively to the first input of the delay line connected with the second output of the multiplier, the first input of the frequency meters, to the second inputs of which the corresponding output of the bandpass filter is connected through a nonlinear element and the first and a second low pass filter.

The signs of this analogue, which coincide with the essential features of the claimed device, are an autocorrelator including a delay line, a multiplier, a low-pass filter. The reasons that impede the achievement of the technical result include the lack of the possibility of opening the spectral-temporal structure of the FMS in communication systems with TDMA.

Also known is an autocorrelation meter of pseudo-random FMS parameters [2], containing two multipliers, a delay line, a low-pass filter, a valve, a pulse counter, a base meter, a delay line tuner, a band-pass filter, a nonlinear element, low-pass filters, frequency and duration meters parcels, registration block.

The signs of this analogue, which coincides with the essential features of the claimed device, are an autocorrelator including a delay line, a multiplier, a low-pass filter.

The reasons that impede the achievement of the technical result include the low reliability of opening the spectral-temporal structure of the FMS in communication systems with TDMA.

Of the known inventions suitable for opening the spectral-temporal structure of the FMS, the closest in technical essence is the method of autocorrelation receiving a noise-like signal [3], implemented by an autocorrelator with quadrature processing, containing two multipliers, a delay line, two phase shifters, two integrators, two quadrators, an adder, a square root extractor, an amplitude limiter and a trigger, while the input of the device is connected to the inputs of the first multiplier, delay line and the first phase switch, the output of the first phase shifter is connected to the first input of the second multiplier, the output of the delay line is connected to the second input of the first multiplier and through the second phase shifter with the second input of the second multiplier, the outputs of the first and second multipliers are connected to the first and second input of the adder through series-connected integrator and quadrator, and the output of the adder through a series-connected square root extraction unit and an amplitude limiter is connected to the trigger input.

The signs of the prototype, which coincides with the essential features of the claimed device, are two multipliers, a delay line, a phase shifter, two integrators, an adder, a square root extraction device.

The disadvantages of the prototype include: 1) the lack of autopsy of the spectral composition of the FMS; 2) low noise immunity when evaluating the time parameters of FMS fragments.

The tasks to be solved by the claimed device:

1) increasing the reliability when opening the spectral composition of the FMS;

2) increased noise immunity when evaluating the temporal parameters of the FMS.

The technical result is achieved by the fact that in the known device introduced:

a) the second channel of the autocorrelator with quadrature processing (AKO), providing a high level of reliability of opening the spectral composition of the FMS due to the simultaneous processing of two ordinates of the correlation functions of the fragments of the FMS;

b) the manager, providing increased noise immunity of the autocorrelation device by adjusting the parameters of autocorrelators with quadrature processing and the transition from analog to discrete integration;

c) a clock pulse generator, two threshold devices, two counters and a deciding device, which ensure the adoption of hypotheses about the spectral composition of FMS fragments and the formation of estimates of the temporal parameters of packets and FMS frames.

To achieve a technical result, a device including multipliers (1, 2), a phase shifter (3), a delay line (4), switched integrators (5, 6), quadrators (7, 8), an adder (9), an extraction device square root (26), the input of the autocorrelation device is connected to the inputs of the phase shifter (3) and the delay line (4), as well as to the first input of the multiplier (2), the output of the phase shifter (3) is connected to the first input of the multiplier (1), the output of the delay line (4) connected to the second inputs of the multipliers (1, 2), the output of the multiplier (1) is connected through the last In addition, the switched integrator (5) and the quadrator (7) with the first input of the adder (9) are connected, the output of the multiplier (2) is connected via a series-connected switching integrator (6), the quad -rator (8) with the second input of the adder (9), the output of which is connected with the input of the square root extraction device (26), the following are additionally introduced: a second autocorrelator with quadrature processing (29), which includes multipliers (12, 13), a phase shifter (15), a delay line (14), switched integrators (16, 17) , quadrators (18, 19), adder (20), device removed square root (27), as well as threshold devices (10, 21), counters (11, 22), a ruler (23), a clock generator (24), and a solving device (25).

The drawing shows a structural diagram of an autocorrelation device (AU), where 1, 2, 12, 13 are multipliers (P ₁ , P ₂ , P ₃ , P ₄ ); 3, 15 - phase shifters (Фв ₁ , Фв ₂ ); 4, 14 - delay lines (LZ ₁ , LZ ₂ ); 5, 6, 16, 17 - switched integrators (And ₁ , And ₂ , And ₃ , And ₄ ); 7, 8, 18, 19 - quadrators (Qu ₁ , Qu ₂ , Qu ₃ , Qu ₄ ); 9, 20 - adders (C ₁ , C ₂ ); 10, 21 - threshold devices (PU ₁ , PU ₂ ); 11, 22 - counters (SCh ₁ , SCh ₂ ); 23 - ruler (Upr); 24 - a clock generator (GTI); 25 - a decisive device (RU); 26, 27 — square root extraction device (PEC ₁ ), PEC ₂ ); 28, 29 - autocorrelators with quadrature processing (AKO ₁ , AKO ₂ ).

The ability to achieve the objectives of the invention is confirmed by the following analysis of the operation of the AU.

To ensure high noise immunity, radio monitoring (RM) of phase-shifted signals (PMS) in communication systems with TDMA is carried out in several stages.

и ширина спектра ФМС

. Поскольку для ФМС имеем Δf_s=2/T_э, где T_э - длительность элемента манипулирующей последовательности, то на первом этапе также оценивается

.At the first stage, the detection and preliminary estimation of the average frequency of the FMS

and FMS spectrum width

. Since for FMS we have Δf _s = 2 / T _e , where T _e is the duration of the element of the manipulating sequence, then at the first stage it is also estimated

.

The second stage is designed to open the spectral-temporal structure of the FMS packets, which then allows you to go to the third stage of the RM, dedicated to intercepting selected channels for transmitting information of communication systems with TDMA.

In this invention, in the framework of the second stage of the RM, we consider the principles of building AC.

The additive mixture is fed to the input of the AU

y (t) = S (t) + n (t) for t ₀ ≤t≤t ₀ + T _c ;

S (t) = U _ms P (t) cos [2πf _s t + φ _s ],

where S (t) is the FMS; U _ms , f _s , φ _s - amplitude, frequency and phase of the FMS; P (t) is the manipulating function; n (t) is the Gaussian stationary noise; t ₀ , T _c - the beginning and duration of the PM session.

FMS in systems with TDMA consist of packets and frames, which use both deterministic and stochastic manipulating functions.

For the case when a determinate sequence of elements with coefficients α _{k of} the same sign is used in P (t), the energy spectrum G (f) corresponds to a fragment in the form of harmonic oscillation (G):

при t₁≤t≤t₁+T_г,

at t ₁ ≤t≤t ₁ + T _g ,

where t ₁ , T _g - the moment of onset and the duration of the fragment of harmonic oscillation.

The autocorrelation function of the FMS fragment R ₁ (τ), corresponding to harmonic oscillation, has the form:

, при τ∈[0,T_гк];

, at τ∈ [0, T _hk ];

; ω_s=2πf_s,

; ω _s = 2πf _s ,

where r ₁ (τ) is the normalized envelope of the autocorrelation function R ₁ (τ).

For the case when a deterministic sequence of elements according to the law of the meander is used in P (t), the energy spectrum of the two-position FMS with manipulation [0, π] has the form:

at t ₂ ≤t≤t ₂ + T _m ,

- круговая частота манипуляции, t₂, T_м - момент начала и длительность элемента ФМС с манипуляцией по закону меандра (М).Where

- circular frequency of manipulation, t ₂ , T _m - the moment of start and duration of the FMS element with manipulation according to the law of the meander (M).

The autocorrelation function of the FMS fragment with a manipulating function according to the law of the meander R ₂ (τ) has the form:

при (i-1)Т_э≤τ≤iT_э;

,

at (i-1) T _e ≤τ≤iT _e ;

,

.

.

For cases where the manipulating sequence P (t) is a pseudorandom sequence of elements (PSP), the energy spectrum and the FMS autocorrelation function for equiprobable phase jumps Δφ∈ [0, π] have the form:

;

;

при t₃≤t<t₃+T_и;

,

at t ₃ ≤t <t ₃ + T _and ;

,

where t ₃ , T _and - the moment of the beginning and the duration of the FMS fragment with the manipulation of the SRP (I).

In the general case, in systems with TDMA, the FMS incorporates all of the above fragments.

Since during PM, as a rule, FMS is processed with an unknown shape, the presence of fragments of various types in the FMS results in an unsteady nature of the energy spectrum, the parameters of which depend on a priori unknown composition, duration, and the law of formation of the manipulating function.

To reveal the spectral-temporal structure of such FMS, as well as locally stationary random processes, under conditions of great a priori uncertainty about the time parameters, the simplest hardware implementation is provided using adaptive multi-channel correlation analysis.

An analysis of the laws of variation of the normalized envelopes of the autocorrelation functions of the above FMS fragments shows that for their approximation it suffices to have information about the two ordinates of the envelope of the autocorrelation coefficient r _s (τ). To classify such FMS fragments as G, M, and I, it is enough to use a set of ordinates:

,

, для М имеем

,

, для И имеем

Кроме того, имеется возможность классификации защитного промежутка (3), так как при отсутствии ФМС имеем

,

.since for Γ we have

,

, for M we have

,

, for And we have

In addition, it is possible to classify the protective gap (3), since in the absence of FMS we have

,

.

The typical structure of the FMS package is shown in Fig. 1, Pr - preamble; G is the harmonic fragment; M - meander fragment; And - the information part; Z is the protective gap; P - package; K - frame; t ₁ , t ₂ , t ₃ , t ₄ - the moments of the beginning of G and Pr, M, I, Z; t ₅ , t ₆ - the moments of the end of the packet (P) and frame (K); ΔТ - time uncertainty interval.

From the package structure it follows that

ΔT = t ₁ -t ₀ <T _b , T _g = t ₂ -t ₁ = n _g T _e ; T _m = t ₃ -t ₂ = n _m T _e ;

T _and = t ₄ -t ₃ = n _and T _e ; T _ol = T _g + T _m ; T _s = t ₅ -t ₄ = n _s T _e

T _p = t ₅ -t ₁ = T _g + T _m + T _m + T _and + T _s ; T _to = t ₆ -t ₁ = n _to T _p ;

n _p = n _g + n _m + n _and + n _s ,

where T _g , T _m - the duration of the harmonic and meander fragments; T _ol - the duration of the preamble; T _and - the duration of the information part of the package; T _s - the duration of the protective gap; T _p , T _to - the duration of the packet and frame; n _g , n _m , n _and , n _s , n _p , n _k - the number of elements in G, M, I, 3, P, K.

To open the spectral-temporal structure of the FMS, it is necessary to classify the components G, M, I, 3, P, and also evaluate the time parameters t ₁ , T _g , T _m , T _u , T _p , T _k .

In view of the above, when constructing UE can confine two channels autocorrelators quadrature processing (ACH), one of which time-shift introduced by LZ ₁ equals τ ₁ = _^1/2 T _e, but in another time shift introduced LZ ₂ is τ ₂ = T _e .

при t₀+(j-1)T≤t_j≤t₀+jT; j∈[1, N_c]; T≥3T_э, где Т - постоянная времени КИ; N_c - количество циклов интегрирования, соответствующих длительности сеанса РМ.To reduce the error in estimating the time parameters of FMS fragments in AC, it is proposed to switch from averaging over time to averaging over a set due to the use of discrete averaging instead of analog averaging due to the use of switched integrators (CI) with impulse response

at t ₀ + (j-1) T≤t _j ≤t ₀ + jT; j∈ [1, N _c ]; T≥3T _e , where T is the time constant of KI; N _c is the number of integration cycles corresponding to the duration of the PM session.

When the additive mixture y ₂ (1) is exposed to the input AKO ₁ and AKO ₂ at the output of the switched integrators (5, 6) and (16, 17), we have quadrature components

;

;

;

;

where U _s1 ( _t ), U _s2 (t) are the cosine and sine components of the components due to the signal-to-signal interaction; U _sn1 (t), U _sn2 (t) - cosine and sine components of the components due to the interaction of the "signal-to-noise"; U _n1 (t), U _n2 (t) are the cosine and sine components of the components due to the interference-noise type of interaction.

When using KI, the signal-to-signal components are as follows:

,

,

where K _p - gear ratio with a dimension of 1 / V.

To ensure the invariance of the amplitude of the output effect from the frequency of the FMS

f _s in AKO, quadrature processing is used, and at the same time, at the output of the square root extraction device (PEC), we obtain:

при t₀+(j-1)T≤t≤t₀+jT.

at t ₀ + (j-1) T≤t≤t ₀ + jT.

The greatest noise immunity of the AC is provided when a threshold device (PU) is supplied to the input along with the voltage U _s (t), the gating voltage

Where

where U _{with the} amplitude; rect [x] - time window; τ _c - the duration of the strobe.

In the time interval when the condition U _s (t)> U _{pop is satisfied} , where U _pore is the threshold voltage, at the output of the PU we obtain a sequence of pulses with a normalized amplitude

при t_н+(i-1)T≤t≤t_н+(i-1)T+τ_c;

at t _n + (i-1) T≤t≤t _n + (i-1) T + τ _c ;

i∈ [1, N _f ]; N _f = T _f / T;

where t _n , T _f - the beginning and duration of the detected FMS fragment; N _f - the number of pulses at the output of the PU.

,

.After counting the number of pulses in the counter (MF) in the decider (RU), the time parameters of the FMS fragment are estimated:

,

.

To ensure synchronization during the operation of KI, PU and RU, a clock pulse generator (GTI) is used, the start time of which t ₀ and the pulse repetition period T are set according to the commands received from Ex. In addition, Upr provides tuning LZ ₁ to τ ₁ = ^l / ₂ T _e and LZ ₂ to τ ₂ = T _e .

At the output of each AU channel for various FMS fragments, we have:

- when taking a fragment with HA

,

,

;j∈ [1, n _g ];

;

- when receiving a fragment with FM-M

,

,

;j∈ [1, n _m ];

;

- when receiving a fragment with FM-PSP

,

,

;j∈ [1, n _and ];

;

where U _s1k (t), U _s2k (t) are the voltage at the output of AKO ₁ and AKO ₂ , due to the interaction of the signal-to-signal type.

In the protective gap, due to the absence of a signal, we have

,

,

.j∈ [1, n _s ];

.

Given the action of interference n (t), the detection algorithms “n” from “n” are used to open the spectral-temporal structure of the packet, characterized by the following relationships:

при

at

при

at

при

at

при

at

,

,

,

- оценки моментов начала фрагментов Г, М, И, З;

,

,

,

- оценки длительности фрагментов Г, М, И, З;

,

,

,

- оценки количества импульсов при обнаружении фрагментов Г, М, И, З.where N _g , N _m , N _and , And _s - the hypothesis of the presence of fragments G, M, I, Z; U _yk1 (T), U _y2k (T) - output effect in AKO ₁ and AKO ₂ taking into account the interference n (t);

,

,

,

- estimates of the moments of the beginning of fragments G, M, I, Z;

,

,

,

- estimates of the duration of fragments G, M, I, Z;

,

,

,

- estimates of the number of pulses when detecting fragments G, M, I, Z.

и длительность пакета

.Based on information about the structure of the package, the moment of start

and packet duration

.

If it is necessary to open the structure and parameters of the frames, information about the set of packets is accumulated and then a decision is made according to similar algorithms.

The reliability of opening the spectral-temporal structure of fragments and the FMS packet when using detection algorithms of type “n” from “n” is characterized by the following relationships:

P _n = 1- (P _Osh1 + P _Osh2 );

P _OSH = P _OSH + P _LTM + P _OSH + P _LTZ ;

P _OSH = P _OSH + P _OSHM + P _LTI + P _LTS ;

R _oshg = 1-P _og ; P _osm = 1-P _ohm ; P _osh = 1-P _oi ;

; β_г=1-D_г;

; β _g = 1-D _g ;

; β_м=1-D_м;

; β _m = 1-D _m ;

; β_и=1-D_и;

; β _and = 1-D _and ;

P _ltz = n _s α; P _ltm = n _m α; P _lti = n _and α,

where R _OSH , R _OSH - the probability of erroneous opening of the spectral composition of the FMS package in AKO ₁ and AKO ₂ ; R _p - the probability of the correct opening of the spectral composition of the FMS package in AC; R _oshg , R _oshm , R _oshi - the probability of erroneous decisions when detecting FMS fragments: G, M, I; R _og , R _om , R _oi , - probabilities of the correct detection of fragments G, M, I; D _g , β _g - the probability of correct detection and skipping of the segments of the fragment F of the FMS for the time T of one integration cycle; D _m

β _m - the probability of correct detection and skipping of segments of the fragment M FMS for the time T of one integration cycle; D _and , β _and are the probabilities of the correct detection and omission of segments of the fragment And the FMS for the time T of one integration cycle; α is the probability of false alarms during the time T of one integration cycle; R _LTM , R _LTI , R _LTZ - the probability of false alarms during the duration of the FMS fragments: M, I, Z.

When using multi-cycle discrete integration in AC, the mean square errors of estimating the moments of the beginning of FMS fragments σt _i and the duration of fragments of FMS σT _{j are} determined by the following relations:

при i∈[1, 5];

при j∈[1, 5].

when i∈ [1, 5];

for j∈ [1, 5].

The value of the time constant T KI is selected on the basis of a compromise between the reliability parameters of opening the spectral composition of the FMS packet and the error in estimating the time parameters of fragments of the FMS packet.

When Δf _n T _e ≤10, that is, when, given that Δf _n = 2 / T _e , we have T≤5T _e , the distribution of the effect at the output of the PEC corresponds to the Rayleigh-Rice law, and then the noise immunity characteristics of the AKO can be calculated as follows :

;

; j∈[1, 3];

;

; j∈ [1, 3];

;

;

j = 1≡Г; j = 2≡M; j = 3≡I;

;

;

,

;

;

,

where D _j is the probability of the correct detection of elements of the jth fragment of the FMS during time T; Q (g _j , g _p ) is the Markum function; I ₀ (...) is the zero-order Bessel function; g _j is the signal-to-noise ratio for the voltage at the output of the PEC when receiving the j-th FMS fragment in one integration cycle; g _p - normalized threshold; g ² _in - input signal-to-noise ratio in power; P _s is the power of the FMS at the input of AC; σ ² _n is the dispersion of the noise n (t) at the input of the AC; Δf _n is the noise band at the input of the AU; r _sj (τ) is the envelope of the FMS autocorrelation coefficient when receiving the jth fragment; N _n is the spectral density of interference n (t) at the input of AC.

Given that AKO ₁ τ ₁ = _^1/2 T _e, then with r _s1 (τ ₁₎ → 1; r _s2 (τ ₁ ) → 0; r _s3 (τ ₁ ) → 0.5; and in AKO ₂ τ ₂ = T _e and at the same time r _s1 (τ ₂ ) → 1; r _s2 (τ ₂ ) → 1; r _s3 (τ ₂ ) → 0.

To illustrate the obtained relations, we determine the characteristics of the reliability of opening the spectral composition and time parameters of fragments of the FMS packets with the following initial data: T _e = 2 · 10 ^-6 s; T _g = 3-10 ^-5 s; T = 10 ^-5 s; T _m = 3-10 ^-5 s; T _and = 10 ^-4 s; T _s = 2 · 10 ^-5 s; T _p = 1.8 · 10 ^-4 s.

When g _p = 4,2 we have α = 10 ^-4 and P _lt = 10 ^-2 , P _ltm = 3 · 10 ^-3 , P _ltz = 2 · 10 ^-3 .

To ensure _Pn ≥0,9 highest noise immunity requirements for stage opening MBF packet structure corresponding fragment treatment and, since r _s3 (τ ₁₎ = _^1/2 r _s1 (τ ₁₎ that is equal to the energy loss of 6dB , and in addition, the number of PMS elements n _and >> n _g and is n _and = 10 ² .

.To ensure R _osh = 8 · 10 ^-2 it is necessary that β _and = 8 · 10 ^-4 . In the AKO ₁ when τ ₁ = ¹ / _{2T e} require that g _{and h} = g = 8.5; but

.

Then, since for the same data g _1r = 17, g _2r ⁼ 17 and g _2m = 17, we obtain β _g = β _m = 10 ^-4 , and P _osr = 3 · 10 ^-7 ; R _OSHM = 3 · 10 ^-7 .

Thus, to ensure P _n > 0.9, it is necessary to have g _bx ≥4. The root-mean-square errors of estimating the moments of the onset of fragments of the FMS package σt _i and the duration of the fragments of the FMS package σT _i are

σt _i = 2.9 T _e = 5.8 · 10 ^-6 s; σT _i = 5.8 T _e = 11.6 · 10 ^-6 s,

that is, the errors σt _i and σT _{i are} much less than T, _etc.

In conclusion, we present a comparison of the reliability of opening the structure of the FMS package when using the proposed AU and autocorrelator based on the prototype [3]. Since in the prototype the indication of the moments of jumps in the phase of the FMS is carried out on the basis of fixing short pulses with a duration of τ _and = T _e -τ ₁ ≤0.5 T _e , for their fixation it is necessary that the integration constant of the prototype T _{0 be} selected from the condition T ₀ ≤T _and .

, то есть энергетический выигрыш АУ составляет 10 дБ.In this case, the prototype is inferior to the AU in noise immunity in proportion

, that is, the energy gain of the AC is 10 dB.

In the case when in the prototype, as in AU, g _bx = 4, due to a significant increase in Р _Ош and Р _{лт, the} opening of the FMS structure becomes unreliable, and the fixation of the moments of phase jumps is difficult due to the large number of additional pulses corresponding to false alarms .

Thus, the proposed device provides prompt opening of the spectral-temporal structure of the FMS with TDMA with high reliability and low error in estimating the temporal parameters of FMS fragments.

The implementation of the device is not difficult. All its functional units are typical and can be performed on a modern element base.

Information sources

1. USSR Copyright Certificate No. 560343. 1976. Dikarev V.I., Romanenko V.A. Autocorrelation meter of pseudo-random phase-manipulated signal parameters.

2. USSR copyright certificate No. 921104. 1982. Dikarev V.I., Romanenko V.A. Autocorrelation meter of pseudo-random phase-manipulated signal parameters.

3. RF patent No. 2309550. 2007 Ipatov A.V., Dikarev V.I. A method for autocorrelation receiving a noise-like signal.

Claims (1)

An autocorrelation device for opening the spectral-temporal structure of signals of digital communication systems containing an autocorrelator with quadrature processing (28), including multipliers (1, 2), a phase shifter (3), a delay line (4), switched integrators (5, 6), quadrators (7, 8), adder (9), square root extraction device (26), the input of the autocorrelation device is connected to the inputs of the phase shifter (3) and delay line (4), as well as to the first input of the multiplier (2), the output of the phase shifter ( 3) connected to the first input of the multiplier (1), output l delay (4) is connected to the second inputs of the multipliers (1, 2), the output of the multiplier (1) is connected through a series-connected switching integrator (5) and quadrator (7) to the first input of the adder (9), the output of the multiplier (2) is connected through series-connected switching integrator (6), a quadrator (8) with a second input of the adder (9), the output of which is connected to the input of the square root extraction device (26), characterized in that a second quadrature processing autocorrelator is additionally introduced into it (29), including lane factors (12, 13), phase shifter (15), delay line (14), switched integrators (16, 17), quadrators (18, 19), adder (20), square root extractor (27), and threshold devices (10, 21), counters (11, 22), controller (23), clock (24), solver (25), the input of the autocorrelation device is connected to the inputs of the phase shifter (15) and the delay line (14), and with the first input of the multiplier (12), the output of the phase shifter (15) is connected to the first input of the multiplier (13), the output of the delay line (14) is connected to the second inputs of the alternator inhabitants (12, 13), the output of the multiplier (12) is connected through a series-connected switching integrator (16) and a quadrator (18) to the first input of the adder (20), the output of the multiplier (13) is connected through a series-connected switching integrator (17) and a quadrator (19) with the second input of the adder (20), the output of which is connected to the input of the square root extraction device (27), the output of the square root extraction device (26) through the threshold device (10) connected in series and the counter (11) is connected to the first input of the decision devices (2 5), the output of the square root extraction device (27) through the threshold device (21) connected in series and the counter (22) is connected to the second input of the resolving device (25), the controller (23) generates commands based on external target designations, from the first output of the controller ( 23) a command is sent to the control input of the delay line (4), from the second output of the controller (23) a command is sent to the control input of the delay line (14), from the third output of the controller (23) commands are sent to the clock generator (24), the output of the generator clock pulses (24) soy Inonii to the control inputs of switched integrator (5, 6, 16, 17) and threshold devices (10, 21) and the third input of the decision unit (25).