RU2480783C1 - Method for radiolocation of non-linear-inertial objects - Google Patents

Abstract

FIELD: radio engineering, communication.

SUBSTANCE: radiated signal is a sequence of harmonic oscillations G_n, of frequency f₁, n=1, 2,…, N, with stepped dependence of their capacity on "n", and the physical essence of processing of signals reflected from non-linear-scattering objects on double frequency f₂=2f₁ and corresponding to emitted oscillations G_n, is reduced to measurement of period-to-period increment of a phase of reflected signals (φ_n=1-φ_n), detection of a module of these increments [φ_n=1-φ_n], calculation of their average value and its comparison with a threshold.

EFFECT: improved characteristics of detection of those of them, equivalent electric circuits of which, apart from non-linear resistor elements, contain inertial elements - inductances and capacitances.

1 dwg

Description

The invention relates to techniques for short-range radar, and in particular to methods and means of search, detection and recognition of nonlinearly scattering objects (NRA).

Known methods of nonlinear radar [1; 2], in which the NRA is irradiated with a probing signal (ZS) in the form of one or two harmonic oscillations and registration of the energy of the reflected signal at multiple and / or combination frequencies in the radar receiver.

Many NRA can be divided into two groups that differ in the electrophysical properties of the NRA:

- RLC-objects, the model of which in the form of an equivalent electric circuit contains non-linear resistor (R) elements and inertial elements - inductive (L) and / or capacitive (C). A typical representative of RLC-objects are mobile phones, electronic listening devices, receiving devices of radio-controlled fuses, etc .;

- R-objects, the electrical model of which contain only non-linear resistor elements. Typical representatives of R-objects are elements of steel structures, welded or riveted, susceptible to corrosion and rust [3].

It is assumed that in the method under consideration, it is the RLC objects that represent the NRA of interest to us.

The disadvantage of analogues is the low efficiency of the RLC detection procedure. This is explained by the presence in the search area of a large number of R-objects, the signal flow from which at the input of the non-linear radar significantly increases the level of false alarms and thereby reduces the efficiency of the search and detection of RLC-objects. At the same time, the energy released in the analogs of the signal reflected from the HPO does not have sufficient information content to reveal reliable signs of recognition of RLC objects and R objects.

The closest among the analogues is the method of nonlinear radar [1; 2], which consists in emitting a harmonic oscillation G (t) of frequency f ₁ in the direction of the NRA, receiving the reflected signal R (t), extracting oscillations from it at the double frequency f ₂ = 2f ₁ and deciding on the detection of the NRO by comparing the energy of the received signal with a threshold. To bring the prototype as close as possible to the proposed method, we introduce the following refinements to its model:

- denote by T the time allotted for the detection of NRA at a given point in space;

- ZS prototype present as a sequence of harmonic oscillations G _n , n = 1, 2, ..., N, emitted alternately, in accordance with the index "n", during the corresponding time periods T _n :

each of duration τ = T / N. All G _n have their own coherence during the time intervals T _n (1), which can be called periods of coherence of the GL. Between themselves, the oscillations G _n do not have to be coherent. The powers p _{n of} all the oscillations G _{n are} assumed to be the same, independent of “n”;

- similarly reflected from the NRA signal R (t) can be represented as a sequence of signals R _n , n = 1, 2, ..., N, arriving at the receiver of the nonlinear radar also alternately, in accordance with the index "n". Given the small size of the desired NRA, the duration of each of R _n coincides with τ. In addition, assuming a short range to the desired NRA, and therefore a small delay of the reflected signal compared to the duration τ = T / N, the existence intervals R _n can be chosen to coincide with the coherence periods T _n (1);

- the selection of oscillations at the double frequency f ₂ = 2f _{1 is} carried out by the method of synchronous detection of each of R _n during the corresponding coherence period T _n , i.e. the entire time interval 0≤t≤T consists of N periods of synchronous detection, each of duration τ.

The disadvantage of the prototype is the low detection efficiency of RLC-objects against the background of a stream of "false" signals from R-objects.

The aim of the invention is to increase the detection efficiency of RLC objects.

To achieve the goal in the prototype method, which consists in generating a sequence of harmonic oscillations G _n , n = 1, 2, ..., N, frequency f ₁ , alternately, in accordance with the index "n", the radiation G _n in the direction of the desired object, receiving the corresponding reflected signals R _n , n = 1, 2, ..., N with highlighting the double frequency signals f ₂ = 2f ₁ in them, determining the complex envelopes Z _{n of the} received signals R _n , measuring the total power of the received signals P _Σ = ΣZ _n Z _n * and deciding the detection of nonlinear scattering object by comparison Ia output value processing unit with a threshold, the present invention further carried out the change in power generated oscillations G _n, forming a sequence of reference signals S _{n, n} = 1, 2, ..., N by limiting G _n and allocating them to twice the oscillation frequency f _2, definition complex envelopes W _{n of} reference signals S _n , calculation of complex inter-period correlation coefficients K _{Z, n} = Z _{n + 1} Z _n * and K _{W, n} = W _{n + 1} W _n * envelopes Z _n and W _n, respectively, calculation of the mutual coefficients inter-period correlation K _n = Im (K _{Z, n} K _{W, n} *) and their mode values of [K _n ], the calculation of the resulting correlation coefficient K _Σ = Σ [K _n ], and K _{Σ is} used as the output value of the processing device, and a value proportional to P _{Σ is} used as the threshold.

Figure 1 shows a possible diagram of a nonlinear radar that implements the proposed method, elements 1-11 of which carry the following technical contents: 1 - frequency synthesizer f ₁ and f ₂ = 2f ₁ ; 2 - power amplifier; 3 - power control unit of the emitted signal; 4 - receiver of the reflected signals; 5 - a device for generating reference signals of a double frequency f ₂ = 2f ₁ ; 6 - synchronous detector of the receiver; 7 - synchronous detector of the transmitter; 8 - arithmetic device; 9 is a first threshold circuit; 10 is a key diagram; 11 is a second threshold circuit. Elements 1, 2, and 3 form a radar transmitter.

The functioning of the proposed method is convenient to consider, referring to the nonlinear radar diagram, figure 1. First we make the following remark. Among the various models of RLC objects, we will be interested only in those whose nonlinear resistor elements (R) have the following distinctive feature: their current-voltage characteristic must contain a quadratic partial component that doubles the frequency f ₁ GC, as a result of which R is reflected in the reflected signals _{n there} will be a component at the double frequency f ₂ = 2f ₁ .

Synthesizer 1 generates a continuous harmonic oscillation of the frequency f ₁ , which from its first output goes to the main input of the amplifier 2, the auxiliary input of which receives a signal from the block 3, which controls the output power of the amplifier 2 so that the power p _{n of the} partial harmonic oscillation G _n at the output of the transmitter depended on "n", for example, changed according to step law

with a constant step Δр.

In the proposed method, a situation is not excluded when variations in the power of the ES can be accompanied by distortions of the phase of the ES

where ψ _n is a random phase varying from one coherence period T _n to another. Even if the frequency oscillation f ₁ from the output of the synthesizer 1 maintains its coherence during the entire time T is detected, the oscillations G _n cease to be coherent with each other after varying the power of the ES.

The signals R _n are received by the receiving antenna and, having passed the receiver 4, in which they are pre-filtered at a double frequency f ₂ , are fed to the main input of the synchronous detector 6. The complex envelopes Z _{n of the} received signals R _n corresponding to the oscillations G _n ЗС

have an amplitude A _n and a phase consisting of two terms:

- φ _n , the phase of the NRA, characterizing the electrophysical properties of the NRA;

- 2ψ _n , random phase, “imposed” by the phase of the ZS (3).

The device 5 generates reference signals S _n in each n-th coherence period T _n . For this, the output oscillations of the transmitter G _{n of the} frequency f _{1 are} limited in amplitude with their subsequent filtering, which suppresses the frequency f ₁ and all multiples of f ₁ frequencies, except for the component at the frequency f ₂ = 2f ₁ , as a result of which a reference sequence is formed at output 5 harmonic signals S _n , n = 1, 2, ..., N, at a double frequency f ₂ , with a constant - independent of "n" - amplitude and with a phase changing from one coherence period T _n to another according to a random law and equal to doubled oscillation phase G _n . The complex envelopes W _{n of the} reference signals S _n are of the form:

It is assumed that the filter time constant of device 5 is significantly less than the duration T / N. Essentially, device 5 simulates a frequency multiplication process occurring in an HPO resistor element. Signals S _n are fed to the main input of the synchronous detector 7.

The quadrature oscillations Cos2πf ₂ t and Sin2πf ₂ t of frequency f ₂ are supplied to the double auxiliary inputs of synchronous detectors 6 and 7 from the second output of synthesizer 1; as a result, at the double outputs of synchronous detector 6, we obtain the quadrature components X _n and Y _{n of the} complex envelopes Z _n accepted Signals R _n :

and at the double outputs of the synchronous detector 7, the quadrature components U _n and V _{n of the} complex envelopes W _{n of the} reference signals S _n :

At the end of each of T _{n, the} quadrature components X _n , Y _n , U _n , V _{n are} converted to digital form and fed to the inputs of the arithmetic device 8, where it is calculated:

- power samples of the received signals R _n :

- complex coefficients of inter-period correlation of envelopes Z _n :

- complex coefficients of inter-period correlation of envelopes W _n :

- coefficients of mutual inter-period correlation

- the total power of the received signals R _n :

- the resulting correlation coefficient:

where "*" is the sign of complex conjugation, "Im" is the operator for extracting the imaginary part of the complex number, [K _n ] are the modular values of K _n .

The essence of the calculations performed in device 8 follows from the amplitude-phase representations (4), (5) of the complex envelopes Z _n and W _n after substituting them in (8), (9) and then substituting (8), (9) in ( 10), (11):

According to (15), the coefficients K _{n are} proportional to the weighted values of the sine of the phase difference of the NRA (φ _{n + 1} -φ _n ) in adjacent periods of the emission of the CS. Assuming in (15) the smallness between the periodic increments of the phases of the NRA:

Algorithm (11) can be considered as the suppression of correlated components in the HPO {φ _n } phase sequence and the selection of weakly correlated values in it. In addition, algorithm (11) makes it possible to neutralize the negative influence of the phase uncertainty of the GL on the efficiency of the nonlinear radar, which follows from the fact that random phases ψ _n CS are absent in (15).

The value of P _{Σ is} compared in the first threshold circuit 9 with the internal noise power of the radar, where a positive or negative decision is made to detect an NRF, regardless of whether it is an R-object or an RLC-object. Note that the calculation of P _Σ and comparison of P _Σ with a threshold are also performed in the prototype. In the case of a positive decision, the key circuit 10 gives permission for comparison in the second threshold circuit 11 of the coefficient K _Σ with a threshold proportional to the value of P _Σ . If this threshold is exceeded, a decision is made to detect the RLC object.

We give a physical interpretation of the entire algorithm for detecting an RLC object. The function performed by the threshold circuit 11 can be considered as measuring the ratio K _Σ / P _Σ and comparing it with the threshold. Given the fact that:

the entire algorithm for detecting an RLC object can be interpreted as a comparison with the threshold of the average value of the HPO phase increment modulus:

where averaging is carried out with weights A _{n + 1} A _n .

We proceed to justify the effectiveness of the proposed method. The initial prerequisite for the creation of the proposed method is the well-known property of currents in nonlinear inertial electric circuits [4, 5], which consists in the fact that not only the amplitudes, but also the phases of the currents functionally depend on the amplitude of the harmonic source of electromotive force in these circuits. In relation to the situation under consideration, this means the dependence of not only the amplitude A of the signal reflected from the RLC object, but also its phase φ on the power of the ES. At the same time, when observing R-objects, the phase φ of the reflected signal does not depend on the power of the CS, although the dependence of its amplitude A on the power of the CS remains. Therefore, phase increments (φ _{n + 1} -φ _n ) included in (15), which determine the value of K _Σ , will have noticeable deviations from the zero value only when observing RLC objects, and when observing R objects, these increments, and therefore the value of K _Σ will be close to zero values. As a result, the K _Σ / P _Σ ratio will be much larger in the case of observing RLC objects than in the case of observing an R object, which allows one to consider the K _Σ / P _Σ ratio as a parameter by which it is possible to recognize RLC objects and R- objects.

It was implicitly assumed above that the relative location of the desired RLC object and radar during the detection time T remains unchanged. Obviously, when the proposed location of the desired RLC object is changed, the above-mentioned process of emission of ES and processing of the reflected signal is repeated.

The value of T lies approximately in the range from fractions to several seconds, N≈5-10, and the values of the initial ES power p ₀ and step Δp are determined by the results of standard tests to detect typical RLC objects.

In conclusion, we note that the proposed method can be used in various nonlinear radars that extract not only the second, but also any other multiple harmonics from the reflected signal when the NRA is irradiated with one harmonic oscillation or the combination frequencies in the case of HPO irradiation with two or more harmonic oscillations.

The necessary changes to the circuit of the proposed device are obvious.

Information sources

1. Musabekov P.M., Panychev S.N. Non-linear radar: methods, techniques and scope. Foreign electronics. Successes of modern radio electronics, 2000, No. 5, p. 54-61.

2. Belyaev V.V., Mayunov A.T., Razinkov S.N. Status and development prospects of "non-linear" radar. Successes of modern radar, 2002, No. 6, p. 59-78.

3. Steinshleiger VB Nonlinear scattering of radio waves by metal objects. Uspekhi Fizicheskikh Nauk, 1984, Volume 142, Issue 1, pp. 131-145.

4. Bessonov L.A. Nonlinear electrical circuits. High School, 1977

5. Danilov L.V., Mathanov P.N., Filippov E.S. Theory of nonlinear electric circuits. Energoatomizdat, 1990

Claims (1)

The method of nonlinear radar tracking of nonlinear inertial objects, which consists in generating a sequence of harmonic oscillations G _n , n = 1, 2, ..., N, frequency f ₁ , alternately, in accordance with index n, radiation G _n in the direction of the desired object, receiving the corresponding reflected signal R _{n, n} = 1, 2, ..., N, with separation therein of doubled frequency signal f ₂ = 2f _1, determining the complex envelopes of the received signals Z _n R _n, measuring the total power of received signals P _Σ = ΣZ _n Z _n *, where the * sign is a sign of complex conjugation, and making a decision to find enii nonlinear scattering object by comparing an output value processing device with a threshold, characterized in that the change in power is further carried generated oscillations G _n, forming a sequence of reference signals S _{n, n} = 1, 2, ..., N, by limiting G _n and isolation the oscillations of the doubled frequency f ₂ in them, the determination of the complex envelopes W _{n of the} reference signals S _n , the calculation of the complex inter-period correlation coefficients K _{Z, n} = Z _{n + 1} Z _n * and K _{W, n} = W _{n + 1} W _n * of the envelopes Z _n and W _n, respectively, calculation of coefficients the inter-period correlation K _n = Im (K _{Z, n} K _{W, n} *) and their modular values [K _n ], where Im is the operator of extracting the imaginary part of the complex number, calculating the resulting correlation coefficient K _∑ = ∑ [K _n ], moreover, K _{∑ is} used as the output value of the processing device, and a value proportional to P _{используется is} used as the threshold.

Images