RU2255354C2 - Device for selecting signals from moving targets - Google Patents

Abstract

FIELD: radio communications.

SUBSTANCE: device has coherent heterodyne, phase detector, analog-digital converter, first memory block, second memory block and subtracting block, second input of phase detector is input for signals from radiolocation receiver, while also provided are interpolation block and permanent memory block.

EFFECT: higher efficiency.

8 dwg

Description

The invention relates to radar technology and can be used for selection (detection) of echo signals reflected by moving air objects (IN). You can use the invention in other areas of technology where the selection of signals is carried out by phase or frequency, for example, in radio communications and in measuring equipment.

The selection of moving targets is one of the most important and complex tasks of the theory and technique of radar. Known devices for moving targets selection (SDC) are implemented in the form of inter-period compensation devices (CPC) and coherent accumulation-filter system for moving targets selection (FSDS) [1, 2].

The FSDC provides optimal inter-period filtering of signals by Doppler frequency. Under the conditions of a priori uncertainty of this parameter and under the conditions of intense interference caused by fixed objects, the implementation of such devices leads to a complex multichannel device.

CPC systems are simpler. However, they have inherent disadvantages associated with the non-optimal inter-period processing, the unevenness of their frequency (speed) characteristics, the presence of transients, during which the maximum (high) suppression of the signals of stationary objects has not yet been achieved, and the reduction of the suppression of signals of stationary objects when scanning the radiation pattern in azimuth.

CPC and FSDC systems are used together, which improves the SDC device as a whole [4]. However, the above disadvantages of the CPC devices are significant in this case as well.

Methods to address these shortcomings are known. For example, to stabilize the suppression coefficient during antenna rotation, additional antennas with special radiation patterns should be used [1, 3].

To improve the uniformity of the speed characteristics in the CNC, feedbacks are used [2].

However, all these devices are either rather complicated, such as [1, 3], or are associated with a compromise, since the improvement of the speed characteristics occurs due to an increase in the duration of transients [2].

Closest to the claimed is a device taken as a prototype. It contains a coherent local oscillator, a phase detector, an analog-to-digital converter, storage devices, an adder, multipliers, an integrator, a shift register, a subtractor, and an adder. The device operates as follows: in a phase detector, to which, in addition to the signal, a coherent reference oscillation is applied, the signal is detected and then converted into a digital code using an analog-to-digital converter. Further, this signal is delayed in two sequential memory devices and fed to a subtractor. At the same time, a signal is received to the subtractor not only after a twofold delay, but also after a one-time delay and its correlation with the output signal of the device in the multiplier and integrator and undetected in the second multiplier.

The output of the subtracting device is the output of the entire SDS device. Circuits containing multipliers, an integrator, and a shift register form feedback. Using this circuit, the optimal ratio of the echo signals (uncontrolled, delayed and twice delayed) is automatically established and as a result of this, compensation and separation of signals of moving VOs are achieved. The prototype is characterized by the disadvantages mentioned above, i.e. the speed characteristic is uneven, there is a transition process when the suppression of the signals of stationary objects has not yet occurred, etc.

The purpose of the invention is to improve the compensation of signals of stationary objects and increase the likelihood of detection of signals of moving targets.

This goal is achieved by the fact that in the device adopted for the prototype and containing a coherent local oscillator connected to a phase detector connected to an analog-to-digital converter, according to the invention, two sequentially connected memory devices and a subtracting device are introduced, to the second input of which an interpolating device connected with the output of the first storage device and a source of support functions in the form of read-only memory.

The use of subtracting and delaying devices in PCV systems is known [1, 2, 5]. When subtracting, they use unrestrained and delayed signals. In the claimed device for subtraction, a signal is used which is not delayed for a period of (or several) repetitions, but rather a signal synthesized from an undetected signal based on a reference to a special spectrum of it (unreserved signal). In this case, the spectrum is determined only in the region of zero frequencies and, as a result, the signal generated from the spectrum contains only signals from stationary objects. In the subtractor, they are compensated, and the signals of moving VOs are saved. Thus, the delayed signal generation circuit in the claimed device has fundamental differences from the known devices and this is the first feature of the claimed device.

The second feature of the claimed device in comparison with the known is associated with the presence of a second storage device in the source signal circuit. The need to turn it on is determined by the peculiarity of the delayed signal generation circuits. Indeed, the formation of this signal occurs on the basis of the determination of their spectra. It is known that this operation is related to integration and is generally performed for the signal as a whole, i.e. the entire signal (the entire packet). Therefore, the signal formed on the basis of the spectrum is delayed with respect to the initial (generating) signal. The second storage device delays the original signal, combines in time the signals at the inputs of the subtractor, and thereby enables the selection of signals from moving VOs.

It is these main features of the claimed device that allow us to achieve the goal and, as the comparative analysis shows, are absent in the known devices.

This allows us to conclude that the claimed device meets the criteria of "novelty" and "significant differences".

The invention is illustrated graphic material. Figure 1, 2, 3, 7 depict, respectively, structural diagrams of the prototype, the claimed and interpolating device. The following notation was used on them: 1 - coherent local oscillator, 2 - phase detector, 3 - analog-to-digital converter, 4 - storage device, 5 - subtracting device, 6 - interpolating device, 7 - read-only memory device, 8 - multiplier, 9 - integrator, 10 - shift register, 12 - phase shifter by π / 2, 11 - adder, 13 - quantization pulse input, 14 - address inputs, 15 - write-read control signal input, 16 - reset pulse input, 17 - transfer pulse input .

Figure 4, 5, 6, 8 depicts a plot of signals explaining the operation of the device. The following notation was used: t - current time; T _G - signal repetition period; S ₁ - echo signals; β is the azimuth; D is the range; D _max - maximum range; 1, 2, ... M - echo signals of M - pulse burst; Δ β is the width of the radiation pattern; β ₁ , β ₂ - azimuthal coordinates of the end and beginning of M - impulse pack; P ₀ , P ₁ , P ₂ , P - basis functions; S _in - restored signal; S ₁₁ - delayed signal; S _1Δ is the signal after compensation; C ₀ , C ₁ , C ₂ are the spectral coefficients for the basis functions P ₀ , P ₁ , P _2, respectively; C ₀ P ₀ , C ₁ P ₁ , C ₂ P ₂ - components of the restored signal; ω _D is the Doppler frequency; K _{1 is the} transmission coefficient of the signals; R _g - shift register; u ₁ , u ₂ - pulses control shift registers; S _1i is the echo signal from the i- ^th element of the range; S _2i - delayed by T _D echo signal from the i ^th element of the range; S _ij - echo signal j ^th element range; Δ t ₁ - the interval of reading signals to the interpolating device (Δ t ₁ ≤ T _and / M). T _and - the duration of the signal.

As follows from figure 2, the claimed device contains a serially connected coherent local oscillator 1, a phase detector 2, an analog-to-digital converter 3, a first and second storage device 4 and a subtractor 5. An interpolating device 8 is connected to the output of the read-only memory 4 by the first input the other m inputs are connected to the outputs of the read-only memory 7, and the output interpolating device 6 is connected to the second input of the subtractor 5. The output of the coherent local oscillator 1 is connected It is connected to the second input of the phase detector 2, the first input of which is connected to the output of the radar receiver. In addition to the memory devices 4 are fed control signals write-read and, as well as read-only memory 7, address signals. In addition, the control signals are supplied to the analog-to-digital Converter 3 and the interpolating device 6.

The device operates as follows: the signal at the frequency of the coherent local oscillator 1 is fed to the first input of the phase detector 2 from the radar receiver. The reference oscillation from the coherent local oscillator 1 is fed to the second input of the phase detector 2. From the output of the phase detector 2, the video signals are fed to an analog-to-digital converter 3, where they are sampled by time and converted to digital form. The sampling times of the analog video signal are determined by the control signals at input 13. These signals are periodic and their repetition period is selected in accordance with the Kotelnikov theorem [5]. Next, the echo signals in digital code are supplied to the first storage device 4 (memory). The address control signals, 14 provide the recording of signals of all N range elements in each repetition period of the M-pulse burst and the reading of these signals. For this, write-read control signals are also used, which determine the state of the memory at each current point in time. In the process of recording and reading signals in the memory, the usual combination of different samples of the received signal in time occurs in the optimal reception technique [5]. For clarity, we can talk about the method of filling the memory when writing and clearing it when reading. The filling of the memory is carried out first at all ranges in one repetition period, then in the next repetition period, etc. Thus, the memory is quickly filled in by the range elements and slow by the azimuth elements.

Signals are read from different azimuth elements and one range element. Then the transition to the next element of the range, etc. As a result, fast reading is performed on azimuth elements and slow on range elements. Thus, the read signal contains M samples of the digitized signal from one element of range learned in M adjacent repetition periods. This signal is supplied to the interpolating device 6. In this device, the signals are decomposed into a spectrum according to several (m = 2 ÷ 7) low-frequency basis functions and the low-frequency components of the original signal are restored from this spectrum. To this end, signals of basic functions m, which are preselected and stored digitally in read-only memory 7 (ROM), are supplied to the interpolating device 6. The order of reading signals from the ROM is determined by the control pulses of address 14 and is a synchronous process of reading signals from the first memory 4.

When a low-frequency component of a similar signal is formed in the interpolator 6, time is inevitable. Therefore, this signal is delayed with respect to the input signal, and before subtraction, it is necessary to eliminate the non-simultaneity of signals. For this, the input echo signal from the first memory 4 is supplied to the second memory 4, in which it is delayed. Next, the simultaneous delayed input signal from the second memory 4 and the converted signal from the interpolator 6 are fed to a subtractor 5.

Since the interpolating device 6 uses several (2 ÷ 7) low-frequency basis functions, the signal converted in it repeats only the low-frequency component of the input signal. It is known that such components are formed by reflections from stationary objects [1, 2]. Due to the Doppler effect, signals of moving targets have non-zero carrier frequencies and therefore practically do not contribute to the spectrum using the basic functions used in the interpolator. As a result, the signals of moving targets are not contained in the signal of the interpolator, which is converted from a spectral signal.

Thus, in the subtractor 5, only the signals of the stationary objects are compensated, and the signals of moving targets are output from the device. Therefore, the device really selects the signals of moving targets.

Important features of the claimed device are associated with the use of a fixed object reconstructed based on the spectrum of the signal. Therefore, it should be considered in more detail (the structure and operation of the interpolating device 6, in which the signal recovery is performed. The interpolating device 6 consists of m (2 ÷ 7) parallel channels, each of which consists of a series-connected multiplier 8, an integrator 9, a shift register 10 and a second multiplier 8. The outputs m of the second multipliers 8 are connected to the inputs of the adder 11, the output of which is the output of the interpolating device 6. The inputs m of the first multipliers 8, to which the signal is supplied, are interconnected, and The second m inputs of these multipliers are supplied with m signals of the basis functions from ROM 7. The second inputs of the multipliers 8 are connected to each other in parallel. Pulses are sent to the integrator 9 and shift register 10 to reset 16 and transfer signals 17.

The interpolating device 6 operates as follows: at its input, i.e. all m inputs of the first multipliers 8 receive a signal from the first memory 4. As indicated earlier, this is a signal from the VO from one range and consists of M digitized pulses that follow continuously one after another. At the same time, the second inputs of m multipliers 8 also receive the digitized signals of m basis functions. The result of multiplication in each channel is fed to integrator 9, where all M samples are summed. To eliminate errors associated with the result of previous integration, the signal 16 is used to set the integrator 9 in the initial state with a zero value of the output signal. The accumulation result from the output of the integrator is transferred to the shift register 10 using the transfer pulse 17, where it is stored until the arrival of the next pulse 17. Thus, the sequence of pulses 16 and 17 must be determined. At the end of the accumulation cycle, after the action of the last Mth pulse, the transfer pulse 17 and then the reset pulse 16 operate. The signals received in the integrators 9 and stored in the registers. 10 are m samples of the spectrum of the BO signal from one range element. Further, these signals in the second multipliers 8 are again multiplied by the basic functions and summed in the adder 11. As a result of this, a time signal corresponding to the spectrum of m samples is generated. It was previously indicated that for small m (2–7) and the choice of optimal basis functions, the received signal with high accuracy corresponds to low-frequency signals from stationary objects.

для примера, приведенного на фиг.4б. Затем в таком же порядке происходит считывание эхо-сигналов со следующего азимутального направления и т.д. Следовательно на интерполирующее устройство поступает как бы сжатая во времени пачка эхо-сигналов.The signal diagrams given in the application materials explain the operation of the claimed device. Figure 4 shows the plot of the echo signals. Figa corresponds to the reception of signals S ₁ from VO at a certain distance D ₁ . Echoes follow with a repetition period of T _G. Using memory 4, these signals are stored. In the process of storing and reading signals are combined in time. For processes in the memory, it is therefore convenient to present the echo signals in the range-azimuth coordinates, as is done in Fig. 4b. Here D _max = cT _G / 2 is the maximum for a given T _G uniquely determined range to VO. The signals from one VO (S ₁ ) on this plot are separated by a time interval (T _G ). The azimuthal coordinate of these signals is almost the same, and its difference is determined only by the width of the radiation pattern (Δ β). The signals to be stored in the first memory 4 come sequentially in time. As can be seen from figa, the signals from one IN (S ₁ ) come through the time interval T _G. When reading, the situation is different, since signals from the same ranges are read sequentially. Thus, M signals are first read from the first range element, then from the second, and so on through the range element D ₁ to the last. In this case, the azimuthal direction of these echo signals is the same and equal

for the example shown in figb. Then, in the same order, the echo signals are read from the next azimuthal direction, etc. Consequently, a packet of echo signals, as it were, compressed in time, enters the interpolating device.

On figv shows the plot of the signals at the main points of the claimed device. S ₁ is the input signal after the first memory 4. Three different situations are depicted. The first two signals correspond to the signal of a stationary object, variously oriented relative to the processing interval. The latter case corresponds to the reception of a signal from a stationary and moving object. This is expressed in modulation of the envelope due to the superposition of the signals. All signals contain M digitized samples. Discreteness is not shown on the diagram. The basic functions P are created in ROM 7, synchronous with the signals S ₁ and also contain M samples. The diagram shows two basis functions of the zero and first orders. The following shows the signal at the output of the interpolator 6, recovered from the spectrum of S _in . We see that this signal is delayed by the duration of the burst signal S ₁ . As indicated earlier, this is due to the integration process in the interpolator 6. The signal S ₁₁ follows the shape of the signal S ₁ , is created by the second memory 4 and is input to the subtractor 5. The last diagram shows the result of the compensation of the signals S ₁₁ and S _in the subtractor 5. We can see that the signals of stationary objects are compensated, since the shape of the signals S ₁₁ and S _in this case is the same. The signals of moving HEs have “high-frequency” spectra and cannot be reconstructed based on the basis functions P ₀ and P ₁ . Therefore, they are not compensated. This reflects the plot for the third signal. As a result, the device has a certain speed characteristic with a narrow notch area near the zero Doppler frequencies. This characteristic is depicted in figa.

Legendre and Chebyshev polynomials and their modifications [6] can be used as basis functions. When processing discretized signals, the so-called polynomials may be more convenient.

Chebyshev. Some of them are given below [7]:

P ₀ = 1

P ₁ = 1-2μ / M-1 = 1-6μ / M-1 + 6 [μ / M-1] · [μ -1 / M-2]

Here M is the number of pulses in the packet.

These sequences are smooth and, therefore, sufficiently low-frequency functions. By increasing the number of basic functions, it is possible to more accurately transmit the shape of the envelope of the echo signals and thereby improve the degree of compensation of the signals of stationary objects in the conditions of scanning the antenna in azimuth, and hence the selection of moving HEs.

Figure 6 shows diagrams explaining the operation of the interpolating device 6. S ₁ is an input signal containing M digitized samples. P ₀ , P ₁ , P ₂ - basis functions, including quadratic components. C ₀ , C ₁ , C ₂ - spectral coefficients corresponding to the basis of P ₀ ÷ P ₂ . It can be seen that during the signal there is a process of establishing these coefficients. This is explained by the presence of an integrator 9 in the composition of the interpolator 6. The spectral coefficients necessary for the formation of signal S _{in the} value of the spectral coefficients correspond to the moment of the end of signal S ₁ . The diagram shows precisely such situations for the components of the reconstructed signal (C ₀ P ₀ , C ₁ P ₁ , C ₂ P ₂ ). We see that the shape of this signal repeats the basic functions, and the value is determined by the final value of the signals C ₀ , C ₁ and C ₂ . For example, for a symmetric signal S ₁ (second signal in the first diagram), the reconstructed signal component for the basis function C ₁ is equal to zero. This reflects the C ₁ plot, where it can be seen that the result of integration is zero, as is the component C ₁ P ₁ .

After summing the components of the SR, a restored signal S _{c is formed} . We see that it repeats the shape of the input signal. It differs from the latter only in a delay in the signal duration.

The implementation of the claimed device should not be in doubt, because As part of it, elements common to SDS devices were used. These include phase detectors and coherent local oscillators [1, p. 190, 197, 206]. The remaining elements are the main ones used in digital technology. They are universal, designed and meet the uniform requirements of GOST. Such elements include delay devices (matrix or on shift registers), adders and subtractors, multipliers and read-only memory devices [8, p. 489 - adders, p. 490 - shift registers, p. 494-memory, etc. , [9].

Regarding the use of the claimed device, it should be noted that it is single-quadrature. In this regard, its use in SDC systems with two quadratures should be carried out in a known manner, i.e. SDS system should be included in each quadrature [5, p.255]. In Fig.7 shows a diagram of the inclusion of the selection device during the quadrature processing. We see that the device has another similar system for selecting moving targets. Its peculiarity and difference is that in it the coherent oscillation at the phase detector 2 is shifted by π / 2. This is ensured by the inclusion in the circuit of the reference oscillation of the phase shifter 12. Elements circled by a dotted line in this diagram include devices 4, 4, 7, 6 and are omitted for simplification.

It seems necessary to dwell on the work of the first memory 4. Earlier it was indicated that its implementation is possible on the basis of matrix memory or shift registers. Let us illustrate, for example, the case of using shift registers. On figa shows a diagram of such a memory. It contains M-1 series-connected shift registers R _g1 . The output of the first and all subsequent registers R _g1 connected registers R _g2 . The registers R _g1 and R _{g2 are} controlled by periodic pulses of U ₁ and U ₂ multiple frequencies. In this case, pulses of a higher repetition rate are used to control R _g2 . The number of shifts in the register R _g1 is determined by the number of range elements and can reach thousands (several thousand) units. For this purpose, for example, you can use the chip 505RU6. In the registers R _{g2, the} number of shifts is small and is determined by the number of pulses in the packet (M). For this purpose, you can use, for example, chip 153IR31. The output signal is removed from the registers R _g2 . In this case, the tap number for the output signal monotonically increases with a constant step - 1, 2, 3, length 2, 4, 6 ... etc. The operation of such a memory device is inherently close to the operation of the analog delay line. Therefore, a situation arises in it when an echo signal from one VO appears at all outputs R _g1 . On Fig depicts the situation when after R _g1 at some point in time there is a signal IN from the i- ^th element of the range (S _1i , S _2i , ... S _Mi ). At another point in time, a BO signal from the jth range element can be observed. These signals are perceived by the registers R _g2 and delayed by them. As a result of this, the signals at the output of the memory appear sequentially in time. First, the signal from the input R _g1 , then from the output of the first R _g2 , the output of the second R _g2 , etc. and the last M signal is from the output of the (M-1) th R _g2 . The time interval Δ t ₁ with which these signals follow is determined by the pulse repetition period and the delay discrete in the registers R _g2 . It should be noted that the interval Δ t ₁ determines the frequency of the address signals 14 introduced earlier in the description of the claimed device. Thus, using the claimed device control pulses are synchronous and differ only in the repetition period. The methods for generating such pulses are well known and widely used in radar and in television technology, for example [10].

Thus, the use of known elements of a unified elemental base ensures the implementation of all nodes used in the claimed device, and therefore, the implementation of the claimed device as a whole.

The use of the claimed device is associated with certain advantages. So, in contrast to the known SDC devices, and in particular, SDC devices with feedback, the use of the claimed device is not associated with limitations and compromises due to the presence of transients. As a result of this, an improvement in the uniformity of the frequency (speed) characteristic, and hence the probability of detecting signals, does not lead to loss of signal-to-noise ratio in the claimed device. Moreover, the use of the claimed device allows to increase the signal-to-noise ratio, which for the conditions specified in [4], will be 2 dB.

When using the inventive device, it becomes possible to reduce the influence of azimuthal scanning of the radiation pattern on the efficiency of the SDS. The use of a larger number of basic functions in the interpolating filter will allow more accurately reproducing signals in it, the envelope of which changes when the antenna is scanned in azimuth. With an increase in the accuracy of reproduction, the degree of compensation of the signals of stationary objects increases, and hence the effectiveness of the SDS in terms of detecting signals against interference, sub-visibility. Estimates [1] show that the elimination of the influence of changes in the envelope of the signal when scanning the antenna on the reduction of compensation in the CDS, and this is what is achieved in the claimed device, allows to increase the compensation of signals of stationary objects by 17 ÷ 36 dB.

LITERATURE

1. P.A. Bakulev "Radar of moving targets", Moscow, Sov. Radio, 1964.

2. Guide to radar, p / ed. M. Skolnik, vol. 3, Moscow, Sov. Radio, 1979.

3. Grisetti R., Santa M., Kirkpatrick J. "Influence of fluctuations and the review of space on the suppression of reflections from local objects in radars with indicators of moving targets", Questions of radar technology, No. 3, 1956.

4. Express information "Radar, television, radio communications", No. 26, July 1977, VINITI, 5028 PTP.

5. L.A. Vainshtein, V.D. Zubakov "Isolation of signals against the background of random interference, Moscow, Sov. Radio, 1960.

6. VL Goncharov "Interpolation and approximation of functions", Moscow, Sov. Radio, 1960.

7. VL Danilov and others. "Mathematical analysis", Moscow, State. ed. Phys.-Math. literature, 1961.

8. "Integrated circuits", Handbook ed. B.V. Tarabrina, Moscow, Radio and Communications, 1984.

9. Yu.M. Kutyrkin et al. "Foreign integrated circuits of widespread use." Directory. Moscow, Energoatomizdat, 1984 10. M. G. Garb, “Synchronization in Television,” Moscow, Radio and Communications, 1982.

Claims (1)

A moving target signal selection device comprising a coherent local oscillator, a phase detector, an analog-to-digital converter, a first storage unit, a second storage unit and a subtracting unit, a second input of the phase detector is a signal input from a radar receiver, characterized in that, in order to increase the probability of detecting signals of moving targets, an interpolating unit and a permanent storage unit are introduced, while the output of the first storage unit is connected to the first input of the interpolator unit, m other inputs of which are connected to m outputs of a permanent storage unit, and the output - to the second input of a subtracting unit, the output of which is the output of a moving target selection device, and the other inputs of the analog-to-digital converter, memory units and interpolating unit are inputs of control pulses .

Images