RU2384860C2 - Method of detecting people and moving objects behind barrier and device for realising said method - Google Patents

Abstract

FIELD: physics, navigation.

SUBSTANCE: method and device for realising said method are designed for detecting and determining location of stationary people and identifying a moving object behind a barrier. The method involves emission of a probing signal towards the barrier, reception of the reflected signal, its subtraction from the signal received after a given compensation period, and thresholding the signal. The reflected signals are received and processed in several parallel identical channels. Before recording the received signal in each channel, selection is done according to range and detected moving objects are located using signal thresholding results in all channels. Also signals accumulate in each channel after selection according to range. The spectrum of the accumulated signal sequence in the frequency band corresponding to human respiratory frequency is calculated. The obtained spectrum undergoes thresholding for each range channel on which presence of frequencies corresponding to human respiratory frequency is determined and location of detected stationary people is determined using the spectrum thresholding results in all channels. The proposed method is realised using a device made in a defined way.

EFFECT: emission of a continuous monochromatic signal towards the search area and corresponding processing of the reflected signal enables identification of a moving object with a living person.

9 cl, 1 dwg

Description

The invention relates to radar and can be used to detect and determine the location of motionless people who are behind the barrier, as well as to detect, determine the location and identification of a moving object located behind the barrier.

An obstruction is an optically opaque medium that transmits radio waves. This can be, for example, a wall of wood, brick or concrete.

The technical solution described in the article by Bugaev A.S. et al. Detection and remote diagnostics of people behind obstacles using radar tools. J. "Radio engineer", 2003, No. 7. It is a modified sensor type "Raskan" (Patent RU 2121671), designed to detect and remotely diagnose people behind obstacles and contains a transmitter, spatially combined transmit and receive antennas, an amplitude detector, a bandpass filter, an amplifier, an interface unit, and a computer. A continuous monochromatic signal is emitted into space. The signal reflected from the human body and surrounding objects is received by the antenna, demodulated, filtered out in the band of the breathing signal, amplified, sampled, and fed to the computer. The measurement results are displayed on the screen of the interface unit in real time. The device responds to changes in the amplitude of the received signal caused by a person’s breathing and heartbeat.

This device is not able to select objects in range, which does not allow to determine their location and to identify moving objects, it also does not allow to automate the measurement process.

A known method and device for detecting moving objects behind a wall, selected as a prototype (patent US 6466155). The device contains a radar, which includes a transmitter and a receiver, connected by means of a switching device, respectively, to the input and output of the antenna array of transceiver antenna pairs, serially connected analog-to-digital converter (ADC), background subtraction unit, radar image forming unit, signal processor, that implements the algorithms for the operation of background subtraction units and the formation of a radar image, a display, as well as a random access memory (RAM) connected to the bl com radar imaging. Transceiver antenna pairs are connected to the radar in series using a switching device. Each of them radiates into space and receives a reflected pulse signal. The signal is converted to digital form, the signal reflected from the wall is subtracted from it. For all antenna pairs, a radar image is formed, it is recorded in RAM and subtracted from the image obtained in the next sensing step. By the presence of residual signals using the threshold processing algorithm, which is embedded in the signal processor, it is concluded that moving objects are detected. Using the signal processor, the location of the targets is determined. Further on the display, a motion map is displayed, which shows the places where activity was detected.

This device is not able to detect motionless people and to identify moving objects.

The technical task of this invention is to provide a method and device for its implementation, capable of automatically detecting and determining the location of stationary people behind an obstacle, as well as automatically detecting, determining a location and identifying a moving object, i.e. establish whether he is a living person or an inanimate object - a robot, a cart, etc.

The problem is solved in a method for detecting people and moving objects behind an obstacle, which consists in emitting an ultra-wideband probing signal in the direction of the obstacle, receiving the reflected signal, storing it, subtracting it from the reflected signal received after a predetermined compensation period, and then threshold processing the signal , in which according to the invention, the reception and processing of reflected signals is carried out in parallel in several identical channels, before storing the received reflected signal and each of its channel selection is performed in range and threshold the signal processing results in all channels determine the location of detected moving objects.

Additionally, in each channel, after selection by range, the signals are accumulated, the spectrum of the accumulated signal sequence is calculated in the frequency band corresponding to the person’s respiration rate, then the spectrum is threshold processed for each range channel, which determines the presence of frequencies corresponding to the person’s breathing and the threshold spectrum processing in all channels determine the location of the detected still people.

In addition, a continuous monochromatic signal is emitted in the direction of the search zone, a reflected signal is received, detected, its spectrum is determined, then a threshold processing of the obtained spectrum is carried out, by which the presence of frequencies corresponding to human breathing is determined, and thereby the detected moving object is identified with a living person.

The task is also achieved in a device for detecting people and moving objects behind an obstacle, containing an antenna unit connected to the output of a transmitter of ultra-wideband sounding signals, a receiving unit, the output of which is connected to the input of the processing channel, and a data display unit, while the processing channel contains an analog a digital converter included at the input of the processing channel, and a random-access memory block and an inter-period compensation block, in which according to the invention a receiving The th block is multi-channel, additional processing channels are introduced that are identical to the first, the inputs of which are connected respectively to the outputs of the receiving block, and the location unit, the inputs of which are connected respectively to the outputs of the processing channels, a range selector is inserted into each processing channel, connected between the analog output digital converter and the input of the RAM block, and the threshold signal processing block, the input of which is connected to the output of the inter-period compensation block, and the output is the first output of the corresponding processing channel.

The task is also achieved by the fact that in each processing channel a series-connected block of accumulation of digital samples, a low-pass filter, a block for calculating the signal spectrum, a block for threshold processing of the signal spectrum are introduced, while the input of the block of accumulation of digital samples is connected to the second output of the range selector, and the output of the threshold signal spectrum processing unit is the second output of the corresponding processing channel.

The transmitter of ultra-wideband probe signals can be made in the form of at least one generator of video pulses of nanosecond or subnanosecond duration, while the receiving unit can be made in the form of a multi-channel stroboscopic receiver.

In this case, the antenna unit consists of at least one transmitting antenna and several receiving antennas.

The transmitter of ultra-wideband probing signals can also be made in the form of a frequency-modulated signal generator with a step change in frequency, while the receiving unit can be made in the form of a multi-channel superheterodyne receiver.

The task is also achieved by the fact that the device is additionally introduced in series with a continuous monochromatic microwave signal generator, an antenna unit consisting of a receiving and transmitting antenna, a detector, an additional analog-to-digital converter, an additional low-pass filter, an additional signal spectrum calculation unit, and an additional unit threshold processing, the output of which is connected to an additional input of the location unit.

The introduction of additional operations, namely, parallel processing of the reflected signal received by all the receiving antennas, and range selection allow us to determine the location of people and moving objects beyond the barrier.

Performing operations of accumulating a sequence of signal samples for each range channel, calculating the spectrum of the accumulated sequence of samples, threshold processing of the obtained spectrum in order to analyze it for the presence of frequencies corresponding to human breathing, allows detecting the presence of motionless people beyond the barrier.

Additional generation and emission of a continuous monochromatic sounding signal, reception of the reflected signal, isolation of the oscillation modulating the amplitude or phase of the received reflected signal, discretization of the received low-frequency signal after detection, calculation of the signal spectrum, threshold processing of the obtained spectrum in order to analyze it for the presence of frequencies corresponding to human breathing , allow the identification of a moving object, namely to determine whether a moving object is alive or inanimate

The drawing shows a structural diagram of a device for detecting moving objects and people over an obstacle.

A device for detecting moving objects and people behind an obstacle contains an antenna unit 1 connected respectively to the output of a transmitter 2 of ultra-wideband sounding signals, a receiving unit 3 made of multi-channel, n outputs of which are connected respectively to the inputs of n identical processing channels 4-1, 4-2, ..., 4-n.

Each processing channel 4 contains an analog-to-digital converter (ADC) 5, included at the input of the processing channel, a range selector 6, which is a range-gated block taking into account the time delay of the signal reflected from the target for each range channel, a random access memory block ( BOP) 7, block of inter-period compensation (BCH) 8, block of threshold signal processing 9, block of accumulation of digital samples 10, low-pass filter (LPF) 11 with a passband from 0 to 1 Hz, block for calculating the spectrum of signal 12, pore processing block signal spectrum 13.

The device also includes a positioning unit 14 and a data display unit 15.

The transmitter 2 of ultra-wideband sounding signals can be made in the form of a generator of video pulses of nanosecond or subnanosecond duration or in the form of a generator of a frequency-modulated signal with a step change in frequency. To increase the power of the probe signal, there may be several generators. In this case, the multi-channel receiving unit 3 can be made, respectively, either as a multi-channel stroboscopic receiver or a multi-channel superheterodyne receiver.

The antenna unit 1 consists of one or more, in the case of using several generators, transmitting antennas and several receiving antennas. To determine the location of the targets, the number of receiving antennas is selected corresponding to the required number of detected targets or greater (to eliminate the ambiguity of the location). The number n of processing channels 4 corresponds to the number of receiving antennas.

The device also contains an additional channel, consisting of a series-connected generator 16 of a continuous monochromatic microwave signal, an antenna unit 17 containing a receiving and transmitting antenna, a detector 18, while in the case of using a phase detector, the signal from the generator 16, an additional ADC 19, additional low-pass filter 20 with a passband from 0 to 1 Hz, an additional block for calculating the spectrum of signal 21 and an additional block for threshold processing of the signal spectrum 22, the output of which is connected to an additional input of the location unit 14.

The device operates as follows.

A video pulse signal or an FM signal with a step change in frequency, generated by the transmitter 2, is emitted by the transmitting antennas of the antenna unit 1. The signal reflected from various objects enters the receiving antennas of the antenna unit 1, from the output of each receiving antenna the signal goes to the input of the corresponding channel of the receiving unit 3, the number of channels of which corresponds to the number of receiving antennas. In the case of a video-pulse sounding signal, a stroboscopic receiver is used, which produces a large-scale increase in the signal duration, i.e. translates it into the region of lower frequencies, which facilitates the subsequent analog-to-digital conversion. In this case, all frequency components of the signal are saved. An FM probe signal uses a multichannel superheterodyne receiver, which translates the signal with a step change in frequency to the corresponding intermediate frequency, which facilitates the subsequent analog-to-digital conversion to ADC 5 in each processing channel 4. Then, in each of n processing channels 4, the signal goes to the input of the selector in range 6, where depending on the time of arrival, the range channel in which the object from which the signal is reflected is located is assigned to it. This procedure is carried out by gating the range selector 6 taking into account the time delay of the signal reflected from the target for each range channel.

To detect motionless people from the spectrum of the received signal, it is necessary to select components in the range from 0 to 1 Hz with an acceptable resolution of at least 0.1 Hz for this purpose, due to fluctuations (movements) of the human chest with a breathing frequency of 0.2 ... 0.8 Hz The signal of heart fluctuations is not allocated because its power is an order of magnitude less due to the fact that the area of the oscillating heart is much smaller than the area of the oscillating chest of a person. For this, in each processing channel 4 for each range channel, digital samples of the signal received from the output of the ADC 5 and then from the first output of the selector in range 6 are sequentially accumulated for at least 10 seconds in the accumulation unit of digital samples 10, which will provide spectral resolution no worse than 0.1 Hz. The accumulated sequence passes through the low-pass filter 11 with a passband from 0 to 1 Hz, to the signal spectrum calculating unit 12, in which its spectrum is determined using the discrete Fourier transform (DFT) or fast Fourier transform (FFT) operation. Detection of motionless people is carried out in the block of threshold processing of the spectrum of the signal 13 by analyzing the spectrum of the sequence of samples accumulated for each range channel obtained from the output of the spectrum calculation unit 12 for the presence of frequencies corresponding to human breathing. This analysis is performed by threshold processing of the spectrum of the sequence of signal samples accumulated for each channel: the value of the spectral samples in each range channel is compared with a threshold whose value is selected based on the acceptable level of probability of false alarm. If the threshold is exceeded, a decision is made to detect a stationary person in the channel under consideration.

As a result, for each channel for receiving and processing a signal corresponding to its receiving antenna, an array of data “range channel - human presence” will be obtained. Such arrays are the initial data for the subsequent determination of the location of a person.

The detection of a moving object is carried out using the algorithm of inter-period compensation (NPK).

At the first stage, the reflected signal received by each receiving antenna of the antenna unit 1, from the second output of the range selector 6 is recorded in memory (BOP) 7.

The second stage corresponds to a probe pulse radiated into space through a time interval equal to the compensation period. The value of this period is set based on the possible speed of movement of the objects (0.3 ... 3 m / s), so that during the period of the NPC the object is likely to fall into another range channel, which is necessary for the algorithm to work correctly. At this stage, the received and digitized signal in each processing channel 4 is also recorded in the BOP memory 7. Next, the signal is fed to the input of the PLC 8, where the signal obtained at the first stage of the algorithm operation and stored in the BOP 7 memory is subtracted. Subtraction is performed for each channel range. The result of the subtraction is a radar image of the room under study, from which information about stationary objects is removed, and the movement is nothing but residual signals. The detection of these signals is carried out automatically in the threshold processing unit 9 using the threshold processing algorithm: the value of the samples of the difference signal coming from the output of the PLC 8 for each range channel is compared with a threshold whose value is selected based on the acceptable level of probability of false alarm. If the threshold is exceeded, a decision is made to detect the range of a moving object in the channel under consideration.

As a result, for each channel corresponding to its receiving antenna, an array of data “range channel - the presence of a moving object” will be obtained. Such arrays are the source data for the subsequent location of the object.

The signals from the outputs of each processing channel 4, i.e. from the outputs of the threshold signal processing units 9 and the threshold spectrum processing units of the signal 13 in each channel, are fed to the inputs of the location determination unit 14, where the coordinates of the detected stationary people and moving objects are calculated. The location is determined, for example, by the total rangefinder method, which has a software implementation stored in the memory of the location unit 14. So, knowing the range channel in which the target is detected, i.e. the total propagation time of the signal from the transmitting antenna to the target and from the target to the receiving antenna, for all receiving positions, the total distance traveled by the signal from the transmitting antenna to the target and from the target to the receiving antenna is calculated. For each receiving position, an equation of the position line — an ellipse — is compiled and, by solving a system of such equations, the intersection points of all position lines are found. To combat the appearance of false targets, redundancy can be introduced into the device: the number of receiving antennas is chosen to be larger than the required number of detected targets. The true target will correspond to some area in which the intersection points of all position lines are concentrated, while the false targets appear as a result of the intersection of a smaller number of position lines and their intersection points are dispersed.

When a moving object is detected, its location and identification are made, while observing it, i.e. detection and positioning are ongoing. To ensure joint operation, the parts of the radar that work with pulsed and continuous signals are separated from each other by a screen, and a notch filter that suppresses a continuous signal can be introduced into each channel of the signal processing of the pulsed part of the radar.

To identify a moving object, an additional channel is used, consisting of blocks 16-22.

The continuous monochromatic microwave signal generator 16 generates a signal with a frequency of 1.5 to 3 GHz, which is fed to the input of the transmitting antenna of the antenna unit 17, from where it is radiated into space. The signal reflected from various objects falls on the receiving antenna of the antenna unit 17. The amplitude or phase detector 18 detects a vibration modulating the amplitude or phase of the received signal, respectively, by the frequency of which it is possible to judge the presence of a breathing person. For subsequent processing, the resulting low-frequency signal (LF) is sampled using ADC 19, and from its output it is fed to the input of the low-pass filter 20 with a bandwidth from 0 to 1 Hz and then to the input of an additional block for calculating the spectrum of signal 21, where, using the DFT or FFT operation the spectrum of the obtained modulating LF signal is determined in the range from 0 to 1 Hz with a resolution of no worse than 0.1 Hz. The conclusion about the nature of the moving object is taken in the threshold processing unit of the spectrum 22 by analyzing the spectrum obtained from the output of the spectrum spectrum calculation unit of the signal 21, for the presence of frequencies corresponding to human breathing. This analysis is carried out by threshold processing of the spectrum of the low-frequency signal obtained after the detection of the received continuous monochromatic signal. The threshold processing algorithm consists in comparing the value of each frequency reference with a threshold whose value is selected based on the acceptable level of probability of false alarm. In this case, spectral readings corresponding to the respiratory rate of a stationary person, if one is detected, are not involved in the processing. Information about the nature of the moving object from the output of the threshold processing unit 22 is supplied to the additional input of the location determining unit 14 in order to further display whether the moving object is a living person or something else.

When the coordinates of the targets are determined, on the screen of the data display unit 15, an image of the studied area is displayed with the coordinate grid deposited on it and marks that correspond to the detected targets. In this case, marks corresponding to moving objects are displayed in colors different from those corresponding to living people.

Claims (9)

1. A method for detecting people and moving objects behind an obstacle, which consists in emitting an ultra-wideband probing signal in the direction of the obstacle, receiving the reflected signal, storing it, subtracting it from the reflected signal received after a predetermined compensation period, and then performing threshold processing of the signal, which differs the fact that the reception and processing of the reflected signals is carried out in parallel in several identical channels, before storing the received reflected signal in each channel, it is selected In terms of range and according to the results of threshold signal processing, moving objects are detected in all range channels, in addition, in parallel, after selection in range, signals are accumulated in each range channel, the spectrum of the accumulated signal sequence is calculated in the frequency band corresponding to the human respiratory rate, and then the threshold processing the obtained spectrum for each range channel, which determines the presence of frequencies corresponding to human breathing, and the threshold value spectrum processing in all range channels detect motionless people, and according to the results of threshold processing, the coordinates of motionless people and moving objects are calculated in the location unit, then, when the coordinates of the targets are determined, an image of the studied area with a coordinate grid applied to it is displayed on the screen of the data display unit and marks that correspond to detected targets, while marks corresponding to moving objects and living people are displayed different from Ruga flowers. 2. The method according to claim 1, characterized in that, in addition, a continuous monochromatic signal is emitted in the direction of the search zone, a reflected signal is received, detected, its spectrum is determined, then a threshold processing of the obtained spectrum is carried out, by which the presence of frequencies corresponding to human breathing is determined at This information about the nature of the moving object is sent to the location unit to determine whether the moving object is a living person or other moving object, while the spectral readings s corresponding to the fixed frequency of respiration person if such is detected, in processing is not involved. 3. A device for detecting people and moving objects over an obstacle, comprising an antenna unit connected to the output of a transmitter of ultra-wideband sounding signals, a receiving unit, the output of which is connected to the input of the processing channel, and a data display unit, while the processing channel contains an analog-to-digital converter, included at the input of the processing channel, and a series-connected block of random access memory and an inter-period compensation unit, characterized in that the receiving unit is multi-channel, additional eel processing channels identical to the first, the inputs of which are connected respectively to the outputs of the receiving unit, and the location unit, the inputs of which are connected respectively to the outputs of the processing channels, a range selector is inserted into each processing channel, connected between the output of the analog-to-digital converter and the input of the operational unit memory, and a threshold signal processing unit, the input of which is connected to the output of the inter-period compensation block, and the output is the first output of the corresponding processing channel and, while in each processing channel, a series-connected block of accumulation of digital samples, a low-pass filter, a block for calculating the signal spectrum, a block for threshold processing of the signal spectrum are introduced, while the input of the block of accumulation of digital samples is connected to the second output of the range selector, and the output of the threshold block signal spectrum processing is the second output of the corresponding processing channel. 4. The device according to claim 3, characterized in that the transmitter of ultra-wideband sounding signals is made in the form of at least one generator of video pulses of nanosecond or subnanosecond duration. 5. The device according to claim 4, characterized in that the antenna unit consists of at least one transmitting antenna and several receiving antennas. 6. The device according to claim 4, characterized in that the receiving unit is made in the form of a multi-channel stroboscopic receiver. 7. The device according to claim 3, characterized in that the transmitter of ultra-wideband probing signals is made in the form of a frequency-modulated signal generator with a step change in frequency. 8. The device according to claim 7, characterized in that the receiving unit is made in the form of a multi-channel superheterodyne receiver. 9. The device according to claim 3, characterized in that a continuous monochromatic microwave signal generator, an antenna unit consisting of a receiving and transmitting antenna, a detector, an additional analog-to-digital converter, an additional low-pass filter, an additional spectrum calculating unit are introduced into it a signal and an additional block of threshold processing of the spectrum, the output of which is connected to an additional input of the location unit.