RU2768227C1 - Method of operational and technical protection of objects and borders - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: invention relates to the field of fibre-optic metrology and technical means of security (TMS) of important objects and boundaries based on detectors, controllers or sensors using a fibre-optic cable (FOC) as a sensitive element, and as measuring devices - reflectometers and methods of obtaining vibroacoustic measurement information. Method of operational and technical protection of boundaries of objects and boundaries, based on use of fibre-optic cables as a distributed sensitive element and reflectometers, which form the basis for constructing fibre-optic security equipment (FOSE), and generating signals by them, by which the fact of violation of protection zone is identified, from them, identifying the signs associated with the detection of the intruder, comparing their values with the established value in the form of comparison thresholds and determining the moment of issuing an alarm signal from the comparison results, characterized by that signals generated by fibre-optic security equipment, are subjected to additional processing using algorithms of structural-algorithmic transformations (SAT-2) and then adaptive nonlinear filtering (ANF), at that, first, the low-frequency component (trend) is eliminated or reduced in the signal generated by the controller, sensor or detector, and then an adaptive nonlinear filtering algorithm is used.

EFFECT: high signalling reliability of security equipment and obtaining additional information contained in signals generated by reflectometers.

1 cl, 5 dwg

Description

The invention relates to the field of fiber-optic metrology and technical means of protection (TSO) of important objects and boundaries based on detectors, sensors or sensors using a fiber-optic cable (FOC) as a sensitive element, and reflectometers as measuring tools and methods for obtaining vibroacoustic measurement information related to the Raman effect.

It can be used to improve the signaling reliability of the existing ones and those being developed under various operating conditions, including those under the influence of interference of various origins. The technical result of using the invention is to reduce the likelihood of false positives leading to the formation of false alarms.

Known "Method of protecting the perimeter of the object" ([1], RU 2263968, MPI ⁷ G08B 13/02). It provides protection for all types of fences from climbing, destruction and undermining. This method involves the installation of a fence, the installation of a sensitive element (SE) on the flank of an extended fence, the determination of the amplitude-frequency characteristics of the signal received from the SE, which are typical for the intruder to overcome the fence by climbing, destroying or digging, processing the current value of the signal from the SE and as it receives previously defined characteristics of the signal - the issuance of an alarm message.

The disadvantages of this method of protection include:

- high costs for the installation and configuration of equipment, caused by the complexity of the work performed to ensure the protection of important objects using engineering structures of the barrage type;

- high costs if it is necessary to change the boundaries of the protected area;

- the lack of a function to determine the method of overcoming the fence by the intruder;

- reducing the probability of detection and increasing the probability of false alarms under the influence of climatic factors: wind loads, icing, snow drifts and ambient temperature.

Also known is the "Method of protecting the perimeter of the object" ([2], patent 2473970 C1, MPI ⁷ G08B 13/02, published on January 27, 2013, bull. No. 3), which consists in installing a sensitive element (SE) of the detector on section of the fence, set the signs of the signal that determine the overcoming of the fence by the intruder, fix the received values of signals from the sensitive elements and extract from them the signs characteristic of the intruder, compare the value with their set value and, based on the results of the comparison, determine the moment of issuing the alarm signal, which differs in that it is distinguished constructively the finished element of the fence, based on its spatial parameters, determine the zones of influence of the intruder on it in the process of overcoming the climbing method, form independent sensitivity zones on the fence canvas, determine the sequence of appearance of signals from the generated sensitivity zones and the limit values of the time between the signals of this sequence, obtained receive signals from the sensitivity zones and compare the obtained order of signal formation with the previously determined sequence, as well as the time intervals between signals from independent sensitivity zones, in case the received order coincides with a certain sequence and the time between signals from independent sensitivity zones is found in initially defined time intervals, issue an alarm.

The method [2] according to claim 1 also differs in that it determines the zones of influence of the intruder on the fence in the process of overcoming the method of controlling the destruction of an engineering structure of the barrier type and establishes the order in which signals appear in the previously defined sensitivity zones.

In this case, a frequency generator is used as a detector with a capacitive SE included in its frequency setting target, which affects the frequency of the generator. A description of the operation of such a detector is given in patents ([3], patent RU 2126173, IPC G08B 13/26, [4], patent RU 2379759, IPC G08B 13/26 and [5], patent RU 2491646, IPC G08C 15/08, G08B 13/26).

The disadvantages of this method of protecting the perimeter include:

the need for additional control of the state of the space surrounding the barrier;

- the need to monitor the state of sensitive elements and detectors and promptly assess the effect of weather conditions on them in order to carry out the necessary adjustment of thresholds, the excess of which leads to the formation of alarms;

- the impossibility of determining the place of penetration of the intruder within the zone of action of one sensitive element;

- insufficient signaling reliability of the detector, to increase which a signaling fence is used, which is not possible when using technical security equipment (TSO) of a non-obstructive type.

Also, among the general shortcomings of the existing methods and technical means of protection, using various sensitive elements (SE) and detectors, based on the signals of which alarm messages are generated, should include:

- focus on the security systems of barrage-type objects, involving the installation of barriers, which is not always possible, primarily due to the need for significant costs and time for their construction, however, in the absence of barriers, the requirements for signaling reliability of detectors are significantly increased, which, with the existing processing system generated signals and received information cannot be provided;

- the impossibility of maintaining the signaling reliability of the SE and detectors at the required level, which is associated with an increase in the number of impacts on the SE and detectors that are perceived by them as interference and lead, as a result, to a decrease in the probability of detecting an intruder and to a reduction in the time intervals between false alarms ( this drawback becomes one of the main ones with artificially formed effects on the SE, detectors, information collection systems and communication channels);

- the presence of situations in which there are limited possibilities for detecting the facts of intruder penetration into protected areas and territories when using SE and detectors of the same type, the lack of sufficiently effective methods for combining SE signals and detectors using different physical principles for detecting violations;

- the lack of the ability to quickly monitor the correct functioning of the detectors, evaluate the reliability indicators of the data and signals generated by them and ensure self-adjustment of the detectors, information collection systems and decision-making devices for changed operating conditions, determined by a change in the state of the environment under the influence of various weather influences and the resulting interference, which, ultimately, leads to an increase in the probability of generating false alarm messages (only in a number of cases, self-control is used in the form of self-testing of the SE and detectors, which are carried out after a specified time or operation cycles [4]);

- insufficiently effective use of modern digital signal processing methods that provide an increase in the signaling reliability of SE and detectors (data and signal processing, if used, then after collecting information from detectors via communication channels (wired or wireless), which do not have the required noise immunity indicators, from - which increases the uncertainty in decision-making).

, характеризующей мощность внешнего возмущения,The closest analogue is the "Method of operational and technical protection of the boundaries of objects and borders", patent RU No. 2705770, application No. 2018119750 with priority dated May 29, 2018, publ. November 11, 2019, bul. No. 32 [6]. It consists in the fact that the sensitive elements (SE) are installed, which form the basis for the construction of technical means of protection (TSO), the most informative signs of the signal generated by the sensitive elements (SE) of the detectors are selected, by which the fact of overcoming the security zone by the intruder is identified, the received signal values are recorded from the sensitive elements and extract from them the signs characteristic of the intruder, compare their values with the set value and, based on the results of the comparison, determine the moment of issuing an alarm signal, characterized in that the signals generated by the detector are subjected to additional processing using adaptive nonlinear filtering algorithms (ANF), as a result which determine indirect estimates of the variance of the shaping noise

characterizing the power of external disturbance,

, где k - индекс наблюдаемого временного отсчета, а (k-1) - индекс предшествующего временного отсчета измерений, которые используют в качестве информационного признака при формировании сигнала оповещения о нарушении, формируемого на основе сравнения полученных оценок дисперсии формирующего шума

с выбранными порогами, u₁, u₂ и u₃, где u₁ - порог, ниже которого вычисленные значения оценок дисперсии формирующего шума, рассматривают как помеховый фон, не связанный с нарушением охраняемых рубежей объектов и границ, при этом нахождение сформированного сигнала, характеризующего изменение оценок дисперсии формирующего шума между порогами u₁ и u₂ рассматривают, как помехи, которые связаны с повышенным влиянием внешних воздействий на чувствительный элемент извещателей, но не свидетельствуют о нарушении зоны охраны, и только факт превышения сформированным сигналом оценок дисперсии формирующего шума порога u₃ используют для формирования сигнала тревоги, при этом для повышения вероятности обнаружения нарушителя по вычисленным оценкам дисперсии формирующего шума используют различные алгоритмы компенсации помех в сформированном сигнале, характеризующем изменение во времени оценок дисперсии формирующего шума.which is an indirect information sign of the intruder's impulse action on the detector's sensitive element, leading to a change in its output signal, then the calculated indirect estimates of the shaping noise dispersion are used

, where k is the index of the observed time reference, and (k-1) is the index of the previous time reference of the measurements that are used as an information feature when generating a violation alert signal generated based on a comparison of the obtained estimates of the variance of the forming noise

with the selected thresholds, u ₁ , u ₂ and u ₃ , where u ₁ is the threshold, below which the calculated values of the estimates of the variance of the shaping noise are considered as an interference background that is not associated with a violation of the protected boundaries of objects and boundaries, while finding the generated signal characterizing a change in the estimates of the variance of the shaping noise between the thresholds u ₁ and u ₂ is considered as interference that is associated with an increased influence of external influences on the sensitive element of the detectors, but does not indicate a violation of the protection zone, and only the fact that the generated signal exceeds the variance estimates of the shaping noise threshold u ₃ is used to generate an alarm signal, while to increase the probability of detecting an intruder according to the calculated estimates of the variance of the shaping noise, various algorithms are used to compensate for interference in the generated signal, which characterizes the change in time of the estimates of the variance of the shaping noise.

с выбранными порогами, u₁, u₂ и u₃, как этого показано на фиг. 1 (А, Б, В, Г) на графиках, приведенных внизу под сигналами, формируемыми датчиками, сенсорами и извещателями на основе распределенного ЧЭ в виде ВОК, позволяет повысить сигнализационную надежность датчиков, сенсоров и извещателей. При этом под сигнализационной надежностью понимается одновременное выполнение предъявляемых требований, относящихся к вероятности обнаружения нарушителя Р_н и к интервалам времени между ложными тревогами. При этом сигнал тревоги формируют при превышении значений дисперсий формирующего шума

последнего порога u₃.Illustrations showing the technical effect obtained when using the method [1] are shown in Fig. 1 (A, B, C, D). In the same place, on the left, there are various methods (A, B, C, D) that the intruder uses to overcome the security zone (ZO), and on the right - the generated signal u(t), which is obtained in existing practice when using the FOC as distributed sensing element (SE). In this case, the selection of a useful signal against the background of interference is carried out using the set threshold of restrictions - U _og . The problem to be solved is what value of the generated signal u(t) should correspond to U _og . Not only does this choice, as follows from the graphs shown in Figs. 1 do not contribute to the peculiarities of the generated signal u(t), which, in the general case, is a non-stationary random process with time-varying averaged values representing a trend. So it also depends on the method of overcoming 30, which is used by the offender. If the set threshold U _limit for the case of a single intruder, overcoming 30, on foot, as shown in Fig. 1(A, B) may be acceptable, it will lead to a condition that is defined as the generation of false alarms, in the event of crossing 30 by crawling or as part of a group of violators of FIG. 1(C, D). Therefore, the signaling reliability of such a variant of the operation of technical means of protection (TSO) will not meet modern requirements and threat models. Using as a result of applying the method [1] the obtained estimates of the variance of the shaping noise

with selected thresholds, u ₁ , u ₂ and u ₃ , as shown in FIG. 1 (A, B, C, D) in the graphs shown below under the signals generated by sensors, sensors and detectors based on a distributed SE in the form of a FOC, allows you to increase the signaling reliability of sensors, sensors and detectors. At the same time, signaling reliability is understood as the simultaneous fulfillment of the requirements related to the probability of detecting an intruder P _n and to the time intervals between false alarms. In this case, an alarm signal is generated when the values of the dispersions of the shaping noise are exceeded

the last threshold u ₃ .

оказываются малочувствительными к изменению медленноменяющейся составляющей нестационарного случайного процесса, представляющей собой тренд u_T(t). Однако при этом наблюдается повышенная чувствительность к импульсным помехам, в результате чего растет количество откликов, которые превышают выставленный порог ограничения u₂ (фиг. 1 (А, Б, В, Г)). Их наличие увеличивает риск формирования техническим средством охраны (ТСО) сигнала «ложной» тревоги.From those shown in Figs. 1 of the illustrations it follows that the main technical positive effect from the application of the method [1] is that the values of the variances of the shaping noise

turn out to be insensitive to a change in the slowly changing component of a non-stationary random process, which is a trend u _T (t). However, in this case, an increased sensitivity to impulse noise is observed, as a result of which the number of responses that exceed the set limit threshold u ₂ increases (Fig. 1 (A, B, C, D)). Their presence increases the risk of generating a “false” alarm signal by a technical security device (TSO).

оказываются наиболее чувствительными по отношению к выбросам нестационарного случайного процесса, к которому относится и сигнал, формируемый техническим средством охраны (ТСО), использующим ВОК в качестве распределенного ЧЭ. При этом выставленные три порога u₁, u₂ и u₃, которые используют для принятия решения о наличии нарушителя, определяют реакцию оценок дисперсии формирующего шума

на шум (u₁), незначительные импульсные помехи (u₁) и наличие нарушителя (u₃) (фиг. 1 (А, Б, В, Г)).Using the method [1] allows you to significantly reduce the influence of the low-frequency component of the signal (trend) (Fig. 1 (A, B, C, D), Fig. 2(A)), formed on the basis of the FOC used as a distributed SE. This is due to the fact that the estimates of the variance of the shaping noise determined as a result of its processing

turn out to be the most sensitive in relation to emissions of a non-stationary random process, which includes the signal generated by the technical means of protection (TSO) using the FOC as a distributed SE. In this case, the set three thresholds u ₁ , u ₂ and u ₃ , which are used to make a decision about the presence of an intruder, determine the reaction of the estimates of the variance of the shaping noise

noise (u ₁ ), slight impulse noise (u ₁ ) and the presence of an intruder (u ₃ ) (Fig. 1 (A, B, C, D)).

Also, the conducted in-depth theoretical and experimental studies have shown that in order to increase the signaling reliability of sensors, sensors and detectors, the sensitive element of which is the FOC.

определять не по формуле (10), используемой в прототипе [6]:This requires a current estimate of the shaping noise variance

determined not by the formula (10) used in the prototype [6]:

where τ _k - is the interval between two successive time readings (t _k and t _k-1 ) measuring the values of the controlled and processed signal generated by the detectors: τ _k =t _k -t _k-1 , k is the number (index) of the time reading ;

- оценка максимальной дисперсии случайных отклонений ε(k) измеренного значения у(k) от прогнозируемой оценки х(k/k-1):

- estimation of the maximum dispersion of random deviations ε(k) of the measured value y(k) from the predicted estimate x(k/k-1):

- минимальная дисперсия случайной помехи, а α и β - весовые коэффициенты рекурсивного фильтра экспоненциального скользящего среднего;where

is the minimum random noise variance, and α and β are the weights of the exponential moving average recursive filter;

- оценка максимальной дисперсии изменений медленно меняющегося систематического смещения ξ(k),

- estimation of the maximum variance of changes in the slowly changing systematic bias ξ(k),

and based on the summation of the obtained values with the weights α and β, which are set for the averaging filter:

This is the new changes that form the basis of the essential characteristics of the invention when using ANF methods. Also, the essential characteristics of the invention lie in the preliminary use of the operations of structural-algorithmic transformations, called in [7] SAP-2 and in the definition of a single fiber-optic technology for improving the signaling reliability of signals generated by reflectometers.

Also, the studies have shown that the use of only one method [6] in the preliminary processing of the signal generated by the TCO may not be sufficient to provide the required indicators of signaling reliability, determined by the probability of detecting an intruder (Р _n ≥ 0.95) and the time interval _{ΔТ lt} between signals "false" alarm ( _{ΔТ lt} ≥ 2000 hours). Therefore, it needs to be combined with other methods.

Also known is the "Method of operational and technical protection of the boundaries of objects and borders", patent RU No. 2674809, application No. 2017124869 with priority dated 12.07.2017, publ. December 13, 2018, bul. No. 35 [7]. In it, as in the method [1], the signaling reliability of sensors, sensors and detectors is also increased by pre-processing the signals generated at their output, with elements of adaptation to various conditions of their use and the artificial intelligence implemented in this case, which contributes to a consistent increase in information features. disturbances in the initial generated physical signals. At the same time, preliminary processing of signals generated by sensors, sensors and detectors is carried out using the conveyor principle based on distributed structural-algorithmic transformations (SAP-i), i=1,2,3,4 - their serial number of implementation. At the same time, the most suitable for the implementation of the proposed idea of sharing methods [6 and 7] is the structural-algorithmic transformation (SAP-2).

The method [7] according to claim 1 differs in that at the second stage of distributed SAP (SAP-2) included in (SAP-i), i=1,2,3,4…, the sequence of pulse signals generated by each of the same type of detectors are inverted and shifted relative to the original by a time Δτ _k equal to half the estimated time T _prk that will be required for the intruder to cross the protected area of the territory or security zone on which the detectors of technical means of protection are installed (Δτ _k ≈ 1/2Т _prk ), which is then summed with the original sequence of pulse signals, the summation results are differentiated for additional suppression of the noise and interference background, as a result of which the signal/noise ratio and signal/(noise + interference) are increased, and the delay Δτ _k , which can change when repeating the above processing operations , including the possibility of parallel processing with different values of Δτ _k corresponding to certain signs of violations that were obtained on the preliminary at the stage of training adaptive and intelligent technical means of protection, on the basis of which an additional information sign is introduced - Δτ _k , which makes it possible to distinguish the fact of crossing the protected area of the territory where detectors of technical means of protection are installed from other interference caused by the use of seismic piezoelectric detectors and geophones by falling branches , stones, snowballs, the passage of vehicles near the protected area and important objects, the influence of rains and showers on radio-beam technical means of protection, an additional information sign Δτ _k and time T formed at the same time _and the impulse signal exceeds one or more threshold levels set for it , including those exhibited simultaneously to identify various event phenomena, are used to classify the alleged violation - single, group, determine the proposed way to overcome the security zone - in transport, running, walking m, crawling, rolling.

The main technical effect of the pre-processing of signals generated by sensors, sensors and detectors, is manifested, as shown in Fig. 2(B) in that the use of the SAP-2 algorithm leads to a decrease in the influence of the trend on the choice of restriction thresholds (U _lim ) and, as a result, to an increase in the energy sign of the selected signal about the violation of the protection zone. As a result, the signaling reliability of the TSO is increased.

The disadvantage of the method [2] is that in its pipeline pre-processing procedure there are no operations that are associated with the implementation of adaptive nonlinear filtering of signals generated by sensors, sensors and detectors. The disadvantage is also associated with the need to take into account the specific features of the signals generated by reflectometers when using the FOC as SE.

In FIG. 2(B) and in FIG. 3 shows various options for the joint use of SAP-2 algorithms and adaptive nonlinear filtering (ANF). In FIG. 2(B) shows the case where the SAP-2 processing algorithm is used after the ANF algorithm. In this case, the processing algorithm based on SAP-2 contributes to additional suppression of the interference background, as a result of which the limit threshold (U _lim ) can be set lower, as a result of which the energy sign of the processed signal related to the fact of violation of the security zone becomes more noticeable. At the same time, a different order of using processing algorithms based on SAP-2 and ANF, when they follow in a written sequence, leads to a more impressive positive technical effect. This is shown in the illustrations shown in Fig. 3 and FIG. 4. Wells providing access to the cable network in Moscow were selected for carrying out field experimental studies. This choice was dictated by the existing frequent cases of cutting and theft of cables. In FIG. 3 and FIG. 4 shows the main operations that an attacker must perform: open hatches No. 1 and No. 2, closing the entrance to the cable well; 2) lower the ladder; and 3) go down yourself. In FIG. 3 and FIG. Figure 4 shows the pre-processed signals generated by a fiber optic cable located next to other cables and used as a SE. In this case, the upper graphical display of the imaginary TCO response corresponds to the results of pre-processing of a distributed sensor, sensor or detector when using the ANF algorithm. A variant of its presentation on the middle graph was obtained for the case when, at first, the signal generated by the reflectometer was processed based on the SAP-2 algorithm, and then the resulting intermediate variant was improved using the ANF algorithm. The bottom graph shown in Fig. 3 and FIG. 4 shows how much the ratios of the estimates of the variances of the shaping noise changed in this case. It indicates that when using the SAP-2 + ANF and ANF algorithms separately, the ratios of the estimates of the variances of the shaping noise increased, on average, by 2 times.

At the same time, impacts became more noticeable, which are identified as interference, and were caused by a tram passing 10 meters from the well (Fig. 4). The distinctive features of signal generation by a reflectometer, which is a measuring instrument, is that it allows you to obtain results in the form of three-dimensional graphics, as shown in Fig. 5(A) and FIG. 5 B). At the same time, three-dimensional graphics are presented in the upper part, where three-dimensional images of the signals generated by reflectometers are shown. The third axis of 3D graphics, going up, is represented by different colors. On the right side of the graphs, there is a correspondence between colors and their corresponding numbers. The 2D plots shown in Fig. 5(A) and FIG. 5(B) at the bottom correspond to a section of three-dimensional graphs, called cognitive ones, which is indicated by a solid light line. As a result, a three-dimensional image is scanned in the form of a two-dimensional graph (Fig. 5(A) and Fig. 5(B) below), which shows the results of pre-processing obtained using the proposed method.

Therefore, the graphs (Fig. 5(A) and Fig. 5(B) below) are clearly visible: along the y-axis, the amplitude of the response from the effect on the SE, which is used as a WOC, and along the abscissa, their temporal sequence. As a result of this, it becomes possible to identify ongoing events: displays of steps are visible that differ in the frequency of the repetition of pulses, as well as the results of the work of the shovel, which appear in the form of a certain irregular sequence of pulses (Fig. 5(A) and Fig. 5(B) below). If there were no processing, then the impulses would be fuzzy, as a result of which the indicator of identification uncertainty would also increase.

The essential characteristics of the proposed invention are that the uncertainty of identification and interpretation of the observed sequence of pulse signals is significantly reduced.

The sequence of operations on material objects, which are represented by discrete values of the measured value y(k) from the predicted estimate x(k/k-1), is determined as:

1) the sequence of impulse signals generated in each time section of the cognitive graph (Fig. 5(A) and Fig. 5(B) at the top) is inverted and shifted relative to the original by a predetermined delay time Δτ _k ;

2) then it is summed with the original sequence of pulse signals, the results of summation are differentiated for additional suppression of noise and interference background, as a result of which the signal-to-noise ratio and signal/(noise + interference) are increased (with a delay Δτ _k , which can change with repetition given processing operations, including the possibility of parallel processing with different values of Δτ _k , provide the best adjustment (automatic adaptation) to the observed situation;

3) predict the value of the controlled parameter and the dispersions of the estimation errors:

случайных отклонений4) find the deviations ε(k) of the predicted estimate from the measured value and the estimate of the maximum variance

random deviations

- минимальная дисперсия формирующего шума, α, β - весовые коэффициенты рекурсивного фильтра скользящего среднего;where

- the minimum variance of the shaping noise, α, β - weight coefficients of the recursive moving average filter;

5) calculate the current estimate of the slowly changing systematic bias and its variance

6) determine the current estimate of the dispersion (power) of the shaping noise

7) carry out forecasting of the adjusted variances of estimation errors:

8) calculate the gains:

9) calculate the current estimates of the controlled parameter:

10) clarify estimates of the dispersion of random noise present in the measurement results due to an inaccurate choice of the sampling frequency of the controlled parameter:

where α _ν , β _ν are the weights of the recursive moving average filter, exclude.

с априорно заданными пороговыми уровнями u_j,

и формируют сигналы оповещения о воздействии помех и нарушителе:11) perform a comparison of the shaping noise power estimate

with a priori specified threshold levels u _j ,

and generate warning signals about the impact of interference and the intruder:

M

The above sequence of operations allows you to go to the description of the claims.

, характеризующей мощность внешнего возмущения, которая является косвенным информационным признаком импульсного воздействия нарушителя на чувствительный элемент датчика, сенсора или извещателя, приводящего к изменению его выходного сигнала, затем используют вычисленные косвенные оценки дисперсии формирующего шума

, где k - индекс наблюдаемого временного отсчета, а (k-1) - индекс предшествующего временного отсчета измерений, которые используют в качестве информационного признака при формировании сигнала оповещения о нарушении, формируемого на основе сравнения полученных оценок дисперсии формирующего шума

, которые получают на основе суммирования ранее полученных значений с весами α и β, которые устанавливают для усредняющего фильтра:

вычисленные обновленные значения дисперсии формирующего шума

сравнивают с выбранными порогами, u₁, u₂ и u₃, где u₁ - порог, ниже которого вычисленные значения оценок дисперсии формирующего шума, рассматривают как помеховый фон, не связанный с нарушением охраняемых рубежей объектов и границ, при этом нахождение сформированного сигнала, характеризующего изменение оценок дисперсии формирующего шума между порогами u₁ и u₂ рассматривают, как помехи, которые связаны с повышенным влиянием внешних воздействий на чувствительный элемент в виде ВОК, но не свидетельствуют о нарушении зоны охраны, и только факт превышения сформированным сигналом оценок дисперсии формирующего шума порога u₃ используют для формирования сигнала тревоги, при этом для повышения вероятности обнаружения нарушителя по вычисленным оценкам дисперсии формирующего шума используют различные алгоритмы компенсации помех в сформированном сигнале, характеризующем изменение во времени оценок дисперсии формирующего шума.The method of operational and technical protection of the boundaries of objects and borders, based on the use of fiber-optic cables as a distributed sensitive element and reflectometers, which form the basis for the construction of fiber-optic technical means of protection (FOTSO), and the formation of signals by them, which identify the fact of violation of the protection zone , from which the signs associated with the detection of the intruder are distinguished, their values are compared with the set value in the form of comparison thresholds, and based on the results of the comparison, the moment of issuing an alarm signal is determined, characterized in that the signals generated by the fiber-optic technical means of protection are subjected to additional processing using algorithms of structural-algorithmic transformations (SAP-2) and then adaptive nonlinear filtering (ANF), while first eliminating or reducing the low-frequency component (trend) in the signal generated by the sensor, sensor or detector, and then using the hell algorithm passive nonlinear filtering (ANF), as a result of which indirect estimates of the variance of the shaping noise are determined

, which characterizes the power of an external disturbance, which is an indirect information sign of the intruder's impulse action on the sensitive element of the sensor, sensor or detector, leading to a change in its output signal, then the calculated indirect estimates of the variance of the forming noise are used

, where k is the index of the observed time reference, and (k-1) is the index of the previous time reference of the measurements that are used as an information feature when generating a violation alert signal generated based on a comparison of the obtained estimates of the variance of the forming noise

, which are obtained based on the summation of previously obtained values with the weights α and β, which are set for the averaging filter:

computed updated shaping noise variance values

compared with the selected thresholds, u ₁ , u ₂ and u ₃ , where u ₁ is the threshold, below which the calculated values of the estimates of the variance of the shaping noise are considered as an interference background that is not associated with a violation of the protected boundaries of objects and boundaries, while finding the generated signal, characterizing the change in the estimates of the variance of the shaping noise between the thresholds u ₁ and u ₂ are considered as interferences that are associated with an increased influence of external influences on the sensitive element in the form of a FOC, but do not indicate a violation of the protection zone, and only the fact that the generated signal exceeds the estimates of the variance of the shaping noise the threshold u ₃ is used to generate an alarm signal, while to increase the probability of detecting an intruder according to the calculated estimates of the variance of the shaping noise, various algorithms for compensating for interference in the generated signal characterizing the change in time of the variance of the shaping noise are used.

The essential characteristics lie in the fact that the algorithms of the methods [1] and [2] are combined, thus forming a single sequence of operations of the new invention, characterized by novelty. At the same time, a significant increase in the signaling reliability of fiber-optic technical security equipment (FOTSO) was provided, which is illustratively shown in Fig. 2, fig. 3, fig. 4 and FIG. five.

Under the signaling reliability understand the complex indicator P _sig.nad =F(P _obt , _Δτ lt ), including: the probability of detecting the intruder (P _ob ) and the duration of the time intervals between false alarms ( _{Δτ lt} ). The more P and ( _{Δτ lt} ), the higher the signaling reliability of the detector.

The tests carried out showed that the required values of the signaling reliability of fiber-optic technical security equipment (FOTSO) were achieved.

As a conclusion, we consider it necessary to note the following promising direction, related, among other things, to fiber optic technologies, which is associated with the use of new methods of applied mathematics, which form the basis of the proposed invention and the solution of numerous existing problems.

The cognitive plots shown at the top of FIG. 5 (A and B), testify to their high saturation with various information, the useful (semantic) part of which is distorted by various kinds of interference and noise. In this case, the main problem is how to extract it from the total signal generated by the reflectometer, clean it from interference, and then how to interpret the results. If we take into account that the physical principles that form the basis of the operation of reflectometers, interferometers and other fiber-optic devices are always limited by the laws in force, then it turns out that the problem formulated above must be solved using mathematical methods. They form the basis for the emergence of various "tricks" that allow you to bypass many obstacles, including those determined by physical laws. But the well-known, traditional methods of applied mathematics [10,11] cannot solve the problem of increasing the level of return on the labor incurred in laying fiber optic cables and manufacturing more advanced fiber-optic devices. This, in particular, is evidenced by the world practice of using FOCs as distributed sensors, sensors and annunciators. Applied methods of a higher level are needed, on the basis of which, as the results show, it is possible to provide adaptation to solving complex applied problems of existing abstract mathematical theories [6–9,12]. Among them is the theory of finite fields by E. Galois [7,8,12] and the theory of adaptive nonlinear filtering [6,7-12].

Literature

1. Method for protecting the perimeter of an object (patent RU 2263968, MPI ⁷ G08B 13/02).

2. A method for protecting the perimeter of an object (patent 2473970 C1, MPI ⁷ G08B 13/02, published on January 27, 2013, bull. No. 3).

3. Method for protecting the perimeter of an object (patent RU 2379759, IPC G08B 13/26).

4. Telemechanical system "LILANA-VIA" and an alarm barrier with a detector by V.I. Yatskov (patent RU 2379759, IPC G08B 13/26).

5. Method for protecting the perimeter of an object (patent RU 2491646, IPC G08C 15/08, G08B 13/26).

6. The method of operational and technical protection of the boundaries of objects and borders (patent RU No. 2705770, application No. 2018119750 with priority dated May 29, 2018, published on November 11, 2019, bull. No. 32).

7. The method of operational and technical protection of the boundaries of objects and borders (patent RU No. 2674809, application No. 2017124869 with priority dated July 12, 2017, published on December 13, 2018, bull. No. 35).

8. A method for transmitting telemetric information adapted to various situations that occur during testing of rocket and space technology, and a system for its implementation (patent RU No. 26571584, MPI G08C19 / 28, publ. 10.10.2015, bull. No. 35) .

9. Kukushkin S.S. Finite field theory and informatics: vol. 1 Methods and algorithms, classical and non-traditional, based on the use of the constructive remainder theorem - M: MO RF, 2003. - 281 p.

10. Kalman R., Falb P., Arbib M. Essays on the Mathematical Theory of Systems / Per. from English. - Mir, 1971. - 400 p.

11. Sage E., Mels J. Estimation theory and its application in communication and management / Per. from English. - Communication, 1976. - 496 p.

12. Kukushkin S.S., Kuznetsov V.I. Analysis of the influence of noisy interference in the results of telemetric measurements - M .: "Izmeritelnaya tekhnika", 2016. No. 10. pp. 21-23.

Claims (1)

The method of operational and technical protection of the boundaries of objects and borders, based on the use of fiber-optic cables as a distributed sensitive element and reflectometers, which form the basis for the construction of fiber-optic technical means of protection (FOTSO), and the formation of signals by them, which identify the fact of violation of the protection zone , from which the signs associated with the detection of the intruder are distinguished, their values are compared with the set value in the form of comparison thresholds, and based on the results of the comparison, the moment of issuing an alarm signal is determined, characterized in that the signals generated by the fiber-optic technical means of protection are subjected to additional processing using algorithms of structural-algorithmic transformations (SAP-2) and then adaptive nonlinear filtering (ANF), while first eliminating or reducing the low-frequency component (trend) in the signal generated by the sensor, sensor or detector, and then using the hell algorithm passive nonlinear filtering (ANF), as a result of which indirect estimates of the variance of the shaping noise are determined

, which characterizes the power of an external disturbance, which is an indirect information sign of the intruder's impulse action on the sensitive element of the sensor, sensor or detector, leading to a change in its output signal, then the calculated indirect estimates of the variance of the forming noise are used

, where k is the index of the observed time reference, and (k-1) is the index of the previous time reference of the measurements that are used as an information feature when generating a violation alert signal generated based on a comparison of the obtained estimates of the variance of the forming noise

, which are obtained based on the summation of previously obtained values with the weights α and β, which are set for the averaging filter:

computed updated shaping noise variance values

compared with the selected thresholds, u ₁ , u ₂ and u ₃ , where u ₁ is the threshold, below which the calculated values of the estimates of the variance of the shaping noise are considered as an interference background that is not associated with a violation of the protected boundaries of objects and boundaries, while finding the generated signal, characterizing the change in the estimates of the variance of the shaping noise between the thresholds u ₁ and u ₂ are considered as interferences that are associated with an increased influence of external influences on the sensitive element in the form of a FOC, but do not indicate a violation of the protection zone, and only the fact that the generated signal exceeds the estimates of the variance of the shaping noise the threshold u ₃ is used to generate an alarm signal, while to increase the probability of detecting an intruder according to the calculated estimates of the variance of the shaping noise, various algorithms for compensating for interference in the generated signal characterizing the change in time of the variance of the shaping noise are used.

Images