CN104459637B - Processing method and system of microwave correlation radar signals - Google Patents

Abstract

The invention provides a processing method and system of microwave correlation radar signals. By means of the processing method and system of the microwave correlation radar signals, radar echo data are obtained, period processing is carried out on the data, a period average value of time periods is obtained through calculation, and a relative error of each time period is calculated according to the period average value and a corresponding predicted value. Whether the relative errors of the time periods are larger than or equal to an abnormal threshold value is judged in sequence, wherein the number of the time periods before the current time period is preset; when the relative errors are larger than or equal to the abnormal threshold value, the predicted values of the corresponding time periods are stored into FIFO, the period average value of the corresponding time periods is stored into the FIFO when the relative errors are smaller than the abnormal threshold value, and therefore a period average value set is obtained. A predicted value of a next period is calculated according to the period average value and the period average value set of the current time period, and the predicted value of the next period is output. Due to the fact that the influences of weather conditions on amplitude changing of radar echo data are taken into account when microwave correlation radar signals are processed and the predicted value of the next time period is predicted, false alarms caused by weather variations are avoided, and accuracy is improved.

Description

Microwave correlation radar signal processing method and system

Technical Field

The invention relates to the technical field of radar detection, in particular to a microwave correlation radar signal processing method and system.

Background

Compared with other safety protection devices, the microwave-pair radar has the advantages of all weather, long detection distance and the like, thereby being a safety protection material widely applied.

The traditional microwave correlation radar signal processing method is that an average value of echo amplitudes is estimated to be used as a reference for judging whether a radar has a fault or not, specifically, the average value of the echo amplitudes is counted and compared with a preset value to judge whether data are abnormal or not, if the data are abnormal, an intrusion event occurs in a safety protection area, alarm information is output, and the preset value used for alarm judgment is fixed and unchanged. In practical application, the amplitude of radar echo can be changed when an intrusion event occurs in a safety protection area or the weather condition changes, and the traditional microwave correlation radar signal processing method has the defect of low accuracy because alarm judgment is carried out by using a fixed preset value.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a method and a system for processing microwave correlation radar signals with high accuracy.

A microwave correlation radar signal processing method comprises the following steps:

acquiring radar echo data, performing segmentation processing, and calculating to obtain a segmentation average value of each time period;

calculating the relative error of each time period according to the segmented average value and the corresponding predicted value;

sequentially judging whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the segment average value of the corresponding time period into FIFO when the relative error is less than the abnormal threshold value to obtain a segment average value set;

and calculating and outputting a predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

A microwave correlation radar signal processing system comprises

The data segmentation module is used for acquiring radar echo data, performing segmentation processing and calculating to obtain a segmentation average value of each time period;

the error calculation module is used for calculating the relative error of each time period according to the segmented average value and the corresponding predicted value;

the data processing module is used for sequentially judging whether the relative error of each time period is greater than or equal to the abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the sectional average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value to obtain a sectional average value set;

and the average value prediction module is used for calculating and outputting the predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

According to the microwave correlation radar signal processing method and system, radar echo data are obtained and are subjected to segmentation processing, the segmentation average value of each time period is obtained through calculation, and the relative error of each time period is calculated according to the segmentation average value and the corresponding predicted value. And sequentially judging whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the segment average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value to obtain a segment average value set. And calculating and outputting a predicted value of the next time period according to the segment average value of the current time period and the segment average value set. When the segment average value set is established, the corresponding segment average value or the predicted value is selected as data in the segment average value set according to the comparison result, abnormal data generated by invasion of the safety protection area are eliminated, and influence on data prediction is avoided. Because the influence of weather conditions on the amplitude change of radar echo data is considered when microwave correlation radar signal processing is carried out and the predicted value of the next time slot is predicted, false alarm caused by weather change is avoided, and compared with the traditional microwave correlation radar signal processing method, the accuracy is improved.

Drawings

FIG. 1 is a flow chart of a microwave correlation radar signal processing method according to an embodiment;

FIG. 2 is a flow chart of a microwave correlation radar signal processing method in another embodiment;

FIG. 3 is a block diagram of a microwave correlation radar signal processing system in accordance with an embodiment;

fig. 4 is a block diagram of a microwave correlation radar signal processing system in another embodiment.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

A microwave correlation radar signal processing method, as shown in fig. 1, includes the following steps:

step S120: and acquiring radar echo data, performing segmentation processing, and calculating to obtain a segmentation average value of each time period.

The radar echo data is electromagnetic wave signal data received by the radar, and the received radar echo data is continuous because the radar constantly emits electromagnetic waves. The radar echo data are segmented, the segmentation can be carried out according to preset time length, or the segmentation can be carried out according to the preset number of segments after the data are obtained.

The segment average value is the average value of the echo amplitude values at all times in the index data segment, and the echo amplitude values at all times can be known according to the radar echo data, so that the segment average value of the corresponding time segment can be directly calculated according to all the data segments.

In one embodiment, step S120 specifically includes steps 122 to 126.

Step 122: and acquiring radar echo data. And acquiring radar echo data received by the radar in real time.

Step 124: and carrying out segmentation processing on the radar echo data according to the preset time length to obtain data segments in each time period.

The preset duration can be adjusted according to actual conditions. The radar echo data are processed in a segmented mode by setting preset time length, and the obtained time length of each data segment is guaranteed to be the same, so that subsequent data processing is facilitated.

In order to enhance the real-time performance of safety protection, the sampling rate of a radar is high, in general, an intrusion event can cause the amplitude change of radar echoes at a plurality of connected time points, and the segmented average value is obtained by dividing the echoes at the connected time points into data segments and calculating. In this embodiment, an average value obtained by counting the time span of the echo amplitude change caused by the intrusion may be used as a specific value of the preset duration.

And performing segmentation processing on the radar echo data according to a preset time length, wherein the segmentation processing can be continuous segmentation, interval segmentation or partial continuous and partial interval segmentation. Continuous segmentation means that the resulting data segments are continuous, and intermittent segmentation means that the resulting data segments are discontinuous. In the embodiment, the radar echo data are continuously segmented according to the preset duration, and the obtained continuous data segments are used as follow-up steps for analysis, so that the influence on fault detection caused by data omission is avoided, and the fault detection accuracy is improved.

Step 126: and respectively calculating the segment average value of each time segment according to the data segments.

In one embodiment, step 126 is embodied as

$y_n=\frac{1}{M}\underset{m=\mathrm{nM}}{\overset{(n+1)M-1}{\mathrm{\Σ}}}{x}_m$

Wherein x is_mAmplitude values, y, representing radar echo data at the m-th moment_nAnd M is a preset time length, namely the time length of each time segment.

In the embodiment, the segment average value of the corresponding time period is calculated according to the echo data values of all the moments in each time period, so that the data accuracy is improved. It is understood that in other embodiments, the echo data values at a part of the time in each time period may be extracted to calculate the corresponding segment average value.

Step S130: and calculating the relative error of each time segment according to the segment average value and the corresponding predicted value.

The prediction value differs for different time periods, and may be calculated and stored in advance. In one embodiment, step S130 is embodied as

$r_n=\left|\frac{y_n-{\hat{y}}_n}{{\hat{y}}_n}\right|\×100\%$

Wherein r is_nDenotes the relative error of the nth time segment, y_nRepresents the segment average of the nth time segment,indicating the predicted value of the nth time segment.

Step S140: and sequentially judging whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the segment average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value to obtain a segment average value set.

The preset number can be adjusted according to actual conditions, and since the radar receives the radar echo data in real time, the radar echo data is also segmented in real time in step S120 to obtain a new segment average value of the time period. And taking the newly acquired time period as the current time period, establishing a segment average value set according to the data of the time periods with the preset number, ensuring that the segment average value set always stores the data closest to the current time period, avoiding the influence caused by historical data which is too long away from the current time period when data processing is carried out in the subsequent steps, and improving the data processing accuracy.

Because the weather condition changes sooner, the change of the echo amplitude is also quicker, the preset number in the embodiment is selected as the minimum time span of the echo amplitude change caused by the weather condition, and the accuracy of subsequently judging whether the weather condition changes is improved.

Specifically, whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of a continuous preset number before the current time period is sequentially judged according to the time sequence, if yes, the safety protection area in the time period is invaded, the obtained segmentation average value is abnormal data, and at the moment, the predicted value of the corresponding time period is stored into FIFO (First Input First output, First in First out queue). If not, the segment average value obtained in the time period is indicated to be normal data, and the segment average value of the corresponding time period is stored in FIFO to obtain a segment average value set. The length of the FIFO corresponds to the preset number, and the minimum time span of the echo amplitude change caused by the weather condition is also taken.

For example, in L consecutive time periods before the current time period, the segment average value of each time period is y₁，y₂，…，y_LThe predicted value of each time period isComparing the relative error of the first time period with an abnormal threshold value, and if the relative error is greater than or equal to the abnormal threshold value, determining that the relative error is greater than or equal to the abnormal threshold valueStoring into FIFO, if the relative error is less than the abnormal threshold value, then y is₁Storing the error in FIFO, comparing the relative error of the second time segment with the abnormal threshold value, and repeating the steps to obtain the final product with length LAnd FIFOs are used for completing the construction of the segmented average value set.

When the segmented average value set is constructed, the relative error of each time period is compared with the abnormal threshold value, and when the predicted value of the next time period is calculated in the subsequent steps, the influence of abnormal data on data processing can be avoided, the data calculation accuracy is improved, and the accuracy of microwave correlation radar signal processing is improved.

In addition, the related data is stored in the form of FIFO to obtain a segmented average value set, and when new data is stored each time to update the FIFO, the earliest stored data is derived, so that when a segmented average value of a new time period (i.e., the current time period) is obtained in step S120, the data stored in the FIFO is always related data of a preset number of time periods before the current time period, and can be directly used as a predicted value for calculating the next time period in a subsequent step, thereby improving the data processing speed.

Step S160: and calculating and outputting a predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

And calculating and outputting a predicted value of the next time period according to the segment average value and the segment average value set of the current time period, wherein the predicted value is used as a standard for performing alarm judgment of the next time period. In one embodiment, step S160 is embodied as

${\stackrel{\‾}{y}}_{n+1}={\stackrel{\‾}{y}}_{n+L-1}+\frac{y_{n+L}-y_n}{L}$

${\mathrm{\β}}_{n+L}={\mathrm{\β}}_{n+L-1}+6\frac{(L-1)y_{n+L}+(L+1)y_n-2L{\stackrel{\‾}{y}}_{n+L-1}}{L^3-L}$

${\hat{y}}_{n+L+1}={\stackrel{\‾}{y}}_{n+L}+{\mathrm{\β}}_{n+L}\·L$

Wherein,respectively representing the average values of the data in the n + L-1 and n + L time FIFOs, β_n+L-1、β_n+LRespectively representing the data change rate in the n + L-1 and n + L sections of time FIFO, y_nIs a segment average or predicted value, y, of the nth time segment in the FIFO_n+LIs the segment average of the n + L time period, L is the length of the FIFO,indicates the predicted value of the (n + L + 1) th period. Specifically, let the data in FIFO be y_n、y_n+1、…、y_n+L-1The segment average value of the current time segment is y_n+LAccording to the formula, the predicted value of the next time period can be calculated

In this embodiment, the least square method is used to calculate the updated expression of the data average value and the data change rate in the FIFOs in the adjacent time periods, and only the initial value of the data average value and the initial value of the data change rate in the FIFOs need to be calculated and stored in advance, for example, the most initial segment average value of L data segments can be directly stored in the FIFOs, and the data average value and the data change rate in the FIFOs can be calculated and stored as corresponding initial values respectively. And calculating to obtain the data average value and the data change rate after the FIFO is updated according to the segment average value of the new time period, the data average value and the data change rate before the FIFO is updated every time when the segment average value of a new time period is obtained. And directly calculating the predicted value of the next time period according to the updated data average value and the data change rate of the FIFO, and using the predicted value as the alarm judgment standard of the next time period. The average value and the change rate of the data after the FIFO updating are directly calculated through the average value and the change rate of the data before the FIFO updating, the algorithm complexity is reduced, and the data processing efficiency is improved. The predicted value of each time segment can be calculated by the expression.

In addition, when the FIFO is updated, the corresponding segment average value or the corresponding predicted value is selected as the data in the FIFO according to the comparison result, so that the data invaded by the safety protection area is eliminated, the influence on the calculation of the predicted value in the next time period can be avoided, and the prediction accuracy is improved.

It can be understood that, in other embodiments, the corresponding data average value and the data change rate may also be directly calculated according to the data updated by the FIFO each time, so as to calculate the predicted value of the next time period.

According to the microwave radar signal processing method, when the segmented average value set is established, the corresponding segmented average value or the corresponding predicted value is selected as data in the segmented average value set according to the comparison result, abnormal data caused by invasion of a safety protection area are eliminated, and influence on data prediction is avoided. Because the change rate of the data is estimated by adopting the least square method when the microwave correlation radar signal processing is carried out and the predicted value of the next time slot is predicted, the influence of weather conditions on the amplitude change of radar echo data is considered, false alarm caused by weather change is avoided, and compared with the traditional microwave correlation radar signal processing method, the accuracy is improved.

In one embodiment, as shown in fig. 2, after step S140 and before step S160, the microwave correlation radar signal processing method further includes the following steps:

step S150: and judging whether the number of the time periods of which the relative error is greater than or equal to the abnormal threshold value is less than or equal to the fault threshold value in the time period of which the number is continuously preset before the current time period.

The specific values of the abnormal threshold and the fault threshold may also be adjusted according to the actual situation, corresponding to the specific value of the preset duration that is the average value of the time span of the echo amplitude change caused by the intrusion in step S120, and the fault threshold is set to be greater than the time span of the echo amplitude change caused by the intrusion and smaller than the length of the segment average value set, and the average value of the two values may be taken.

And counting the number of time periods with relative errors larger than or equal to the abnormal threshold value in the time periods with continuous preset number before the current time period, and judging whether the relative errors are smaller than or equal to the fault threshold value. If yes, the data abnormality is in the allowable range, the microwave radar does not have a fault, step S160 is performed, and a predicted value of the next time period is calculated and output according to the segment average value of the current time period and the segment average value set, and is used as a standard for performing alarm judgment of the next time period.

In one embodiment, if the number of time periods in which the relative error is greater than or equal to the abnormal threshold value is greater than the fault threshold value in a time period of which the number is continuously preset before the current time period, the microwave correlation radar signal processing method further includes a step of outputting fault alarm information when the data change rate in the segment average value set is less than the change trend threshold value.

The data change rate in the segment mean value set can be calculated by using a least square method. For example, a functional relation between the data y in the segment average value set and the corresponding time x may be "a + bx", and the data in the segment average value set and the corresponding time are substituted into the functional relation to perform iterative operation, so that the values of the finally obtained parameters a and b are within an allowable error range, and the final value of the parameter b is the data change rate in the segment average value set. It is to be understood that the specific manner of calculating the data change rate in the segment mean value set is not exclusive, and other functional relationships may be used.

The threshold value of the variation trend can be adjusted according to the actual situation, and the selected basis is the statistical average of the variation trend when the weather condition is stable. In step S140, when the segment average value set is established, the corresponding segment average value or the predicted value is selected as the data in the segment average value set according to the comparison result, and the abnormal data invaded by the safety protection area is removed, so that the influence caused by the abnormal data can be avoided when the data change rate is calculated, and the calculation accuracy of the data change rate is improved.

If the number of the time periods with the relative error larger than or equal to the abnormal threshold value is larger than the fault threshold value and the data change rate in the segmented average value set is smaller than the change trend threshold value in the time periods with the continuous preset number before the current time period, the segmented average value is suddenly changed, the microwave fails to the radar, and fault alarm information is output so that a worker can timely overhaul the fault alarm information. The output fault alarm information can be specifically a sound alarm, a luminous alarm or a simultaneous sound and light alarm.

In addition, if the number of time periods in which the relative error is greater than or equal to the abnormal threshold value is greater than the fault threshold value and the data change rate in the segment average value set is greater than the change trend threshold value in the time periods continuously preset by the current time period, the process may return to step S160, and the predicted value of the next time period is calculated and output according to the segment average value of the current time period and the segment average value set.

If the number of the time periods with the relative error larger than or equal to the abnormal threshold value is larger than the fault threshold value in the time periods with the continuous preset number before the current time period, and the data change rate in the segmented average value set is larger than the change trend threshold value, it is indicated that the echo amplitude changes because the weather condition changes, and at this moment, the fault is not reported, and the average value of the next time period is continuously predicted.

The invention also provides a microwave correlation radar signal processing system, as shown in fig. 3, which includes a data segmentation module 120, an error calculation module 130, a data processing module 140, and a mean value prediction module 160.

The data segmentation module 120 is configured to obtain radar echo data, perform segmentation processing, and calculate a segment average value of each time period.

The radar echo data is electromagnetic wave signal data received by the radar, and the received radar echo data is continuous because the radar constantly emits electromagnetic waves. The radar echo data are segmented, the segmentation can be carried out according to preset time length, or the segmentation can be carried out according to the preset number of segments after the data are obtained.

The segment average value is the average value of the echo amplitude values at all times in the index data segment, and the echo amplitude values at all times can be known according to the radar echo data, so that the segment average value of the corresponding time segment can be directly calculated according to all the data segments.

In one embodiment, the data segmentation module 120 includes a first unit, a second unit, and a third unit.

The first unit is used for acquiring radar echo data. And acquiring radar echo data received by the radar in real time.

The second unit is used for carrying out segmentation processing on the radar echo data according to the preset duration to obtain data segments in each time period.

The preset duration can be adjusted according to actual conditions. The radar echo data are processed in a segmented mode by setting preset time length, and the obtained time length of each data segment is guaranteed to be the same, so that subsequent data processing is facilitated.

In order to enhance the real-time performance of safety protection, the sampling rate of a radar is high, in general, an intrusion event can cause the amplitude change of radar echoes at a plurality of connected time points, and the segmented average value is obtained by dividing the echoes at the connected time points into data segments and calculating. In this embodiment, an average value obtained by counting the time span of the echo amplitude change caused by the intrusion may be used as a specific value of the preset duration.

And performing segmentation processing on the radar echo data according to a preset time length, wherein the segmentation processing can be continuous segmentation, interval segmentation or partial continuous and partial interval segmentation. Continuous segmentation means that the resulting data segments are continuous, and intermittent segmentation means that the resulting data segments are discontinuous. In the embodiment, the radar echo data are continuously segmented according to the preset duration, and the obtained continuous data segments are used as follow-up steps for analysis, so that the influence on fault detection caused by data omission is avoided, and the fault detection accuracy is improved.

And the third unit is used for respectively calculating the segment average value of each time segment according to the data segments.

In one embodiment, the third unit calculates segment averages of the time segments according to the data segments, specifically

$y_n=\frac{1}{M}\underset{m=\mathrm{nM}}{\overset{(n+1)M-1}{\mathrm{\Σ}}}{x}_m$

Wherein x is_mAmplitude values, y, representing radar echo data at the m-th moment_nAnd M is a preset time length, namely the time length of each time segment.

In the embodiment, the segment average value of the corresponding time period is calculated according to the echo data values of all the moments in each time period, so that the data accuracy is improved. It is understood that in other embodiments, the echo data values at a part of the time in each time period may be extracted to calculate the corresponding segment average value.

The error calculation module 130 is configured to calculate a relative error of each time segment according to the segment average value and the corresponding predicted value.

The prediction value differs for different time periods, and may be calculated and stored in advance. In one embodiment, the error calculation module 130 calculates a relative error of each time segment according to the segment average value and the corresponding predicted value, specifically, the relative error of each time segment is calculated

$r_n=\left|\frac{y_n-{\hat{y}}_n}{{\hat{y}}_n}\right|\×100\%$

Wherein r is_nDenotes the relative error of the nth time segment, y_nRepresents the segment average of the nth time segment,indicating the predicted value of the nth time segment.

The data processing module 140 is configured to sequentially determine whether a relative error of each time period is greater than or equal to an abnormal threshold value in a time period that is a preset number of times continuously before a current time period, store a predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, and store a segment average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value, so as to obtain a segment average value set.

The preset number can be adjusted according to actual conditions, and since the radar receives the radar echo data in real time, the data segmentation module 120 also performs segmentation processing on the radar echo data in real time to obtain a data segment in a new time period. And taking the newly acquired time period as the current time period, establishing a segment average value set according to the data of the time periods with the preset number, ensuring that the segment average value set always stores the data closest to the current time period, avoiding the influence caused by historical data which is too long away from the current time period when data processing is carried out in the subsequent steps, and improving the data processing accuracy.

Because the weather condition changes sooner, the change of the echo amplitude is also quicker, the preset number in the embodiment is selected as the minimum time span of the echo amplitude change caused by the weather condition, and the accuracy of subsequently judging whether the weather condition changes is improved.

Specifically, whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of a continuous preset number before the current time period is sequentially judged according to the time sequence, if yes, the safety protection area of the time period is invaded, the obtained segmentation average value is abnormal data, and at the moment, the predicted value of the corresponding time period is stored into FIFO. If not, the segment average value obtained in the time period is indicated to be normal data, and the segment average value of the corresponding time period is stored in FIFO to obtain a segment average value set. The length of the FIFO corresponds to the preset number, and the minimum time span of the echo amplitude change caused by the weather condition is also taken.

When the segmented average value set is constructed, the relative error of each time period is compared with the abnormal threshold value, and when the predicted value of the next time period is calculated in the subsequent steps, the influence of abnormal data on data processing can be avoided, the data calculation accuracy is improved, and the accuracy of microwave correlation radar signal processing is improved.

In addition, the related data is stored in an FIFO form to obtain a segmented average value set, and when new data is stored each time to update the FIFO, the data stored earliest is exported, so that when the data segmentation module 120 obtains a segmented average value of a new time period (i.e., the current time period), the data stored in the FIFO is always related data of a preset number of time periods before the current time period, and can be directly used as a predicted value for calculating the next time period in subsequent steps, thereby improving the data processing speed.

The mean value predicting module 160 is configured to calculate and output a predicted value of the next time period according to the segment mean value of the current time period and the segment mean value set.

And calculating and outputting a predicted value of the next time period according to the segment average value and the segment average value set of the current time period, wherein the predicted value is used as a standard for performing alarm judgment of the next time period. In one embodiment, the mean prediction module 160 calculates a predicted value of the next time period, specifically, the predicted value of the next time period according to the segment mean of the current time period and the segment mean set

${\stackrel{\‾}{y}}_{n+1}={\stackrel{\‾}{y}}_{n+L-1}+\frac{y_{n+L}-y_n}{L}$

${\mathrm{\β}}_{n+L}={\mathrm{\β}}_{n+L-1}+6\frac{(L-1)y_{n+L}+(L+1)y_n-2L{\stackrel{\‾}{y}}_{n+L-1}}{L^3-L}$

${\hat{y}}_{n+L+1}={\stackrel{\‾}{y}}_{n+L}+{\mathrm{\β}}_{n+L}\·L$

Wherein,respectively represent the n +Average value of data in FIFO of L-1, n + L sections of time β_n+L-1、β_n+LRespectively representing the data change rate in the n + L-1 and n + L sections of time FIFO, y_nIs a segment average or predicted value, y, of the nth time segment in the FIFO_n+LIs the segment average of the n + L time period, L is the length of the FIFO,indicates the predicted value of the (n + L + 1) th period. Specifically, let the data in FIFO be y_n、y_n+1、…、y_n+L-1The segment average value of the current time segment is y_n+LAccording to the formula, the predicted value of the next time period can be calculated

In this embodiment, the least square method is used to calculate the updated expression of the data average value and the data change rate in the FIFOs in the adjacent time periods, and only the initial value of the data average value and the initial value of the data change rate in the FIFOs need to be calculated and stored in advance, for example, the most initial segment average value of L data segments can be directly stored in the FIFOs, and the data average value and the data change rate in the FIFOs can be calculated and stored as corresponding initial values respectively. And calculating to obtain the data average value and the data change rate after the FIFO is updated according to the segment average value of the new time period, the data average value and the data change rate before the FIFO is updated every time when the segment average value of a new time period is obtained. And directly calculating the predicted value of the next time period according to the updated data average value and the data change rate of the FIFO, and using the predicted value as the alarm judgment standard of the next time period. The average value and the change rate of the data after the FIFO updating are directly calculated through the average value and the change rate of the data before the FIFO updating, the algorithm complexity is reduced, and the data processing efficiency is improved. The predicted value of each time segment can be calculated by the expression.

In addition, when the FIFO is updated, the corresponding segment average value or the corresponding predicted value is selected as the data in the FIFO according to the comparison result, so that the data invaded by the safety protection area is eliminated, the influence on the calculation of the predicted value in the next time period can be avoided, and the prediction accuracy is improved.

It can be understood that, in other embodiments, the corresponding data average value and the data change rate may also be directly calculated according to the data updated by the FIFO each time, so as to calculate the predicted value of the next time period.

According to the microwave-pair radar signal processing system, when the segmented average value set is established, the corresponding segmented average value or the corresponding predicted value is selected as data in the segmented average value set according to the comparison result, abnormal data caused by invasion of a safety protection area are eliminated, and influence on data prediction is avoided. Because the change rate of the data is estimated by adopting the least square method when the microwave correlation radar signal processing is carried out and the predicted value of the next time slot is predicted, the influence of weather conditions on the amplitude change of radar echo data is considered, false alarm caused by weather change is avoided, and compared with the traditional microwave correlation radar signal processing method, the accuracy is improved.

In one embodiment, as shown in fig. 4, the microwave correlation radar signal processing system further includes a determining module 150. The judging module 150 is configured to sequentially judge whether the relative error of each time period is greater than or equal to the abnormal threshold value in the time periods of the preset number of consecutive time periods before the current time period in the data processing module 140, store the predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, store the segment average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value, and after obtaining the segment average value set, judge whether the number of the time periods of the preset number of consecutive time periods before the current time period in which the relative error is greater than or equal to the abnormal threshold value is less than or equal to the fault threshold value.

And when the number of the time periods with the relative error larger than or equal to the abnormal threshold value is smaller than or equal to the fault threshold value in the time period with the continuous preset number before the current time period, the average value prediction module 160 calculates and outputs the predicted value of the next time period according to the segment average value and the segment average value set of the current time period.

The specific values of the abnormal threshold and the fault threshold can also be adjusted according to the actual situation, corresponding to the specific value taking the average value of the time span causing the echo amplitude change to the intrusion as the preset time length, the fault threshold is set to be larger than the time span causing the echo amplitude change to the intrusion and smaller than the length of the segmented average value set, and the average value of the two values can be taken.

And counting the number of time periods with relative errors larger than or equal to the abnormal threshold value in the time periods with continuous preset number before the current time period, and judging whether the relative errors are smaller than or equal to the fault threshold value. If yes, the data abnormality is in the allowable range, and the microwave-pair radar does not have a fault. The average prediction module 160 calculates and outputs a predicted value of the next time period according to the segment average value and the segment average value set of the current time period, and the predicted value is used as a standard for performing alarm judgment of the next time period.

In one embodiment, the microwave correlation radar signal processing system further includes a fault alarm module, and the fault alarm module is configured to output fault alarm information when the number of time periods in which the relative error is greater than or equal to the abnormal threshold value is greater than a fault threshold value and the data change rate in the segment average value set is less than the change trend threshold value in a time period in which a preset number of times are continuously performed before the current time period.

The data change rate in the segment mean value set can be calculated by using a least square method. The microwave correlation radar signal processing method is explained in detail, and is not described herein again.

The threshold value of the variation trend can be adjusted according to the actual situation, and the selected basis is the statistical average of the variation trend when the weather condition is stable. When the data processing module 140 establishes the segment average value set, the corresponding segment average value or the predicted value is selected as the data in the segment average value set according to the comparison result, and abnormal data invaded by the safety protection area is eliminated, so that the fault alarm module can avoid the influence of the abnormal data when calculating the data change rate, and the calculation accuracy of the data change rate is improved.

If the number of the time periods with the relative error larger than or equal to the abnormal threshold value is larger than the fault threshold value and the data change rate in the segmented average value set is smaller than the change trend threshold value in the time periods with the continuous preset number before the current time period, the segmented average value is suddenly changed, the microwave fails to the radar, and fault alarm information is output so that a worker can timely overhaul the fault alarm information. The output fault alarm information can be specifically a sound alarm, a luminous alarm or a simultaneous sound and light alarm.

In addition, if the number of time periods in which the relative error is greater than or equal to the abnormal threshold value is greater than the fault threshold value and the data change rate in the segment average value set is greater than the change trend threshold value in the time period of the continuous preset number before the current time period, the mean value predicting module 160 calculates and outputs the predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

If the number of the time periods with the relative error larger than or equal to the abnormal threshold value is larger than the fault threshold value in the time periods with the continuous preset number before the current time period, and the data change rate in the segmented average value set is larger than the change trend threshold value, it is indicated that the echo amplitude changes because the weather condition changes, and at this moment, the fault is not reported, and the average value of the next time period is continuously predicted.

The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A microwave correlation radar signal processing method is characterized by comprising the following steps:

acquiring radar echo data, performing segmentation processing, and calculating to obtain a segmentation average value of each time period;

calculating the relative error of each time period according to the segmented average value and the corresponding predicted value;

sequentially judging whether the relative error of each time period is greater than or equal to an abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the segment average value of the corresponding time period into FIFO when the relative error is less than the abnormal threshold value to obtain a segment average value set;

and calculating and outputting a predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

2. The microwave correlation radar signal processing method according to claim 1, wherein the step of obtaining radar echo data, performing segmentation processing, and calculating a segment average value of each time period comprises the following steps:

acquiring radar echo data;

performing segmentation processing on the radar echo data according to preset time to obtain data segments in each time segment;

and respectively calculating the segment average value of each time segment according to the data segments.

3. The microwave correlation radar signal processing method according to claim 2, wherein the step of calculating a segment average value for each time segment from the data segments is performed, specifically

$y_n=\frac{1}{M}\underset{m=\mathrm{nM}}{\overset{(n+1)M-1}{\mathrm{\Σ}}}{x}_m$

Wherein x is_mAmplitude values, y, representing radar echo data at the m-th moment_nAnd M is a preset time length.

4. The microwave correlation radar signal processing method according to claim 1, wherein the step of calculating the relative error of each time segment from the segment average value and the corresponding predicted value is specifically performed

$r_n=\left|\frac{y_n-{\hat{y}}_n}{{\hat{y}}_n}\right|\×100\%$

Wherein r is_nDenotes the relative error of the nth time segment, y_nRepresents the segment average of the nth time segment,indicating the predicted value of the nth time segment.

5. The microwave correlation radar signal processing method according to claim 1, wherein the step of calculating a predicted value of a next time period according to the segment mean of the current time period and the segment mean set is specifically a step of calculating a predicted value of a next time period according to the segment mean of the current time period and the segment mean set

${\stackrel{\‾}{y}}_{n+L}={\stackrel{\‾}{y}}_{n+L-1}+\frac{y_{n+L}-y_n}{L}$

${\mathrm{\β}}_{n+L}={\mathrm{\β}}_{n+L-1}+6\frac{(L-1)y_{n+L}+(L+1)y_n-2L{\stackrel{\‾}{y}}_{n+L-1}}{L^3-L}$

${\hat{y}}_{n+L+1}={\stackrel{\‾}{y}}_{n+L}+{\mathrm{\β}}_{n+L}\·L$

Wherein,respectively representing the average values of the data in the n + L-1 and n + L time FIFOs, β_n+L-1、β_n+LRespectively representing the data change rate in the n + L-1 and n + L sections of time FIFO, y_nIs a segment average or predicted value, y, of the nth time segment in the FIFO_n+LIs the segment average of the n + L time period, L is the length of the FIFO,indicates the predicted value of the (n + L + 1) th period.

6. The microwave correlation radar signal processing method according to claim 1, wherein, after the step of sequentially determining whether a relative error of each time period is greater than or equal to an abnormal threshold value in a time period which is a preset number of consecutive time periods before a current time period, and when the relative error is greater than or equal to the abnormal threshold value, storing a predicted value of the corresponding time period into the FIFO, and when the relative error is less than the abnormal threshold value, storing a segment average value of the corresponding time period into the FIFO to obtain a segment average value set, and before the step of calculating and outputting the predicted value of a next time period according to the segment average value of the current time period and the segment average value set, the method further comprises the following steps:

judging whether the number of time periods with relative errors larger than or equal to the abnormal threshold value is smaller than or equal to the fault threshold value in the time periods with the continuous preset number before the current time period;

and if so, calculating and outputting a predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

7. The microwave correlation radar signal processing method according to claim 6, wherein if the number of time periods in which the relative error is greater than or equal to the abnormal threshold value is greater than the fault threshold value in a preset number of time periods before the current time period, the microwave correlation radar signal processing method further includes a step of outputting fault alarm information when the data change rate in the segment average value set is less than the change trend threshold value.

8. A microwave correlation radar signal processing system is characterized by comprising

The data segmentation module is used for acquiring radar echo data, performing segmentation processing and calculating to obtain a segmentation average value of each time period;

the error calculation module is used for calculating the relative error of each time period according to the segmented average value and the corresponding predicted value;

the data processing module is used for sequentially judging whether the relative error of each time period is greater than or equal to the abnormal threshold value in the time periods of which the number is continuously preset before the current time period, storing the predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, and storing the sectional average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value to obtain a sectional average value set;

and the average value prediction module is used for calculating and outputting the predicted value of the next time period according to the segment average value of the current time period and the segment average value set.

9. The microwave correlation radar signal processing system according to claim 8, further comprising a determining module, configured to determine, by the data processing module, whether a relative error of each time period is greater than or equal to an abnormal threshold value in a time period of a preset number of consecutive time periods before a current time period in sequence, store a predicted value of the corresponding time period into the FIFO when the relative error is greater than or equal to the abnormal threshold value, store a segment average value of the corresponding time period into the FIFO when the relative error is less than the abnormal threshold value, and after a segment average value set is obtained, determine whether the number of time periods of the preset number of consecutive time periods before the current time period in which the relative error is greater than or equal to the abnormal threshold value is less than or equal to a fault threshold value;

and the average value prediction module calculates and outputs the predicted value of the next time period according to the segment average value of the current time period and the segment average value set when the number of the time periods with the relative error larger than or equal to the abnormal threshold value is smaller than or equal to the fault threshold value in the time period with the continuous preset number before the current time period.

10. The microwave correlation radar signal processing system of claim 9, further comprising a fault alarm module, wherein the fault alarm module is configured to output fault alarm information when, in a time period of a preset number of consecutive time periods before a current time period, the number of time periods in which a relative error is greater than or equal to an abnormal threshold value is greater than a fault threshold value, and a data change rate in the segment average value set is less than a change trend threshold value.