A method for correcting antenna angle deviation based on the collaboration of multiple meteor radars on the same site

Abstract

The present invention provides a method for correcting antenna angle deviation based on the co-location of multiple meteor radars. The method is based on the situation that any two meteor radars among the multiple meteor radars are deployed in close proximity and have overlapping observation periods. After the two meteor radars have accumulated a certain amount of data, the meteors observed by the two are selected, the difference in the corresponding angles of the meteor echoes is counted, and then the angle data is corrected using the statistical mean of Gaussian fitting, so as to obtain and correct the angle deviation of the reference directions of the two meteor radars. The present invention can correct the relative angle deviation of the two meteor radar receiving systems without the need for additional hardware and calibration data. By correcting the angle deviation of the two meteor radar systems, the present invention can qualitatively and quantitatively compare the true consistency and performance difference between the two according to the inverted atmospheric parameters, which is conducive to selecting the most suitable meteor radar model according to the parameters to be measured in the future, and is also conducive to the verification of atmospheric parameters such as tides and gravity waves.

Description

Technical Field

[0001] The invention belongs to the technical field of meteor radars, and in particular relates to a method for correcting antenna angle deviation based on the co-location of multiple meteor radars. Background Art

[0002] The middle and upper atmosphere usually refers to the atmospheric region with an altitude of 60 to 120 kilometers. Although there are no severe weather phenomena such as wind, rain and thunderstorms in the troposphere, the complex physical, chemical and dynamic processes in the atmosphere are closely related to human production and life and aerospace military engineering, so it is of great research value. Because the middle and upper atmosphere is thin, it is difficult to detect using conventional atmospheric detection technology (such as sounding balloons, sounding rockets, etc.). With the emergence of radar remote sensing technology, the detection and research of the middle and upper atmosphere has become possible.

[0003] Meteor radar is one of the main means of detecting the middle and upper atmosphere. Meteor radar uses the radio signals reflected by the meteor trails in the mesosphere-lower thermosphere region (60-120km) to invert the average wind field, temperature, density and other atmospheric parameters of the background atmosphere. Meteor radar can conduct continuous observations in all weather conditions, and has low cost, simple equipment and convenient deployment. The networking of multiple meteor radars can achieve large-scale and refined detection of the middle and upper atmosphere.

[0004] In 1950, Manning et al. of Stanford University in the United States first calculated the upper atmospheric wind field from the radial velocity of meteor trails. Subsequently, the middle and upper atmospheric research group of the University of Adelaide in Australia first used meteor radar to observe the upper atmospheric wind field in 1953, which opened the prelude to meteor radar observation of the dynamic characteristics of the mesosphere-lower thermosphere atmosphere. After that, the development of meteor radar in the field of middle and upper atmospheric detection was relatively slow. From the 1990s to the early 21st century, with the development of radar hardware, especially computer technology, the efficiency of meteor observation has been greatly improved. In 1997, Jones et al. improved the antenna array of the all-sky meteor radar. By spatially distributing the receiving antenna array, a more accurate estimation of the meteor arrival angle can be obtained, while minimizing the mutual coupling between antennas. At present, the EMDR series of all-sky meteor radars of the Australian ATRAD Atmospheric Radar Company are commonly used internationally. They are usually used to detect physical parameters such as horizontal wind field, temperature, density, etc. in the middle and upper atmosphere, as well as astronomical phenomena such as meteor velocity and meteor shower flux. my country's first meteor radar was built in Wuhan in 2002. After nearly 20 years of development, my country has deployed more than 10 meteor radars in Mohe, Beijing, Mengcheng, Kunming and other places, providing important observation data for real-time monitoring of my country's middle and upper atmosphere.

[0005] The middle and upper atmosphere is very clean and has a very low density. It is very difficult to detect this area directly. Generally, this area is detected through remote sensing. Among them, meteor radar is an important detection method of radio remote sensing. System parameters such as radar transmitter power, transmission frequency, and operation mode can affect the detection of meteor echoes. Cross-comparison and verification of observation data from different detection equipment is the main way to determine the accuracy of observation data. However, there are obvious differences in observation principles and equipment operation between different detection equipment. Two meteor radars in the same area with the same principle and similar detection mode, but operating independently, provide a valuable opportunity to verify the reliability of meteor radars and the accuracy of observations.

[0006] The existing meteor radar receiving antenna array consists of five three-unit Yagi antennas, which are often in the shape of 'X', 'T' or 'L' in space. The arrival angle can be determined by the interference array. The arrival angle information includes the zenith angle and the azimuth angle. After the meteor radar is installed, the installer will use external equipment to test and roughly correct the system error to ensure the accuracy of the meteor trail position information. The user will assume that the data is reliable within a certain range, so as to be used for other subsequent inversion work of atmospheric parameters. In fact, in daily observations, due to the influence of human or natural factors, the receiving antenna array will have an angle offset or phase drift, so that the position information of the meteor echo will be biased, which will eventually affect the accuracy of the inversion result and interfere with the judgment of space weather phenomena. In order to reduce the influence of the angle offset of the receiving antenna array, Chen et al. compared the position of the known orbiter recorded by the ground camera with the position recorded by the radar system in 2001, and obtained the phase difference deviation between the receiving channels of the radar system. In 2007, Chen Xiaobo et al. estimated the phase difference deviation between the receiving channels of the meteor radar system by using the geometric relationship between the original data of the meteor radar and the receiving antenna array. However, the existing methods are all aimed at reducing the phase difference deviation between the receiving channels of a single meteor radar system. With the increase in the number of meteor radar deployments, joint observation of multiple meteor radars has become a trend. Calibrating only a single system will still result in angle deviations, because there may be angle deviations between the receiving systems of each two meteor radars. Therefore, it is necessary to estimate the angle deviation between the two systems and correct the reference angle deviation.

[0007] After the existing meteor radar is installed, the installer will conduct a test to obtain some error information, and then input the correction parameters into the data processing software to ensure the accuracy of the meteor trail position information. Generally speaking, the azimuth angle obtained by the receiving matrix is ​​the angle of counterclockwise rotation relative to the due east (north) direction, and the zenith angle is the angle of rotation toward the ground relative to the vertical ground, the direction of the local zenith, and the obtained meteor trail distance is half of the electromagnetic wave signal distance. When the meteor radar is installed and tested, it is relatively simple to correct the distance parameters obtained by the receiving matrix, and the angle will cause the actual reference direction to be not due east (north) and zenith due to the deviation of the reference direction. This is not obvious when a single-station meteor radar measures the atmosphere, but when multiple-station meteor radars measure the atmosphere in coordination, due to the accumulation of the reference direction deviations of the multi-station receiving matrices, the meteor positions calculated by multiple meteor radar systems are inconsistent, and the wind field heights and sizes calculated by the meteor echo parameters are different, which leads to different results of parameters such as tides and gravity waves calculated based on the wind field calculation, which may cause misjudgment of atmospheric physical phenomena. The existing method is to correct the phase difference deviation between the receiving channels of a single meteor radar, and the angles between the reference directions of the receiving arrays of multiple meteor radars are basically ignored. Summary of the invention

[0008] The technical problem to be solved by the present invention is to correct the relative deviation of the reference directions of the receiving antenna arrays of multiple meteor radars so that the multiple meteor radars are networked and cooperate based on the same reference direction, thereby ensuring that the measured atmospheric physical parameters approach the true value.

[0009] In order to solve the above technical problems, the present invention proposes the following technical solutions:

[0010] A method for correcting antenna angle deviation based on the co-location of multiple meteor radars, wherein the multiple meteor radars are respectively used to observe meteors and are relatively close to each other and have a common observation range. The method comprises: firstly, screening meteors observed jointly by any two meteor radars; secondly, taking a difference between the meteor angle data measured by the two screened meteor radars, and performing histogram fitting to obtain a fitting mean, which is identified as the angle between the angle reference directions of the two meteor radars; and finally, adjusting the meteor angle data of one of the two meteor radars or the two meteor radars according to the above fitting mean.

[0011] Furthermore, the screening criteria are: setting a threshold based on the resolution of the meteor radar, selecting meteors whose meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference are all less than the threshold, and at the same time, if one of the meteor radars that meets the above conditions detects more than one meteor, the meteor screened this time will be discarded.

[0012] Furthermore, the method specifically comprises the following steps:

[0013] 1) arbitrarily selecting two meteor radars from the plurality of meteor radars;

[0014] 2) setting screening conditions for multiple parameters of meteors observed by the two meteor radars according to the resolution of the two meteor radars, and screening meteors according to the screening conditions, wherein the multiple parameters include meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference; executing step 3);

[0015] 3) Determine whether the number of meteors before and after the screening is of the same order of magnitude. If so, end the screening and the angular reference directions of the two meteor radars can be considered parallel. If not, execute step 4);

[0016] 4) Based on the screening conditions described in step 2), remove the conditions related to the zenith angle and perform re-screening; execute step 5);

[0017] 5) Determine whether the number of screened meteors has increased in magnitude, if so, execute step 8), if not, execute step 6);

[0018] 6) Based on the screening conditions in step 2), remove the conditions related to the azimuth angle and perform re-screening; execute step 7);

[0019] 7) Determine whether the number of screened meteors has increased in magnitude, if so, execute step 8), if not, determine that the screening has failed, reconsider the screening conditions and execute step 2);

[0020] 8) Difference is made between the meteor angle data of the two meteor radars selected, and a histogram of the difference is obtained; and step 9 is executed;

[0021] 9) Perform Gaussian fitting on the histogram to obtain the fitting mean; execute step 10);

[0022] 10) Correcting the meteor angle data of one of the two or both meteor radars according to the fitting mean; executing step 1).

[0023] The present invention proposes a method for correcting the system angle deviation based on the cooperation of multiple meteor radars in the same place. The method is mainly based on the meteors observed by two meteor radars together. For any two meteor radar systems in the same place, the data of the file with the suffix .met can be used in theory to select the meteors observed by the two together according to certain criteria. The selection criteria are: setting a threshold based on the resolution of the meteor radar, selecting meteors whose meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference are all less than the threshold, and at the same time, if any of the meteor radars that meet the above conditions detects more than one meteor, the selected meteor will be discarded.

[0024] In fact, since there may be a certain angle between the reference directions of the azimuth angles of the two meteor radar systems, the number of screened meteors may be very small or even zero, so the selection method needs to be adjusted. This method first removes the zenith angle screening condition based on the above selection criteria to observe whether the number of screened meteors meets the requirements. If so, proceed to the next step. If not, add the zenith angle condition, remove the azimuth angle condition, and re-screen the jointly observed meteors; then, make a difference between the meteor angle data measured by the two screened meteor radars, perform histogram fitting, and obtain the fitting mean. This mean is identified as the angle between the angle reference directions of the two meteor radar systems; finally, adjust the angle data of one or two meteor radars appropriately according to the fitting mean, and perform angle difference histogram fitting again so that the fitting mean is close to 0, and the angle deviation is corrected at this time. The corrected data can be used for subsequent comparative calculations of atmospheric parameters.

[0025] The present invention has the following advantages and positive effects:

[0026] The method for correcting antenna angle deviation based on the co-location of two meteor radars described in the present invention can accurately measure the deviation of the reference direction of the two meteor radar receiving antennas and correct the angle deviation of the receiving antennas of the two meteor radar systems. By correcting the angle of the receiving antenna, the mean deviation of the reference direction of the two meteor radar receiving systems can be reduced to less than 1°, the two systems can be regarded as arranged in parallel, and the meteor radar system can obtain the atmospheric wind field more accurately and reliably.

[0027] The present invention is based on the situation that two meteor radar systems are deployed in close proximity. No additional hardware support and calibration data are required. Only a few days of data need to be accumulated to correct the azimuth deviation by screening the meteors observed together. The present invention does not waste observation time and can still perform calibration based on historical data.

[0028] Since the deviation of the angle reference direction can be corrected by screening the meteors observed together, the present invention can qualitatively and quantitatively compare the differences between two meteor radars according to the inverted atmospheric parameters, which is conducive to selecting the most suitable meteor radar model according to the parameters to be measured in the future, and is also conducive to the observation and verification of atmospheric tides, atmospheric gravity waves and other mid- and high-level atmospheric dynamic processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 These are the three-day meridional atmospheric wind field maps retrieved by two meteor radars;

[0030] Figure 2 is the correlation coefficient of the atmospheric wind fields of the two meteor radars;

[0031] The dotted lines marked with triangles correspond to the latitudinal components, and the solid lines marked with circles correspond to the longitudinal components.

[0032] Figure 3 is a flow chart of the method of the present invention;

[0033] Figure 4 The statistical histogram and fitting curve of the azimuth difference between two meteor radars;

[0034] Among them, the left one is the result after correction, and the right one is the result before correction;

[0035] Figure 5 The three-day meridional atmospheric wind field map retrieved by two meteor radars after correction;

[0036] Figure 6 is the corrected wind field correlation coefficient;

[0037] The dotted lines marked with triangles correspond to the latitudinal components, and the solid lines marked with circles correspond to the longitudinal components. DETAILED DESCRIPTION

[0038] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0039] The present invention relates to a method for correcting antenna angle deviation based on the co-location cooperation of multiple meteor radars, wherein the multiple meteor radars are respectively used for observing meteors, and the method comprises: firstly, screening meteors observed by any two meteor radars together; secondly, taking a difference between meteor angle data measured by the two screened meteor radars, and performing histogram fitting to obtain a fitting mean, wherein the fitting mean is identified as the angle between the angle reference directions of the two meteor radars; and finally, adjusting the meteor angle data of one of the two meteor radars or the two meteor radars according to the fitting mean.

[0040] The specific embodiment of the present invention mainly uses data measured by two meteor radars at Kunming Station with operating frequencies of 37.5 MHz and 53.1 MHz respectively from November 1, 2013 to December 31, 2014. Figure 1 The meridional wind field is obtained by inverting the data of two meteor radars from November 8 to 10, 2013. It can be seen that due to the different working frequencies, the wind field range that can be inverted is different, and there are also differences in the wind field values ​​within the inversion range that overlaps the two. From this, we can roughly draw a conclusion that the observation height range of the 53.1MHz meteor radar is lower than that of the 37.5MHz, and the inverted wind field is similar to that of the 37.5MHz, but the absolute value of the wind speed is lower than that of the 37.5MHz. Figure 2The correlation coefficients of the wind fields of the two meteor radars at different heights for 14 months are shown in Figure 2. The correlation coefficients are roughly in the range of [0.75, 0.95], indicating that the two meteor radars have a high correlation. The correlation coefficient of the latitudinal component hardly changes with height, while the correlation coefficient of the longitudinal component changes significantly with height, indicating that there may be errors between the two meteor radar systems.

[0041] Figure 3 It is a flow chart of the method of the present invention, which specifically includes the following steps:

[0042] 1) arbitrarily selecting two meteor radars from the plurality of meteor radars;

[0043] 2) setting screening conditions for multiple parameters of meteors observed by the two meteor radars according to the resolution of the two meteor radars, and screening meteors according to the screening conditions, wherein the multiple parameters include meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference; executing step 3);

[0044] 3) Determine whether the number of meteors before and after the screening is of the same order of magnitude. If so, end the screening and the angular reference directions of the two meteor radars can be considered parallel. If not, execute step 4);

[0045] 4) Based on the screening conditions described in step 2), remove the conditions related to the zenith angle and perform re-screening; execute step 5);

[0046] 5) Determine whether the number of screened meteors has increased in magnitude, if so, execute step 8), if not, execute step 6);

[0047] 6) Based on the screening conditions in step 2), remove the conditions related to the azimuth angle and perform re-screening; execute step 7);

[0048] 7) Determine whether the number of screened meteors has increased in magnitude, if so, execute step 8), if not, determine that the screening has failed, reconsider the screening conditions and execute step 2);

[0049] 8) Difference is made between the meteor angle data of the two meteor radars selected, and a histogram of the difference is obtained; and step 9 is executed;

[0050] 9) Perform Gaussian fitting on the histogram to obtain the fitting mean; execute step 10);

[0051] 10) Correcting the meteor angle data of one of the two or both meteor radars according to the fitting mean; executing step 1).

[0052] According to the process, it can be determined that the main angle deviation in the specific embodiment of the present invention is the azimuth deviation. The azimuth difference histogram and fitting curve obtained by screening the commonly observed meteor statistics are shown in Figure 4 , the right side is the azimuth difference obtained by statistics before the angle deviation is corrected, and the left side is the result after correction. As can be seen from the figure, the mean of the Gaussian fitting curve before correction is 17.159°, so at this time, the azimuth data in the 53.1MHz meteor radar data is subtracted by 17.159°, and the azimuth difference fitting is performed again, and the Gaussian fitting mean is reduced to -0.005°. Figure 5 The meridional wind fields of the two meteor radar systems from November 8 to 10, 2013, obtained by inverting the corrected azimuth data and other time position data, can be similar to the previous conclusions, but the wind fields are closer. As a quantitative comparison, the correlation coefficients before and after the correction are also compared, see Figure 6 .from Figure 6 It can be seen that in the range of [76 km, 94 km], the correlation coefficients of the two are greater than 0.9. In the range of 80-92 km where there are many meteors, the correlation coefficient of the meridional component is ≥0.95, indicating that the two meteor radar systems are actually very consistent. In addition, the correlation coefficients of the meridional and latitudinal components have the same trend of changing with altitude, indicating that this method effectively solves the systematic error caused by the reference angle deviation of the two meteor radar systems.

[0053] The above specific embodiments show that, except for the large difference in the detection height range, the wind fields inverted by the two meteor radar systems with different frequencies are almost the same, which illustrates the reliability of the inversion results of the two systems and can be used for subsequent research on tides, gravity waves, etc.

Claims

1. A method for correcting antenna angle deviation based on the cooperation of multiple meteor radars in the same location, wherein the multiple meteor radars are used to observe meteors respectively and are close to each other and have a common observation range. It is characterized in that The method comprises: firstly, screening meteors observed by any two meteor radars; secondly, taking a difference between the meteor angle data measured by the two screened meteor radars, and performing histogram fitting to obtain a fitting mean, wherein the fitting mean is identified as the angle between the angle reference directions of the two meteor radars; finally, adjusting the meteor angle data of one or both of the two meteor radars according to the fitting mean; the screening criterion is: setting a threshold value based on the resolution of the meteor radar, selecting meteors whose meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference are all less than the threshold value, and at the same time, if more than one meteor is detected by one of the meteor radars that meets the above conditions, the screened meteor will be discarded; The method specifically comprises the following steps: 1) arbitrarily selecting two meteor radars from the plurality of meteor radars; 2) setting screening conditions for multiple parameters of meteors observed by the two meteor radars according to the resolutions of the two meteor radars, and screening meteors according to the screening conditions, wherein the multiple parameters include meteor echo time difference, echo height difference, zenith angle difference and azimuth angle difference; executing step 3); 3) Determine whether the number of meteors before and after the screening is of the same order of magnitude. If so, end the screening and the angular reference directions of the two meteor radars can be considered parallel. If not, proceed to step 4); 4) Based on the screening conditions described in step 2), remove the conditions related to the zenith angle and perform re-screening; proceed to step 5); 5) Determine whether the number of screened meteors has increased in magnitude, if so, execute step 8), if not, execute step 6); 6) Based on the screening conditions in step 2), remove the conditions related to the azimuth angle and perform a new screening; execute step 7); 7) Determine whether the number of screened meteors has increased in magnitude. If so, execute step 8). If not, determine that the screening has failed, reconsider the screening conditions and execute step 2); 8) Difference is made between the meteor angle data of the two selected meteor radars, and a histogram of the difference is obtained; go to step 9); 9) Perform Gaussian fitting on the histogram to obtain the fitting mean; proceed to step 10); 10) Correcting the meteor angle data of one of the two or both meteor radars according to the fitting mean; executing step 1).

Images