Method and system for identifying migrated insects in air based on insect radar and high-altitude lamp

Abstract

The invention provides a method and a system for identifying migrated insects in the air based on an insect radar and a high-altitude lamp. The method comprises the following steps of: 1, receiving four biological parameters of flapping frequency, body length, body width and body weight of each migrated insect in the air monitored by the insect radar in real time; 2, the received four biological parameters are input into an optimal insect type classification model, the type of each aerial migratory fly insect is obtained, and the optimal insect type classification model is determined according to the counted types of the insects trapped by the high-altitude lamp within the recent T3 time. According to the invention, real-time accurate identification of insect individual types can be realized for target insects captured by an insect radar; the real insect condition information is monitored in real time according to the received data of the insects trapped by the high-altitude lamp, the optimal insect type classification model is determined according to the real insect condition information, it is guaranteed that the optimal insect type classification model is matched with the real insect condition, and good expansibility and broad-spectrum adaptability are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of insect migration, and in particular to a method and system for identifying aerial migratory insects based on insect radar and high-altitude lights. Background Art

[0002] Migratory pests are characterized by sudden onset, rapid spread, and difficulty in control. High-resolution insect radar, with its wide range, all-weather capabilities, and high spatiotemporal resolution, has become an important tool for real-time monitoring of aerial migratory insects. However, the radar's lack of direct species recognition capabilities limits its effectiveness in accurate identification and species-specific monitoring. Accurate classification of radar target insects is one of the core technical bottlenecks that must be overcome in building an efficient migratory pest monitoring and early warning system. In recent years, insect radar has made significant progress in identifying aerial migratory insect species, and a variety of identification methods have been developed.

[0003] For example, the Chinese invention patent application with application number 202410981019.2 discloses a method and system for identifying target insects based on insect radar. The method includes the following steps: obtaining the radar echo signal of the insect based on insect radar monitoring technology; trapping the insect based on insect trapping technology, comparing the number of the insect radar echo signal and the number of trapped insects, and obtaining the target insects and the number of target insect species at the radar echo point; measuring the flight ability and measurement quality of the target insect; analyzing existing meteorological data and the radar echo signal of the target insect to obtain the target insect's own flight speed and calculation quality; performing cluster analysis on the target insect's flight ability and measurement quality with its own flight speed and calculation quality, and identifying the target insect species based on the number of target insect species. This invention integrates insect radar monitoring technology and insect trapping technology to identify target insects, solving the problem that it is difficult to identify target insects by relying solely on radar echo point analysis technology. However, it has the problems of complex operation and inability to achieve real-time identification of migratory insect species in the air.

[0004] For example, Chinese invention patent application number 202510104597.2 discloses a method for insect species identification based on a multidimensional radar electromagnetic scattering feature sequence. This method constructs a multidimensional feature sequence based on a fully polarized radar scattering matrix and uses an insect recognition neural network for feature learning. The insect recognition neural network extracts spatial features of different scales through multi-scale convolution and effectively captures the temporal dependencies in the feature sequence through state updates based on time series inputs. Ultimately, insect species are identified based on multi-scale spatial features and their temporal dependencies. This invention effectively improves the accuracy and robustness of insect species identification and has broad application prospects. It is particularly suitable for insect migration monitoring, agricultural pest control, and ecological research. However, the types of migratory pests it can identify are limited and its scalability is insufficient.

[0005] Another example is the Chinese invention patent application number 202510111754.2, which discloses an intelligent identification method for migratory insects based on insect-attracting light-assisted radar database construction. The method first integrates insect monitoring data from insect-attracting light and radar during the same nighttime period, and establishes an insect dataset by matching the two. This effectively overcomes the data collection difficulties of traditional insect dataset database construction methods. Secondly, a CL-GC-Swin Transformer network model is proposed. By introducing a contrastive learning module and a global context block into the Swin Transformer and using the insect's full polarization time-frequency feature map for training, the network's ability to represent insect time-frequency feature maps is significantly enhanced, effectively improving the accuracy of insect species identification. This invention directly utilizes the flight characteristics of insects for identification, no longer relying on complex scattering feature calculations and precise radar calibration, greatly reducing the requirements for radar system accuracy and calibration, and improving the adaptability and robustness of identification. However, it is only suitable for large-scale insect swarm identification and cannot achieve individual species identification. Summary of the Invention

[0006] In order to solve at least one of the above technical problems, the present invention provides a method and system for identifying aerial migratory insects based on insect radar and high-altitude lights, which can realize real-time and accurate identification of individual species of target insects captured by the insect radar.

[0007] A first aspect of the present invention provides a method for identifying aerial migratory insects based on insect radar and high-altitude lights, comprising: Step 1: Receive four biological parameters of each migratory insect, including wing beat frequency, body length, body width, and weight, monitored in real time by the insect radar; Step 2: Input the received four biological parameters into the optimal insect species classification model to obtain the species of each aerial migratory insect; In step 2, the optimal insect species classification model is determined by repeatedly performing the following steps: Step 201: receiving data on insects trapped by a high-altitude light according to a set first time interval T1; Step 202: Based on the received data on insects trapped by the high-altitude lights, the species of insects trapped by the high-altitude lights within the most recent time period T3 are counted according to the set second time interval T2, and compared with the insect species counted in the previous time. If the insect species has changed, step 203 is executed; if the insect species has not changed, step 204 is executed. Step 203: updating the optimal insect species classification model; Step 204: Do not update the optimal insect species classification model.

[0008] Preferably, in step 1, the industrial control computer of the insect radar performs real-time inversion calculation of four biological parameters, namely, wing beat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar.

[0009] Preferably, in any of the above solutions, in step 201, the data on the insects trapped by the high-altitude lamp includes data on the species of each trapped insect.

[0010] Preferably, in any of the above solutions, in step 201, the data of insects trapped by the high-altitude light includes image data of each trapped insect.

[0011] Preferably, in any of the above schemes, when the data of the insects trapped by the high-altitude lamp includes image data of each trapped insect, in step 202, the type of each insect is identified based on the image data.

[0012] Preferably, in any of the above solutions, the value of the second time interval T2 is not less than the value of the first time interval T1.

[0013] Preferably, in any of the above solutions, the value of T3 is not less than 48 hours.

[0014] Preferably, in any of the above solutions, in step 203, updating the optimal insect species classification model includes: Step 2031: Based on the insect species most recently counted in step 202, corresponding insect biological parameter data are read from the insect biological parameter database to form a sample set; Step 2032: Using the sample set, apply XGBoost, random forest and BP neural network classification algorithms to respectively construct three insect species classification models, and select the insect species classification model with the highest F1 score as the optimal insect species classification model.

[0015] Preferably, in any of the above schemes, in step 2031, each piece of insect biological parameter data in the insect biological parameter database includes the insect's wing flapping frequency, body length, body width, weight and species.

[0016] Any of the above schemes is preferably that in step 2031, after reading the corresponding insect biological parameter data from the insect biological parameter database based on the latest statistical insect species, the four data items of wing flapping frequency, body length, body width and weight in the read insect biological parameter data are detected and eliminated for outliers and standardized to form a sample set.

[0017] Preferably, in any of the above solutions, in step 2032, the sample set is divided into a training set and a test set according to a set ratio.

[0018] Preferably, in any of the above schemes, in step 2, if the optimal insect species classification model is updated, the four biological parameters of wingbeat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar received during the past T2 time are input into the latest optimal insect species classification model, and the species of each aerial migratory insect is reclassified.

[0019] A second aspect of the present invention provides an aerial migratory insect identification system based on insect radar and high-altitude lights, which is used to execute the aerial migratory insect identification method based on insect radar and high-altitude lights.

[0020] Preferably, the aerial migratory insect identification system based on insect radar and high-altitude light is communicatively connected to the high-altitude light and the insect radar monitoring station respectively.

[0021] Preferably, in any of the above schemes, the high-altitude light is arranged within the L distance range of the insect radar monitoring station.

[0022] The method and system for identifying aerial migratory insects based on insect radar and high-altitude lights of the present invention have the following beneficial effects: 1. Ability to accurately identify individual species of target insects captured by insect radar in real time; 2. Monitor the actual insect situation in real time based on the insect data received from high-altitude light traps, and determine the optimal insect species classification model based on the actual insect situation information, ensuring that the optimal insect species classification model matches the actual insect situation, with good scalability and broad-spectrum adaptability; 3. The identified target insect species can be combined with the insect biological parameters and flight parameters calculated by the radar industrial control computer to provide data support for the construction of an agricultural and forestry migratory pest monitoring and early warning system, with good prospects for implementation; 4. It can provide accurate data support for the simulation of migration trajectories of different types of insects, population dynamics monitoring, and the formulation of prevention and control strategies, which is conducive to improving the scientificity and practicality of insect monitoring and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The figure is a flow chart of a preferred embodiment of the method for identifying aerial migratory insects based on insect radar and high-altitude lights according to the present invention.

[0024] Figure 2 According to the method for identifying aerial migratory insects based on insect radar and high-altitude lights of the present invention, Figure 1 A schematic diagram of a process for determining an optimal insect species classification model in the illustrated embodiment.

[0025] Figure 3 According to the method for identifying aerial migratory insects based on insect radar and high-altitude lights of the present invention, Figure 1A schematic diagram of the process of updating the optimal insect species classification model in the illustrated embodiment. DETAILED DESCRIPTION

[0026] In order to better understand the present invention, the present invention is described in detail below with reference to specific embodiments.

[0027] Example 1 like Figure 1 As shown, a method for identifying aerial migratory insects based on insect radar and high-altitude lights includes: Step 1: Receive four biological parameters of each migratory insect, including wing beat frequency, body length, body width, and weight, monitored in real time by the insect radar; Step 2: Input the received four biological parameters into the optimal insect species classification model to obtain the species of each aerial migratory insect; Among them, such as Figure 2 As shown, in step 2, the optimal insect species classification model is determined by repeatedly performing the following steps: Step 201: receiving data on insects trapped by a high-altitude light according to a set first time interval T1; Step 202: Based on the received data on insects trapped by the high-altitude lights, the species of insects trapped by the high-altitude lights within the most recent time period T3 are counted according to the set second time interval T2, and compared with the insect species counted in the previous time. If the insect species has changed, step 203 is executed; if the insect species has not changed, step 204 is executed. Step 203: updating the optimal insect species classification model; Step 204: Do not update the optimal insect species classification model.

[0028] Preferably, in step 1, what is received are four biological parameters of wing beat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar, which are inverted and calculated in real time by the industrial control computer of the insect radar.

[0029] Further preferably, in step 201, the data on insects trapped by the high-altitude light includes species data of each trapped insect, or image data of each insect. If the data on insects trapped by the high-altitude light includes image data of each trapped insect, in step 202, the species of each insect is identified based on the image data. It should be noted that determining the species data of each trapped insect after trapping by the high-altitude light can be accomplished using existing technologies, and identifying the species of each insect based on image data of each insect trapped by the high-altitude light can also be accomplished using existing technologies, which are not described in detail in this application.

[0030] In this embodiment, the second time interval T2 is preferably not less than the first time interval T1. This ensures that the latest insect information is used each time the species of trapped insects are counted and whether the species has changed. The value of T3 can be set as needed and is generally not less than 48 hours.

[0031] like Figure 3 As shown, in step 203, updating the optimal insect species classification model includes: Step 2031: Based on the insect species most recently counted in step 202, corresponding insect biological parameter data are read from the insect biological parameter database to form a sample set; Step 2032: Using the sample set, apply XGBoost, random forest and BP neural network classification algorithms to respectively construct three insect species classification models, and select the insect species classification model with the highest F1 score as the optimal insect species classification model.

[0032] In step 2031, each piece of insect biological parameter data in the insect biological parameter database includes the insect's wing flapping frequency, body length, body width, weight and species; according to the latest statistical insect species, the corresponding insect biological parameter data is read from the insect biological parameter database, and the four data items of wing flapping frequency, body length, body width and weight in the read insect biological parameter data are detected and eliminated for outliers and standardized to form a sample set.

[0033] In step 2032, the sample set is divided into a training set and a test set according to a set ratio. It should be noted that the application of XGBoost, random forest and BP neural network three classification algorithms to construct three insect species classification models can be achieved using existing technologies and will not be described in detail in this application.

[0034] In step 2, if the optimal insect species classification model is updated, the four biological parameters of wingbeat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar within the past T2 time are input into the latest optimal insect species classification model, and the species of each aerial migratory insect is reclassified.

[0035] It should be understood that the process of counting insect species and updating the optimal insect species classification model according to the T2 time interval will not affect the process of receiving data on insects trapped by high-altitude lights according to the T1 time interval; the process of receiving the four biological parameters of wingbeat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar will not be affected by other processes.

[0036] It should be noted that the insect bio-parameter database covers all known migratory insect species, including some non-migratory insect species. While high-altitude lights trap migratory insects, non-migratory insects inevitably fall into them. Considering that the update of the optimal insect species classification model in step 203 relies on data in the insect bio-parameter database, the counting of insect species trapped by the high-altitude lights in step 202 is limited to the insect species recorded in the insect bio-parameter database. This does not affect the optimal insect species classification model.

[0037] It should be further explained that the initial optimal insect species classification model can be determined by the insect species initially counted within the T3 time, or it can be directly determined based on all the data in the insect biological parameter database.

[0038] Example 2 A system for identifying migratory insects using an insect radar and a high-altitude light is used to implement the method for identifying migratory insects using an insect radar and a high-altitude light. The system is communicatively connected to the high-altitude light and an insect radar monitoring station. The high-altitude light is located within a distance L from the insect radar monitoring station to ensure consistency between the insect species trapped by the high-altitude light and those detected by the insect radar, thereby ensuring accurate identification of the species of each migratory insect detected by the insect radar.

[0039] Example 3 Prior to the implementation of this example, the insect biological parameter database already contained nearly 15,000 data items for 54 species of migratory insects, each including data on the insect's wingbeat frequency, body length, body width, weight, and species. The data in this insect biological parameter database was sourced from the following: starting in 2023, healthy, active insects were selected daily from insects trapped by multiple insect trapping devices (e.g., high-altitude lights, traps, etc.) in Area A. After species identification, their wingbeat frequency, body length, body width, and weight were measured in the laboratory. These four key biological parameters, wingbeat frequency, body length, body width, weight, and species, were then entered into the database. It should be noted that the data source for the insect biological parameter database described in this example is not restrictive; the database can also include data from insects trapped by intelligent high-altitude lights in other areas, or data on insects identified through other means. The database supports continuous entry of insect data to enrich the data volume.

[0040] In this embodiment, a high-altitude light is located within 300 meters (i.e., L is set to 300 meters) of a high-resolution insect radar monitoring station in area A. This high-altitude light can take photos of captured insects at a set first time interval T1, perform species identification based on the captured photos using a trained neural network, and transmit the identification results (the species and number of insects in the photos) via a communication connection to an aerial migratory insect identification system based on the insect radar and high-altitude light. It should be understood that the aerial migratory insect identification system based on the insect radar and high-altitude light receives data on insects captured by the high-altitude light at the set first time interval T1, including data on the species of each captured insect. In this embodiment, preferably, T1 is set to 2 minutes.

[0041] After receiving the insect species data transmitted by the high-altitude lights, the aerial migratory insect identification system based on insect radar and high-altitude lights calculates the number of insect species captured by the high-altitude lights within the most recent time period T3, based on the data and at a set second time interval T2. In this embodiment, T2 is set to 1 hour, and T3 is set to 72 hours. For example, if at 21:00 on July 18, 2024, the number of insect species captured by the high-altitude lights within the last 72 hours, including 29 species listed in the insect biological parameter database, is compared with the data collected at 20:00 on July 18, 2024, and a change in insect species is detected (in this embodiment, an increase in insect species), the optimal insect species classification model is updated.

[0042] The specific method for updating the optimal insect species classification model is as follows.

[0043] First, based on the 29 insect species counted at 21:00 on July 18, 2024, all data for these 29 insect species were retrieved from the insect biological parameter database. In this embodiment, a total of 8509 insect biological parameter data items were retrieved. It should be understood that each insect biological parameter data item retrieved includes the insect's wingbeat frequency, body length, body width, weight, and species. After outlier detection and standardization, the 8509 insect biological parameter data items, including wingbeat frequency, body length, body width, and weight, were all processed. A sample set of 8189 insect biological parameter data items was obtained. It should be noted that, in this embodiment, when performing outlier detection and removal, the quartile method is used, that is, if a certain data in a certain data is higher than its upper boundary value upper=Q3+1.5×(Q3-Q1) or lower than its lower boundary value lower=Q1-1.5×(Q3-Q1), then the data is removed, where Q1 and Q3 are the 0.25 quantile and 0.75 quantile of the quartile of a certain data respectively; when performing standardization, Z-score standardization is used, and the standardization process can eliminate the impact of the dimensional differences of each data and balance the contribution of each data. It should be further noted that other methods disclosed in the prior art can also be used for outlier detection and removal and standardization, which are not limited in this application.

[0044] Then, using the sample set, it was divided into a training set and a test set according to a set ratio. Three classification algorithms, XGBoost, random forest, and BP neural network, were used to construct three insect species classification models, respectively. The insect species classification model with the highest F1 score was selected as the optimal insect species classification model. It should be noted that in this embodiment, the sample set was divided into a training set and a test set according to a ratio of 7:3. The four biological parameters of the insects, namely, wingbeat frequency, body length, body width, and body weight, were used as the input of the model, and the insect species was used as the output of the model. In order to improve the performance of the model, a hyperparameter grid search optimization was performed on each model to obtain the optimal hyperparameter combination and the model accuracy performance on the test set. In this embodiment, for the 29 insect species counted at 21:00 on July 18, 2024, the F1 scores of the three insect species classification models constructed using the three classification algorithms of XGBoost, random forest and BP neural network were 0.912, 0.874 and 0.843 respectively. Therefore, the insect species classification model constructed using XGBoost was selected as the optimal insect species classification model, and was used to identify the species of each aerial migratory insect monitored by the insect radar at 21:00 on July 18, 2024 and thereafter, until it was detected again that the insect species trapped by the high-altitude lights in the last 72 hours had changed compared with the insect species counted last time, and the optimal insect species classification model was updated.

[0045] In this embodiment, after 21:00 on July 18, 2024, and again at 23:00 on July 20, 2024, the species of insects trapped by the high-altitude lights in the last 72 hours have changed compared with the insect species counted at 22:00 on July 20, 2024. The optimal insect species classification model is updated based on the species of insects trapped by the high-altitude lights in the last 72 hours detected at 23:00 on July 20, 2024, and the updated optimal insect species classification model is used to identify the species of each migratory insect in the air monitored by the insect radar at 23:00 on July 20, 2024 and thereafter until the optimal insect species classification model is updated again.

[0046] It should be noted that the insect species trapped by the high-altitude lights at 23:00 on July 20, 2024 have changed compared with the insect species counted last time (22:00 on July 20, 2024). This shows that the species of migratory insects have changed between 22:00 and 23:00 on July 20, 2024. During this period, the optimal insect species classification model has not been updated based on the insect species counted at 23:00 on July 20, 2024. That is, in this period, During this time period, the optimal insect species classification model, determined based on insect species counted at 21:00 on July 18, 2024, is still used. This model may be inaccurate, so the optimal insect species classification model, updated based on insect species counted at 23:00 on July 20, 2024, is used to re-identify the species of each migratory insect detected by insect radar within the past T2 time period (1 hour, i.e., between 22:00 and 23:00 on July 20, 2024). It should be understood that once the optimal insect species classification model is updated, it is necessary to re-identify the species of each migratory insect detected by insect radar within the past T2 time period using the updated optimal insect species classification model.

[0047] The insect radar monitoring station is equipped with a radar industrial computer. This computer can inversely calculate in real time the four biological parameters of each migratory insect detected by the insect radar: wingbeat frequency, body length, body width, and weight. The computer then transmits these calculated parameters to the migratory insect identification system based on the insect radar and high-altitude lights. The system then inputs the four biological parameters into the optimal insect species classification model to determine the species of each migratory insect. It should be noted that the computer can also inversely calculate in real time the flight parameters of each migratory insect detected by the insect radar, such as flight altitude, flight direction, vertical speed, and horizontal speed. By integrating the target insect species identified by the system with the biological and flight parameters calculated by the computer, the computer can provide data support for the construction of a monitoring and early warning system for migratory pests in agriculture and forestry. The computer can also provide accurate data support for simulating the migration trajectories of different insect species, monitoring population dynamics, and formulating prevention and control strategies, thereby enhancing the scientific and practical nature of insect monitoring and early warning systems.

[0048] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the above embodiments describe the present invention in detail, those skilled in the art should understand that they can modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein, and these replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A method for identifying aerial migratory insects based on insect radar and high-altitude lights, comprising: Step 1: Receive four biological parameters of each migratory insect, including wing beat frequency, body length, body width, and weight, monitored in real time by the insect radar; Step 2: Input the received four biological parameters into the optimal insect species classification model to obtain the species of each aerial migratory insect; It is characterized in that: in step 2, the optimal insect species classification model is determined by repeatedly performing the following steps: Step 201: receiving data on insects trapped by a high-altitude light according to a set first time interval T1; Step 202: Based on the received data on insects trapped by the high-altitude lights, the species of insects trapped by the high-altitude lights within the most recent time period T3 are counted according to the set second time interval T2, and compared with the insect species counted in the previous time. If the insect species has changed, step 203 is executed; if the insect species has not changed, step 204 is executed. Step 203: updating the optimal insect species classification model; Step 204: Do not update the optimal insect species classification model.

2. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, characterized in that: In step 1, the industrial control computer of the insect radar performs real-time inversion calculations on the four biological parameters of wingbeat frequency, body length, body width and weight of each migratory insect monitored by the insect radar in real time.

3. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, characterized in that: In step 201, the data of insects trapped by the high-altitude lights includes data on the species of each trapped insect.

4. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, wherein: In step 201, the data of the insects trapped by the high-altitude light includes image data of each trapped insect. In step 202, the species of each insect is identified based on the image data.

5. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, wherein: The second time interval T2 is not less than the first time interval T1, and the value of T3 is not less than 48 hours.

6. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, characterized in that: In step 203, the optimal insect species classification model is updated, including: Step 2031: Based on the insect species most recently counted in step 202, corresponding insect biological parameter data are read from the insect biological parameter database to form a sample set; Step 2032: Using the sample set, apply XGBoost, random forest and BP neural network classification algorithms to respectively construct three insect species classification models, and select the insect species classification model with the highest F1 score as the optimal insect species classification model.

7. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 6, characterized in that: Each piece of insect biological parameter data in the insect biological parameter database includes the insect's wing flapping frequency, body length, body width, weight and species; in step 2031, according to the latest statistical insect species, the corresponding insect biological parameter data is read from the insect biological parameter database, and the four data items of wing flapping frequency, body length, body width and weight in the read insect biological parameter data are detected and eliminated for outliers and standardized to form a sample set.

8. The method for identifying aerial migratory insects based on insect radar and high-altitude lights as claimed in claim 1, wherein: In step 2, if the optimal insect species classification model is updated, the four biological parameters of wingbeat frequency, body length, body width and weight of each aerial migratory insect monitored in real time by the insect radar within the past T2 time are input into the latest optimal insect species classification model, and the species of each aerial migratory insect is reclassified.

9. A system for identifying migratory insects in the air based on insect radar and high-altitude lights, characterized by: Used to execute the aerial migratory insect identification method based on insect radar and high-altitude lights as described in any one of claims 1-8.

10. The aerial migratory insect identification system based on insect radar and high-altitude light according to claim 9, characterized in that: It is respectively connected to the high-altitude light and the insect radar monitoring station for communication. The high-altitude light is arranged within the L distance range of the insect radar monitoring station.

Images