Method and system for establishing a communication link in a many-to-many fashion

Abstract

The application provides a method and system for establishing a communication link in a many-to-many mode, which performs grayscale, smoothing and histogram equalization on an environmental image, extracts texture features and then performs normalization, generates an SVM classification model for identifying a peer device, takes a position of the peer device as a scanning area, acquires a beacon light spot image in the scanning area, performs grayscale, smoothing and histogram equalization on the spot image, performs secondary otsu adaptive segmentation and morphological processing, segments the spot and calculates a centroid thereof, takes the centroid of the spot as an alignment point, generates an optimal scanning path based on the alignment points of the peer devices, performs dynamic scanning based on the scanning path under the premise of a unified clock, completes automatic alignment, and performs adaptive alignment error compensation in real time. The application can still stably and reliably, omnidirectionally and flexibly establish a communication link in a many-to-many mode under complex and changeable conditions, and realizes concurrent communication of multiple targets.

Description

Technical Field

[0001] The present invention relates to the technical field of establishing communication links for space laser terminals, and in particular to a method and system for establishing many-to-many communication links. Background Art

[0002] Space laser communication technology uses air as the medium and lasers as the carrier for data transmission. With its abundant bandwidth resources, high transmission rates, large communication capacity, strong anti-interference capabilities, and excellent confidentiality, space laser communication technology is becoming an important technology in the fields of drone formations, ship formations, satellite communications, and intelligent transportation.

[0003] Automatic scanning, capture, and tracking technologies are key components of laser communication terminals. They are used to locate and quickly capture the long-distance transmitted beacon light emitted by the other terminal device, ensuring that the two terminals can automatically track and lock onto each other while in motion, establishing and maintaining communication links. The point-to-point and point-to-many communication methods of laser communication terminals cannot meet the needs of complex and changing communication environments and situations with multiple targets communicating concurrently. There is a need for a method and system for establishing many-to-many communication links that can achieve comprehensive and flexible multi-target concurrent communication while ensuring stability and reliability. Summary of the Invention

[0004] The present invention provides a method and system for establishing a many-to-many communication link, which solves the problem in the prior art of "being unable to stably, reliably, comprehensively and flexibly establish a many-to-many communication link".

[0005] The present invention solves the technical problem through the following technical solutions:

[0006] A method for establishing a many-to-many communication link comprises the following steps:

[0007] (1) For any laser communication terminal, determine the scanning area and alignment point

[0008] (11) Determine the scanning area

[0009] (111) The laser communication terminal currently rotates and scans to obtain an environmental image;

[0010] (112) Preprocessing the environment image to obtain a preprocessed environment image; the environment image preprocessing includes grayscale conversion, smoothing, and histogram equalization;

[0011] (113) extracting texture features from the environment preprocessing image and performing normalization processing to obtain a normalized LBP texture feature map;

[0012] (114) generating an SVM classification model using the normalized LBP texture feature map as a feature vector;

[0013] (115) identifying a peer device using the SVM classification model, and using the location of the peer device as a scanning area;

[0014] (12) Determine the alignment point

[0015] (121) acquiring a beacon light spot image in the scanning area determined in step (1);

[0016] (122) Preprocessing the light spot image to obtain a light spot preprocessed image; the light spot image preprocessing includes grayscale, smoothing, and histogram equalization;

[0017] (123) performing secondary OTSU adaptive segmentation on the spot preprocessing image to obtain a binary image;

[0018] (124) performing morphological processing on the binary image to segment the light spots; calculating the centroid of the light spots; using the centroid as an alignment point; recording each alignment point, the corresponding received power, and the corresponding code disk angle;

[0019] (2) Laser communication terminals dynamically establish communication links from one to many

[0020] (21) Synchronizing the clocks of each laser communication terminal; for any laser communication terminal, generating a scanning path based on the alignment points of each opposite terminal device;

[0021] (22) Dynamic scanning is performed according to the scanning path under the same clock, and automatic alignment is performed based on the corresponding received power and code disk angle, and adaptive alignment error compensation is performed in real time.

[0022] Furthermore, in step (112), the grayscale conversion is to calculate the grayscale value of each pixel:

[0023] G=0.299*R+0.587*G+0.114*B, where R is the red channel pixel value, G is the green channel pixel value, B is the blue channel pixel value, and G is the grayscale value of the environment image.

[0024] Furthermore, in step (112), the environment image is smoothed using Gaussian filtering:

[0025] The environment image is smoothed by presetting the Gaussian kernel size and variance, and the smoothing effect is improved by adjusting the variance. When the smoothing effect reaches the preset target, a convolution operation is performed on each pixel to obtain a smoothed image.

[0026] Furthermore, in step (112), histogram equalization is performed on the smoothed image:

[0027] The probability density function of the gray value of each pixel is calculated, and each probability density function is accumulated to obtain a cumulative distribution function. The smoothed image is subjected to gray value mapping based on the cumulative distribution function to obtain an environmental preprocessing image.

[0028] Furthermore, in step (113), texture features are extracted from the environment preprocessed image.

[0029] The steps are:

[0030] (1131) Define a 3*3 texture window;

[0031] (1132) taking pixels in a non-boundary pixel area of ​​the environment pre-processed image according to a texture window size;

[0032] (1133) using the grayscale value of the center pixel in the texture window as a grayscale threshold, traversing eight pixels outside the center pixel in the texture window in a clockwise or counterclockwise manner, marking the current pixel as 1 when the grayscale value of the current pixel is greater than the grayscale threshold, otherwise marking the current pixel as 0, thereby obtaining an 8-bit binary number;

[0033] (1134) traverse the non-boundary pixel area in the environmental preprocessing image where texture feature calculation is not performed, and execute steps (1132) and (1133);

[0034] (1135) Convert each obtained 8-bit binary number into a decimal number, and each decimal number is the LBP texture feature map of the environment preprocessing image.

[0035] Furthermore, in step (113), the LBP texture feature map is normalized:

[0036] The number of occurrences of each decimal number in the LBP texture feature map is counted to obtain an LBP histogram; the horizontal axis of the LBP histogram is each decimal number, and the vertical axis of the LBP histogram is the number of occurrences of the corresponding decimal number; for each decimal number on the horizontal axis of the LBP histogram, the decimal number is divided by the sum of the decimal numbers to obtain a corresponding new value, and the decimal number in the original position is replaced by the new value to obtain a normalized LBP texture feature map.

[0037] Furthermore, in step (124), the centroid of the light spot is calculated:

[0038] Where x c 、y c is the coordinate of the centroid of the light spot, x and y are the pixel coordinates associated with the centroid of the light spot, and I(x, y) is the pixel value under the x and y coordinates.

[0039] Furthermore, in step (22), each laser communication terminal performs dynamic scanning at a rotation speed of 720 rpm according to the scanning path at the same clock.

[0040] A system for establishing a many-to-many communication link by executing the above-mentioned method for establishing a many-to-many communication link comprises three or more laser communication terminals; each laser communication terminal has the same structure and function;

[0041] When any laser communication terminal is used as the current terminal, the remaining laser communication terminals are the opposite terminal devices;

[0042] Each laser communication terminal includes a signal light processing unit, a signal light transmitting unit, a signal light receiving unit, a control unit, a beacon light transmitting unit, a GPS time system unit, a camera unit (10) and a rotating unit;

[0043] The signal light processing unit, the signal light emitting unit, the signal light receiving unit, the control unit, the beacon light emitting unit, the GPS time system unit, and the camera unit (10) are installed on the rotating unit;

[0044] The rotating unit rotates under the control of the control unit; the signal light emitting unit emits a laser beam under the control of the signal light processing unit, and the emission azimuth and pitch angle of the laser beam are controlled by the control unit; the signal light receiving unit receives the laser beam under the control of the signal light processing unit; the beacon light emitting unit emits beacon light under the control of the control unit, and the emission azimuth and pitch angle of the beacon light are controlled by the control unit; the GPS timing unit provides a unified clock for the control unit; the camera unit (10) provides the control unit with an environment image and a light spot image; and the control unit executes steps (1) and (2).

[0045] The advantages and effects of the present invention are:

[0046] For any laser communication terminal, rotate and scan the surrounding environment, obtain the environmental image through the camera module, grayscale, smooth, and histogram equalization the environmental image, extract the texture features and perform normalization, and then generate an SVM classification model to identify the opposite device. The location of the opposite device is used as the scanning area, and the beacon light spot image is obtained in the scanning area. The light spot image is grayscale, smoothed, and histogram equalization is performed, and then secondary OTSU adaptive segmentation and morphological processing are performed to segment the light spot and calculate the light spot centroid. The light spot centroid is used as the alignment point. For the current laser communication terminal, the remaining opposite devices are used. The alignment point of the device generates an optimal scanning path, performs dynamic scanning according to the scanning path under the premise of a unified clock, completes automatic alignment, and performs adaptive alignment error compensation in real time. Any laser communication terminal of the present invention can perform rotational scanning to achieve 360° all-round scanning, and obtain the scanning area, alignment point and optimal scanning path, and realize automatic alignment under the premise of a unified clock, and perform adaptive alignment error compensation in real time, so that any two laser communication terminals can establish a communication link. Under complex and changeable conditions, it can still stably, reliably, comprehensively and flexibly realize the establishment of many-to-many communication links, and realize concurrent communication of multiple targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a block diagram of the principle of establishing communication links for multiple pairs according to the present invention.

[0048] Figure 2 This is a block diagram of the structural principles of the laser communication terminal of the present invention.

[0049] Figure 3 This is the appearance structure diagram of the laser communication terminal of the present invention.

[0050] Figure 4 Schematic diagram of the process of rotational scanning, automatic alignment, and establishing a communication link according to an embodiment of the present invention.

[0051] Figure 5 This is a block diagram of the principle of automatic alignment implemented in one embodiment of the present invention.

[0052] The numbers in the figure are: 1. First laser communication terminal; 2. Second laser communication terminal; 3. Third laser communication terminal; 4. Scanning path of the third laser communication terminal; 5. Scanning path of the second laser communication terminal; 6. Scanning path of the first laser communication terminal; 7. Optical port of the signal light emitting unit; 8. Optical port of the signal light receiving unit; 9. Optical port of the beacon light emitting unit; 10. Camera unit; 11. Communication link between the first laser communication terminal and the second laser communication terminal; 12. Communication link between the first laser communication terminal and the third laser communication terminal; 13. Communication link between the second laser communication terminal and the first laser communication terminal; 14. Communication link between the second laser communication terminal and the third laser communication terminal; 15. Communication link between the third laser communication terminal and the second laser communication terminal; 16. Communication link between the third laser communication terminal and the first laser communication terminal. DETAILED DESCRIPTION

[0053] The present invention will be further described below with reference to the following examples, but the present invention is not limited to these examples.

[0054] A method for establishing a many-to-many communication link comprises the following steps:

[0055] (1) For any laser communication terminal, determine the scanning area and alignment point

[0056] (11) Determine the scanning area

[0057] (111) The laser communication terminal currently rotates and scans to obtain an environmental image;

[0058] (112) Preprocessing the environment image to obtain a preprocessed environment image; the environment image preprocessing includes grayscale conversion, smoothing, and histogram equalization;

[0059] (113) extracting texture features from the environment preprocessing image and performing normalization processing to obtain a normalized LBP texture feature map;

[0060] (114) generating an SVM classification model using the normalized LBP texture feature map as a feature vector;

[0061] (115) identifying a peer device using the SVM classification model, and using the location of the peer device as a scanning area;

[0062] (12) Determine the alignment point

[0063] (121) acquiring a beacon light spot image in the scanning area determined in step (1);

[0064] (122) Preprocessing the light spot image to obtain a light spot preprocessed image; the light spot image preprocessing includes grayscale, smoothing, and histogram equalization;

[0065] (123) performing secondary OTSU adaptive segmentation on the spot preprocessing image to obtain a binary image;

[0066] (124) performing morphological processing on the binary image to segment the light spots; calculating the centroid of the light spots; using the centroid as an alignment point; recording each alignment point, the corresponding received power, and the corresponding code disk angle;

[0067] (2) Laser communication terminals dynamically establish communication links from one to many

[0068] (21) Synchronizing the clocks of each laser communication terminal; for any laser communication terminal, generating a scanning path based on the alignment points of each opposite terminal device;

[0069] (22) Dynamic scanning is performed according to the scanning path under the same clock, and automatic alignment is performed based on the corresponding received power and code disk angle, and adaptive alignment error compensation is performed in real time. The principle block diagram of the method for establishing a communication link for multiple pairs of the present invention is as follows Figure 1 shown.

[0070] In step (111), the laser communication terminal performs rotation scanning at a low speed to acquire an environmental image. The low speed may be as low as 10 rpm.

[0071] In step (112), the grayscale is to calculate the grayscale value of each pixel:

[0072] G=0.299*R+0.587*G+0.114*B, where R is the red channel pixel value, G is the green channel pixel value, B is the blue channel pixel value, and G is the grayscale value of the environment image.

[0073] In step (112), the environment image is smoothed using Gaussian filtering:

[0074] The environment image is smoothed by presetting the Gaussian kernel size and variance, and the smoothing effect is improved by adjusting the variance. When the smoothing effect reaches the preset target, a convolution operation is performed on each pixel to obtain a smoothed image.

[0075] In step (112), the smoothed image is subjected to histogram equalization:

[0076] The probability density function of the gray value of each pixel is calculated, and each probability density function is accumulated to obtain a cumulative distribution function. The smoothed image is subjected to gray value mapping based on the cumulative distribution function to obtain an environmental preprocessing image.

[0077] In step (113), the step of extracting texture features from the environment preprocessing image is as follows:

[0078] (1131) Define a 3*3 texture window;

[0079] (1132) taking pixels in a non-boundary pixel area of ​​the environment pre-processed image according to a texture window size;

[0080] (1133) using the grayscale value of the center pixel in the texture window as a grayscale threshold, traversing eight pixels outside the center pixel in the texture window in a clockwise or counterclockwise manner, marking the current pixel as 1 when the grayscale value of the current pixel is greater than the grayscale threshold, otherwise marking the current pixel as 0, thereby obtaining an 8-bit binary number;

[0081] (1134) traverse the non-boundary pixel area in the environmental preprocessing image where texture feature calculation is not performed, and execute steps (1132) and (1133);

[0082] (1135) Convert each obtained 8-bit binary number into a decimal number, and each decimal number is the LBP texture feature map of the environment preprocessing image.

[0083] In step (113), the LBP texture feature map is normalized:

[0084] The number of occurrences of each decimal number in the LBP texture feature map is counted to obtain an LBP histogram; the horizontal axis of the LBP histogram is each decimal number, and the vertical axis of the LBP histogram is the number of occurrences of the corresponding decimal number; for each decimal number on the horizontal axis of the LBP histogram, the decimal number is divided by the sum of the decimal numbers to obtain a corresponding new value, and the decimal number in the original position is replaced by the new value to obtain a normalized LBP texture feature map.

[0085] Support vector machine (SVM) is a supervised machine learning algorithm that classifies data by finding the optimal line or hyperplane to maximize the distance between each category in N-dimensional space. In step (114), the method for generating an SVM classification model is as follows: using the normalized LBP texture feature map as a feature vector and forming a training data set with corresponding labels (such as object type, texture type); selecting an SVM model (such as linear SVM, polynomial SVM, radial basis function SVM); using the training data set to train the SVM model to obtain the optimal hyperplane that distinguishes samples of different categories (such as the peer device and other items within the scanning range); cross-validating and evaluating the trained SVM model. If the performance meets the requirements, the SVM model is determined as a qualified SVM classification model. The SVM classification model can be used to classify environmental images and determine the category to which it belongs, such as determining which object is the peer device.

[0086] In step (123), the spot pre-processed image is subjected to secondary OTSU adaptive segmentation:

[0087] The Otsu algorithm is an existing technology that performs initial threshold segmentation on the pre-processed image of the spot, dividing the image into foreground and background, resulting in an initial binary image. The foreground portion may contain multiple different objects or regions, requiring secondary segmentation to distinguish them. The Otsu algorithm is then applied to the portion requiring further segmentation, performing secondary threshold segmentation to obtain a final binary image. The segmentation threshold is determined by counting the number of pixels in the entire image at each grayscale value; calculating the probability distribution of each pixel within the entire image; performing a traversal search of the grayscale values ​​to calculate the inter-class probability of foreground and background at the current grayscale value; and calculating the threshold corresponding to the intra-class and inter-class variances using an objective function, which is used as the segmentation threshold. When segmenting the image, pixels above the segmentation threshold are set to 255, while pixels below the threshold are set to 0, resulting in a binary image.

[0088] Step (124) performs morphological processing on the binary image: morphological processing is performed in the order of erosion-erosion-dilation-dilation to effectively remove small spots and false targets and accurately segment the light spots in the light spot image. Among them, the erosion processing is to delete the edge pixels of the foreground object in the light spot image, so as to reduce the foreground area in the light spot image, thereby eliminating small burrs in the image and disconnecting narrow connections. The dilation processing is to expand the foreground object in the light spot image, connect disconnected areas, and fill holes. The present invention performs secondary OTSU adaptive segmentation on the light spot preprocessing image and then performs morphological processing, which can greatly remove the noise area and achieve accurate segmentation of the light spot.

[0089] In step (124), the center of mass of the light spot is calculated:

[0090] Where x c 、y c is the coordinate of the centroid of the light spot, x and y are the pixel coordinates associated with the centroid of the light spot, and I(x, y) is the pixel value under the x and y coordinates.

[0091] In step (21), each laser communication terminal performs high-precision clock synchronization through the GPS time unit to ensure that the time error between each laser communication terminal is controlled within the microsecond level and is highly synchronized in time, thereby ensuring strict alignment and precise coordination of the laser beam in transmission and reception, and ensuring the synchronization and reliability of the communication link.

[0092] In step (21), a scanning path is generated based on the alignment points of each opposite-end device: when the alignment points are obtained in step (124), the signal light receiving power corresponding to the alignment points and the code disk angles in the horizontal and pitch directions of the rotating unit are recorded, so that the optimal scanning path is generated based on the code disk angles in the horizontal and pitch directions in the automatic alignment link. The laser communication terminal uses the signal light receiving power as a judgment basis, performs rotation scanning along the optimal scanning path, and accurately establishes a communication link with the opposite-end device.

[0093] In step (22), at the same clock, the rotating unit of each laser communication terminal rotates at a high speed of greater than or equal to 720 rpm, driving other units of the laser communication terminal to rotate rapidly on the scanning path, performing dynamic scanning, and ensuring that the alignment points provided by each opposite device can be quickly and accurately aligned during the scanning process.

[0094] Step (22) performs adaptive alignment error compensation: the control unit monitors the position and posture information of the rotating unit in real time, and when the rotating unit scans the alignment point position, the control unit controls the signal light transmitting unit to align with the signal light receiving unit of the opposite device; the control unit monitors the receiving power and the code disk angle of the communication link in real time and performs closed-loop control, specifically comparing the receiving power and the code disk angle recorded when the alignment point is obtained in step (124), obtaining the corresponding deviation value, and using the deviation value as a basis for accurately adjusting the code disk angle, and in real time, fine-tuning the position of the alignment point up, down, left, and right with the alignment point as the center, to ensure that the position of the alignment point is always in the optimal state, thereby ensuring the stability and reliability of the communication link.

[0095] A system for establishing a many-to-many communication link by executing the above method:

[0096] Comprising three or more laser communication terminals; each laser communication terminal has the same structure and function;

[0097] When any laser communication terminal is used as the current terminal, the remaining laser communication terminals are the opposite terminal devices;

[0098] Each laser communication terminal includes a signal light processing unit, a signal light transmitting unit, a signal light receiving unit, a control unit, a beacon light transmitting unit, a GPS time system unit, a camera unit 10 and a rotation unit;

[0099] The signal light processing unit, signal light emitting unit, signal light receiving unit, control unit, beacon light emitting unit, GPS time system unit, and camera unit 10 are installed on the rotating unit;

[0100] The rotating unit rotates under the control of the control unit; the signal light emitting unit emits a laser beam under the control of the signal light processing unit, and the emission azimuth and pitch angle of the laser beam are controlled by the control unit; the signal light receiving unit receives the laser beam under the control of the signal light processing unit; the beacon light emitting unit emits beacon light under the control of the control unit, and the emission azimuth and pitch angle of the beacon light are controlled by the control unit; the GPS timing unit provides a unified clock for the control unit; the camera unit 10 provides the control unit with an environmental image and a light spot image; the control unit executes steps (1) and (2). The structural principle block diagram of the laser communication terminal of the present invention is as follows: Figure 2 The appearance and structure of the laser communication terminal is shown in Figure 3 As shown, in Figure 3 Among them, 7 is the optical port of the signal light emitting unit, 8 is the optical port of the signal light receiving unit, 9 is the optical port of the beacon light emitting unit, and 10 is the camera unit.

[0101] The rotation unit mainly includes two mutually perpendicular rotation axes, which are responsible for precise angle adjustment in the horizontal and vertical directions respectively. It realizes high-precision position, speed and acceleration control through the closed-loop control system of FOC (Field-Oriented Control) algorithm. Once the beacon light spot emitted by the opposite device is captured, it is converted into a code disk feedback signal, and the deviation angle is calculated to control the azimuth and pitch direction of the signal light emission, thereby achieving alignment with the opposite signal light receiving unit. The addition of the rotation unit enables the laser communication terminal to achieve 360° rotation scanning, thereby achieving all-round coverage of the laser beam; the signal light receiving unit supports time-sharing communication between any two transmitting and receiving ends, which can achieve the effect of concurrent communication, thereby meeting the needs of multi-target concurrent communication in complex communication environments. The signal light processing unit is responsible for the modulation and demodulation of the laser signal.

[0102] The method of establishing a many-to-many communication link is explained using three laser communication terminals. The process diagram of realizing rotation scanning, automatic alignment, and establishing a communication link is shown in the following figure. Figure 4 As shown in the figure, the principle block diagram of automatic alignment is as follows Figure 5 After the first laser communication terminal 1, the second laser communication terminal 2, and the third laser communication terminal 3 are powered on and initialized and calibrated, steps (1) and (2) are executed.

[0103] The process of determining the scanning area and alignment point is explained by taking the establishment of a communication link between the first laser communication terminal 1 and the second laser communication terminal 2 and the third laser communication terminal 3 as an example:

[0104] The scanning area of ​​the first laser communication terminal 1 is pre-divided, and the rotation unit of the first laser communication terminal 1 drives the terminal to rotate 360 ​​degrees at a low speed and scan in the preset area;

[0105] The camera unit acquires an image of the surrounding environment, and the control unit pre-processes the image by grayscale, smoothing, and histogram equalization, extracts texture features, performs normalization processing, and then generates an SVM classification model. The SVM classification model is used to identify the opposite device within the scanning area, thereby determining the scanning area, that is, the location area where the second laser communication terminal 2 and the third laser communication terminal 3 are located;

[0106] The camera unit 10 of the first laser communication terminal 1 obtains the beacon light spot image in the position area of ​​the second laser communication terminal 2 and the third laser communication terminal 3 respectively. For the spot image of the beacon light from the second laser communication terminal 2, the control unit grayscales, smoothes, and histogram equalizes the spot image, and then performs secondary OTSU adaptive segmentation and morphological processing to segment the spot, calculate the center of mass of the spot, and use the center of mass of the spot as the alignment point. The same processing method is applied to the spot image from the third laser communication terminal 3. The optimal scanning path can be generated based on the alignment points of the second laser communication terminal 2 and the third laser communication terminal 3; each alignment point, the corresponding receiving power, and the corresponding code disk angle are recorded, which can be used as a reference value when establishing a communication link.

[0107] The process by which the second laser communication terminal 2 determines the scanning area and alignment point of the first laser communication terminal 1 and the third laser communication terminal 3, and the process by which the third laser communication terminal 3 determines the scanning area and alignment point of the first laser communication terminal 1 and the second laser communication terminal 2 is the same as the above method.

[0108] When establishing a communication link, the first laser communication terminal 1, the second laser communication terminal 2, and the third laser communication terminal 3 unify their clocks and simultaneously rotate 360° at high speed driven by their respective rotation units; the first laser communication terminal 1 scans along the scanning path 6, and the control unit uses the receiving power stored in step (124) as a reference and the code disk angle as a basis to control the signal light emitting unit through the rotation unit to align with the signal light receiving units of the second laser communication terminal 2 and the third laser communication terminal 3, thereby establishing a communication link 11 between the first laser communication terminal 1 and the second laser communication terminal 2, and a communication link 12 between the first laser communication terminal 1 and the third laser communication terminal 3, and monitors the position and posture information of the rotation unit in real time, monitors the receiving power and the code disk angle in real time and performs closed-loop control, and performs adaptive alignment error compensation in real time. Similarly, the second laser communication terminal 2 scans along the scanning path 5, establishing a communication link 13 between the second laser communication terminal 2 and the first laser communication terminal 1, and a communication link 14 between the second laser communication terminal 2 and the third laser communication terminal 3; the third laser communication terminal 3 scans along the scanning path 4, establishing a communication link 15 between the third laser communication terminal 3 and the second laser communication terminal 2, and a communication link 16 between the third laser communication terminal 3 and the first laser communication terminal 1.

[0109] In addition, the time to establish communication on time is the communication time slot, combined with Figure 4 and Figure 5 It can be seen that the three laser communication terminals generate time slot 1, time slot 2 and time slot 3. During time slot 1, the first laser communication terminal 1 and the second laser communication terminal 2 establish a communication link. During time slot 2, the first laser communication terminal 1 and the third laser communication terminal 3 establish a communication link. During time slot 3, the second laser communication terminal 2 and the third laser communication terminal 3 establish a communication link.

Claims

1. A method for establishing a many-to-many communication link, characterized in that: The steps include: (1) For any laser communication terminal, determine the scanning area and alignment point (11) Determine the scanning area (111) The laser communication terminal currently rotates and scans to obtain an environmental image; (112) Preprocessing the environment image to obtain a preprocessed environment image; the environment image preprocessing includes grayscale conversion, smoothing, and histogram equalization; (113) extracting texture features from the environment preprocessing image and performing normalization processing to obtain a normalized LBP texture feature map; (114) generating an SVM classification model using the normalized LBP texture feature map as a feature vector; (115) identifying a peer device using the SVM classification model, and using the location of the peer device as a scanning area; (12) Determine the alignment point (121) acquiring a beacon light spot image in the scanning area determined in step (1); (122) Preprocessing the light spot image to obtain a light spot preprocessed image; the light spot image preprocessing includes grayscale, smoothing, and histogram equalization; (123) performing secondary OTSU adaptive segmentation on the spot preprocessing image to obtain a binary image; (124) performing morphological processing on the binary image to segment the light spots; calculating the centroid of the light spots; using the centroid as an alignment point; recording each alignment point, the corresponding received power, and the corresponding code disk angle; (2) Laser communication terminals dynamically establish communication links from one to many (21) Synchronizing the clocks of each laser communication terminal; for any laser communication terminal, generating a scanning path based on the alignment points of each opposite terminal device; (22) Dynamic scanning is performed according to the scanning path under the same clock, and automatic alignment is performed based on the corresponding received power and code disk angle, and adaptive alignment error compensation is performed in real time.

2. The method for establishing a many-to-many communication link according to claim 1, wherein: In step (112), the grayscale is to calculate the grayscale value of each pixel: G=0.299*R+0.587*G+0.114*B, where R is the red channel pixel value, G is the green channel pixel value, B is the blue channel pixel value, and G is the grayscale value of the environment image.

3. The method for establishing a many-to-many communication link according to claim 2, wherein: In step (112), the environment image is smoothed using Gaussian filtering: The environment image is smoothed by presetting the Gaussian kernel size and variance, and the smoothing effect is improved by adjusting the variance. When the smoothing effect reaches the preset target, a convolution operation is performed on each pixel to obtain a smoothed image.

4. The method for establishing a many-to-many communication link according to claim 3, wherein: In step (112), the smoothed image is subjected to histogram equalization: The probability density function of the gray value of each pixel is calculated, and each probability density function is accumulated to obtain a cumulative distribution function. The smoothed image is subjected to gray value mapping based on the cumulative distribution function to obtain an environmental preprocessing image.

5. The method for establishing a many-to-many communication link according to claim 4, wherein: In step (113), the step of extracting texture features from the environment preprocessing image is as follows: (1131) Define a 3*3 texture window; (1132) taking pixels in a non-boundary pixel area of ​​the environment pre-processed image according to a texture window size; (1133) using the grayscale value of the center pixel in the texture window as a grayscale threshold, traversing eight pixels outside the center pixel in the texture window in a clockwise or counterclockwise manner, marking the current pixel as 1 when the grayscale value of the current pixel is greater than the grayscale threshold, otherwise marking the current pixel as 0, thereby obtaining an 8-bit binary number; (1134) traverse the non-boundary pixel area in the environmental preprocessing image where texture feature calculation is not performed, and execute steps (1132) and (1133); (1135) Convert each obtained 8-bit binary number into a decimal number, and each decimal number is the LBP texture feature map of the environment preprocessing image.

6. The method for establishing a many-to-many communication link according to claim 5, wherein: In step (113), the LBP texture feature map is normalized: The number of occurrences of each decimal number in the LBP texture feature map is counted to obtain an LBP histogram; the horizontal axis of the LBP histogram is each decimal number, and the vertical axis of the LBP histogram is the number of occurrences of the corresponding decimal number; for each decimal number on the horizontal axis of the LBP histogram, the decimal number is divided by the sum of the decimal numbers to obtain a corresponding new value, and the decimal number in the original position is replaced by the new value to obtain a normalized LBP texture feature map.

7. The method for establishing a many-to-many communication link according to claim 1, wherein: In step (124), the center of mass of the light spot is calculated: Where x c 、y c is the coordinate of the centroid of the light spot, x and y are the pixel coordinates associated with the centroid of the light spot, and I(x, y) is the pixel value under the x and y coordinates.

8. The method for establishing a many-to-many communication link according to claim 1, wherein: In step (22), each laser communication terminal performs dynamic scanning at a rotation speed of 720 rpm according to the scanning path at the same clock.

9. A system for establishing a many-to-many communication link by executing the method for establishing a many-to-many communication link according to claims 1-8, characterized in that: Comprising three or more laser communication terminals; each laser communication terminal has the same structure and function; When any laser communication terminal is used as the current terminal, the other laser communication terminals are the opposite terminal devices; Each laser communication terminal includes a signal light processing unit, a signal light transmitting unit, a signal light receiving unit, a control unit, a beacon light transmitting unit, a GPS time system unit, a camera unit (10) and a rotating unit; The signal light processing unit, the signal light emitting unit, the signal light receiving unit, the control unit, the beacon light emitting unit, the GPS time system unit, and the camera unit (10) are installed on the rotating unit; The rotating unit rotates under the control of the control unit; the signal light emitting unit emits a laser beam under the control of the signal light processing unit, and the emission azimuth and pitch angle of the laser beam are controlled by the control unit; the signal light receiving unit receives the laser beam under the control of the signal light processing unit; the beacon light emitting unit emits beacon light under the control of the control unit, and the emission azimuth and pitch angle of the beacon light are controlled by the control unit; the GPS timing unit provides a unified clock for the control unit; the camera unit (10) provides the control unit with an environment image and a light spot image; and the control unit executes steps (1) and (2).

Images