Visible light communication noise reduction method based on adaptive optical antenna and intelligent model

Abstract

The invention discloses a visible light communication noise reduction method based on an adaptive optical antenna and an intelligent model, and relates to the technical field of visible light communication noise reduction. Comprising the following steps: S1, integrating a power amplifier, a master controller and a power supply in an interlayer of a protection model, and placing a PD detector in the center of the bottom to realize uniform and efficient illumination of LED light on a target area; the bottom adopts an aluminum alloy heat dissipation bearing plate; and S2, the interior of the lens is vacuumized, water vapor is thoroughly eliminated, and internal condensation is prevented. According to the invention, an optical signal transmission mechanism is greatly improved through optimization, the signal-to-noise ratio and the signal intensity are remarkably enhanced, and the effectiveness and the stability of remote visible light communication are greatly improved; especially in complex light environments such as natural light interference, the field angle can be automatically adjusted to maintain high gain and large wide angle; through cooperation and comprehensive optimization of all parts, the size of light spots output by the system is smaller, energy distribution is more uniform, optical signals efficiently focus on a detector, and signal-to-noise ratio loss caused by reflection or diffusion is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of visible light communication noise reduction, and in particular to a visible light communication noise reduction method based on an adaptive optical antenna and an intelligent model. Background Art

[0002] In recent years, the leap forward in mobile Internet technology has improved people's travel convenience, and the popularity of GPS and BDS has made it even better. However, in confined spaces such as underground parking lots, GPS and BDS cannot effectively provide stable communication system services due to limited signals. To solve this problem, researchers have proposed a visible light communication system based on VLC, but VLC faces challenges such as reduced signal-to-noise ratio when used in confined spaces. Problems such as emission, reflection or loss in optical signal transmission will reduce the signal-to-noise ratio and affect the quality of electrical signals. The complex structure and thick walls of confined spaces will block or weaken optical signals, resulting in uneven signal coverage, further reducing the signal-to-noise ratio and affecting system stability and accuracy.

[0003] In response to these challenges, the present invention designs an adaptive large-field-of-view curved mother-and-child compound eye optical antenna and an intelligent protection model visible light communication noise reduction technology. The new protection model solves the problem of lens fogging and sets up a black pipe artificial channel model to eliminate the influence of diffuse and reflected noise on optical signal transmission, ensuring stable operation of the system. The large-field-of-view curved mother-and-child compound eye optical antenna can efficiently process optical signals from non-Lambertian light beam LED light sources. Through innovative optical design, it improves the edge field of view image quality, optimizes light processing and system layout, expands the theoretical field of view range, increases the effective receiving area, and gathers enough light signals to the detector to maximize the optical gain, thereby further increasing the signal-to-noise ratio gain. The adaptive ray tracing zoom system can ensure that the field of view angle is automatically adjusted in complex situations, ensuring the high gain and wide angle of the communication system, and enhancing the robustness of the communication system.

[0004] The purpose of this invention is to optimize the optical signal transmission mechanism, improve the signal-to-noise ratio and signal quality, solve the problems of weak signal and susceptibility to interference in VLC environments in confined spaces such as underground parking lots, resulting in reduced signal-to-noise ratio and poor signal quality, and provide users with stable and reliable communication services. By introducing innovative technologies such as new protection models and large-field-of-view curved compound eye optical antennas, this invention is expected to provide efficient and accurate visible light communication services for confined spaces such as underground parking lots. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a visible light communication noise reduction method based on adaptive optical antenna and intelligent model.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A visible light communication noise reduction method based on an adaptive optical antenna and an intelligent model comprises the following steps:

[0008] S1: The amplifier, main control and power supply are integrated in the interlayer of the protective model, and the PD detector is placed in the center of the bottom to achieve uniform and efficient illumination of the target area by LED lights; the bottom adopts an aluminum alloy heat dissipation load-bearing plate;

[0009] S2: Evacuate the inside of the lens to completely eliminate water vapor and prevent internal condensation; then, coat the outside of the lens with an optical film to reduce the reflection of light signals transmitted to the lens surface and increase transmission; and design ventilation holes to use the excellent thermal conductivity of titanium alloy to quickly heat up the antenna object itself to the outside temperature;

[0010] S3: Place the large-field-of-view mother-and-child compound eye optical system and the light adaptive zoom tracking system in the card slot of the protective device, and reduce the diffuse and reflection noise by dynamically reducing the field of view angle of the reflective lens;

[0011] S4: An equal-diameter black pipe is integrated at the rear end of the protection model focusing lens, directly connected to the photoelectric detector and COMS sensor, forming a closed optical path channel and constructing an artificial optical channel.

[0012] Preferably: the tracking system is based on light spot detection and intelligent algorithm operation, the light spot detection part of which is composed of a clock and an image acquisition module, and uses an OV5640 CMOS camera and an IIC communication protocol configuration to output a signal to the FPGA, and uses an algorithm to identify the light spot after quantization compensation processing; during the tracking process, the center of the light spot is determined according to the change in the number of pixels and compared with the center of the screen, and the servo and motor are driven to achieve tracking; the motor zoom is controlled according to the comparison between the light spot size and a preset threshold.

[0013] Preferably: the method also involves an adaptive zoom function of the spherical parabolic sub-eye, and the adaptive tracking zoom function is implemented based on a light spot detection system. The light spot detection system includes a clock module, an image acquisition module, a DDR3 module, a light spot detection module, and an LCD drive module. The image acquisition module uses a CMOS camera of model OV5640.

[0014] Preferably: in the light spot detection system, after configuring its register through the IIC communication protocol, the camera outputs 4 signals to the FPGA, HSYNC is used as a field synchronization signal to indicate the start or end of a frame of image data, HREF is used as a line synchronization signal to indicate the start or end of a line of image data, D[9:0] represents the 10-bit video data stream output by the final camera, and the first eight bits are valid data; then the format conversion is performed using the quantization compensation method so that the low bit of the data is no longer 0, and after the light spot is moved to the cross position of the screen, the FPGA color extraction button is pressed, and the FPGA calculates the average value of the three color channel components of R, G, and B of 32x32 pixels around the cross position and stores it in the register variable, and each subsequent pixel data is compared with the register variable for difference, and the light spot pixel data is binary processed and marked, when Binary_out is 1, the pixel is identified as a light spot pixel and displayed as green on the LCD screen, and the remaining non-light spot pixels are displayed normally, and the image is filtered using the opening operation in morphological filtering.

[0015] Preferably, the spot detection system is specifically as follows in terms of spot tracking:

[0016] In the vth frame image, the number of pixels marked as spot data is recorded as Sum. In the v+1th frame image data, when the number of identified spots is equal to half of Sum, the pixel field counter at this time is saved as center_v, and this value is the vertical coordinate of the center position of the pixel point; for the vth frame image, the center_v value of the v-1th frame image and the number of spot pixels in the center_v row center_line_nun can be obtained; then the count H_num_cnt is set, and the number of pixels of the target monochrome object in the center_vth row of the vth frame image is counted. When the count value is equal to half of center_line_num, the column count value of the pixel is saved as the required horizontal coordinate center_h; the spot pixel data is binarized and marked:

[0017]

[0018] After determining the center position of the light spot, compare it with the cross center coordinates (XO, YO) of the center of the screen; if the horizontal coordinate center_h of the light spot is less than XO, output a right turn command to the servo controlling the horizontal direction; if it is greater than XO, output a left turn command, and the same applies to the vertical direction, so that the center of the screen (XO, YO) approaches the center coordinates of the light spot to achieve light spot tracking.

[0019] Preferably: after the light spot detection system realizes light spot tracking, the light spot size is measured by the number of marked light spot pixels, and compared with the number after setting a threshold value, as a basis for controlling the rotation distance and direction of the motor; if the number of light spot pixels on the screen is less than the preset threshold, a left turn drive pulse signal is output to the motor, and the spherical parabolic sub-eye distance is lengthened to achieve amplification; if it is greater than the preset threshold, a right turn drive pulse signal is output to make the number of light spot pixels on the screen close to the threshold.

[0020] Preferably: the method also involves a large-field-of-view curved optical mother-and-child compound eye system, which includes a plurality of subsystems, each subsystem corresponding to an independent image plane, each subsystem having a field of view of 40°, 9 spherical parabolic sub-eyes arranged on the meridian plane, and a curved base radius of 100 cm.

[0021] Preferably: in the large field of view curved optical mother-and-child compound eye system, the spherical parabolic sub-eye adopts a polygonal structure, and its single field of view is θmax; the free-form surface lens is placed behind the spherical parabolic sub-eye array, and is responsible for turning the light of each channel of the sub-compound eye system so that the imaging is basically perpendicular to the image plane, the focal length of each channel is set to 20mm, the aperture array is set in the post-correction group, and the array base shape structure is determined by the obtained ΔL value.

[0022] Preferably: in the large field of view curved optical mother-and-child compound eye system, the evenly distributed spherical parabolic sub-eyes are divided into 4 groups in the meridian plane, each group is equipped with a dedicated integration lens, a rear correction lens group and an image plane; the sub-eye light first passes through the integration lens, and the light of different subsystems converges at the center of the sphere and enters the corresponding aberration correction lens group respectively, and finally forms an image on the corresponding subsystem image plane.

[0023] Preferably: in the large field of view curved optical mother-and-child compound eye system, the spherical parabolic sub-eye is located in the sandwich spherical surface, and the sandwich spherical surface is equipped with a servo and a motor; the sub-eye is responsible for receiving light in a field of view angle range of a specific size in a specific direction in space, and is imaged on the same plane image plane after passing through the free-form surface and the aberration correction lens group; wherein the free-form surface lens and the aberration correction lens group are precisely located in the field of view fan-shaped area of ​​the corresponding subsystem, avoiding light obstruction and lens installation interference, ensuring that each subsystem is independent of each other, and each channel is imaged on a larger and same image plane without overlap.

[0024] The beneficial effects of the present invention are:

[0025] 1. The optimization of the present invention greatly improves the optical signal transmission mechanism, significantly enhances the signal-to-noise ratio and signal strength, and greatly improves the effectiveness and stability of long-distance visible light communication; especially in complex light environments such as natural light interference, it can automatically adjust the field of view to maintain high gain and wide angle; through the coordination and comprehensive optimization of various parts, the system outputs a smaller spot size and more uniform energy distribution, and the optical signal is efficiently focused on the detector, which greatly reduces the signal-to-noise ratio loss caused by reflection or diffusion, and significantly improves the signal-to-noise ratio gain, effectively solving the communication problem in confined spaces and ensuring stable and efficient communication.

[0026] 2. The present invention optimizes the optical antenna structure through innovative design, thereby achieving an expansion of the system's field of view angle and an improvement in optical receiving efficiency, while reducing the field of view angle of the reflective lens to reduce diffuse or reflected noise, thereby ensuring high gain while providing a larger system field of view angle, ultimately obtaining a light spot with a smaller size and more uniform energy distribution. This comprehensive optimization significantly improves the overall performance and technical value of the optical antenna.

[0027] 3. The present invention In the present invention, the adaptive zoom technology of the spherical parabolic sub-eye has significant advantages over the traditional optical system; in terms of accuracy, its advanced spot detection and tracking algorithm can quickly lock the spot position in a complex light environment, ensure the maximum focusing ratio and field of view, greatly improve the positioning accuracy through pixel-level processing and comparison mechanism, effectively prevent tracking deviation and signal-to-noise ratio fluctuation caused by ambient light interference and small displacement of the target, and ensure the operation of subsequent zoom and optical signal transmission; to ensure high response speed and high flexibility, each module works closely and efficiently, and the signal acquisition, processing to command output process is fast and smooth, which can respond to spot changes in real time, so that the optical system can be stably in the best working state in dynamic scenes; the adaptive algorithm can intelligently adjust the light size according to the light intensity, spot size and target motion state without manual intervention, and is suitable for various scenarios, such as underground parking lots of different sizes and complex indoor environments, to promote the optimization of noise reduction performance of visible light communication systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A flowchart of the visible light communication noise reduction method based on adaptive optical antenna and intelligent model proposed by the present invention;

[0029] Figure 2 A schematic diagram of a spherical parabolic sub-eye of a visible light communication noise reduction method based on an adaptive optical antenna and an intelligent model proposed in the present invention;

[0030] Figure 3 This is a schematic diagram of FPGA spot detection of the visible light communication noise reduction method based on adaptive optical antenna and intelligent model proposed in the present invention;

[0031] Figure 4A meridian plane schematic diagram of a large-field-of-view curved optical mother-and-child compound eye system of a visible light communication noise reduction method based on an adaptive optical antenna and an intelligent model proposed in the present invention;

[0032] Figure 5 A schematic diagram of the structure of the protection model in the visible light communication noise reduction method based on the adaptive optical antenna and the intelligent model proposed in the present invention;

[0033] Figure 6 This is a light path diagram of the visible light communication noise reduction method based on adaptive optical antenna and intelligent model proposed in the present invention. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described in detail below in conjunction with specific implementation methods.

[0035] Embodiment 1:

[0036] A visible light communication noise reduction method based on adaptive optical antenna and intelligent model, based on large field of view curved surface mother-child compound eye optical antenna, protection model and light adaptive zoom tracking system, to achieve visible light communication noise reduction; including: construction of optical antenna, construction of protection model and construction of tracking system;

[0037] The method specifically comprises the following steps:

[0038] S1: The amplifier, main control and power supply are integrated in the interlayer of the protective model, and the PD detector is placed in the center of the bottom to achieve uniform and efficient illumination of the target area by LED lights; the bottom adopts an aluminum alloy heat dissipation bearing plate to effectively dissipate heat, prevent noise or distortion caused by overheating of components, improve the signal-to-noise ratio, and ensure the robustness and stability of the system operation;

[0039] S2: The inside of the lens (i.e. the black tube protecting the model) is evacuated to completely eliminate water vapor and prevent internal condensation; then, an optical film is coated on the outside of the lens to reduce the reflection of the light signal transmitted to the lens surface and increase the transmission; ventilation holes are designed to use the excellent thermal conductivity of titanium alloy to quickly heat up the antenna object itself to the outside temperature, further reducing the reflection noise caused by the attachment of external water vapor to the lens surface;

[0040] S3: The large-field-of-view mother-and-child compound eye optical system and the light adaptive zoom tracking system are placed in the card slot of the protective device. With the help of its structure, not only the system field of view is dynamically maximized, but also the optical receiving efficiency is improved; at the same time, by dynamically reducing the field of view of the reflective lens, the diffuse and reflection noise is effectively reduced, ensuring that a wide system field of view can be provided under high gain and high concentration ratio conditions. At the same time, through light adaptive zoom tracking, high gain and wide angle can be guaranteed even in complex situations. This series of comprehensive optimization measures ultimately obtains a continuously stable output light spot with a smaller size and more uniform energy distribution, thereby greatly improving and stabilizing the signal-to-noise ratio gain of visible light communication;

[0041] S4: An equal-diameter black pipe is integrated at the rear end of the protection model focusing lens, which directly connects the photoelectric detector and COMS sensor to form a closed optical path and build an artificial optical channel. This design eliminates multipath transmission caused by internal reflection of the lens, ensures direct link transmission and maximizes link gain, while significantly suppressing interference from external light sources and improving the purity and sensitivity of optical signals;

[0042] These optimizations greatly improve the optical signal transmission mechanism, significantly enhance the signal-to-noise ratio and signal strength, and greatly improve the effectiveness and stability of long-distance visible light communications.

[0043] Among them, the tracking system is based on light spot detection and intelligent algorithm operation. The light spot detection part is composed of modules such as clock and image acquisition. It uses OV5640 CMOS camera and IIC communication protocol configuration to output signals to FPGA, and uses algorithms to identify light spots after processing such as quantization compensation. During the tracking process, the center of the light spot is determined according to the change in the number of pixels and compared with the center of the screen, and the servo and motor are driven to achieve tracking. The motor zoom is controlled according to the comparison between the light spot size and the preset threshold.

[0044] Especially in complex light environments such as natural light interference, the field of view can be automatically adjusted to maintain high gain and wide angle. Through the coordination and comprehensive optimization of various parts, the system outputs a smaller spot size and more uniform energy distribution. The optical signal is efficiently focused on the detector, greatly reducing the signal-to-noise ratio loss caused by reflection or diffusion, significantly improving the signal-to-noise ratio gain, effectively solving the communication problem in confined spaces, and ensuring stable and efficient communication.

[0045] The method also involves a spherical parabolic sub-eye system, such as Figure 2As shown in the figure, CB and C1F are two parabolas with symmetrical characteristics. They rotate around the central axis to construct a three-dimensional parabolic concentrator in the optical antenna. θmax, as a key field of view angle parameter of the optical antenna, plays a vital role. Specifically, when the angle of the incident light is less than or equal to θmax, these light rays can directly reach the lower opening of the optical antenna, or be effectively captured and utilized by the receiver after reflection. However, once the angle of the incident light exceeds θmax, these light rays will experience multiple reflections and eventually be reflected from the incident port of the optical antenna.

[0046] Therefore, the field of view angle θmaxx is not only an important characteristic parameter of the optical antenna, but also directly related to and affects the focusing performance of the optical antenna.

[0047] In three-dimensional space, the geometric concentration ratio is

[0048] C g =1 / sin 2 θ max

[0049] From the formula, we can see that the smaller the field of view angle, the greater the geometric focusing ratio, which means that the focusing ability of the parabolic concentrator is correspondingly stronger.

[0050] The total height H is:

[0051]

[0052] The height H of the reflection arc is:

[0053]

[0054] The focal length f of the optical antenna is:

[0055]

[0056] (n is the refractive index of the lens material, R1 and R2 are the two radii of curvature of the lens, and T is the total height minus the distance of the reflection arc)

[0057] Construct equations for parabolas CB and C1F such as:

[0058]

[0059] The value range of tc is:

[0060]

[0061] in:

[0062]

[0063] (θ is the half field of view of the optical antenna, f is the focal length of the parabola, tc is the parameter of point C, and H is the height of the reflection arc.)

[0064] By optimizing the optical antenna structure through innovative design, the system's field of view is expanded and the optical receiving efficiency is improved. At the same time, the field of view of the reflective lens is reduced to reduce diffuse or reflected noise, ensuring high gain while providing a larger system field of view, ultimately obtaining a smaller light spot with more uniform energy distribution. This comprehensive optimization significantly improves the overall performance and technical value of the optical antenna.

[0065] The focusing gain is:

[0066]

[0067] (n is the refractive index of the optical receiving end, and θ is the field of view angle of the receiving end).

[0068] The method also involves an adaptive zoom function of a spherical parabolic sub-eye. The realization of its adaptive tracking zoom function is based on a specific mechanism. First, there is a spot detection system, which consists of five parts: a clock module, an image acquisition module, a DDR3 module, a spot detection module, and an LCD driver module, in which a CMOS camera of model OV5640 is used. After configuring its registers through the IIC communication protocol, the camera will output 4 signals to the FPGA. HSYNC is used as a field synchronization signal to indicate the start or end of a frame of image data, HREF is used as a line synchronization signal to indicate the start or end of a line of image data, and D[9:0] represents the 10-bit video data stream output by the final camera, and the first eight bits are valid data. Then, the quantization compensation method is used to convert the format so that the low bit of the data is no longer 0, which reduces the loss of image data accuracy and improves utilization. After moving the light spot to the cross position of the screen, press the FPGA color extraction button, the FPGA will calculate the average value of the three color channel components R, G, and B of the 32x32 pixels around the cross position and store it in the register variable. Each subsequent pixel data is compared with the register variable and subtracted. The light spot pixel data is binarized and marked according to a specific formula. When Binary—out is 1, the pixel is identified as a light spot pixel and displayed in green on the LCD screen. The other non-light spot pixels are displayed normally. At the same time, the opening operation in morphological filtering (erode first and then expand) is used to filter the image.

[0069] In terms of spot tracking, it is as follows: in the vth frame image, the number of pixels marked as spot data is recorded as Sum. Then in the v+1th frame image data, when the number of identified spots is equal to half of Sum, the pixel field counter at this time is saved as center_v, and this value is the vertical coordinate of the center position of the pixel. Due to the above method of calculating the vertical coordinate, for the vth frame image, the center_v value of the v-1th frame image and the number of spot pixels center_line_nun in the center_v row must be obtained. Then set the count H_num_cnt, count the number of pixels of the target monochrome object in the center_vth row of the vth frame image, and when the count value is equal to half of center_line_num, save the column count value of the pixel as the required horizontal coordinate center_h. Binarize the spot pixel data and mark it:

[0070]

[0071] After determining the center position of the spot (center_h, center_v), compare it with the cross center coordinates (XO, YO) of the center of the screen. If the horizontal coordinate center_h of the spot is less than XO, a right turn command is output to the servo that controls the horizontal direction; if it is greater than XO, a left turn command is output, and the same applies to the vertical direction, so that the center of the screen (XO, YO) approaches the center coordinates of the spot (center_h, center_v) to achieve spot tracking. After tracking is achieved, it is relatively simple to zoom in and out the spot. The size of the spot is measured by the number of pixels of the marked spot. After setting a suitable threshold, it is compared with it as the basis for controlling the rotation distance and direction of the motor. If the number of pixels of the spot on the screen is less than the preset threshold, a left turn drive pulse signal is output to the motor to lengthen the spherical parabolic sub-eye distance to achieve enlargement; if it is greater than the preset threshold, a right turn drive pulse signal is output to make the number of pixels of the spot on the screen close to the threshold.

[0072] In the present invention, the adaptive zoom technology of the spherical parabolic sub-eye has significant advantages over the traditional optical system. In terms of accuracy, its advanced spot detection and tracking algorithm can quickly lock the spot position in a complex light environment, ensuring the maximum focusing ratio and field of view. The positioning accuracy is greatly improved through pixel-level processing and comparison mechanisms, effectively preventing tracking deviations and signal-to-noise ratio fluctuations caused by ambient light interference and small displacements of the target, and ensuring the operation of subsequent zoom and optical signal transmission. To ensure high response speed and high flexibility, each module works closely and efficiently, and the signal acquisition, processing to command output process is fast and smooth, which can respond to spot changes in real time, so that the optical system can be stable in the best working state in dynamic scenes. The adaptive algorithm can intelligently adjust the light size according to the light intensity, spot size and target motion state without manual intervention. It is suitable for various scenarios, such as underground parking lots of different sizes and complex indoor environments, and promotes the optimization of noise reduction performance of visible light communication systems.

[0073] The method also involves a large-field-of-view curved optical mother-and-child compound eye system, including multiple subsystems, each subsystem corresponding to an independent image plane, and a field of view angle of each subsystem of 40°. Nine spherical parabolic sub-eyes are arranged on the meridian plane, and the radius of the curved surface base is 100 cm. In order to eliminate the field of view blind spot of the entire system, the spherical parabolic sub-eye adopts a polygonal structure, and its single field of view is θmax. The free-form surface lens is placed after the spherical parabolic sub-eye array, and is responsible for turning the light of each channel of the sub-compound eye system so that the imaging is basically perpendicular to the image plane. The focal length of each channel is set to 20 mm, and the aperture array is set in the post-correction group. The array substrate shape structure is determined by the obtained ΔL value, and the size and position of the subsequent aberration correction group are strictly controlled to prevent exceeding the field of view and blocking the light of other subsystems.

[0074] In the meridian plane, the evenly distributed spherical parabolic sub-eyes are divided into 4 groups, each of which is equipped with a dedicated integration lens, a rear correction lens group and an image plane. The sub-eye light first passes through the integration lens, and the light from different subsystems converges at the center of the sphere and enters the corresponding aberration correction lens group, and finally forms an image on the image plane of the corresponding subsystem. This design greatly reduces the turning angle of the edge field of view light, and by selecting optical elements with smaller focal lengths and reducing the size of the image plane of each channel, the field of view of each sub-eye system is effectively reduced.

[0075] The spherical parabolic sub-eye is located in the sandwich spherical surface, which is equipped with a servo and a motor. The sub-eye is responsible for receiving light with a specific field of view angle range in a specific direction in space, and imaging it on the same plane image plane after passing through the free-form surface and aberration correction lens group. Among them, the free-form surface lens and the aberration correction lens group are precisely located in the fan-shaped area of ​​the field of view angle of the corresponding subsystem to avoid light blocking and interference with lens installation, ensuring that each subsystem is independent of each other, and each channel is imaged on a larger and same image plane without overlap.

[0076] Compared with the traditional compound eye system, this structure has a stronger aberration correction capability in the edge field of view of the compound eye, which greatly improves the imaging quality of the edge field of view, and the theoretical field of view can reach 360°.

[0077] The protection model:

[0078] To address the lens fogging problem caused by changes in ground and air temperature, the inside of the lens is evacuated to eliminate water vapor, thereby preventing internal condensation. At the same time, an optical film is coated on the outside of the lens, and ventilation holes are designed to quickly heat up using the excellent thermal conductivity of titanium alloy, and to solve the heat dissipation problem of the bottom parts and photoelectric detectors.

[0079] When the protective model (assembly) is embedded in underground concrete, in order to prevent the device from cracking due to increased concrete shrinkage stress caused by weather changes, a layer of new nano-ceramic material is added to the outside of the device to reduce stress and provide heat insulation. A new mechanical structure is designed inside to resist increased stress, thereby further enhancing the durability and stability of the equipment.

[0080] The protection model:

[0081] A hemispherical lens is used at the lens, and an equal-diameter black pipe is integrated behind the hemispherical lens, which directly connects to the photoelectric detector to form a closed optical path and construct an artificial optical channel. If light shines on the black pipe, it will be completely absorbed, and the DC gain of the reflected light will no longer be considered. Therefore, the direct light gain at this time is:

[0082]

[0083] Where i is the i-th LED, m is the number of radiation modes, D is the distance between the detector and the light source, is the radiation angle of the light source, φ is the incident angle, A is the effective receiving area of ​​the detector, Ψi is the half field of view of the detector, and g is the optical antenna gain.

[0084] Receive power:

[0085] P LOS =pH LOS (0)

[0086] Signal Power:

[0087]

[0088] Noise reduction processing:

[0089] Shot Noise SNR:

[0090]

[0091] Where PR(signal) is the received signal light power, B is the equivalent noise bandwidth, PR(isi) is the power of inter-symbol interference, I2 is the noise bandwidth factor, Ibg is the dark current, and γ is the responsivity of the photodetector.

[0092] Thermal Noise:

[0093]

[0094] Where G is the open-loop voltage gain, F is the channel noise factor of the field effect transistor (FET), η is the capacitance per unit area of ​​the photodetector, k is the Boltzmann constant, gm is the transconductance of the FET, and Tk is the absolute temperature.

[0095] Preamplifier Noise:

[0096] N_amp=F*KT*B-KT*B

[0097] Where kT*B is the thermal noise power at the amplifier input (k is the Boltzmann constant, T is the absolute temperature, and B is the bandwidth), and F*kT*B is the total noise power at the amplifier output (including the input noise and the noise generated by the amplifier).

[0098] Signal-to-Noise Ratio:

[0099]

[0100] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A visible light communication noise reduction method based on adaptive optical antenna and intelligent model, characterized in that: The steps include: S1: The amplifier, main control and power supply are integrated in the interlayer of the protective model, and the PD detector is placed in the center of the bottom to achieve uniform and efficient illumination of the target area by LED lights; the bottom adopts an aluminum alloy heat dissipation load-bearing plate; S2: Evacuate the inside of the lens to completely eliminate water vapor and prevent internal condensation; then, coat the outside of the lens with an optical film to reduce the reflection of light signals transmitted to the lens surface and increase transmission; and design ventilation holes to use the excellent thermal conductivity of titanium alloy to quickly heat up the antenna object itself to the outside temperature; S3: Place the large-field-of-view mother-and-child compound eye optical system and the light adaptive zoom tracking system in the card slot of the protective device, and reduce the diffuse and reflection noise by dynamically reducing the field of view angle of the reflective lens; S4: An equal-diameter black pipe is integrated at the rear end of the protection model focusing lens, directly connected to the photoelectric detector and COMS sensor, forming a closed optical path channel and constructing an artificial optical channel.

2. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 1 is characterized in that: The tracking system is based on light spot detection and intelligent algorithm operation. The light spot detection part is composed of a clock and an image acquisition module. It uses an OV5640CMOS camera and an IIC communication protocol configuration to output signals to the FPGA. After quantitative compensation processing, the algorithm is used to identify the light spot. During the tracking process, the center of the light spot is determined according to the change in the number of pixels and compared with the center of the screen, and the servo and motor are driven to achieve tracking. The motor zoom is controlled according to the comparison between the light spot size and the preset threshold.

3. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 2 is characterized in that: The method also involves an adaptive zoom function of a spherical parabolic sub-eye. The adaptive tracking zoom function is implemented based on a light spot detection system. The light spot detection system includes a clock module, an image acquisition module, a DDR3 module, a light spot detection module, and an LCD drive module. The image acquisition module uses a CMOS camera of model OV5640.

4. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 3 is characterized in that: In the light spot detection system, after configuring its register through the IIC communication protocol, the camera outputs 4 signals to the FPGA, HSYNC is used as a field synchronization signal to indicate the start or end of a frame of image data, HREF is used as a line synchronization signal to indicate the start or end of a line of image data, and D[9:0] represents the 10-bit video data stream output by the final camera, and the first eight bits are valid data; then the format conversion is performed using the quantization compensation method so that the low bit of the data is no longer 0, and after the light spot is moved to the cross position of the screen, the FPGA color extraction button is pressed, and the FPGA calculates the average value of the three color channel components of R, G, and B of 32x32 pixels around the cross position and stores it in the register variable, and each subsequent pixel data is compared with the register variable for difference, and the light spot pixel data is binary processed and marked, when Binary_out is 1, the pixel is identified as a light spot pixel and displayed as green on the LCD screen, and the remaining non-light spot pixels are displayed normally, and the image is filtered by the opening operation in the morphological filtering.

5. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 4 is characterized in that: The light spot detection system is specifically as follows in terms of light spot tracking: In the vth frame image, the number of pixels marked as spot data is recorded as Sum. In the v+1th frame image data, when the number of identified spots is equal to half of Sum, the pixel field counter at this time is saved as center_v, and this value is the vertical coordinate of the center position of the pixel point; for the vth frame image, the center_v value of the v-1th frame image and the number of spot pixels in the center_v row center_line_nun can be obtained; then the count H_num_cnt is set, and the number of pixels of the target monochrome object in the center_vth row of the vth frame image is counted. When the count value is equal to half of center_line_num, the column count value of the pixel is saved as the required horizontal coordinate center_h; the spot pixel data is binarized and marked: After determining the center position of the light spot, compare it with the cross center coordinates (XO, YO) of the center of the screen; if the horizontal coordinate center_h of the light spot is less than XO, output a right turn command to the servo controlling the horizontal direction; if it is greater than XO, output a left turn command, and the same applies to the vertical direction, so that the center of the screen (XO, YO) approaches the center coordinates of the light spot to achieve light spot tracking.

6. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 5 is characterized in that: After the light spot detection system realizes light spot tracking, the size of the light spot is measured by the number of marked light spot pixels, and compared with the number after setting a threshold value, as a basis for controlling the rotation distance and direction of the motor; if the number of light spot pixels on the screen is less than the preset threshold, a left turn drive pulse signal is output to the motor, and the spherical parabolic sub-eye distance is lengthened to achieve amplification; if it is greater than the preset threshold, a right turn drive pulse signal is output to make the number of light spot pixels on the screen close to the threshold.

7. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 6 is characterized in that: The method also involves a large-field-of-view curved optical mother-and-child compound eye system, which includes multiple subsystems, each subsystem corresponds to an independent image plane, each subsystem has a field of view of 40°, 9 spherical parabolic sub-eyes are arranged on the meridian plane, and the radius of the curved surface base is 100 cm.

8. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 7 is characterized in that: In the large-field-of-view curved optical mother-and-child compound eye system, the spherical parabolic sub-eye adopts a polygonal structure, and its single field of view is θmax; the free-form surface lens is placed behind the spherical parabolic sub-eye array, and is responsible for turning the light of each channel of the sub-compound eye system so that the imaging is basically perpendicular to the image plane, the focal length of each channel is set to 20 mm, and the aperture array is set in the post-correction group, and the array base shape structure is determined by the obtained ΔL value.

9. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 8, characterized in that: In the large-field-of-view curved optical mother-and-child compound eye system, the evenly distributed spherical parabolic sub-eyes are divided into 4 groups in the meridian plane, each group is equipped with a dedicated integration lens, a rear correction lens group and an image plane; the sub-eye light first passes through the integration lens, and the light from different subsystems converges at the center of the sphere and enters the corresponding aberration correction lens group respectively, and finally forms an image on the corresponding subsystem image plane.

10. The visible light communication noise reduction method based on adaptive optical antenna and intelligent model according to claim 9, characterized in that: In the large-field-of-view curved optical mother-and-child compound eye system, the spherical parabolic sub-eye is located in the sandwich spherical surface, and the sandwich spherical surface is equipped with a servo and a motor; the sub-eye is responsible for receiving light in a field-of-view angle range of a specific size in a specific direction in space, and is imaged on the same plane image surface after passing through the free-form surface and the aberration correction lens group; wherein the free-form surface lens and the aberration correction lens group are precisely located in the field-of-view fan-shaped area of ​​the corresponding subsystem, avoiding light obstruction and interference with lens installation, ensuring that each subsystem is independent of each other, and each channel is imaged on a larger and same image plane without overlap.

Images