Low-power-consumption dynamic coverage optimization method and system based on distributed intelligent metasurface

Abstract

The invention discloses a low-power-consumption dynamic coverage optimization method and system based on a distributed intelligent metasurface. The system comprises three low-power-consumption 2-bit intelligent metasurfaces and control panels thereof, a binocular camera and a matched upper computer control system. According to the system, a binocular camera is adopted to capture image data in real time, a human body target in an image is recognized by using a deep learning-based target detection algorithm, and relative coordinates of the human body target in a three-dimensional space are calculated. According to coordinates of a grid cell where a target is located, a corresponding intelligent metasurface coding sequence optimized by a conditional sample mean value algorithm is indexed, and the intelligent metasurface coding sequence is converted into a voltage control signal of the intelligent metasurface through a control panel with an FPGA as a main controller chip; and the real-time dynamic enhancement of the distributed intelligent metasurface on the wireless signal coverage of the area where the target is located is realized. According to the system, the real-time positioning of the target user and the dynamic coverage optimization of the wireless signal are realized by utilizing a visual perception combined distributed coding optimization technology, and the signal receiving quality of the target user is ensured.

Description

Technical Field

[0001] The present invention belongs to the field of signal coverage enhancement system design based on intelligent metasurface, which mainly includes a low-power dynamic coverage optimization method and system based on distributed intelligent metasurface. Background Art

[0002] With the deployment of 5G networks and the popularization of IoT applications, users have higher expectations for the coverage and quality of wireless communications. However, how to achieve comprehensive and high-quality network coverage in various environments, such as urban areas with tall buildings, vast rural fields, and complex indoor or underground spaces, remains a difficult problem. Physical obstacles often cause signal attenuation or blockage, forming "blind spots" that are difficult to cover with existing base station networks. In addition, as spectrum resources become increasingly scarce, operators are forced to use higher frequency bands, such as millimeter waves. The penetration and coverage capabilities of these high-frequency signals are significantly lower than those of traditional frequency bands, exacerbating the signal coverage problem. Therefore, how to expand signal coverage while maintaining network performance has become a key issue that needs to be urgently addressed in the field of wireless communications.

[0003] To meet the challenges in the field of wireless communications, researchers have explored a variety of deployment strategies for small base stations and heterogeneous networks. By densely deploying small base stations in urban environments, researchers can effectively fill the blind spots of network coverage and significantly improve the spatial utilization efficiency of the network. At the algorithm level, various signal attenuation and coverage optimization algorithms have been developed based on genetic algorithms, particle swarm optimization (PSO), deep learning and other algorithms. These algorithms are designed to optimize network configuration and resource allocation, striving to achieve the best balance between signal coverage and performance.

[0004] However, although the individual cost of small base stations is low compared to traditional base stations, the total cost of large-scale deployment and subsequent maintenance is still relatively high. In addition, although optimization algorithms can achieve optimal network performance, they usually require a lot of computing resources, especially in real-time processing or large-scale applications. Therefore, a cost-effective and flexible solution is urgently needed to address these challenges.

[0005] In recent years, smart metasurface technology has shown great application potential due to its unique advantages. A metasurface is an artificial surface composed of numerous subwavelength scattering units. By adjusting the active devices embedded in these units, the smart metasurface can intelligently manipulate the incident electromagnetic waves, including changing their reflection direction, focusing, and diffusion. The rational use of smart metasurface technology can effectively enhance signal penetration and coverage in complex environments, thereby solving the problem of signal blind spots and providing novel solutions for expanding network coverage. However, due to the inherent Brewster angle limitation of metasurfaces, a single metasurface may not be able to cover the entire space in some complex non-line-of-sight scenarios.

[0006] To overcome this limitation, the present invention introduces the strategy of distributed intelligent metasurfaces. By rationally deploying multiple intelligent metasurfaces, full-area signal coverage can be optimized in complex environments. Under this setting, signal coverage no longer depends solely on the beam steering capability of a single intelligent metasurface. The collaborative operation between intelligent metasurfaces can not only effectively expand the signal coverage range, but also enhance the anti-interference capability of the network. However, the deployment of multiple intelligent metasurfaces will inevitably lead to a substantial increase in power consumption. Therefore, how to effectively reduce the power consumption of distributed deployment solutions has also become a key consideration.

[0007] On this basis, a low-power dynamic coverage optimization system based on distributed intelligent metasurface is proposed. Aiming at the low-power requirements of distributed deployment, this scheme designs a metasurface unit with microwatt-level operating energy consumption. Then, the conditional sample mean algorithm is used to optimize the blind beamforming codebook of the intelligent metasurface in a given space, and different coding sequences of the codebook are used to enhance the signal coverage of the corresponding area. Finally, the dynamic coverage optimization of spatial signals is realized by combining visual tracking technology. Summary of the invention

[0008] The purpose of the present invention is to design a low-power dynamic coverage optimization system based on a distributed intelligent metasurface, which has the characteristics of low power consumption, high gain, wide coverage and real-time signal enhancement.

[0009] To solve the above technical problems, the specific technical solution of the present invention includes the following four steps:

[0010] Step 1: Design a low-power intelligent metasurface unit with 2-bit phase response and beamforming effect in the 5 GHz frequency band and determine the composition architecture of its array, as well as reasonably design the FPGA drive control board and control strategy of the intelligent metasurface.

[0011] Step 2: Use the three intelligent metasurfaces to build a distributed intelligent metasurface system, and use the conditional sample mean codebook optimization algorithm to calculate the optimal coding for each area in this scenario.

[0012] Step 3: Build a visual perception system and cooperate with the host computer to realize the recognition of human targets. At the same time, the target position information is converted into the corresponding code of the area in real time and transmitted to the FPGA drive control board to control the intelligent metasurface.

[0013] Step 4: Build a video wireless transmission platform through software radio (USRP), and use video transmission to intuitively demonstrate the system's real-time signal coverage enhancement for target users in non-line-of-sight scenarios.

[0014] Furthermore, step 1 specifically includes the following steps:

[0015] The designed low-power intelligent metasurface is composed of 8×12 metasurface units with 2-bit quantized phase. The basic pattern of each unit is composed of an irregular metal patch and two feeding rectangular patches. At the same time, in order to improve the oblique incidence stability of the unit, a row of metal vias is placed on both sides of the unit along the polarization orthogonal direction of the unit. The unit has a relatively simple stacked structure, including a top metal pattern layer, a bottom all-metal backplane layer, and an F4B dielectric substrate layer between them.

[0016] In order to achieve dynamic control of the phase of the reflected electromagnetic wave, it is necessary to embed active adjustable electronic components in the basic structure of the unit. Commonly used adjustable electronic components include PIN diodes, Schottky diodes, etc., but PIN diodes and other devices usually require a current of 5mA when turned on, and the power consumption is relatively large when forming a large-scale array. The metasurface unit designed by the present invention reduces power consumption by embedding a single-pole single-throw (SPST) switch chip based on CMOS technology in a carefully designed position. In order to achieve higher beam control accuracy, a unit with a 2-bit quantized phase is designed. By using a chip with two different state combinations, four reflection phase states with a phase difference of 90 degrees are realized, defined as "00", "01", "10" and "11". At the center frequency of 5GHz, the reflection coefficient amplitudes of the four coding states are all less than -1.5dB, which ensures the good reflection efficiency of the metasurface. At 5GHz, the reflection phases of the four coding states differ by 90 degrees in sequence, realizing four discrete phase states within a range of 360 degrees.

[0017] Furthermore, step 2 specifically includes the following steps:

[0018] The conditional sample mean codebook optimization algorithm used in the present invention is different from the traditional beam pattern coding optimization scheme in that the optimization method is based on the statistical characteristics of a large number of measured received signal energies and performs coding optimization for a specific environment without channel estimation, which is also called blind beamforming. The following is a basic theoretical analysis of the mathematical model of the intelligent metasurface controlled channel under general conditions. Usually, the channel coefficient from the transmitter to the receiver is defined as

[0019]

[0020] in

[0021] β n ∈(0,1),α n ∈[0,2π) (2)

[0022] β n represents the amplitude of the channel coefficient, α n represents the phase of the channel coefficient. In particular, h0∈C represents the sum of other channels from transmission to reception that do not pass through the smart metasurface, which can also be called the background channel.n ∈C,n=1,2,...,N represents the cascade channel from the nth metasurface unit from emission to reception. Define the vector

[0023] θ=(θ1,...,θ N ), (3)

[0024] θ N represents the reflection phase of the Nth metasurface unit, and its optional value set is

[0025] Φ K ={ω,2ω,...,Kω}, (4)

[0026] in

[0027]

[0028] ω represents the minimum interval of phase change in the set, and K is the number of elements in the set. Let X∈C represent the transmitted signal with an average power of P, that is,

[0029] E[|X| 2 ]=P, (6)

[0030] Then the received signal Y∈C can be calculated as

[0031]

[0032] Among them, Z~CN(0,σ 2 ) is used to simulate additive background noise, σ 2 represents the variance of the random distribution. Therefore, the accepted signal-to-noise ratio model can be expressed as

[0033]

[0034] In particular, the baseline signal-to-noise ratio is defined as

[0035]

[0036] Therefore, the signal-to-noise ratio improvement can be expressed as follows

[0037]

[0038] At this point, the optimization goal is already very clear, that is, to find the vector θ that maximizes the signal-to-noise ratio

[0039]

[0040] However, there are two challenges in solving the above problem. First, it is very complicated to numerically optimize the discrete variable θ. Second, from a practical perspective, it is difficult to obtain the channel coefficients {h0,...,h NThe cost of training is very high. The conditional sample mean data training method cleverly avoids these two difficulties and obtains θ that meets the target by calculating the mean of the conditional samples of random coding. This method first generates a set of random coding data samples of number T, and then It is a subset of him, where

[0041] Q nk ={t:θ nt =kω} (12)

[0042] The reflection phase of the nth metasurface unit is the index kω, and then the conditional sample mean of the reflection phase of each metasurface unit is calculated. The calculation formula is

[0043]

[0044] This conditional sample mean essentially reflects the value of each hypersurface unit in θ n = kω, so it is obvious that each θ can be selected based on this value. n ,Right now

[0045]

[0046] There are two main reasons why the present invention adopts the blind beam coding method: first, in the distributed intelligent super surface deployment scenario discussed in the present invention, most of them are accompanied by obvious multipath effects and complex channel models. In such an environment, the blind beamforming coding optimization method can make full use of the electromagnetic waves on each propagation path in the space, so that it can gather enough energy at a specified position. The second point is the joint control method of the distributed intelligent super surface. With the increase in the number of intelligent super surfaces, the traditional beamforming method requires the intelligent super surface and the target to perform multiple coordinate transformations to obtain the electromagnetic wave control angle of each intelligent super surface. The calculation complexity is high and the position accuracy requirements are high. Therefore, it is more convenient and efficient to use the conditional sample mean to optimize the coding of multiple intelligent super surfaces at the same time than the separate beamforming. Through the study of various blind beamforming coding algorithms, it is found that when the number of quantized phases of the intelligent super surface is not greater than 4, the conditional sample mean algorithm performs well in terms of optimization speed and signal-to-noise ratio improvement. The number of intelligent super surfaces used in the present invention just meets the application conditions of the conditional sample mean algorithm, so this method is selected as the basis for dynamic coverage optimization.

[0047] Furthermore, step 3 specifically includes the following steps:

[0048] At present, the hardware foundation of distributed intelligent metasurfaces and the conditional sample mean algorithm for optimizing blind beam coding are already available. However, to achieve dynamic signal coverage optimization, a target tracking platform that can provide real-time feedback on target position information is also required. The present invention builds a platform based on mature visual perception technology. The main hardware device of the platform is a binocular camera, and the software algorithm is based on the YOLO-v5 framework. The device uses dual cameras to simulate the human visual system, capture image information, and perceive three-dimensional structural information by comparing the slightly offset images taken by the two cameras. By continuously processing each frame of image information, the three-dimensional coordinate position of the target user is located in real time. Finally, the codebook pre-trained in step 3 is used to index the code at the corresponding position and send it to the driver board of each metasurface unit to update the status, thereby achieving dynamic signal coverage optimization of the target position and simplifying the process such as Figure 5 In this process, the host computer generates a coding matrix and distributes it to the FPGA control board of each metasurface unit synchronously through the switch. The control board maps the received code to the voltage that drives each metasurface unit to jointly regulate the complex electromagnetic waves in space.

[0049] Furthermore, step 4 specifically includes the following steps:

[0050] In order to intuitively demonstrate the capabilities of the proposed low-power dynamic coverage optimization system, the present invention constructs a wireless video transmission platform based on software radio. The quality of video transmission can clearly reflect the real-time received signal strength in the corresponding area, thereby effectively proving the effectiveness of the proposed system. Three intelligent metasurfaces, a binocular camera, a transmitting antenna and an omnidirectional receiving antenna are placed in fixed positions. The transmitting antenna is placed outside the laboratory wall, which proves that the system can work even in NLoS scenarios. In this experiment, when the target moves in different areas of space, the receiving antenna remains stationary and the change of the communication quality at the receiving end is observed. Two typical locations are selected for illustration. When the target in the green box moves to the receiving antenna area, the signal quality in the area is significantly improved, the video transmission is smooth, and the symbols in the constellation diagram are clustered together. However, when the target moves out of the receiving antenna area, the received signal-to-noise ratio at the antenna position becomes too low, resulting in a high bit error rate and a decrease in video transmission quality, because the distributed intelligent metasurface system guides the main electromagnetic energy to the target position. In addition, the experimental results show that the proposed system has the characteristics of tracking designated targets, which can contain multiple targets. This feature has great potential and scalability in future multi-target, multi-intelligent metasurface scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1Structural design and performance of the metasurface unit of the implementation example of the present invention: (a) exploded view; (b) top view; (c) side view; (d) reflection amplitude of each coding state; (e) reflection phase of each coding state.

[0052] Figure 2 Experimental prototype and measurements of implementation cases of the present invention: (a) measurement environment; (b) fabricated metasurface array prototype; (c) power consumption measurement; (d) normalized reflection amplitude; (e) normalized reflection phase; (f) far-field scattering diagrams for different beam directions

[0053] Figure 3 This is a schematic diagram of blind beam coding training in a distributed intelligent metasurface deployment according to an implementation example of the present invention.

[0054] Figure 4 The blind beam coding training performance of the implementation case of the present invention under distributed intelligent metasurface deployment: (a) the relationship between the number of intelligent metasurfaces and the received power; (b) the relationship between the number of training samples and the optimization effect; (ce) the received power heat map of no intelligent metasurface, one intelligent metasurface, and three intelligent metasurfaces.

[0055] Figure 5 This is the operation process of the signal tracking platform under the distributed intelligent metasurface deployment of the implementation case of the present invention.

[0056] Figure 6 Wireless communication experiment for an implementation case of the present invention: (a) experimental scenario; (b) communication performance when the target is within the receiving area; (c) communication performance when the target is outside the receiving area. DETAILED DESCRIPTION

[0057] In order to better understand the purpose, structure and function of the present invention, the low-power dynamic coverage optimization system based on distributed intelligent metasurface is further described in detail below in conjunction with the accompanying drawings.

[0058] To solve the above technical problems, the specific technical solution of the present invention includes the following four steps:

[0059] Step 1: Design a low-power intelligent metasurface unit with 2-bit phase response and beamforming effect in the 5 GHz frequency band and determine the composition architecture of its array, as well as reasonably design the FPGA drive control board and control strategy of the intelligent metasurface.

[0060] Step 2: Use the three intelligent metasurfaces to build a distributed intelligent metasurface system, and use the conditional sample mean codebook optimization algorithm to calculate the optimal coding for each area in this scenario.

[0061] Step 3: Build a visual perception system and cooperate with the host computer to realize the recognition of human targets. At the same time, the target position information is converted into the corresponding code of the area in real time and transmitted to the FPGA drive control board to control the intelligent metasurface.

[0062] Step 4: Build a video wireless transmission platform through USRP, and realize real-time signal coverage enhancement of the target in non-line-of-sight scenarios based on the low-power dynamic coverage optimization system of the distributed intelligent metasurface.

[0063] Furthermore, step 1 specifically includes the following steps:

[0064] The designed low-power intelligent metasurface is composed of 8×12 metasurface units with 2-bit quantized phase, such as Figure 1 As shown in (b), the basic pattern of each unit is composed of an irregular metal patch and two feeding rectangular patches, and its geometric parameters are as follows: L1 = 20mm, L2 = 0.89mm, L3 = 11.32mm, W1 = 20mm, W2 = 3.8mm, W3 = 2.9mm, W4 = 2.47mm, W5 = 6.2mm, P1 = 6.62mm, P2 = 6.25mm, H = 5mm, R = 0.6mm. At the same time, in order to improve the oblique incidence stability of the unit, a row of metal vias is placed on both sides of the unit along the orthogonal polarization direction of the unit. The stacked structure of the unit is relatively simple, including a top metal pattern layer, a bottom all-metal backplane layer, and an F4B dielectric substrate layer (ε r =2.65andtanδ=0.001).

[0065] In order to achieve dynamic control of the phase of the reflected electromagnetic wave, it is necessary to embed active adjustable electronic components in the basic structure of the unit. Commonly used adjustable electronic components include PIN diodes, Schottky diodes, etc., but devices such as PIN diodes usually require a current of 5mA when turned on, and consume a lot of power when forming a large-scale array. The metasurface unit designed in this study reduces power consumption by embedding a single-pole single-throw (SPST) switch chip based on CMOS technology in a carefully designed position. The bias voltage of the switch chip is +2.5V in the on state and -2.5V in the off state. In addition, the chip consumes only 10uA of current in the on state, and the power consumption of each chip is only 25uW. Compared with PIN diodes, the power consumption is reduced by three orders of magnitude. In order to achieve higher beam control accuracy, the present invention designs a unit with a 2-bit quantized phase. By using a chip with two different state combinations, we realize four reflection phase states with a phase difference of 90 degrees, defined as "00", "01", "10" and "11"

[0066] The present invention performs full-wave simulation in CST Microwave Studio, and obtains the reflection amplitude and phase of the unit under different encoding states by jointly simulating the passive structure of the unit and the S parameter file of the switch chip. Figure 1 (d) It can be seen that at the center frequency of 5 GHz, the reflection coefficient amplitudes of the four coding states are all less than -1.5 dB, which ensures the good reflection efficiency of the metasurface. Figure 1 As shown in (e), at 5 GHz, the reflection phases of the four coding states differ by 90 degrees respectively, realizing four discrete phase states within a range of 360 degrees.

[0067] In order to verify the reliability of the designed metasurface performance, a low-power metasurface array with an effective size of 160×240mm was fabricated using standard printed circuit board (PCB) technology. 2 In addition, a custom driver circuit board with FPGA as the main controller was developed to ensure flexible beam steering. A vector network analyzer (Agilent N5245A) was used in a microwave darkroom to measure the reflection amplitude, phase, and far-field beam pattern of the metasurface. Figure 2 As shown in (a), the transmitting antenna is placed in the far-field region of the metasurface to achieve approximate plane wave excitation when measuring the far-field scattering pattern. The sample image and unit details are shown in Figure 2 (b) As shown. Since the interface between the metasurface array and the driving circuit board is considered in the design layout, the two boards can be easily integrated and used, which is convenient for experiments. In addition, the power consumption of the metasurface array under full power operation is also measured, as shown in Figure 2 (c) The multimeter current value shows that when all chips on the metasurface array are turned on, the total current consumption is 2.27mA, and the average current consumption of each low-power chip is 11uA, which is consistent with the theoretical design. This low power consumption ensures the feasibility of the subsequent deployment of distributed intelligent metasurfaces in the present invention.

[0068] Figure 2 (d) and Figure 2(e) shows the measured metasurface reflection coefficient amplitude and phase, normalized using the measured values ​​of a metal plate of the same size as the metasurface array. The curve shows that at a frequency of 5 GHz, the measured reflection coefficient amplitude of the designed metasurface is approximately 2 dB lower than the simulation result. Compared with the simulation results, this difference is attributed to the potential differences in dielectric constant and substrate loss during the manufacturing process. According to the measured phase data, the four encoding states maintain a 2-bit quantized phase within the operating bandwidth. Overall, the measured metasurface reflection coefficient amplitude and phase are basically consistent with the simulation results. In addition, the present invention designs several phase encodings for different beamforming directions, covering angles from 10° to 40°, and measurements are performed at intervals of 10°. The measured far-field scattering diagram is shown in Figure 2. Figure 2 (f) shows the excellent beam steering capability of the smart metasurface. However, it can be observed that the limited array size during the manufacturing process leads to some side lobes in the reflected beam. Overall, the maximum beam gain direction confirms the reliability of the designed metasurface and lays the foundation for further dynamic signal coverage using distributed smart metasurfaces.

[0069] Furthermore, step 2 specifically includes the following steps:

[0070] The conditional sample mean codebook optimization algorithm used in the present invention is different from the traditional beam pattern coding optimization scheme in that the optimization method is based on the statistical characteristics of a large number of measured received signal energies and performs coding optimization for a specific environment without channel estimation, which is also called blind beamforming. The following is a basic theoretical analysis of the mathematical model of the intelligent metasurface controlled channel under general conditions. Usually, the channel coefficient from the transmitter to the receiver is defined as

[0071]

[0072] in

[0073] β n ∈(0,1),α n ∈[0,2π) (2)

[0074] β n represents the amplitude of the channel coefficient, α n represents the phase of the channel coefficient. In particular, h0∈C represents the sum of other channels from transmission to reception that do not pass through the smart metasurface, which can also be called the background channel. n ∈C,n=1,2,...,N represents the cascade channel from the nth metasurface unit from emission to reception. Define the vector

[0075] θ=(θ1,...,θ N ), (3)

[0076] θ Nrepresents the reflection phase of the Nth metasurface unit, and its optional value set is

[0077] Φ K ={ω,2ω,...,Kω}, (4)

[0078] in

[0079]

[0080] ω represents the minimum interval of phase change in the set. Let X∈C represent the transmitted signal with average power P, that is,

[0081] E[|X| 2 ]=P, (6)

[0082] Then the received signal Y∈C can be calculated as

[0083]

[0084] Among them, Z~CN(0,σ 2 ) is used to simulate additive background noise, σ 2 represents the variance of the random distribution. Therefore, the accepted signal-to-noise ratio model can be expressed as

[0085]

[0086] In particular, the baseline signal-to-noise ratio is defined as

[0087]

[0088] Therefore, the signal-to-noise ratio improvement can be expressed as follows

[0089]

[0090] At this point, the optimization goal is already very clear, that is, to find the vector θ that maximizes the signal-to-noise ratio

[0091]

[0092] However, there are two challenges in solving the above problem. First, it is very complicated to numerically optimize the discrete variable θ. Second, from a practical perspective, it is difficult to obtain the channel coefficients {h0,...,h N The cost of training is very high. The conditional sample mean data training method cleverly avoids these two difficulties and obtains θ that meets the target by calculating the mean of the conditional samples of random coding. This method first generates a set of random coding data samples of number T, and then It is a subset of him, where

[0093] Q nk ={t:θ nt=kω} (12)

[0094] The reflection phase of the nth metasurface unit is the index kω, and then the conditional sample mean of the reflection phase of each metasurface unit is calculated. The calculation formula is

[0095]

[0096] This conditional sample mean essentially reflects the value of each hypersurface unit in θ n = kω, so it is obvious that each θ can be selected based on this value. n ,Right now

[0097]

[0098] There are two main reasons why the present invention adopts the blind beam coding method: first, in the distributed intelligent super surface deployment scenario discussed in the present invention, most of them are accompanied by obvious multipath effects and complex channel models. In such an environment, the blind beam forming coding optimization method can make full use of the electromagnetic waves on each propagation path in the space, so that it can gather enough energy at a specified position. Second, it is the joint control method of the distributed intelligent super surface. With the increase in the number of intelligent super surfaces, the traditional beam forming method requires the intelligent super surface and the target to perform multiple coordinate transformations to obtain the electromagnetic wave control angle of each intelligent super surface. The calculation complexity is high and the position accuracy requirements are high. Therefore, it is more convenient and efficient to use the conditional sample mean to optimize the coding of multiple intelligent super surfaces at the same time than the separate beam forming. Through the study of various blind beam coding algorithms, it can be found that when the number of quantized phases of the intelligent super surface is not greater than 4, the conditional sample mean algorithm performs well in terms of optimization speed and signal-to-noise ratio improvement. The number of intelligent super surfaces used in the present invention just meets the application conditions of the conditional sample mean algorithm, so this method is selected as the basis for dynamic coverage optimization.

[0099] Schematic diagram of the distributed intelligent metasurface blind beam coding training experiment Figure 3 As shown. The experimental space is divided into 20 equal-area areas, each area is numbered, and smart metasurfaces are deployed at multiple locations in the environment. In order to accurately measure the received power at the antenna position, a horizontally polarized omnidirectional antenna with a central operating frequency of 5GHz is designed as a receiving antenna. The antenna is placed at the specified position to receive controlled electromagnetic waves emitted from all directions in the space. The signal power value is then fed back to the host computer in real time using a spectrum analyzer to obtain a set of sample data for the specified position. Repeating the above steps a predetermined number of times can obtain a fairly large data set, which is then used to calculate the blind beamforming code suitable for the position using the conditional sample mean algorithm.

[0100] The received power of the same area without smart metasurface, with one smart metasurface, and with three smart metasurfaces was measured and compared. Figure 4 (a). The measurement data shows that three smart metasurfaces can more effectively enhance the received power at a specified location than one smart metasurface. Of course, deploying a single smart metasurface can also improve the received power compared to no smart metasurface. In addition, the present invention analyzes the impact of different data set sizes on the improvement of system performance. Figure 4 The fitting curve in (b) shows that the received power increases significantly with the increase of sample size when the sample size is less than 150. However, when the sample size exceeds 150, the improvement effect tends to be stable. Figure 4 (c) to 4(e) show the received power heat maps of 20 divided areas under three different deployment schemes. The overall color change can more intuitively observe the significant improvement of the electromagnetic environment brought about by the deployment of distributed intelligent metasurfaces.

[0101] Furthermore, step 3 specifically includes the following steps:

[0102] We currently have the hardware foundation for distributed intelligent metasurfaces and the conditional sample mean algorithm for optimizing the blind beam coding codebook, but to achieve dynamic signal coverage optimization, we also need a target tracking platform that can provide real-time feedback on target location information.

[0103] The present invention builds a signal tracking platform based on mature visual perception technology. The main hardware device of the platform is a binocular camera, and the software algorithm is based on the YOLO-v5 framework. The device uses dual cameras to simulate the human visual system, capture image information, and perceive depth and three-dimensional structure information by comparing the slightly offset images taken by the two cameras. It continuously processes each frame of image information and locates the three-dimensional coordinate position of the target in real time. Finally, the codebook pre-trained in the previous section is used to index the encoding of the corresponding position and send it to the driver board of each hypersurface unit to update the status, thereby realizing dynamic signal coverage optimization of the target position, such as Figure 5 The simplified process is shown in the figure. In this process, the host computer generates a coding matrix and distributes it to the FPGA control board of each metasurface unit synchronously through the switch. The control board maps the received code to the voltage that drives each metasurface unit to jointly regulate the complex electromagnetic waves in space.

[0104] Furthermore, step 4 specifically includes the following steps:

[0105] In order to intuitively demonstrate the capabilities of the proposed low-power dynamic coverage optimization system, the present invention constructs a wireless video transmission platform based on software radio. The quality of video transmission can clearly reflect the real-time received signal strength of the corresponding area, thereby effectively proving the effectiveness of the proposed system. Three smart metasurfaces, a binocular camera, a transmitting antenna and an omnidirectional receiving antenna are placed in fixed positions, such as Figure 6 (a) shows. Placing the transmitting antenna outside the laboratory wall proves that the system can work even in NLoS scenarios. In this experiment, when the target moves in different areas of space, the receiving antenna remains stationary, and the changes in the communication quality at the receiving end can be observed. The present invention selects two typical locations for illustration, such as Figure 6 (b) and Figure 6 (c) As shown. When the target in the green box moves to the receiving antenna area, the signal quality in this area is significantly improved, the video transmission is smooth, and the symbols in the constellation diagram are clustered together. However, when the target moves out of the receiving antenna area, since the distributed intelligent metasurface system guides the main electromagnetic energy to the target location, the received signal-to-noise ratio at the antenna location becomes too low, resulting in a high bit error rate and a decrease in video transmission quality. In addition, the experimental results show that the proposed system has the characteristics of tracking designated targets, which can include multiple targets. This feature has great potential and scalability in future multi-target, multi-intelligent metasurface scenarios.

Claims

1. A low-power dynamic coverage optimization method based on distributed intelligent metasurface, characterized in that: The steps include: Step 1: Design a low-power intelligent metasurface unit with 2-bit phase response and beamforming effect in the 5 GHz frequency band and determine the composition architecture of its array, and reasonably design the FPGA drive control board and control strategy of the intelligent metasurface; Step 2: Use several blocks of the intelligent metasurface to build a distributed intelligent metasurface system, and use the conditional sample mean codebook optimization algorithm to calculate the optimal coding of each area in the scenario; Step 3: Build a visual perception system and cooperate with the host computer to realize the recognition of human targets. At the same time, the target position information is converted into the corresponding code of the area in real time and transmitted to the FPGA drive control board to control the intelligent metasurface.

2. According to claim 1, a low-power dynamic coverage optimization method based on distributed intelligent metasurface is characterized in that: The smart metasurface is composed of 8×12 metasurface units with 2-bit quantized phase, and the basic pattern of each unit is composed of an irregular metal patch and two feeding rectangular patches; the geometric parameters of the unit are: L1=20mm, L2=0.89mm, L3=11.32mm, W1=20mm, W2=3.8mm, W3=2.9mm, W4=2.47mm, W5=6.2mm, P1=6.62mm, P2=6.25mm, H=5mm, R=0.6mm; in order to improve the oblique incidence stability of the unit, a row of metal vias is placed on both sides of the unit along the orthogonal direction of the polarization of the unit; the stacked structure of the unit includes a top metal pattern layer, a bottom all-metal backplane layer, and an F4B dielectric substrate layer between them.

3. According to claim 2, a low-power dynamic coverage optimization method based on distributed intelligent metasurface is characterized in that: Active adjustable electronic components are embedded in the basic structure of the metasurface unit to achieve dynamic control of the phase of the reflected electromagnetic wave; the adjustable electronic components are single-pole single-throw switch chips based on CMOS technology; the bias voltage of the switch chip in the on state is +2.5V, and the bias voltage in the off state is -2.5V.

4. According to claim 2, a low-power dynamic coverage optimization method based on distributed intelligent metasurface is characterized in that: The metasurface unit realizes four reflection phase states with a phase difference of 90 degrees, defined as "00", "01", "10" and "11", by using a switch chip with two different state combinations; at 5GHz, the reflection phases of the four coding states differ by 90 degrees respectively, realizing four discrete phase states within a range of 360 degrees.

5. The low-power dynamic coverage optimization method based on distributed intelligent metasurface according to claim 1, characterized in that: In step 2, the conditional sample mean codebook optimization algorithm is used to calculate the optimal coding for each area in the scenario. The specific steps include: dividing the experimental space into multiple equal-area areas, numbering each area, and deploying intelligent metasurfaces at multiple locations in the environment; using a horizontally polarized omnidirectional antenna with a central operating frequency of 5 GHz as the receiving end, and measuring the received power at the antenna position; placing the antenna at a specified position, receiving regulated electromagnetic waves from all directions in space, and using a spectrum analyzer to feed back the signal power value to the host computer in real time, thereby obtaining a set of sample data for the specified position; repeating the above steps a predetermined number of times to obtain a data set, and then using the data set to calculate the blind beam coding suitable for the position through the conditional sample mean algorithm.

6. The low-power dynamic coverage optimization method based on distributed intelligent metasurface according to claim 1, characterized in that: The visual perception system includes a binocular camera and a target tracking host computer that can provide real-time feedback on the target position information; in step 3, the host computer is used to recognize the human target, and the target position information is converted into the corresponding code of the area in real time and transmitted to the FPGA driving control board to control the intelligent super surface, specifically including: based on the deep learning target detection algorithm, the binocular camera is used to capture image information and perceive the three-dimensional structure information, and the three-dimensional coordinate position of the target is continuously recognized and located in real time; using the pre-trained blind beam codebook, the host computer indexes the code of the corresponding position, and distributes it to the control board of each intelligent super surface synchronously through the switch, and the control board maps the received code to the voltage driving each super surface unit, and jointly regulates the complex electromagnetic waves in the space to achieve dynamic signal coverage optimization of the target position.

7. The low-power dynamic coverage optimization method based on distributed intelligent metasurface according to claim 6, characterized in that: The deep learning-based target detection algorithm is a visual perception algorithm based on the YOLO-v5 framework.

8. A low-power dynamic coverage optimization method based on a distributed intelligent metasurface according to any one of claims 1 to 7, characterized in that: The number of the smart metasurfaces is no more than 4.

9. A low-power dynamic coverage optimization method based on a distributed intelligent metasurface according to any one of claims 1 to 7, characterized in that: It also includes step 4: building a video wireless transmission platform through software radio, and using video transmission to intuitively demonstrate the system's real-time signal coverage enhancement for target users in non-line-of-sight scenarios.

10. A low-power dynamic coverage optimization system based on distributed intelligent metasurface, characterized in that: The system includes several low-power 2-bit intelligent metasurfaces and their control boards, a binocular camera and a supporting host computer control system; the system uses a binocular camera to capture image data in real time, and uses a deep learning-based target detection algorithm to identify human targets in the image and calculate their relative coordinates in three-dimensional space; according to the coordinates of the grid unit where the target is located, the corresponding intelligent metasurface coding sequence optimized by the conditional sample mean algorithm is indexed, and converted into a voltage control signal of the intelligent metasurface through a control board with FPGA as the main controller chip, thereby realizing real-time dynamic enhancement of the wireless signal coverage of the distributed intelligent metasurface in the target area.

Images