5G personal wi-fi signal enhancement method and system

Abstract

The application provides a 5G portable WiFi signal enhancement method and system, which collects surrounding environment data through an environment perception sensor, optimizes device placement scheme in combination with a deep learning algorithm, and improves signal coverage range; dynamically adjusts the strategies of transmission power, signal frequency and modulation mode, and can optimize signal strength according to environmental changes; uses quantum computing to analyze interference signals, and forms a null through beamforming technology, effectively suppresses interference signal reception, and enhances target direction signal strength; in combination with the Internet of Things technology, device state data is uploaded to an edge server in real time for big data analysis, and personalized signal enhancement schemes are pushed according to different user use environments; through accurate placement guidance and dynamic frequency band switching strategies, users enjoy smoother network connection in the use process, and ensure good signal quality in complex environments. The application not only can improve the signal quality of 5G portable WiFi, but also can provide users with more intelligent and personalized network experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of 5G, in particular to a 5G personal WiFi signal enhancement method and system. BACKGROUND

[0002] 5G intelligent personal WiFi devices have gradually become popular in recent years, providing users with more convenient and high-speed network experience. 5G personal WiFi devices can provide faster data transmission speed than traditional 4G devices, with a theoretical downlink rate of 10Gbps, and actual use of several hundred Mbps or even higher, easily supporting high-definition video playback, 4K live streaming and large file downloads; These devices are usually designed to be lightweight and portable, without the need for installation and wiring, providing WiFi signals anytime and anywhere; Many 5G personal WiFi devices support large traffic packages, with monthly traffic up to 1500G or more, meeting the user's long-term and high-traffic usage needs; Most devices can connect to multiple devices at the same time, usually around 32, meeting the needs of multiple people or multiple devices accessing the Internet at the same time. The existing 5G personal WiFi devices have problems such as poor signal, lack of intelligence, etc. due to their variable working environment and simple structure. SUMMARY

[0003] The present application is based on the above problems, and proposes a 5G personal WiFi signal enhancement method and system. The scheme of the present application not only improves the signal quality of 5G personal WiFi, but also provides users with more intelligent and personalized network experience.

[0004] Therefore, one aspect of the present application proposes a 5G personal WiFi signal enhancement method, comprising:

[0005] Turning on the multi-type environment perception sensor built-in the first 5G personal WiFi device to collect first data;

[0006] Real-time acquisition of a first surrounding environment image within a first preset range of a location where a first terminal connected to the first 5G personal WiFi device is located, and a second surrounding environment image within a second preset range of a location where the first 5G personal WiFi device is located;

[0007] Using a target detection algorithm based on deep learning to identify signal blocking object data from the first surrounding environment image and the second surrounding environment image, and determining a placement scheme of the first 5G personal WiFi device according to the signal blocking object data, the first data and a preset device placement model;

[0008] Activating the network monitoring program built-in the first 5G pocket WiFi device to conduct full-band scanning, continuously collecting signal data such as signal strength, phase, polarization characteristics, and noise spectrum distribution of each frequency band to construct an original data sample set;

[0009] For the original data sample set, using an adaptive filtering algorithm to dynamically adjust the filtering parameters, removing noise interference, using high-order statistical analysis method to mine the nonlinear characteristics of signal data, and cooperating with time-frequency analysis tool to extract time-varying characteristics and frequency domain distribution characteristics of the frequency band;

[0010] Using an ensemble learning algorithm, fusing multiple weak classifiers, and taking historical frequency band characteristics as input to construct a frequency band quality evaluation model;

[0011] Combining the frequency band quality evaluation model, the time-varying characteristics and the frequency domain distribution characteristics, the frequency band quality evaluation result is obtained;

[0012] According to the frequency band quality evaluation result, based on deep reinforcement learning algorithm, the frequency band switching action of the first 5G pocket WiFi device is expanded to continuous space;

[0013] With the parallel computing capability of quantum computing, a quantum neural network is constructed for interference signal analysis;

[0014] According to the interference analysis result, the first 5G pocket WiFi device uses coding technology to dynamically adjust the coding rate according to the interference strength;

[0015] Combined with orthogonal frequency division multiplexing technology, subcarriers are allocated, and subcarriers of frequency bands severely affected by interference are used to transmit redundant information to ensure error-free transmission of critical data;

[0016] Introducing beamforming technology, using multiple antenna arrays to form nulls in the direction of interference, suppressing interference signal reception, and enhancing signal strength in the target direction;

[0017] Continuously using semantic segmentation algorithm in image recognition and processing technology to divide the latest environment image obtained in real time, identify the position, contour and material of newly appeared occlusion, and evaluate its signal occlusion degree;

[0018] According to the evaluation result of the signal occlusion degree of the newly appeared occlusion, combined with the temperature and humidity, electromagnetic environment change data feedback by the Internet of Things sensor, the first 5G pocket WiFi device dynamically adjusts the transmission power, signal frequency, modulation mode, and maintains high-frequency interaction with the mobile communication base station, uploads the local environment and signal state, downloads the latest optimization instructions, forms a closed-loop optimization link, and continuously adapts and optimizes the signal enhancement effect;

[0019] The first 5G personal WiFi device key state data is uploaded to the edge server through the Internet of Things, the edge server performs big data analysis according to the data of a plurality of 5G personal WiFi devices, and based on the data analysis result, personalized signal enhancement schemes are pushed to different users.

[0020] Optionally, the step of utilizing a target detection algorithm based on deep learning to identify signal shielding object data from the first and second surrounding environment images, and determining a placement scheme of the first 5G personal WiFi device according to the signal shielding object data, the first data and a preset device placement model, comprises:

[0021] The first and second surrounding environment images are subjected to standardization processing to obtain third and fourth surrounding environment images;

[0022] The third and fourth surrounding environment images are subjected to target detection and instance segmentation using a target detection algorithm based on deep learning, specifically including: extracting image features to generate candidate regions; classifying the candidate regions to identify the type of shielding object; generating a shielding object pixel-level mask; and calculating the accurate contour and spatial position information of the shielding object;

[0023] Based on the principle of binocular vision, the third and fourth surrounding environment images are subjected to three-dimensional reconstruction, specifically including: extracting feature points from the images and matching them; calculating the feature point disparity to construct a depth map; generating a three-dimensional point cloud model of the shielding object; and analyzing the surface features of the shielding object based on a material recognition model to determine the material properties;

[0024] A signal propagation model is constructed in combination with the first environmental parameters in the first data, including: calculating the signal attenuation coefficient based on the three-dimensional position and material of the shielding object; considering the influence of temperature and humidity on signal propagation; simulating the multi-path effect and signal reflection, diffraction phenomena;

[0025] Based on a preset device placement model, multi-objective optimization calculation is performed, including: inputting the shielding object data, first environmental parameters and signal propagation model; considering the device heat dissipation requirement and operation convenience; iteratively optimizing the placement position and angle using a reinforcement learning algorithm; and outputting an optimal placement scheme including specific coordinates and placement posture.

[0026] Optionally, the step of using an ensemble learning algorithm to fuse a plurality of weak classifiers and taking historical frequency band features as input to construct a frequency band quality evaluation model comprises:

[0027] Based on historical frequency band characteristics, a multi-dimensional feature vector is constructed, including: extracting statistical characteristics of signal strength, including mean, variance, skewness, kurtosis; calculating signal stability indicators, including signal jitter rate, frequency of disconnection; generating spectral occupancy, bandwidth utilization and other spectral characteristics; normalizing feature data and detecting outliers;

[0028] A plurality of basic classifiers for different feature subspaces are constructed, including: deploying gradient boosting decision trees for processing continuous features; configuring AdaBoost classifiers to process categorical features; applying random forests to process the correlation of high-dimensional features; assigning initial weights to each weak classifier;

[0029] An adaptive integration strategy is designed to dynamically adjust the weights of the classifiers, including: calculating the prediction accuracy of each weak classifier; dynamically updating the weights of the classifiers based on the accuracy; using a weighted voting mechanism to integrate the prediction results of each classifier; introducing a time decay factor to reduce the weight of historical data;

[0030] A generative adversarial network is deployed to expand the training data, including: constructing a generator network to simulate various interference scenarios; designing a discriminator network to distinguish between real and generated data; iteratively training the GAN network to generate high-quality samples; adding generated samples to the training set to improve model robustness;

[0031] Fine-grained classification of frequency band quality is implemented, including: establishing a multi-level evaluation standard, including main classification and sub-classification; dynamically updating the threshold parameters of each level; combining time series analysis to predict the trend of frequency band quality changes; continuously optimizing the evaluation model parameters.

[0032] Optionally, the step of combining the frequency band quality evaluation model, the time-varying feature, and the frequency domain distribution characteristic to obtain the frequency band quality evaluation result includes:

[0033] The input multi-source features are standardized, specifically including: normalizing the evaluation indicators output by the frequency band quality evaluation model; performing time window segmentation processing on the time-varying features; converting the frequency domain distribution characteristics into a unified feature vector; constructing a feature fusion matrix;

[0034] Dynamic evaluation based on time-varying features, specifically including: calculating the time series fluctuation indicators of signal strength; extracting the periodic variation patterns of signal quality; analyzing the mutation feature points of signal parameters; generating time series evaluation scores;

[0035] In-depth analysis of frequency domain distribution characteristics, including: calculating the spectral energy distribution density; identifying frequency domain interference characteristics; evaluating frequency band utilization efficiency; generating frequency domain evaluation scores;

[0036] The weighted fusion strategy is used to integrate the evaluation results of each dimension, including: designing an adaptive weight calculation method; considering the reliability indicators of different features; fusing the evaluation scores of each dimension; and generating a comprehensive evaluation index;

[0037] Based on the comprehensive evaluation index, the final frequency band quality evaluation result is generated, including: establishing a multi-threshold evaluation standard; calculating the confidence level of quality grade; generating an evaluation result report; and providing optimization suggestion instructions.

[0038] Optionally, the step of expanding the frequency band switching action of the first 5G personal WiFi device to a continuous space based on the frequency band quality evaluation result according to a deep reinforcement learning algorithm, includes:

[0039] The state representation of deep reinforcement learning is constructed, including: encoding the current frequency band quality evaluation result; fusing the historical frequency band switching record; collecting network environment parameters; and constructing a multi-dimensional state vector;

[0040] The discrete frequency band switching is expanded to a continuous action space, including: defining the continuous value range of frequency and bandwidth; designing an action space mapping function; establishing action constraint conditions; and implementing an action smooth transition mechanism;

[0041] The double-network architecture is deployed to implement policy learning, including: constructing an Actor network to output continuous action values; deploying a Critic network to evaluate action values; designing an experience replay buffer; and implementing a target network soft update mechanism;

[0042] A multi-objective reward function is established, including: evaluating the frequency band switching effect; considering the switching time cost; calculating the energy consumption penalty term; and fusing the user experience index;

[0043] Online learning and policy updating are implemented, including: collecting interactive experience data; calculating the time difference error; updating the Actor-Critic network parameters; and optimizing the exploration-exploitation balance.

[0044] Optionally, the step of constructing a quantum neural network for interference signal analysis by means of the parallel computing capability of quantum computing, includes:

[0045] The interference signal data is encoded into a quantum state, including: using an amplitude encoding method to map signal features to quantum bits; constructing a quantum state superposition to represent multi-dimensional features; implementing quantum entanglement of feature space; and establishing a quantum data preprocessing pipeline;

[0046] A special quantum neural network architecture is designed, including: deploying a parameterized quantum circuit layer; constructing a quantum convolution layer to process time-frequency features; designing a quantum pooling operation to compress features; and implementing a quantum-classical hybrid computing interface;

[0047] Feature analysis using quantum parallelism, including: simultaneously processing interference features of multiple frequency bands; parallel computing time-frequency correlation; extracting hidden features through quantum state evolution; constructing quantum feature map;

[0048] Quantum classification of interference patterns, including: designing quantum measurement operators; constructing quantum decision trees; implementing quantum ensemble learning; outputting probability distribution of interference types;

[0049] Optimizing prediction models based on quantum feedback, including: designing quantum gradient descent algorithm; implementing quantum acceleration of parameter update; optimizing quantum-classical data conversion; dynamically adjusting model structure.

[0050] Optionally, the first 5G personal WiFi device dynamically adjusts the coding rate according to the interference analysis result using coding technology, including:

[0051] Multi-dimensional evaluation based on interference analysis results, including: calculating the dynamic change range of signal-to-noise ratio; analyzing the time-frequency characteristics of interference signals; evaluating the impact of interference on channel capacity; generating interference intensity quantization indicators;

[0052] Selecting the optimal coding scheme according to the interference intensity, including: constructing a hybrid coding library of LDPC codes and polar codes; designing coding scheme evaluation criteria; calculating performance indicators of different coding schemes; selecting the coding scheme most suitable for the current interference environment;

[0053] Dynamically adjusting coding parameter configuration, including: calculating target coding rate according to interference intensity; optimizing codeword length and check bit distribution; adjusting interleaving depth and pattern structure; generating optimized coding parameters;

[0054] Implementing real-time switching of coding schemes, including: designing coding switching buffer mechanism; implementing smooth transition of coding parameters; monitoring coding performance indicators; executing dynamic update of coding schemes;

[0055] Establishing a coding effect feedback mechanism, including: collecting decoding error statistics; analyzing throughput trend; evaluating coding overhead and gain ratio; optimizing coding strategy decision model.

[0056] Optionally, the combination of orthogonal frequency division multiplexing technology, subcarrier allocation, and the use of severely interfered frequency band subcarriers for transmitting redundant information to ensure error-free transmission of critical data, including:

[0057] Dynamic evaluation of subcarriers in OFDM systems, including: measuring the signal-to-noise ratio of each subcarrier; evaluating the interference level between subcarriers; calculating subcarrier channel capacity; generating subcarrier quality score;

[0058] Classifying the data to be transmitted, including: identifying key data packets; calculating data transmission priorities; estimating required transmission reliability; generating data transmission strategies;

[0059] Intelligently allocating based on subcarrier states, including: allocating high-quality subcarriers to key data; using interfered subcarriers for redundant transmission; optimizing subcarrier combination schemes; and implementing dynamic bandwidth allocation;

[0060] Designing an adaptive redundant transmission mechanism, including: calculating the proportion of redundant data; selecting a redundant coding scheme; determining the distribution of redundant data; and implementing redundant transmission scheduling;

[0061] Establishing a real-time monitoring and adjustment mechanism, including: tracking data transmission success rates; analyzing transmission delay changes; evaluating resource utilization efficiency; and dynamically optimizing allocation strategies.

[0062] Optionally, the step of introducing beamforming technology, using a multi-antenna array, forms nulls in the direction of interference, suppresses interference signal reception, and enhances the signal strength in the target direction, includes:

[0063] Channel characteristic analysis using a multi-antenna array, including: obtaining the received signal characteristics of each antenna; calculating the signal correlation between antennas; estimating the angle of arrival; and constructing a spatial channel matrix;

[0064] Accurate identification of the spatial position of the interference source, including: performing spatial spectrum scanning; calculating the signal angle of arrival distribution; identifying the direction of the main interference source; and generating an interference source spatial distribution map;

[0065] Designing optimal beamforming weights, including: constructing an adaptive beamforming algorithm; calculating the constraint conditions of the null direction; optimizing the gain of the main lobe direction; and generating a weight vector;

[0066] Implementing precise beam control, including: configuring the phase relationship of the antenna array; adjusting the transmission power of each antenna; synchronizing the timing of the antenna array; and performing beamforming operations;

[0067] Establishing a real-time optimization feedback mechanism, including: monitoring the beamforming effect; analyzing the change in signal-to-interference ratio; evaluating system performance indicators; and adjusting the beamforming strategy.

[0068] Another aspect of the present application provides a 5G portable WiFi signal enhancement system for performing a 5G portable WiFi signal enhancement method, including: a first 5G portable WiFi device, a first terminal and an edge server;

[0069] The first 5G portable WiFi device is configured to: turn on the multi-type environment perception sensor built-in the first 5G portable WiFi device to collect first data;

[0070] The edge server is configured to:

[0071] acquire a first surrounding environment image within a first preset range of a location where a first terminal connected to the first 5G pocket WiFi device is located, and a second surrounding environment image within a second preset range of a location where the first 5G pocket WiFi device is located in real time;

[0072] identify signal shielding object data from the first surrounding environment image and the second surrounding environment image by using a target detection algorithm based on deep learning, and determine a placement scheme of the first 5G pocket WiFi device according to the signal shielding object data, the first data, and a preset device placement model;

[0073] activate a network monitoring program built-in the first 5G pocket WiFi device to perform full-band scanning, continuously collect signal data such as signal strength, phase, polarization characteristics, and noise spectrum distribution of each frequency band to construct an original data sample set;

[0074] for the original data sample set, use an adaptive filtering algorithm to dynamically adjust filtering parameters, remove noise interference, use a high-order statistical analysis method to mine nonlinear characteristics in the signal data, and cooperate with a time-frequency analysis tool to extract time-varying characteristics and frequency domain distribution characteristics of the frequency band;

[0075] use an ensemble learning algorithm to fuse multiple weak classifiers, and use historical frequency band characteristics as input to construct a frequency band quality evaluation model;

[0076] combine the frequency band quality evaluation model, the time-varying characteristics, and the frequency domain distribution characteristics to obtain a frequency band quality evaluation result;

[0077] based on the frequency band quality evaluation result, expand the frequency band switching action of the first 5G pocket WiFi device to a continuous space based on a deep reinforcement learning algorithm;

[0078] with the parallel computing capability of quantum computing, construct a quantum neural network for interference signal analysis;

[0079] The first 5G pocket WiFi device is further configured to:

[0080] according to the interference analysis result, use coding technology to dynamically adjust the coding rate according to the interference strength;

[0081] combine orthogonal frequency division multiplexing technology, allocate subcarriers, and use subcarriers of frequency bands severely affected by interference to transmit redundant information to ensure error-free transmission of critical data;

[0082] introduce beamforming technology, use a multi-antenna array to form a null in the interference direction, suppress interference signal reception, and enhance the signal strength of the target direction;

[0083] The edge server is further configured to:

[0084] The semantic segmentation algorithm in the image recognition and processing technology is continuously used to divide the latest environment image acquired in real time, identify the position, contour and material of the newly appeared shielding object, and evaluate the signal shielding degree of the newly appeared shielding object;

[0085] According to the evaluation result of the signal shielding degree of the newly appeared shielding object, in combination with the temperature and humidity and electromagnetic environment change data fed back in real time by the Internet of Things sensor, the first 5G personal WiFi device is controlled to dynamically adjust the transmission power, signal frequency and modulation mode, and high-frequency interaction is maintained with the mobile communication base station to upload the local environment and signal state and download the latest optimization instruction, so that a closed-loop optimization link is formed to continuously adaptively optimize the signal enhancement effect.

[0086] Key state data of the first 5G personal WiFi device is received through the Internet of Things, and big data analysis is performed according to the data of a plurality of 5G personal WiFi devices, and personalized signal enhancement schemes are pushed to different users based on the data analysis result.

[0087] By using the technical scheme of the present application, the surrounding environment data is collected through the environment perception sensor, and the placement scheme of the device is optimized in combination with the deep learning algorithm, so that the coverage range of the signal can be effectively improved to ensure that the user can obtain stronger signal in different environments; the strategy of dynamically adjusting the transmission power, signal frequency and modulation mode can optimize the signal strength according to the real-time environmental changes, reduce signal attenuation and interference, and thus improve the network experience of the user; the interference signal is analyzed through quantum computing, and a null is formed by using the beamforming technology, so that the reception of the interference signal can be effectively suppressed, the signal strength in the target direction is enhanced, and the stable transmission of the key data is ensured; in combination with the Internet of Things technology, the device state data is uploaded to the edge server in real time for big data analysis, so that personalized signal enhancement schemes can be pushed to different users according to their use environment to realize continuous adaptive optimization; through the precise placement guidance and dynamic frequency band switching strategy, the user can enjoy smoother network connection in the use process, and the connection interruption and delay caused by poor signal can be reduced; the working state of the device can be adjusted in real time according to the static characteristics and dynamic changes of the environment to ensure that good signal quality can still be maintained in a complex environment. In summary, the scheme of the embodiment of the present application not only can improve the signal quality of the 5G personal WiFi, but also can provide more intelligent and personalized network experience for the user. BRIEF DESCRIPTION OF DRAWINGS

[0088] Fig. 1 is a flowchart of a 5G personal WiFi signal enhancement method provided by an embodiment of the present application;

[0089] Fig. 2is a schematic block diagram of a 5G personal WiFi signal enhancement system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0090] In order to enable a more complete understanding of the above-mentioned objects, features and advantages of the present application, the present application will be described in further detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict, if possible.

[0091] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other manners different from those described herein, and therefore the scope of protection of the present application is not limited by the specific embodiments disclosed below.

[0092] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include other steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0093] In this document, reference to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearance of the phrase in various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily mutually exclusive of other embodiments. It will be explicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0094] The 5G personal WiFi signal enhancement method and system provided by some embodiments of the present application will be described below with reference to the accompanying drawings. Figs. 1-2

[0095] As shown in Fig. 1 , an embodiment of the present application provides a 5G personal WiFi signal enhancement method, comprising:

[0096] Turning on the multi-type environment perception sensor built-in the first 5G personal WiFi device to collect first data;

[0097] ​It can be understood that the multi-type environment perception sensor includes infrared, ultrasonic, light sensor and the like, and micro-electro-mechanical system (MEMS) accelerometer, gyroscope and the like; the accelerometer and the gyroscope can perceive the inclination of the device placement plane, assist in judging whether the device is in a stable, beneficial to signal diffusion placement posture; the collected first data not only includes the material properties of the nearby objects, the distance from the obstacles, but also includes the environmental temperature, humidity, and even the air flow direction and speed, because the temperature and humidity will affect the signal transmission medium characteristics, and the airflow may cause signal perturbation. It can be understood that the device is built-in with a basic intelligent analysis module, which is responsible for collecting and analyzing its own running data such as device temperature, signal transmission power, real-time connection speed and the like.

[0098] acquire in real time a first surrounding environment image within a first preset range of a position where a first terminal connected to the first 5G portable WiFi device is located, and a second surrounding environment image within a second preset range of a position where the first 5G portable WiFi device is located;

[0099] It can be understood that in this step, first, ensure that the first terminal (such as a smartphone or tablet) is successfully connected to the first 5G portable WiFi device (including scanning of WiFi signals, sending of connection requests and connection confirmation); use the built-in positioning system of the first terminal (such as GPS, Wi-Fi positioning or Bluetooth positioning) to obtain its current geographical location in real time (the location data will be used to determine the first preset range); after determining the location of the first terminal, start the environment image acquisition module (the module can use a built-in camera or an external camera device) to capture the surrounding environment image of the location of the first terminal in real time; process and analyze the collected first surrounding environment image; use the image recognition algorithm to recognize the image The first 5G portable WiFi device also acquires an image of the environment within a second preset range of its own location (this step is similar to image acquisition for the first terminal, ensuring a comprehensive understanding of the environment surrounding the device). The environmental image data collected by the first terminal and the first 5G portable WiFi device are integrated to form a comprehensive environmental dataset. Using this data, the system provides real-time feedback to optimize signal transmission and connection quality. Based on the analysis results, the system dynamically adjusts the signal transmission parameters (such as frequency and power) of the first 5G portable WiFi device to adapt to current environmental conditions and ensure optimal network connection quality. This solution enables real-time acquisition of surrounding environmental images, enabling the system to better understand and adapt to environmental changes. This analysis and dynamic adjustment improves WiFi signal stability and coverage, reducing signal interference. Users can enjoy smoother network connections and reduce connection interruptions and delays caused by environmental factors. Customized network optimization solutions are provided based on different environmental conditions to meet user needs in different scenarios.

[0100] Using a deep learning-based object detection algorithm, identify signal obstruction data from the first surrounding environment image and the second surrounding environment image, and determine a placement plan for the first 5G portable WiFi device based on the signal obstruction data, the first data, and a preset device placement model;

[0101] It can be understood that the signal shielding object, such as a large metal cabinet, a thick wall and the like, the signal shielding object data includes but is not limited to three-dimensional data, position data, material data and the like of the signal shielding object. In the embodiment of the application, the collected environment data for training is transmitted to the edge server in advance, and the device placement model is trained by means of the edge computing and the deep fusion of artificial intelligence algorithm; the device placement model based on deep learning on the edge server can be a complex architecture fused with convolutional neural network (CNN) and long short-term memory network (LSTM), which comprehensively considers the static characteristics and dynamic changes of the environment, simulates the propagation path and attenuation of the signal in the current environment, and generates the most suitable device placement strategy (including: placement position, such as placement window, open desktop and the like; placement angle; number and angle of external antenna; and the like) through the training mechanism of reinforcement learning, so as to maximize the signal coverage range and intensity, and finally accurately push the placement guidance information to the user to guide the user to complete the placement.

[0102] Activate the network monitoring program (use high sensitivity and high resolution radio frequency receiving module) built in the first 5G personal WiFi device to perform full-band scanning (not only cover the authorized frequency band, but also monitor the potential unlicensed frequency band), continuously collect signal data such as signal strength, phase, polarization characteristics and noise spectrum distribution of each frequency band to construct an original data sample set;

[0103] It can be understood that the network monitoring program built-in in the first 5G portable WiFi device is started (to ensure that the device is in normal working condition and connected to the power supply to support continuous monitoring operation); the high-sensitivity and high-resolution radio frequency receiving module is initialized (which includes calibrating the receiving module to ensure that it can accurately capture signals of different frequency bands); the scanning parameters are configured to ensure coverage of all licensed frequency bands and potential unlicensed frequency bands (set the frequency range and time interval of scanning to facilitate subsequent data collection); start full-band scanning and continuously collect signal data of each frequency band. Specifically, it includes: signal strength (records the signal strength value of each frequency band), phase (measures and records the phase information of the signal), polarization characteristics (analyzes the polarization state of the signal and obtains related data), noise spectrum distribution (monitors and records the frequency spectrum distribution of background noise); the collected signal data is stored in the memory or external storage medium of the device in real time, to build an original data sample set and ensure the integrity and accuracy of the data for subsequent analysis; the original data sample set is preliminarily analyzed to identify the change trend of signal strength, the stability of phase, the influence of polarization characteristics, and the characteristics of noise spectrum, which will provide the basis for subsequent network optimization. In this step, all licensed frequency bands and potential unlicensed frequency bands can be covered to ensure a comprehensive understanding of the signal environment; high-sensitivity and high-resolution radio frequency receiving modules are used to ensure the accuracy and reliability of signal data; multi-dimensional data such as signal strength, phase, polarization characteristics and noise spectrum distribution are collected to provide a rich information base for subsequent network optimization and troubleshooting; through continuous data collection and analysis, network problems can be found and optimized in time to improve user network experience.

[0104] For the original data sample set, the adaptive filtering algorithm is used to dynamically adjust the filtering parameters, remove noise interference, use high-order statistical analysis method (such as high-order cumulant analysis), mine the nonlinear characteristics of signal data, and cooperate with time-frequency analysis tool (such as short-time Fourier transform, wavelet packet transform) to extract the time-varying characteristics and frequency domain distribution characteristics of the frequency band;

[0105] It can be understood that in this step, the original data sample set obtained from the previous step, including signal strength, phase, polarization characteristics and noise spectrum distribution, etc. is sorted and preprocessed for subsequent analysis; a suitable adaptive filtering algorithm (such as LMS algorithm or RLS algorithm) is selected to process the original data sample set; the filtering parameters are dynamically adjusted to remove noise interference and ensure the clarity and accuracy of the signal data; high-order statistical analysis method (such as high-order cumulant analysis) is used to analyze the denoised signal data; the nonlinear characteristics in the signal are extracted to identify the complexity and change pattern of the signal; time-frequency analysis tools (such as short-time Fourier transform or wavelet packet transform) are used to analyze the signal data in time-frequency domain, the specific steps include: short-time Fourier transform (segmenting the signal, calculating the Fourier transform of each segment to obtain the distribution characteristics of the signal in time and frequency), wavelet packet transform (multi-scale analysis of the signal to extract the time-varying characteristics and frequency domain distribution characteristics of the signal, and identify the change of the signal in different frequency bands); key features are extracted from the time-frequency analysis results, including time-varying characteristics of frequency bands, frequency domain distribution characteristics, etc. These features are integrated into a feature vector for subsequent model training and analysis; the extracted features are verified to ensure their accuracy and effectiveness. According to the verification results, the analysis parameters and methods are adjusted to optimize the extraction process of signal characteristics. In this step, the adaptive filtering algorithm can dynamically adjust the parameters to effectively remove the noise interference in the signal and improve the clarity of the signal; the high-order statistical analysis method can extract the nonlinear characteristics in the signal data and reveal the complexity and potential pattern of the signal; combined with the time-frequency analysis tools, the time-varying characteristics and frequency domain distribution characteristics of the signal can be extracted to provide rich information for subsequent signal processing and optimization; by extracting and integrating multi-dimensional features, the understanding ability of the system for signal characteristics is improved to provide data support for intelligent decision-making and optimization.

[0106] An integrated learning algorithm is used to fuse multiple weak classifiers, and historical frequency band characteristics are used as input to build a frequency band quality evaluation model.

[0107] It can be understood that in this step, an integrated learning algorithm is used to fuse multiple weak classifiers (such as gradient boosting tree, AdaBoost algorithm), and historical frequency band characteristics are used as input to build a frequency band quality evaluation model. This model not only distinguishes between high-quality, medium-quality and low-quality frequency bands, but also further subdivides temporary available, long-term stable available, and specific scene available frequency bands into more detailed categories. Combined with the generative adversarial network (GAN) in deep learning, simulated interference scene data is continuously generated to expand the training set and improve the evaluation ability of the frequency band quality evaluation model in extreme environments.

[0108] The frequency band quality evaluation model, the time-varying characteristics and the frequency domain distribution characteristics are combined to obtain the frequency band quality evaluation result.

[0109] According to the frequency band quality evaluation result, based on a deep reinforcement learning algorithm (such as a deep deterministic policy gradient (DDPG) algorithm), the frequency band switching action of the first 5G personal WiFi device is expanded to a continuous space;

[0110] It can be understood that in this step, the current environment state, the historical frequency band switching effect, and the estimated future interference change are comprehensively considered to continuously optimize the frequency band switching strategy; when signal fluctuation is monitored, frequency band switching exploration is initiated, and the optimal frequency band is quickly converged according to real-time feedback; the intelligent pre-synchronization technology is used in the switching process to ensure uninterrupted data transmission in the switching moment and maintain the ultimate smoothness of signal transmission.

[0111] With the parallel computing capability of quantum computing, a quantum neural network is constructed for interference signal analysis;

[0112] It can be understood that the superposition state characteristics of quantum bits allow simultaneous processing of a large number of interference signal samples, accelerating the neural network training process. Through quantum entanglement, interference signal data in different time and space are associated to accurately capture the hidden rules and mutation points of interference signals, quickly update the complex model of interference signals, not only predict short-term interference trends, but also predict medium and long-term interference evolution, providing a basis for subsequent coping strategies.

[0113] The first 5G personal WiFi device uses coding technology to dynamically adjust the coding rate according to the interference strength based on the interference analysis result;

[0114] Combined with orthogonal frequency division multiplexing technology, subcarriers are allocated, and subcarriers of frequency bands severely affected by interference are used to transmit redundant information to ensure error-free transmission of critical data;

[0115] Beamforming technology is introduced to form nulls in the direction of interference using multiple antenna arrays to suppress interference signal reception and enhance signal strength in the target direction;

[0116] It can be understood that in this embodiment, the adaptive modulation and demodulation module built-in the first 5G personal WiFi device uses intelligent coding technology (such as a hybrid coding scheme of low-density parity-check code (LDPC) and polar code) to dynamically adjust the coding rate according to the interference strength based on the interference analysis result; combined with orthogonal frequency division multiplexing (OFDM) technology, subcarriers are intelligently allocated, and subcarriers of frequency bands severely affected by interference (i.e. affected by interference intensity exceeding the preset intensity) are used to transmit redundant information to ensure error-free transmission of critical data; at the same time, intelligent beamforming technology is introduced to form nulls in the direction of interference using multiple antenna arrays to suppress interference signal reception and enhance signal strength in the target direction;

[0117] Continuously apply semantic segmentation algorithms in image recognition and processing technology to divide the latest environment images obtained in real time, identify the position, contour, and material of the newly appeared occlusion, and evaluate the degree of signal occlusion;

[0118] It can be understood that in this step, the camera or other sensors can be used to obtain the environment images in real time, which will be used as input data for subsequent processing; the obtained images are preprocessed, including denoising, contrast enhancement, and image size adjustment, to improve the effect of the subsequent semantic segmentation algorithm; a suitable semantic segmentation algorithm (such as FCN, U-Net, or DeepLab) is selected to segment the preprocessed images, which assigns a class label to each pixel in the image to identify different objects and regions; in the segmentation result, the newly appeared occlusion is identified, and the contour and position of the occlusion are extracted by analyzing the segmented image, which can be achieved by using a contour detection algorithm (such as Canny edge detection); the identified occlusion is classified by material using image feature extraction techniques (such as texture analysis or color histogram), which can help determine the nature of the occlusion (such as metal, wood, plastic, etc.); according to the position and contour of the occlusion, the degree of signal occlusion is evaluated, which can be achieved by establishing an occlusion model to analyze the relative position and size between the occlusion and the signal source; the identification results (including the position, contour, material of the occlusion, and the degree of signal occlusion) are output to the user interface or other system modules for subsequent processing or decision-making; by continuously obtaining new images and feedback results, the semantic segmentation model and occlusion identification algorithm are optimized, and machine learning techniques are used to improve the system's adaptability to new environments. In this step, the system can identify and process newly appeared occlusions in the environment in real time, enhancing the response capability to environmental changes; through the semantic segmentation algorithm, the position, contour, and material of the occlusion are accurately identified, improving the accuracy of identification; it can effectively evaluate the influence of the occlusion on the signal, providing a basis for subsequent signal processing and optimization; through continuous learning and optimization, the system can adapt to different environmental changes, improving the overall performance and reliability.

[0119] According to the evaluation results of the degree of signal occlusion of the newly appeared occlusion, combined with the real-time feedback of the temperature and humidity, electromagnetic environment change data of the Internet of Things sensor, the first 5G personal WiFi device dynamically adjusts the transmission power, signal frequency, modulation mode, and maintains high-frequency interaction with the mobile communication base station, uploads the local environment and signal state, downloads the latest optimization instructions, forms a closed-loop optimization link, and continuously optimizes the signal enhancement effect.

[0120] It can be understood that, according to the newly appearing occlusion identified in the previous step, the degree of occlusion of the signal is analyzed (which includes evaluating the size, position and material of the occlusion) to determine its impact on signal propagation; real-time feedback of environmental data, including temperature and humidity and electromagnetic environmental changes, is provided by Internet of Things sensors, which will provide important basis for subsequent signal adjustment; the occlusion degree evaluation result of the occlusion and the environmental data feedback of the Internet of Things sensor are analyzed by fusion, and the quality and stability of the current signal are determined by comprehensively considering these factors; according to the analysis result, the transmission power, signal frequency and modulation mode of the first 5G personal WiFi device are dynamically adjusted: transmission power (when the signal is severely occluded, increase the transmission power to enhance the signal penetration ability), signal frequency (select the appropriate frequency according to the environmental changes to reduce interference and improve transmission efficiency), modulation mode (select the appropriate modulation mode according to the signal quality to optimize the reliability of data transmission); high-frequency interaction with the mobile communication base station is maintained, and the local environment and signal state are uploaded in real time, which ensures that the base station can obtain the state information of the device in time and make corresponding network optimization; according to the uploaded environmental and signal state, the mobile communication base station will issue the latest optimization instructions. These instructions may include adjusting network configuration, optimizing resource allocation, etc.; the uploaded environmental data, signal state and downloaded optimization instructions form a closed-loop optimization link, and through this mechanism, the system can continuously monitor and adjust the signal parameters to adapt to the changing environment; through the above steps, the first 5G personal WiFi device can continuously adaptively optimize the signal enhancement effect, ensuring good signal quality and transmission efficiency under different environmental conditions. In this embodiment, dynamically adjusting the transmission power and signal frequency can effectively respond to environmental changes and ensure stable transmission of the signal under the influence of the occlusion; real-time feedback of environmental data and signal state analysis enables the system to quickly adapt to different environmental conditions and optimize signal quality; through high-frequency interaction with the mobile communication base station, efficient use of network resources can be realized, and overall communication performance can be improved; the establishment of a closed-loop optimization link enables the system to continuously learn and optimize, improving user experience and the intelligent level of the system.

[0121] The key state data of the first 5G personal WiFi device is uploaded to the edge server through the Internet of Things, and the edge server performs big data analysis based on the data of multiple 5G personal WiFi devices, and pushes personalized signal enhancement schemes to different users based on the data analysis results (for example, for users who often stay in densely built-up areas, more aggressive anti-interference settings are provided).

[0122] It can be understood that in this step, the first 5G personal WiFi device monitors and collects key state data in real time, including signal strength, connection quality, user location, environmental changes and other information; the collected key state data is encrypted to ensure the security and privacy protection of the data during transmission (common encryption algorithms can include AES (Advanced Encryption Standard) and the like); the encrypted data is uploaded to the edge server through the Internet of Things communication protocol (such as MQTT or HTTP), ensuring the stability and reliability of data transmission; the edge server receives and stores encrypted data from multiple 5G personal WiFi devices, establishes a data warehouse for subsequent analysis; using big data analysis technology, the stored data is processed and analyzed. The analysis includes user usage habits, signal quality changes, environmental impact factors, etc. to identify the needs and patterns of different users; based on the data analysis results, the edge server generates personalized signal enhancement schemes, for example, for users who are often in dense building areas, more aggressive anti-interference settings are provided, or signal parameters are optimized for specific environmental conditions; the generated personalized signal enhancement scheme is pushed to the devices of different users through the application or notification, ensuring that users can obtain optimization suggestions in a timely manner; collect user feedback on the pushed scheme to further optimize the signal enhancement strategy, and through continuous data collection and analysis, improve the accuracy and effectiveness of personalized services. In this step, the personalized signal enhancement scheme can significantly improve signal quality and network connection stability according to the actual needs of users and environmental conditions; by uploading encrypted data, user privacy and data security are ensured, and the risk of data leakage is reduced; based on the big data analysis of the personalized scheme push, the service is more intelligent and can adapt to different user scenarios and needs; through user feedback and data analysis, the system can continuously optimize the signal enhancement strategy to improve the overall service quality.

[0123] The embodiment of the present application can effectively improve the coverage range of the signal, ensure that the user can obtain stronger signal in different environments, by collecting surrounding environment data through the environmental perception sensor, combining the deep learning algorithm to optimize the placement scheme of the device; the strategy of dynamically adjusting the transmission power, signal frequency and modulation mode can optimize the signal strength according to the real-time environmental changes, reduce the signal attenuation and interference, thereby improving the network experience of the user; the interference signal is analyzed through quantum computing, and the nulling is formed by using the beam forming technology, which can effectively suppress the reception of the interference signal, enhance the signal strength of the target direction, and ensure the stable transmission of the key data; in combination with the Internet of Things technology, the device state data is uploaded to the edge server for big data analysis in real time, which can push personalized signal enhancement scheme according to the use environment of different users, realize continuous adaptive optimization; through the accurate placement guidance and the dynamic frequency band switching strategy, the user can enjoy smoother network connection in the use process, reduce the connection interruption and delay caused by poor signal; the working state of the device can be adjusted in real time according to the static characteristics and dynamic changes of the environment, and the good signal quality can be ensured in the complex environment. In summary, the scheme of the embodiment of the present application can not only improve the signal quality of the 5G personal WiFi, but also provide more intelligent and personalized network experience for the user.

[0124] In some possible embodiments of the present application, the step of identifying signal shielding object data from the first surrounding environment image and the second surrounding environment image by using a target detection algorithm based on deep learning, and determining the placement scheme of the first 5G personal WiFi device according to the signal shielding object data, the first data and a preset device placement model, comprises:

[0125] The first surrounding environment image and the second surrounding environment image are subjected to standardization processing (including: image light compensation and color balance; uniformly adjusting the image resolution to a preset size; removing image noise and distortion), to obtain a third surrounding environment image and a fourth surrounding environment image;

[0126] The third surrounding environment image and the fourth surrounding environment image are subjected to target detection and instance segmentation by using a target detection algorithm based on deep learning (such as an improved Mask R-CNN network model), specifically including: extracting image features to generate candidate regions; classifying the candidate regions to identify the type of shielding object; generating a shielding object pixel-level mask; calculating the accurate contour and spatial position information of the shielding object;

[0127] Based on the binocular vision principle, the third surrounding environment image and the fourth surrounding environment image are three-dimensionally reconstructed, specifically including: extracting image feature points and matching; calculating feature point parallax and constructing a depth map; generating a three-dimensional point cloud model of the occlusion; based on a material identification model, analyzing the surface characteristics of the occlusion to determine the material properties;

[0128] Combined with the first environment parameters in the first data, a signal propagation model is constructed, including: calculating the signal attenuation coefficient according to the three-dimensional position and material of the occlusion; considering the influence of temperature and humidity on signal propagation; simulating multi-path effect and signal reflection, diffraction phenomenon;

[0129] Based on a preset device placement model, multi-objective optimization calculation is performed, including: inputting occlusion data, first environment parameters and signal propagation model; considering device heat dissipation requirements and operation convenience; using a reinforcement learning algorithm to iteratively optimize the placement position and angle; outputting an optimal placement scheme, including specific coordinates and placement posture.

[0130] The scheme of the embodiment realizes sub-pixel level occlusion contour recognition through deep learning, accurately evaluates the influence degree of different materials on signals, and realizes centimeter-level device placement position optimization; can adjust the placement strategy in real time according to environmental changes, can adapt to the layout characteristics of different indoor scenes, has dynamic compensation ability for environmental factors such as temperature and humidity; reduces the operation difficulty of the user, provides intuitive placement guidance; optimizes the device heat dissipation condition, prolongs the service life; balances the signal strength and use convenience; supports online learning of new occlusion types, can update and optimize the algorithm and model through the edge server, and has multi-device collaborative optimization ability.

[0131] In some possible embodiments of the present application, the step of applying an ensemble learning algorithm, fusing multiple weak classifiers, and taking historical frequency band features as input to construct a frequency band quality evaluation model includes:

[0132] Based on historical frequency band features, a multi-dimensional feature vector is constructed, including: extracting statistical features of signal strength, including mean, variance, skewness, kurtosis; calculating signal stability indicators, including signal jitter rate, frequency of disconnection; generating spectral features such as spectral occupancy rate and bandwidth utilization rate; normalizing and detecting outliers of feature data;

[0133] A plurality of basic classifiers for different feature subspaces are constructed, including: deploying gradient boosting decision trees (GBDT) for processing continuous features; configuring an AdaBoost classifier to process categorical features; applying a random forest to process the correlation of high-dimensional features; assigning an initial weight to each weak classifier;

[0134] An adaptive integration strategy is designed to dynamically adjust the weights of the classifiers, including: calculating the prediction accuracy of each weak classifier; dynamically updating the weights of the classifiers based on the accuracy; using a weighted voting mechanism to fuse the prediction results of each classifier; introducing a time decay factor to reduce the weight of historical data;

[0135] A generative adversarial network is deployed to expand the training data, including: constructing a generator network to simulate various interference scenarios; designing a discriminator network to distinguish between real and generated data; iteratively training the GAN network to generate high-quality samples; adding the generated samples to the training set to improve the robustness of the model;

[0136] A fine-grained classification of frequency band quality is implemented, including: establishing a multi-level evaluation standard, including main classification and sub-classification; dynamically updating the threshold parameters of each level; combining time series analysis to predict the trend of frequency band quality; continuously optimizing the evaluation model parameters.

[0137] The frequency band quality evaluation model constructed in the scheme of the embodiment can accurately distinguish frequency bands of different quality and subdivide them into multiple categories to meet the needs of different application scenarios; the robustness and accuracy of the model are improved through the integration of multiple weak classifiers, which can adapt to complex signal environments; the training set is expanded by using GAN-generated interference scenario data, which significantly enhances the evaluation capability of the model in extreme environments and improves the generalization ability of the model; the deployed model can evaluate the frequency band quality in real time and provide timely frequency band usage recommendations for users, optimizing the configuration of network resources.

[0138] In some possible embodiments of the present application, the step of obtaining a frequency band quality evaluation result by combining the frequency band quality evaluation model, the time-varying feature, and the frequency domain distribution characteristic includes:

[0139] The input multi-source features are standardized, specifically including: normalizing the evaluation indicators output by the frequency band quality evaluation model; performing time window segmentation processing on the time-varying feature; converting the frequency domain distribution characteristic into a unified feature vector; constructing a feature fusion matrix;

[0140] The time-varying feature is dynamically evaluated, specifically including: calculating the time series fluctuation index of signal strength; extracting the periodic variation mode of signal quality; analyzing the mutation feature points of signal parameters; generating a time series evaluation score;

[0141] The frequency domain distribution characteristic is analyzed in depth, including: calculating the spectral energy distribution density; identifying the frequency domain interference characteristic; evaluating the frequency band utilization efficiency; generating a frequency domain evaluation score;

[0142] A weighted fusion strategy is used to integrate the evaluation results of each dimension, including: designing an adaptive weight calculation method; considering the reliability indicators of different features; fusing the evaluation scores of each dimension; generating a comprehensive evaluation indicator;

[0143] The final frequency band quality evaluation result is generated based on the comprehensive evaluation index, including: establishing a multi-threshold evaluation standard; calculating the quality level confidence; generating an evaluation result report; and providing optimization suggestion instructions.

[0144] The scheme of the embodiment combines time-varying characteristics and frequency domain distribution characteristics, and the frequency band quality evaluation model can provide more accurate and detailed evaluation results to help users better understand the use of the frequency band; the evaluation results provide a scientific basis for the use and management of the frequency band, which can effectively support the optimization configuration and dynamic adjustment of network resources; through regular updating and iteration, the model can adapt to the changing environment and needs, and maintain efficient evaluation ability; accurate frequency band quality evaluation results can improve the user experience when using the frequency band, and ensure the stability of communication quality and service.

[0145] In some possible embodiments of the application, the step of expanding the frequency band switching action of the first 5G personal WiFi device to a continuous space based on the frequency band quality evaluation result according to a deep reinforcement learning algorithm includes:

[0146] The state representation of deep reinforcement learning is constructed, including: encoding the current frequency band quality evaluation result; fusing the historical frequency band switching record; collecting network environment parameters; and constructing a multi-dimensional state vector;

[0147] The discrete frequency band switching is expanded to a continuous action space, including: defining the continuous value range of frequency and bandwidth; designing an action space mapping function; establishing action constraint conditions; and realizing an action smooth transition mechanism;

[0148] The double network architecture is deployed to realize policy learning, including: constructing an Actor network to output continuous action values; deploying a Critic network to evaluate action values; designing an experience replay buffer; and realizing a target network soft update mechanism;

[0149] The multi-objective reward function is established, including: evaluating the frequency band switching effect; considering the switching time cost; calculating the energy consumption penalty term; and fusing the user experience index;

[0150] Online learning and policy updating are realized, including: collecting interactive experience data; calculating the time series difference error; updating the Actor-Critic network parameters; and optimizing the exploration-exploitation balance.

[0151] The scheme of the embodiment expands the frequency band switching action to the continuous space, enables the device to switch between frequency bands more flexibly and smoothly, and improves the user experience; based on the deep reinforcement learning algorithm, the device can make intelligent frequency band switching decisions according to real-time frequency band quality evaluation results, and optimize the use of network resources; through continuous learning and model updating, the device can adapt to the changing network environment and user demand, and maintain efficient frequency band switching capability; the optimized frequency band switching strategy will help improve the overall network performance and ensure that users can obtain stable connection quality in different environments.

[0152] In some possible embodiments of the application, the step of constructing a quantum neural network for interference signal analysis by means of the parallel computing capability of quantum computing includes:

[0153] Encoding interference signal data into quantum states includes: using an amplitude encoding method to map signal features to quantum bits; constructing a quantum state superposition to represent multi-dimensional features; realizing quantum entanglement of the feature space; establishing a quantum data preprocessing pipeline;

[0154] Designing a special quantum neural network architecture includes: deploying a parameterized quantum circuit layer; constructing a quantum convolution layer to process time-frequency features; designing a quantum pooling operation to compress features; implementing a quantum-classical hybrid computing interface;

[0155] Performing feature analysis using quantum parallelism includes: simultaneously processing interference features of multiple frequency bands; parallel computing time-frequency correlation; extracting hidden features through quantum state evolution; constructing a quantum feature map;

[0156] Implementing quantum classification of interference patterns includes: designing a quantum measurement operator; constructing a quantum decision tree; implementing quantum ensemble learning; outputting the probability distribution of interference types;

[0157] Optimizing the prediction model based on quantum feedback includes: designing a quantum gradient descent algorithm; implementing quantum acceleration of parameter updates; optimizing quantum-classical data conversion; dynamically adjusting the model structure.

[0158] The scheme of the embodiment, with the parallel computing capability of quantum computing, the quantum neural network can significantly improve the computing speed and efficiency when processing complex interference signals; the quantum neural network can extract richer signal features by utilizing the characteristics of quantum states, thereby improving the accuracy of interference signal analysis; through the flexibility of quantum computing, the model can quickly adapt to different types of interference signals, enhancing its application ability in dynamic environments.

[0159] In some possible embodiments of the application, the first 5G personal WiFi device dynamically adjusts the coding rate according to the interference strength by using an encoding technique according to the interference analysis result, and the step includes:

[0160] Based on the interference analysis results, multi-dimensional evaluation is carried out, including: calculating the dynamic change range of signal-to-noise ratio; analyzing the time-frequency characteristics of the interference signal; evaluating the influence of the interference on the channel capacity; generating an interference intensity quantification index;

[0161] According to the interference intensity, the optimal encoding scheme is selected, including: constructing a hybrid encoding library of LDPC code and polar code; designing an encoding scheme evaluation criterion; calculating the performance index of different encoding schemes; selecting the encoding scheme most suitable for the current interference environment;

[0162] The encoding parameter configuration is dynamically adjusted, including: calculating the target encoding rate according to the interference intensity; optimizing the codeword length and check bit distribution; adjusting the interleaving depth and pattern structure; generating the optimized encoding parameters;

[0163] Real-time switching of the encoding scheme is realized, including: designing a coding switching buffer mechanism; realizing smooth transition of the coding parameters; monitoring the coding performance index; and executing dynamic updating of the coding scheme;

[0164] A coding effect feedback mechanism is established, including: collecting decoding error statistical data; analyzing the throughput change trend; evaluating the coding overhead and gain ratio; and optimizing the coding strategy decision model.

[0165] The scheme of the embodiment can effectively cope with different interference environments by dynamically adjusting the encoding rate, ensure reliable transmission of data in high interference conditions, increase the encoding rate in low interference environments, improve the data transmission rate, and optimize the use of network resources. The device can automatically adjust the coding strategy according to the real-time interference conditions, thereby enhancing the adaptability in complex network environments. Through the optimized data transmission performance, users can obtain more stable and fast network experience when using the first 5G portable WiFi device.

[0166] In some possible embodiments of the present application, the step of combining the orthogonal frequency division multiplexing technology, allocating subcarriers, and using subcarriers in severely interfered frequency bands for transmitting redundant information to ensure error-free transmission of critical data, includes:

[0167] The subcarriers in the OFDM system are dynamically evaluated, including: measuring the signal-to-noise ratio of each subcarrier; evaluating the interference degree between subcarriers; calculating the subcarrier channel capacity; and generating a subcarrier quality score;

[0168] The data to be transmitted is classified and processed, including: identifying critical data packets; calculating the data transmission priority; estimating the required transmission reliability; and generating a data transmission strategy;

[0169] Intelligent allocation is performed based on the state of the subcarriers, including: allocating high-quality subcarriers to critical data; using interfered subcarriers for redundant transmission; optimizing the subcarrier combination scheme; and realizing dynamic bandwidth allocation;

[0170] The adaptive redundancy transmission mechanism is designed, including: calculating the redundancy data ratio; selecting a redundancy coding scheme; determining the redundancy data distribution; and implementing redundancy transmission scheduling;

[0171] The real-time monitoring and adjustment mechanism is established, including: tracking the data transmission success rate; analyzing the transmission delay change; evaluating the resource utilization efficiency; and dynamically optimizing the allocation strategy.

[0172] The scheme of the embodiment ensures that the key data is not easily lost or damaged during transmission by transmitting redundancy information on the interfered subcarriers; reasonably allocates subcarriers to effectively utilize the frequency spectrum resources while ensuring the transmission quality of the key data; even in a serious interference environment, the system can recover the key data through the redundancy information, improving the stability of the overall communication; users can enjoy higher transmission reliability and less data loss during use, thereby improving the overall user experience.

[0173] In some possible embodiments of the application, the step of introducing the beamforming technology, using a multi-antenna array, forms a null in the interference direction, suppresses the reception of interference signals, and enhances the signal strength in the target direction, includes:

[0174] Channel characteristic analysis is performed using a multi-antenna array, including: obtaining the received signal characteristics of each antenna; calculating the signal correlation between antennas; estimating the signal angle of arrival; and constructing a spatial channel matrix;

[0175] Accurate identification of the spatial position of the interference source includes: performing spatial spectrum scanning; calculating the signal angle of arrival distribution; identifying the main interference source direction; and generating an interference source spatial distribution map;

[0176] Designing optimal beamforming weights includes: constructing an adaptive beamforming algorithm; calculating the constraint conditions of the null direction; optimizing the main lobe direction gain; and generating a weight vector;

[0177] Implementing accurate beam control includes: configuring the phase relationship of the antenna array; adjusting the transmission power of each antenna; synchronizing the timing of the antenna array; and performing beamforming operations;

[0178] Establishing a real-time optimization feedback mechanism includes: monitoring the beamforming effect; analyzing the signal-to-interference ratio change; evaluating system performance indicators; and adjusting the beamforming strategy.

[0179] The scheme of the embodiment significantly reduces the impact of interference signals on the receiving system by forming a null in the interference direction, thereby improving the quality of the signals; enhances the signal strength in the target signal direction to ensure that the target signal can still be clearly received in a complex environment; the system can maintain stable performance in various interference environments, enhancing the reliability of communication; through the beamforming technology, the frequency spectrum resources are reasonably utilized, improving the overall efficiency of the system.

[0180] Referring to Fig. 2 Another embodiment of the present application provides a 5G portable WiFi signal enhancement system for performing a 5G portable WiFi signal enhancement method, comprising a first 5G portable WiFi device, a first terminal and an edge server.

[0181] The first 5G portable WiFi device is configured to start the first data collection of the multi-type environment perception sensor built-in the first 5G portable WiFi device.

[0182] The edge server is configured to:

[0183] Real-time acquisition of the first surrounding environment image within the first preset range of the position where the first terminal connected to the first 5G portable WiFi device is located, and the second surrounding environment image within the second preset range of the position where the first 5G portable WiFi device is located;

[0184] Using a target detection algorithm based on deep learning, signal blocking object data is identified from the first surrounding environment image and the second surrounding environment image, and a placement scheme of the first 5G portable WiFi device is determined according to the signal blocking object data, the first data and a preset device placement model;

[0185] Activating the network monitoring program built-in the first 5G portable WiFi device to perform full-band scanning, continuously collecting signal data such as signal strength, phase, polarization characteristics and noise spectrum distribution of each frequency band to construct an original data sample set;

[0186] For the original data sample set, an adaptive filtering algorithm is used to dynamically adjust the filtering parameters, remove noise interference, use a high-order statistical analysis method to mine the nonlinear characteristics of the signal data, and cooperate with a time-frequency analysis tool to extract the time-varying characteristics and frequency domain distribution characteristics of the frequency band;

[0187] Using an ensemble learning algorithm, multiple weak classifiers are fused, and historical frequency band characteristics are used as input to construct a frequency band quality evaluation model;

[0188] Combining the frequency band quality evaluation model, the time-varying characteristics and the frequency domain distribution characteristics, a frequency band quality evaluation result is obtained;

[0189] According to the frequency band quality evaluation result, based on a deep reinforcement learning algorithm, the frequency band switching action of the first 5G portable WiFi device is expanded to a continuous space;

[0190] With the parallel computing capability of quantum computing, a quantum neural network is constructed for interference signal analysis;

[0191] The first 5G personal WiFi device is further configured to:

[0192] According to the interference analysis result, an encoding technology is adopted to dynamically adjust the encoding rate according to the interference intensity.

[0193] In combination with the orthogonal frequency division multiplexing technology, subcarriers are allocated, and the subcarriers of the frequency band seriously interfered are used to transmit redundant information to ensure the error-free transmission of critical data.

[0194] The beamforming technology is introduced, and a null is formed in the interference direction by using a multi-antenna array to suppress the reception of interference signals and enhance the signal strength in the target direction.

[0195] The edge server is further configured to:

[0196] The semantic segmentation algorithm in the image recognition and processing technology is continuously used to divide the latest environment image acquired in real time, identify the position, contour and material of the newly appeared shielding object, and evaluate the signal shielding degree of the shielding object.

[0197] According to the evaluation result of the signal shielding degree of the newly appeared shielding object, in combination with the temperature, humidity and electromagnetic environment change data fed back by the Internet of Things sensor in real time, the first 5G personal WiFi device is controlled to dynamically adjust the transmission power, signal frequency and modulation mode, and high-frequency interaction is maintained with the mobile communication base station to upload the local environment and signal state and download the latest optimization instruction, so as to form a closed-loop optimization link and continuously adaptively optimize the signal enhancement effect.

[0198] The key state data of the first 5G personal WiFi device is received through the Internet of Things, and big data analysis is performed according to the data of multiple 5G personal WiFi devices, and personalized signal enhancement schemes are pushed to different users based on the data analysis result.

[0199] It should be understood that, Fig. 2 The block diagram of the 5G personal WiFi signal enhancement system shown is only schematic, and the number of each module shown does not limit the protection scope of the present application. The 5G personal WiFi signal enhancement system provided in the embodiment can be used to execute each embodiment scheme of the corresponding 5G personal WiFi signal enhancement method. For the specific implementation process, please refer to the description of each method embodiment, which is not described here.

[0200] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action order described, because according to the present application, certain steps can be adopted in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0201] In the above-described embodiments, the description of each embodiment focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0202] In several embodiments provided in the present application, it should be understood that the disclosed apparatus can be implemented by other means. For example, the apparatus embodiments described above are only illustrative, for example, the division of the above-mentioned units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical or other forms.

[0203] The units described above as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0204] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0205] The integrated unit described above, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable memory. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that makes a contribution or the whole or part of the technical solutions can be embodied in the form of a software product, which is stored in a memory and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the above-mentioned method of each embodiment of the present application. The aforementioned memory includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0206] Those skilled in the art can understand that all or part of the steps of various methods in the above embodiments can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer readable memory, which can include a flash disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0207] The above has carried out the detailed introduction to the embodiments of the application, the principle and implementation mode of the application are described in this paper by applying specific examples, the above embodiment explanation is only for helping to understand the method of the application and its core idea; at the same time, for the general technical personnel in the art, according to the idea of the application, the specific implementation mode and application range will have the change, according to the above, the content of the specification should not be understood as the limitation of the application.

[0208] Although the present application is disclosed as above, the present application is not limited to this. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present application, and can make various changes and modifications, including the combination of different functions and implementation steps, including the software and hardware implementation, which are all within the protection scope of the present application.

Claims

1. A 5G portable WiFi signal enhancement method, characterized in that: include: Activate multiple types of environment perception sensors built into the first 5G portable WiFi device to collect first data; Acquire in real time a first surrounding environment image within a first preset range of a location of a first terminal connected to the first 5G portable WiFi device, and a second surrounding environment image within a second preset range of a location of the first 5G portable WiFi device; Using a deep learning-based object detection algorithm, identify signal obstruction data from the first surrounding environment image and the second surrounding environment image, and determine a placement plan for the first 5G portable WiFi device based on the signal obstruction data, the first data, and a preset device placement model; Activate the network monitoring program built into the first 5G portable WiFi device to perform full-band scanning, continuously collect signal data such as signal strength, phase, polarization characteristics, and noise spectrum distribution of each frequency band to construct an original data sample set; For the original data sample set, an adaptive filtering algorithm is used to dynamically adjust the filtering parameters to remove noise interference, a high-order statistical analysis method is used to mine the nonlinear characteristics in the signal data, and a time-frequency analysis tool is used to extract the time-varying characteristics and frequency domain distribution characteristics of the frequency band; Using an ensemble learning algorithm to fuse multiple weak classifiers and taking historical frequency band features as input, a frequency band quality assessment model is constructed. Combining the frequency band quality assessment model, the time-varying characteristics, and the frequency domain distribution characteristics to obtain a frequency band quality assessment result; Based on the frequency band quality assessment result, the frequency band switching action of the first 5G portable WiFi device is expanded to a continuous space based on a deep reinforcement learning algorithm; Leveraging the parallel computing capabilities of quantum computing, we construct a quantum neural network for interference signal analysis. The first 5G portable WiFi device uses coding technology based on interference analysis results to dynamically adjust the coding rate according to the interference intensity; Combined with orthogonal frequency division multiplexing technology, subcarriers are allocated and subcarriers in frequency bands with severe interference are used to transmit redundant information, ensuring the correct transmission of critical data; Introducing beamforming technology, using a multi-antenna array to form a null in the interference direction, suppressing the reception of interfering signals and enhancing the signal strength in the target direction; Continuously apply semantic segmentation algorithms in image recognition and processing technology to segment the latest environmental images acquired in real time, identify the location, outline, and material of newly appearing obstructions, and assess their degree of signal obstruction; Based on the assessment of the degree of signal obstruction caused by newly appeared obstructions and combined with real-time feedback from IoT sensors on temperature, humidity, and electromagnetic environment changes, the first 5G portable WiFi device dynamically adjusts its transmission power, signal frequency, and modulation method. It also maintains high-frequency interaction with mobile communication base stations, uploading local environment and signal status and downloading the latest optimization instructions, forming a closed-loop optimization link and continuously adaptively optimizing the signal enhancement effect. The key status data of the first 5G portable WiFi device is encrypted and uploaded to the edge server through the Internet of Things. The edge server performs big data analysis based on the data of multiple 5G portable WiFi devices, and pushes personalized signal enhancement solutions to different users based on the data analysis results.

2. The 5G portable WiFi signal enhancement method according to claim 1, characterized in that: The step of using a deep learning-based target detection algorithm to identify signal obstruction data from the first surrounding environment image and the second surrounding environment image, and determining a placement plan for the first 5G portable WiFi device based on the signal obstruction data, the first data, and a preset device placement model includes: Normalizing the first surrounding environment image and the second surrounding environment image to obtain a third surrounding environment image and a fourth surrounding environment image; Using a deep learning-based object detection algorithm to perform object detection and instance segmentation on the third surrounding environment image and the fourth surrounding environment image, the method specifically includes: extracting image features to generate candidate regions; classifying the candidate regions to identify the type of occluders; generating pixel-level masks of the occluders; and calculating the precise outline and spatial position information of the occluders. Based on the principle of binocular vision, the third surrounding environment image and the fourth surrounding environment image are subjected to three-dimensional reconstruction, specifically including: extracting and matching image feature points; calculating feature point parallax and constructing a depth map; generating a three-dimensional point cloud model of the obstruction; and analyzing the surface features of the obstruction and determining the material properties based on a material recognition model; Constructing a signal propagation model based on the first environmental parameter in the first data, including: calculating a signal attenuation coefficient based on the three-dimensional position and material of the obstruction; considering the effects of temperature and humidity on signal propagation; and simulating multipath effects and signal reflection and diffraction phenomena; Based on the preset device placement model, multi-objective optimization calculations are performed, including: inputting obstruction data, first environmental parameters, and signal propagation models; considering the device's heat dissipation requirements and operational convenience; using reinforcement learning algorithms to iteratively optimize placement position and angle; and outputting the optimal placement solution, including specific coordinates and placement posture.

3. The 5G portable WiFi signal enhancement method according to claim 2, characterized in that: The steps of using an ensemble learning algorithm to fuse multiple weak classifiers and using historical frequency band features as input to construct a frequency band quality assessment model include: Constructing a multidimensional feature vector based on historical frequency band characteristics, including: extracting statistical features of signal strength, including mean, variance, skewness, and kurtosis; calculating signal stability indicators, including signal jitter rate and disconnection frequency; generating spectrum features such as spectrum occupancy and bandwidth utilization; and performing normalization and outlier detection on feature data. Construct multiple basic classifiers for different feature subspaces, including: deploying gradient boosting decision trees to process continuous features; configuring AdaBoost classifiers to process categorical features; applying random forests to handle the correlation of high-dimensional features; and assigning initial weights to each weak classifier. Design an adaptive ensemble strategy to dynamically adjust classifier weights, including: calculating the prediction accuracy of each weak classifier; dynamically updating classifier weights based on the accuracy; using a weighted voting mechanism to fuse the prediction results of each classifier; and introducing a time decay factor to reduce the weight of historical data. Deploy a generative adversarial network to expand training data, including: building a generator network to simulate various interference scenarios; designing a discriminator network to distinguish between real and generated data; iteratively training the GAN network to generate high-quality samples; and adding generated samples to the training set to improve model robustness. Realize refined grading of frequency band quality, including: establishing multi-level evaluation standards, including main categories and subcategories; dynamically updating threshold parameters at each level; combining time series analysis to predict frequency band quality change trends; and continuously optimizing evaluation model parameters.

4. The 5G portable WiFi signal enhancement method according to claim 3, characterized in that: The step of obtaining a frequency band quality assessment result by combining the frequency band quality assessment model, the time-varying feature, and the frequency domain distribution characteristic includes: The input multi-source features are standardized, specifically including: normalizing the evaluation indicators output by the frequency band quality assessment model; performing time window segmentation processing on the time-varying features; converting the frequency domain distribution characteristics into a unified feature vector; and constructing a feature fusion matrix; Dynamic evaluation based on time-varying characteristics, specifically including: calculating the time series fluctuation index of signal strength; extracting the periodic change pattern of signal quality; analyzing the mutation feature points of signal parameters; generating time series evaluation scores; Conduct in-depth analysis of frequency domain distribution characteristics, including: calculating spectrum energy distribution density; identifying frequency domain interference characteristics; evaluating frequency band utilization efficiency; and generating frequency domain evaluation scores. A weighted fusion strategy is used to integrate the evaluation results of each dimension, including: designing an adaptive weight calculation method; considering reliability indicators with different characteristics; fusing the evaluation scores of each dimension; and generating a comprehensive evaluation index. Generate the final frequency band quality assessment results based on comprehensive assessment indicators, including: establishing multi-threshold assessment standards; calculating quality level confidence; generating an assessment result report; and providing optimization suggestion instructions.

5. The 5G portable WiFi signal enhancement method according to claim 4, characterized in that: The step of extending the frequency band switching action of the first 5G portable WiFi device to a continuous space based on the deep reinforcement learning algorithm according to the frequency band quality assessment result includes: Constructing a state representation for deep reinforcement learning, including: encoding the current frequency band quality assessment results; integrating historical frequency band switching records; collecting network environment parameters; and constructing a multi-dimensional state vector. Expanding discrete frequency band switching into a continuous action space includes: defining the continuous value range of frequency and bandwidth; designing the action space mapping function; establishing action constraints; and implementing a smooth action transition mechanism. Deploy a dual-network architecture to implement policy learning, including: building an Actor network to output continuous action values; deploying a Critic network to evaluate action values; designing an experience replay buffer pool; and implementing a soft update mechanism for the target network. Establish a multi-objective reward function, including: evaluating the effect of frequency band switching; considering the switching time cost; calculating the energy consumption penalty; integrating user experience indicators; Implement online learning and strategy updates, including: collecting interactive experience data; calculating temporal difference error; updating actor-critic network parameters; and optimizing the exploration-exploitation balance.

6. The 5G portable WiFi signal enhancement method according to claim 5, characterized in that: The step of constructing a quantum neural network for interference signal analysis by leveraging the parallel computing capability of quantum computing includes: Encoding interference signal data into quantum states, including: using amplitude encoding to map signal features to quantum bits; constructing quantum state superposition to represent multidimensional features; achieving quantum entanglement in feature space; and establishing a quantum data preprocessing pipeline. Design a dedicated quantum neural network architecture, including: deploying a parameterized quantum circuit layer; building a quantum convolution layer to process time-frequency features; designing quantum pooling operation compression features; and implementing a quantum-classical hybrid computing interface. Utilize quantum parallelism for feature analysis, including: simultaneous processing of interference features in multiple frequency bands; parallel calculation of time-frequency correlations; extraction of hidden features from quantum state evolution; and construction of quantum feature maps. Implement quantum classification of interference patterns, including: designing quantum measurement operators; constructing quantum decision trees; implementing quantum ensemble learning; and outputting probability distribution of interference types. Optimize the prediction model based on quantum feedback, including: designing a quantum gradient descent algorithm; realizing quantum acceleration of parameter updates; optimizing quantum-classical data conversion; and dynamically adjusting the model structure.

7. The 5G portable WiFi signal enhancement method according to claim 6, characterized in that: The first 5G portable WiFi device uses coding technology based on the interference analysis result to dynamically adjust the coding rate according to the interference intensity, including the following steps: Perform multi-dimensional evaluation based on the interference analysis results, including: calculating the dynamic range of signal-to-noise ratio; analyzing the time-frequency characteristics of the interference signal; evaluating the impact of interference on channel capacity; and generating quantitative indicators of interference intensity. Select the optimal coding scheme based on the interference intensity, including: building a hybrid coding library of LDPC codes and polar codes; designing coding scheme evaluation criteria; calculating the performance indicators of different coding schemes; and selecting the coding scheme that best suits the current interference environment; Dynamically adjust coding parameter configuration, including: calculating the target coding rate based on interference intensity; optimizing codeword length and parity bit distribution; adjusting interleaving depth and pattern structure; and generating optimized coding parameters; Implement real-time switching of coding schemes, including: designing a coding switching buffer mechanism; achieving smooth transition of coding parameters; monitoring coding performance indicators; and executing dynamic updates of coding schemes. Establish a coding effect feedback mechanism, including: collecting decoding error statistics; analyzing throughput change trends; evaluating coding overhead and gain ratios; and optimizing coding strategy decision models.

8. The 5G portable WiFi signal enhancement method according to claim 7, characterized in that: The steps of combining orthogonal frequency division multiplexing technology to allocate subcarriers and use subcarriers in frequency bands with severe interference to transmit redundant information to ensure accurate transmission of key data include: Dynamically evaluate subcarriers in OFDM systems, including: measuring the signal-to-noise ratio of each subcarrier; evaluating the degree of interference between subcarriers; calculating subcarrier channel capacity; and generating subcarrier quality scores. Classify and process the data to be transmitted, including: identifying key data packets; calculating data transmission priority; estimating required transmission reliability; and generating data transmission strategies; Intelligent allocation based on subcarrier status, including: allocating high-quality subcarriers to critical data; using interfered subcarriers for redundant transmission; optimizing subcarrier combination schemes; and achieving dynamic bandwidth allocation; Design an adaptive redundant transmission mechanism, including: calculating the redundant data ratio; selecting a redundant coding scheme; determining the redundant data distribution; and implementing redundant transmission scheduling. Establish a real-time monitoring and adjustment mechanism, including: tracking data transmission success rate; analyzing transmission delay changes; evaluating resource utilization efficiency; and dynamically optimizing allocation strategies.

9. The 5G portable WiFi signal enhancement method according to claim 8, characterized in that: The steps of introducing beamforming technology and utilizing a multi-antenna array to form a null in the interference direction, suppressing reception of interference signals, and enhancing signal strength in the target direction include: Channel characteristic analysis using a multi-antenna array, including: obtaining the received signal characteristics of each antenna; calculating the signal correlation between antennas; estimating the signal arrival angle; and constructing a spatial channel matrix. Accurately identify the spatial location of interference sources, including: performing spatial spectrum scanning; calculating the signal arrival angle distribution; identifying the direction of the main interference source; and generating a spatial distribution map of the interference source; Designing optimal beamforming weights, including: constructing an adaptive beamforming algorithm; calculating constraints on the null direction; optimizing the mainlobe direction gain; and generating a weight vector. Achieve precise beam steering, including: configuring antenna array phase relationships; adjusting transmit power of each antenna; synchronizing antenna array timing; and performing beamforming operations. Establish a real-time optimization feedback mechanism, including: monitoring beamforming effects; analyzing signal-to-interference ratio changes; evaluating system performance indicators; and adjusting beamforming strategies.

10. A 5G portable WiFi signal enhancement system, used to execute the 5G portable WiFi signal enhancement method according to any one of claims 1 to 9, characterized in that: include: The first 5G portable WiFi device, the first terminal and edge server; The first 5G portable WiFi device is configured to: enable multiple types of environment perception sensors built into the first 5G portable WiFi device to collect first data; The edge server is configured to: Acquire in real time a first surrounding environment image within a first preset range of a location of a first terminal connected to the first 5G portable WiFi device, and a second surrounding environment image within a second preset range of a location of the first 5G portable WiFi device; Using a deep learning-based object detection algorithm, identify signal obstruction data from the first surrounding environment image and the second surrounding environment image, and determine a placement plan for the first 5G portable WiFi device based on the signal obstruction data, the first data, and a preset device placement model; Activate the network monitoring program built into the first 5G portable WiFi device to perform full-band scanning, continuously collect signal data such as signal strength, phase, polarization characteristics, and noise spectrum distribution of each frequency band to construct an original data sample set; For the original data sample set, an adaptive filtering algorithm is used to dynamically adjust the filtering parameters to remove noise interference, a high-order statistical analysis method is used to mine the nonlinear characteristics in the signal data, and a time-frequency analysis tool is used to extract the time-varying characteristics and frequency domain distribution characteristics of the frequency band; Using an ensemble learning algorithm to fuse multiple weak classifiers and taking historical frequency band features as input, a frequency band quality assessment model is constructed. Combining the frequency band quality assessment model, the time-varying characteristics, and the frequency domain distribution characteristics to obtain a frequency band quality assessment result; Based on the frequency band quality assessment result, the frequency band switching action of the first 5G portable WiFi device is expanded to a continuous space based on a deep reinforcement learning algorithm; Leveraging the parallel computing capabilities of quantum computing, we construct a quantum neural network for interference signal analysis. The first 5G portable WiFi device is further configured to: Based on the interference analysis results, coding technology is used to dynamically adjust the coding rate according to the interference intensity; Combined with orthogonal frequency division multiplexing technology, subcarriers are allocated and subcarriers in frequency bands with severe interference are used to transmit redundant information, ensuring the correct transmission of critical data; Introducing beamforming technology, using a multi-antenna array to form a null in the interference direction, suppressing the reception of interfering signals and enhancing the signal strength in the target direction; The edge server is further configured to: Continuously apply semantic segmentation algorithms in image recognition and processing technology to segment the latest environmental images acquired in real time, identify the location, outline, and material of newly appearing obstructions, and assess their degree of signal obstruction; Based on the assessment of the degree of signal obstruction caused by newly appeared obstructions and combined with real-time feedback from IoT sensors on temperature, humidity, and electromagnetic environment changes, the first 5G portable WiFi device is controlled to dynamically adjust the transmission power, signal frequency, and modulation method. The device also maintains high-frequency interaction with mobile communication base stations, uploads local environment and signal status, and downloads the latest optimization instructions, forming a closed-loop optimization link and continuously adaptively optimizing the signal enhancement effect. Receive key status data of the first 5G portable WiFi device through the Internet of Things, perform big data analysis based on data from multiple 5G portable WiFi devices, and push personalized signal enhancement solutions to different users based on the data analysis results.

Images