A critical area electromagnetic spectrum monitoring system based on communication tower clusters

Abstract

This invention discloses a critical area electromagnetic spectrum monitoring system based on communication tower clusters, comprising a sensor network module, a data transmission module, a data analysis and processing module, a communication and coordination module, a security and compliance module, a decision support and report generation module, a user interface and interaction module, and a system management and maintenance module. The sensor network module is used to collect electromagnetic spectrum data within the critical area; the data analysis and processing module is used to analyze and process the received electromagnetic spectrum data; and the decision support and report generation module is used to generate analysis reports and make decisions based on the monitored data. Based on existing communication towers, this system integrates advanced spectrum monitoring and analysis functions to perform efficient and accurate electromagnetic spectrum monitoring of critical areas. This system can optimize spectrum resource allocation, ensure national and communication security, and promote the quality and efficiency of wireless communication services.

Description

Technical Field

[0001] This invention relates to the field of radio communication technology, and in particular to a critical area electromagnetic spectrum monitoring system based on a network of communication towers. Background Technology

[0002] With the rapid development of wireless communication technology, electromagnetic spectrum resources are becoming increasingly scarce, making their effective management and utilization a critical issue. Especially in key areas such as city centers, military bases, and airports, spectrum interference and illegal occupation occur frequently, severely impacting the normal operation of legitimate communication services. Traditional spectrum monitoring systems often rely on ground-based mobile monitoring vehicles or fixed monitoring stations, which suffer from drawbacks such as limited coverage, slow response times, and high monitoring costs. Summary of the Invention

[0003] The purpose of this invention is to provide an electromagnetic spectrum monitoring system for key areas based on communication tower clusters, which integrates advanced spectrum monitoring equipment and intelligent analysis algorithms to achieve comprehensive and real-time monitoring and management of the electromagnetic environment in key areas.

[0004] To achieve the above objectives, the present invention provides a key area electromagnetic spectrum monitoring system based on a network of communication towers, comprising:

[0005] The spectrum monitoring subsystem is installed on each communication tower in the key area to collect and monitor various wireless communication signals and record the precise geographical location of each monitoring point;

[0006] The data transmission unit is used to transmit the monitoring data collected by the spectrum monitoring subsystem through a wireless network to the central processing and analysis platform.

[0007] The central processing and analysis platform is used to process and analyze the received monitoring data, and send alarm information to the user terminal when the monitoring data is abnormal.

[0008] Furthermore, the spectrum monitoring subsystem includes a first sensor module, a second sensor module, a system management and maintenance module, an intelligent sensing module, and a power supply module;

[0009] The first sensor module is used to monitor various wireless communication signals, current sensor location information, and environmental information near the sensor;

[0010] The second sensor module is used to receive the raw monitoring data sent by the first sensor module and preprocess it;

[0011] The system management and maintenance module includes a monitoring and alarm unit and a maintenance and upgrade unit. The monitoring and alarm unit is used to monitor the system status in real time and send alarm information to the user terminal when a fault is detected. The maintenance and upgrade unit is used for system upgrades and updates.

[0012] The intelligent sensing module is used to automatically adjust the receiver's sensitivity and operating frequency band according to the current ambient noise level and signal type;

[0013] The power supply module is used to supply power to the spectrum monitoring subsystem.

[0014] Furthermore, the first sensor module includes:

[0015] Radio frequency (RF) sensors are used to capture signals in the electromagnetic spectrum;

[0016] GPS / positioning devices are used to determine the location of sensors;

[0017] An environmental sensor is used to detect ambient temperature and humidity near the sensor.

[0018] Furthermore, the second sensor module includes:

[0019] The data acquisition unit is used to collect the raw monitoring data sent by the first sensor network module;

[0020] The data preprocessing unit is used to perform adaptive filtering and noise reduction on the raw monitoring data.

[0021] Furthermore, the workflow of the intelligent sensing module includes:

[0022] S1: Initialize the parameters of the broadband RF transceiver, including the initial operating frequency band and receiver sensitivity;

[0023] S2: Collect the noise level and signal type of the current environment through sensors;

[0024] S3: Analyze the data collected in step S2 using machine learning model algorithms to identify noise characteristics and signal types;

[0025] S4: Adjust the operating parameters of the broadband radio frequency transceiver based on the analysis results of step S3. The operating parameters include the operating frequency band and the receiving sensitivity.

[0026] S5: Monitor the performance indicators of the broadband RF transceiver after adjusting the operating parameters. The performance indicators include bit error rate and signal quality. If the value of the performance indicator is less than the preset threshold, return to step S3 for further adjustment.

[0027] S6: Repeat steps S1-S5 according to the preset cycle, and continuously adjust the receiver parameters based on the new monitoring results.

[0028] Furthermore, the data transmission unit includes:

[0029] The data transmission module includes a wireless transmission unit and an encryption unit; wherein, the wireless transmission unit is used to transmit all data collected by the first sensor network module to the data analysis and processing module through a wireless communication network; the encryption unit is used to encrypt all data monitored by the first sensor network module;

[0030] The data analysis and processing module is used to call the intelligent analysis engine in the central processing and analysis platform to perform pattern recognition, trend prediction, anomaly detection, and interference source location on the acquired monitoring data.

[0031] The communication and coordination module includes an interface unit and a communication protocol unit. The interface unit provides an interface to external systems and the data transmission module to ensure effective communication between the system and various modules. The communication protocol unit manages the protocols required for communication between the internal and external systems of the monitoring system.

[0032] Furthermore, the data analysis and processing module includes:

[0033] The spectrum analysis unit is used to perform spectrum analysis on the monitoring data obtained by the second sensor network module using intelligent spectrum analysis algorithms, to obtain frequency band occupancy and detect illegal emission sources, thereby providing data support for future spectrum allocation;

[0034] The signal classification unit is used to classify the wireless communication signals collected by the first sensor network module;

[0035] The anomaly detection unit is used to detect anomalies in the acquired monitoring data through a rapid interference source localization algorithm, and to discover unusual communication patterns or behaviors.

[0036] Furthermore, the central processing and analysis platform includes:

[0037] The data storage module includes a data storage unit and a data management unit; the data storage unit is used to store the preprocessed data from the second sensor network module; the data management unit is used to perform add, delete, and modify operations on the stored data.

[0038] The intelligent analysis engine integrates multiple machine learning models and various deep learning algorithms to analyze and process monitoring data.

[0039] The decision support and report generation module is used to formulate new monitoring decisions for the monitoring system based on the analysis results of the intelligent analysis engine.

[0040] User interface and interaction modules are used to provide a web- and mobile-friendly visual interface.

[0041] Furthermore, the machine learning models integrated by the intelligent analysis engine include logistic regression models, decision tree models, random forest models, support vector machines, gradient boosting trees, and K-nearest neighbor models; the integrated deep learning algorithms include convolutional neural networks, recurrent neural networks, long short-term memory networks, Transformer networks, generative adversarial networks, autoencoders, and reinforcement learning algorithms.

[0042] Furthermore, the processing steps of the intelligent spectrum analysis algorithm include:

[0043] S1: Acquire the preprocessed electromagnetic spectrum data detected;

[0044] S2: Extract the maximum, average, and standard deviation of signal strength from electromagnetic spectrum data;

[0045] S3: Select a suitable machine learning model in the intelligent analytics engine;

[0046] S4: Train the selected machine learning model using historical monitoring data, adjust the model parameters through cross-validation, and evaluate the machine learning model;

[0047] S5: Update the model periodically using real-time acquired electromagnetic spectrum data;

[0048] S6: Use the updated model to predict the spectrum occupancy at a specified time and location.

[0049] Therefore, the present invention employs the above-mentioned electromagnetic spectrum monitoring system for key areas based on communication tower clusters, which has the following beneficial effects:

[0050] First, by setting up a spectrum monitoring subsystem, the present invention utilizes a high-performance broadband receiver for spectrum monitoring, which can cover a wide frequency band from MHz to GHz, ensuring comprehensive monitoring of various wireless communication signals (such as mobile communication, broadcasting, satellite communication, etc.); in addition, this unit integrates a GPS module, which can ensure the accurate recording of the geographical location information of each monitoring point, facilitating subsequent interference source location.

[0051] Secondly, by introducing adaptive tuning technology, the present invention enables the monitoring unit to automatically adjust the receiving sensitivity and operating frequency band according to the current environmental noise level and signal type, thereby reducing the false alarm rate while maintaining high monitoring accuracy.

[0052] Third, this invention utilizes solar panels or wind power generation devices on communication towers to power the monitoring unit, supplemented by an intelligent power management system, to achieve efficient energy utilization and long-term autonomous operation.

[0053] Fourth, this invention utilizes existing fiber optic networks as the backbone transmission channel, combined with high-speed wireless backhaul technology (such as 5G or private network communication) to ensure real-time and stable data transmission.

[0054] Fifth, this invention improves upon existing communication towers, making full use of existing tower resources, reducing additional infrastructure construction, and lowering implementation costs and complexity.

[0055] Sixth, the present invention employs adaptive filtering technology and signal denoising algorithm, which can improve the ability to capture weak signals in complex electromagnetic environments and ensure high accuracy of spectrum monitoring.

[0056] Seventh, this invention proposes an intelligent scheduling algorithm that can dynamically adjust spectrum usage strategies based on real-time monitoring results, realize spectrum resource sharing and dynamic allocation, maximize spectrum resource utilization efficiency, and at the same time ensure the communication quality of critical services.

[0057] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0058] Figure 1 This is a diagram of the module architecture of this system. Detailed Implementation

[0059] In the description of this invention, it should also be noted that, unless otherwise expressly specified and limited, these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0060] This invention discloses a critical area electromagnetic spectrum monitoring system based on communication tower clusters, which is deployed on existing communication tower infrastructure in critical areas to achieve comprehensive, real-time monitoring and management of the electromagnetic environment in critical areas.

[0061] The electromagnetic spectrum monitoring system includes a spectrum monitoring unit, a data transmission unit, and a central processing and analysis platform.

[0062] The following sections will introduce each module in detail:

[0063] 1. Spectrum Monitoring Unit

[0064] This spectrum monitoring unit is installed on each communication tower in the key area to monitor various wireless communication signals and record the precise geographical location of each monitoring point;

[0065] The spectrum monitoring unit includes a first sensor module, a second sensor module, a system management and maintenance module, an intelligent sensing module, and a power supply module;

[0066] The following sections will introduce each of the above sub-modules in detail:

[0067] (1) First sensor module

[0068] The first sensor module includes a radio frequency sensor, a GPS / positioning device, and an environmental sensor. The radio frequency sensor is used to capture signals in the electromagnetic spectrum, the GPS / positioning device is used to determine the location of the sensor, and the environmental sensor is used to detect environmental factors such as temperature and humidity, which helps to understand the reasons for signal changes.

[0069] The radio frequency sensor covers a wide frequency band from MHz to GHz, ensuring comprehensive monitoring of various wireless communication signals (such as mobile communication, broadcasting, satellite communication, etc.). The broadband radio frequency transceiver is the AD9375 broadband RF transceiver. Furthermore, the broadband radio frequency transceiver is connected to a high-speed signal processor. The high-speed signal processor is responsible for the initial decoding and preprocessing of the signal, and can be a TMS320C55x high-speed signal processor.

[0070] (2) Second sensor module

[0071] The second sensor module includes a data acquisition unit and a data preprocessing unit. The data acquisition unit is used to collect the raw monitoring data sent by the first sensor network module, and the data preprocessing unit is used to perform adaptive filtering and noise reduction on the raw monitoring data.

[0072] Specifically, the data preprocessing unit uses adaptive filtering and signal denoising algorithms to adaptively filter and denoise the raw monitoring data, thereby ensuring high accuracy of spectrum monitoring.

[0073] in:

[0074] The adaptive filtering method includes the following steps:

[0075] ① Data collection

[0076] Signal Acquisition: Sensor nodes capture signals in the electromagnetic spectrum, which can be continuous or intermittent.

[0077] Data recording: Recording information such as signal frequency, power, and arrival time.

[0078] ② Data preprocessing

[0079] Signal separation: Using filtering techniques and related denoising algorithms to separate signals from different sources.

[0080] Data cleaning: Remove invalid or noisy data to ensure data quality.

[0081] Signal denoising methods include the following steps:

[0082] ① Use a bandpass filter to remove noise outside the signal frequency range.

[0083] ② Use wavelet transform to decompose the signal into multiple scales, and apply thresholding at each scale to remove noise.

[0084] ③ If the signal is non-stationary, an adaptive filter can be used for noise reduction.

[0085] (3) System Management and Maintenance Module

[0086] The system management and maintenance module includes a monitoring and alarm unit and a maintenance and upgrade unit. The monitoring and alarm unit is used to monitor the system status in real time and send alarm information to the user terminal when a fault is detected. The maintenance and upgrade unit is used for system upgrades and updates.

[0087] (4) Intelligent sensing module

[0088] The intelligent sensing module is used to automatically adjust the receiving sensitivity and operating frequency band according to the current ambient noise level and signal type, thereby reducing the false alarm rate while maintaining high monitoring accuracy;

[0089] This module introduces adaptive tuning techniques, such as the adaptive tuning method disclosed in the paper "Adaptive Antenna Tuning for Improved Cellular Coverage and Battery Life," [IEEE Transactions on Antennas and Propagation, vol. 61, no. 10, pp. 5230-5238, Oct. 2013]. This method provides feedback signals to maintain the optimal matching state by tracking the antenna's operating state through a closed-loop tuning scheme and a mismatch sensor.

[0090] The workflow of the aforementioned intelligent sensing module is as follows:

[0091] S1: Initialize the parameters of the broadband RF transceiver, including the initial operating frequency band and receiver sensitivity;

[0092] S2: Collect the noise level and signal type of the current environment through built-in or external sensors (such as spectrum analyzers and noise detectors);

[0093] S3: Analyze the data collected in step S2 using machine learning model algorithms to identify noise characteristics and signal types;

[0094] S4: Adjust the operating parameters of the broadband radio frequency transceiver based on the analysis results of step S3. The operating parameters include the operating frequency band and the receiving sensitivity.

[0095] Specifically, the receiver's sensitivity to signals is adjusted by changing gain settings, filter parameters, etc., and the optimal operating frequency or band is selected to avoid frequency bands with greater interference.

[0096] S5: Monitor the performance indicators of the broadband RF transceiver after adjusting the operating parameters. The performance indicators include bit error rate and signal quality. If the value of the performance indicator is less than the preset threshold, return to step S3 for further adjustment.

[0097] Specifically, the performance indicators of broadband radio frequency transceivers are monitored by performing performance tests on the transmitter and receiver. When testing the transmitter performance, it is necessary to measure the transmitter's output power, test frequency stability, evaluate modulation quality, and test spurious radiation. When testing the receiver performance, it is necessary to test the receiver's sensitivity, selectivity, blocking capability, dynamic range, linearity, and noise figure. During the testing, auxiliary equipment is used for performance testing, including signal generators, spectrum analyzers, bit error rate (BER) testers, noise sources, and vector network analyzers (VNAs).

[0098] S6: Establish a continuous monitoring mechanism to check for changes in environmental conditions regularly or as needed, and continue to adjust receiver parameters based on new monitoring results.

[0099] The monitoring mechanism includes:

[0100] 1) Define monitoring objectives: Define the key performance indicators that need to be monitored, including bit error rate and signal quality;

[0101] 2) Select monitoring equipment: Use a spectrum analyzer and bit error rate tester for monitoring;

[0102] 3) Data Acquisition: Install sensors and collect data;

[0103] 4) Data transmission: Set up a data transmission channel to ensure that data is transmitted to the central server in real time.

[0104] 5) Data analysis and processing: Data is stored and analyzed on a central server for anomaly detection;

[0105] 6) Alarms and notifications: Set thresholds to trigger alarms when performance metrics exceed the thresholds;

[0106] 7) User interface: Provides real-time monitoring and historical data query functions through the user interface.

[0107] 8) On-demand monitoring: Allows users to manually trigger specific monitoring tasks as needed.

[0108] 9) Regular maintenance: Regularly check the status of monitoring equipment and update it.

[0109] 10) Security: Implement access control policies to ensure data security.

[0110] (5) Power supply module

[0111] The power supply module connects to the output of solar panels or wind power generation devices on the communication tower to power the spectrum monitoring unit, enabling efficient energy utilization and long-term autonomous operation.

[0112] 2. Data transmission unit

[0113] The data transmission unit includes a data transmission module, a data analysis and processing module, and a communication and coordination module;

[0114] The following sections will introduce each sub-module in detail.

[0115] (1) Data transmission module

[0116] The data transmission module includes a wireless transmission unit and an encryption unit; wherein, the wireless transmission unit is used to transmit all data collected by the first sensor network module to the data analysis and processing module through a wireless communication network; the encryption unit is used to encrypt all data monitored by the first sensor network module;

[0117] Specifically, the wireless transmission unit utilizes existing fiber optic networks as the backbone transmission channel, combined with high-speed wireless backhaul technology (such as 5G or private network communication). This involves deploying wireless access points in areas not covered by the fiber optic network and connecting these access points to the optical network backbone using high-speed wireless backhaul technology, ensuring real-time and stable data transmission. Furthermore, this transmission network employs a redundancy mechanism, deploying multiple physical links to connect the same two points. When one link fails, traffic automatically switches to another link, ensuring that a single point of failure does not affect the overall data flow. Moreover, personally identifiable information is removed during data transmission through the network, protecting user privacy.

[0118] The encryption unit efficiently compresses the data before transmission and encrypts it using industry-standard encryption algorithms to prevent data interception or tampering during transmission, thereby improving transmission efficiency and ensuring data security. Specifically, the compression method used in this invention is LZ77 / LZ78 or Deflate. The data encryption process uses the AES encryption algorithm. The sender generates an AES key and uses it to encrypt the compressed data. The encrypted data and the AES key are then sent to the receiver. The receiver generates an RSA key pair (public and private keys), decrypts the AES key using its own RSA private key, and finally decrypts the original data using the AES key.

[0119] (2) Data Analysis and Processing Module

[0120] The data analysis and processing module is used to call the intelligent analysis engine in the central processing and analysis platform to perform pattern recognition, trend prediction, anomaly detection, and interference source location on the acquired monitoring data.

[0121] The data analysis and processing module includes a spectrum analysis unit, a signal classification unit, and an anomaly detection unit. The spectrum analysis unit uses intelligent spectrum analysis algorithms to perform spectrum analysis on the monitoring data acquired by the second sensor network module, obtaining frequency band occupancy information and identifying illegal transmission sources, thus providing data support for future spectrum allocation. The signal classification unit classifies the wireless communication signals collected by the first sensor network module. The anomaly detection unit uses a rapid interference source localization algorithm to detect anomalies in the acquired monitoring data, identifying unusual communication patterns or behaviors.

[0122] in:

[0123] ① Intelligent spectrum analysis algorithm

[0124] The intelligent spectrum analysis algorithm is based on an integrated deep learning model to predict spectrum occupancy.

[0125] It uses historical monitoring data to train various models integrated in the intelligent analysis engine to predict spectrum demand at specific times and locations, assisting spectrum management departments in planning and adjusting spectrum allocation in advance.

[0126] The specific steps of the intelligent spectrum analysis algorithm are as follows:

[0127] S1: Data collection, collecting preprocessed electromagnetic spectrum data.

[0128] S2: Feature selection, extracting the maximum, average, and standard deviation of signal strength from electromagnetic spectrum data.

[0129] S4: Model selection. Choose a suitable machine learning algorithm from the deep learning models integrated in the intelligent analysis engine, such as support vector machine (SVM) or neural network (such as LSTM).

[0130] S5: Model training, using historical data to train the selected model, adjusting model parameters through cross-validation, and evaluating the selected model.

[0131] S6: Model optimization, collecting current new electromagnetic spectrum data and updating the model regularly to improve prediction accuracy.

[0132] S7: Model Prediction: Predict spectrum demand at specific times and locations using the updated model.

[0133] ② Fast Interference Source Localization Algorithm

[0134] This invention combines Time Difference of Arrival (TDoA), Angle of Arrival (AoA) positioning technology with a multi-base station joint positioning algorithm to achieve rapid and accurate location of interference sources.

[0135] The working principle of the fast interference source localization algorithm is as follows:

[0136] In a typical TDOA / AoA joint positioning scenario, suppose we have three receiving sites A, B and C, and each site is equipped with an antenna array capable of measuring AoA.

[0137] First, perform TDOA calculation: calculate the time difference ΔtAB between the arrival of the signal at A and B, and the time difference ΔtAC between the arrival of the signal at A and C.

[0138] Next, perform AoA measurement: measure the angles θA, θB, and θC at which the signal arrives at stations A, B, and C.

[0139] Secondly, data fusion: the least squares method is used to fuse these time differences and angle information.

[0140] Finally, position calculation is performed: based on TDOA information, a series of hyperbolas are obtained, and the intersection of these hyperbolas is the potential position. AoA information is used to further limit the possibility of the position, and finally the accurate position of the signal source is determined.

[0141] Based on the above working principle, the fast interference source localization algorithm adopted in this invention includes the following steps:

[0142] S1: Data Acquisition

[0143] - Multiple receiving stations collect signal data simultaneously.

[0144] - Each station records the timestamp of signal arrival and signal strength.

[0145] - Record the angle information of the signal arrival.

[0146] S1: Time Difference Calculation

[0147] Calculate the time difference of arrival (TDOA) of signals between different receiving stations.

[0148] S3: Angle Measurement

[0149] The angle at which the signal arrives at each receiving station is measured.

[0150] S4: Data Fusion

[0151] - Integrate TDOA and AoA data.

[0152] -Weightedly combine these two types of information.

[0153] S5: Positioning Calculation

[0154] - Calculate the distance difference between the signal source and the receiving station based on the TDOA information.

[0155] - Further narrow down the location range by combining AoA information.

[0156] - Use the least squares method or maximum likelihood estimation to solve for the location coordinates of the signal source.

[0157] S6: Output Results

[0158] - Output the location of the interference source calculated by S5.

[0159] S7: Error Correction

[0160] - Based on the actual application scenario and requirements, perform error correction or iterative optimization to improve positioning accuracy.

[0161] S8: Real-time updates

[0162] In dynamic environments, positioning requires constantly updating the data from the receiving station and recalculating the location.

[0163] (3) Communication and Coordination Module

[0164] The communication and coordination module includes an interface unit and a communication protocol unit. The interface unit provides an interface to external systems and the data transmission module, ensuring effective communication between the system and various modules. The communication protocol unit manages the protocols required for internal and external communication, ensuring that data is encapsulated and decapsulated in the correct protocol format, ensuring data integrity and accuracy, and improving the overall system performance.

[0165] 3. Central Processing and Analysis Platform

[0166] The central processing and analysis platform is used to analyze and process the data sent from the data transmission unit;

[0167] It includes a data storage module, a security and compliance module, an intelligent analysis engine, a decision support and report generation module, and a user interface and interaction module;

[0168] The above sub-modules will be introduced below.

[0169] (1) Data storage module

[0170] The data storage module uses a cloud service platform provided by an external cloud service provider to receive spectrum monitoring data sent from the data transmission network and store the spectrum monitoring data.

[0171] The data storage and management module includes a data storage unit and a data management unit; the data storage unit is used to store the data preprocessed by the second sensor network module; the data management unit is used to perform add, delete, and modify operations on the stored data.

[0172] (2) Intelligent Analysis Engine

[0173] It integrates multiple machine learning models and various deep learning algorithms to perform pattern recognition, trend prediction, anomaly detection, and interference source localization on massive monitoring data. The model continuously learns and self-optimizes, improving analysis accuracy and response speed.

[0174] Specifically, its integrated machine learning models include:

[0175] ① Logistic Regression Model

[0176] This model is a linear model used for binary or multi-class classification tasks. It is used to predict the probability of an event occurring and to determine which features have the greatest impact on the classification result.

[0177] ② Decision Trees Model

[0178] This model is a tree-based model that classifies or regresses data using a series of rules. It is used to predict outcomes based on different branches of feature values ​​and generate easy-to-understand decision rules.

[0179] ③ Random Forest model

[0180] This model is an ensemble learning method based on multiple decision trees, which improves prediction accuracy through majority voting or averaging prediction results. It is used to improve generalization ability by integrating multiple decision trees and to evaluate the importance of features.

[0181] ④ Support Vector Machines (SVM)

[0182] This model is a supervised learning model for classification and regression tasks that partitions data by finding the hyperplane with the maximum margin. It is used to handle both linear and nonlinear datasets, mapping the data to a high-dimensional space to solve nonlinear problems.

[0183] ⑤ Gradient Boosting Trees

[0184] This model progressively improves the prediction model by sequentially adding weak learners (usually decision trees). It is used to reduce bias and variance and improve prediction accuracy by iteratively adding models; and to handle imbalanced datasets by giving more weight to the minority class.

[0185] ⑥ K-Nearest Neighbors (KNN)

[0186] KNN is an instance-based learning method that predicts results by calculating the distance between test samples and samples in the training set. It is used for classification and regression, as well as for feature scaling through standardization or normalization.

[0187] Its integrated deep learning algorithms include:

[0188] ① Convolutional Neural Networks (CNN)

[0189] It is a deep learning model specifically designed for processing input data with a grid structure, such as images and videos. It can be used for image classification, object detection, and semantic segmentation.

[0190] ② Recurrent Neural Networks (RNN)

[0191] It is a deep learning model for processing sequential data (such as time series data, speech, and text). It can be used for speech recognition, natural language processing (NLP), and time series prediction.

[0192] ③ Long Short-Term Memory (LSTM) network

[0193] LSTM is a special form of RNN that can solve the problem of long-term dependencies. It can be used for text generation, speech synthesis, and sequence-to-sequence (Seq2Seq) learning.

[0194] ④Transformer

[0195] Transformer is an architecture based on self-attention mechanisms, originally designed for natural language processing tasks. It can be used for machine translation, text summarization, and question answering.

[0196] ⑤ Generative Adversarial Networks (GANs)

[0197] GANs consist of two networks—a generator and a discriminator—which are trained in a competitive manner. They are commonly used for image generation, image type conversion, and converting low-resolution images to high-resolution images.

[0198] ⑥ Autoencoders

[0199] It is a type of neural network used for unsupervised learning, which learns efficient data representations by reconstructing input data. It is commonly used for data dimensionality reduction and anomaly detection.

[0200] ⑦ Reinforcement Learning (RL)

[0201] RL is a learning strategy that uses trial and error, where agents take actions in the environment to maximize cumulative rewards.

[0202] The steps for continuous learning and self-optimization of all the integrated models above are as follows:

[0203] S1: Data Collection: Continuously collect electromagnetic spectrum data.

[0204] S2: Online learning: Using online learning algorithms (such as SGD) to update the model.

[0205] S3: Experience Replay: Store recent data and periodically sample from it for retraining.

[0206] S4: Regularization: Add a regularization term to the loss function to prevent the model from overfitting to the latest data.

[0207] S5: Multi-task learning: Simultaneously learns recent spectral characteristics to improve the accuracy of recommendations.

[0208] (3) Security and Compliance Module

[0209] The security and compliance module includes a security audit unit and a compliance check unit; the security audit unit is used to analyze system log files to find potential security vulnerabilities; the compliance check unit is used to perform compliance checks on user identities in the system.

[0210] The compliance check unit uses access control methods to perform compliance checks on user identities to prevent unauthorized access and operations;

[0211] The access control method includes:

[0212] 1) User authentication

[0213] The system verifies a user's identity by using one of the following methods: username / password verification, multi-factor authentication (MFA), or biometric authentication, ensuring that only authorized users can access the system.

[0214] 2) Access Tokens

[0215] Certificates are issued to authenticated users via JSON Web Tokens (JWT) or OAuth methods to prove the user's identity in subsequent requests.

[0216] 3) Password Policies

[0217] Based on the requirements for password complexity and password expiration policies, rules are specified regarding password complexity, validity period, and other related aspects.

[0218] (4) Decision support and report generation module

[0219] The decision support and report generation module includes a report generation unit and a decision support unit. The report generation unit summarizes abnormal data and signal frequency band usage, and generates corresponding reports. The decision support unit formulates new monitoring decisions for the monitoring system based on the analysis results of the intelligent analysis engine. These monitoring decisions include:

[0220] Resource allocation: guides the effective allocation of frequency resources.

[0221] Task scheduling: Determine the priority and execution order of monitoring tasks.

[0222] System adjustment: Adjust the operating parameters of the monitoring equipment to improve monitoring quality.

[0223] Alarm management: Respond to alarms based on their severity, such as notifying relevant personnel to take action;

[0224] (5) User Interface and Interaction Module

[0225] The user interface and interaction modules include a web interface and a mobile app, providing a user-friendly visual interface for both web and mobile devices, displaying real-time spectrum occupancy graphs, spectrum usage statistics reports, interference event history records, etc. It can also automatically send SMS, email, or app push notifications to relevant personnel according to preset rules.

[0226] Example

[0227] 1. Sensor Network Module

[0228] Function: Responsible for collecting electromagnetic spectrum data within key areas.

[0229] composition:

[0230] Radio frequency (RF) sensors: used to capture signals in the electromagnetic spectrum.

[0231] GPS / Positioning Devices: Used to determine the location of sensors.

[0232] Environmental sensors: used to monitor environmental factors such as temperature and humidity, which helps to understand the reasons for signal changes.

[0233] Workflow: The sensor network module continuously monitors signals in the electromagnetic spectrum and records information such as frequency and power.

[0234] 2. Data Storage and Management Module

[0235] Function: Store and manage collected data in a high-performance cloud computing infrastructure.

[0236] composition:

[0237] Data storage unit: Utilizes high-performance storage systems (such as object storage and block storage) to store data.

[0238] Data Management Unit: Provides data indexing, querying and other functions to facilitate subsequent data analysis.

[0239] Workflow: The data storage and management module is responsible for storing the collected data and providing data management functions such as indexing and querying.

[0240] 3. User Interface and Interaction Module

[0241] Function: Provides a user interface and interactive features to facilitate user access to system functions and data viewing.

[0242] composition:

[0243] Web interface: Provides users with a browser-based access method.

[0244] Mobile application: Supports access on mobile devices.

[0245] Workflow: The user interface and interaction module provide users with an interface to access system functions and view data.

[0246] 4. Communication and Coordination Module

[0247] Function: Responsible for communication and coordination with other systems (such as data centers).

[0248] composition:

[0249] Interface unit: Provides a data exchange interface with external systems.

[0250] Communication protocol unit: Supports multiple communication protocols to ensure data compatibility and interoperability.

[0251] Workflow: The communication and coordination module is responsible for data exchange and coordination with other systems.

[0252] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A key area electromagnetic spectrum monitoring system based on communication tower clusters, characterized in that, include: The spectrum monitoring subsystem is installed on each communication tower in the key area to collect and monitor various wireless communication signals and record the precise geographical location of each monitoring point; The data transmission unit is used to transmit the monitoring data collected by the spectrum monitoring subsystem through a wireless network to the central processing and analysis platform. The central processing and analysis platform is used to process and analyze the received monitoring data, and send alarm information to the user terminal when the monitoring data is abnormal. The spectrum monitoring subsystem includes a first sensor module, a second sensor module, a system management and maintenance module, an intelligent sensing module, and a power supply module. The first sensor module is used to monitor various wireless communication signals, current sensor location information, and environmental information near the sensor; The second sensor module is used to receive the raw monitoring data sent by the first sensor module and preprocess it; The system management and maintenance module includes a monitoring and alarm unit and a maintenance and upgrade unit. The monitoring and alarm unit is used to monitor the system status in real time and send alarm information to the user terminal when a fault is detected. The maintenance and upgrade unit is used for system upgrades and updates. The intelligent sensing module is used to automatically adjust the receiver's sensitivity and operating frequency band according to the current ambient noise level and signal type; The power supply module is used to power the spectrum monitoring subsystem; The workflow of the intelligent sensing module includes: S1: Initialize the parameters of the broadband RF transceiver, including the initial operating frequency band and receiver sensitivity; S2: Collect the noise level and signal type of the current environment through sensors; S3: Analyze the data collected in step S2 using machine learning model algorithms to identify noise characteristics and signal types; S4: Adjust the operating parameters of the broadband radio frequency transceiver based on the analysis results of step S3. The operating parameters include the operating frequency band and the receiving sensitivity. S5: Monitor the performance indicators of the broadband RF transceiver after adjusting the operating parameters. The performance indicators include bit error rate and signal quality. If the value of the performance indicator is less than the preset threshold, return to step S3 for further adjustment. S6: Repeat steps S1-S5 according to the preset cycle, and continuously adjust the receiver parameters based on the new monitoring results.

2. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 1, characterized in that, The first sensor module includes: Radio frequency (RF) sensors are used to capture signals in the electromagnetic spectrum; GPS / positioning devices are used to determine the location of sensors; An environmental sensor is used to detect ambient temperature and humidity near the sensor.

3. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 2, characterized in that, The second sensor module includes: The data acquisition unit is used to collect the raw monitoring data sent by the first sensor network module; The data preprocessing unit is used to perform adaptive filtering and noise reduction on the raw monitoring data.

4. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 3, characterized in that, The data transmission unit includes: The data transmission module includes a wireless transmission unit and an encryption unit; wherein, the wireless transmission unit is used to transmit all data collected by the first sensor network module to the data analysis and processing module through a wireless communication network; the encryption unit is used to encrypt all data monitored by the first sensor network module; The data analysis and processing module is used to call the intelligent analysis engine in the central processing and analysis platform to perform pattern recognition, trend prediction, anomaly detection, and interference source location on the acquired monitoring data. The communication and coordination module includes an interface unit and a communication protocol unit. The interface unit provides an interface to external systems and the data transmission module to ensure effective communication between the system and various modules. The communication protocol unit manages the protocols required for communication between the internal and external systems of the monitoring system.

5. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 4, characterized in that, The data analysis and processing module includes: The spectrum analysis unit is used to perform spectrum analysis on the monitoring data obtained by the second sensor network module using intelligent spectrum analysis algorithms, to obtain frequency band occupancy and detect illegal emission sources, thereby providing data support for future spectrum allocation; The signal classification unit is used to classify the wireless communication signals collected by the first sensor network module; The anomaly detection unit is used to detect anomalies in the acquired monitoring data through a rapid interference source localization algorithm, and to discover unusual communication patterns or behaviors.

6. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 5, characterized in that, The central processing and analysis platform includes: The data storage module includes a data storage unit and a data management unit; the data storage unit is used to store the preprocessed data from the second sensor network module; the data management unit is used to perform add, delete, and modify operations on the stored data. The intelligent analysis engine integrates multiple machine learning models and various deep learning algorithms to analyze and process monitoring data. The decision support and report generation module is used to formulate new monitoring decisions for the monitoring system based on the analysis results of the intelligent analysis engine. User interface and interaction modules are used to provide a web- and mobile-friendly visual interface.

7. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 6, characterized in that, The intelligent analysis engine integrates machine learning models including logistic regression, decision tree, random forest, support vector machine, gradient boosting tree, and K-nearest neighbor model; and integrates deep learning algorithms including convolutional neural network, recurrent neural network, long short-term memory network, Transformer network, generative adversarial network, autoencoder, and reinforcement learning algorithm.

8. The electromagnetic spectrum monitoring system for key areas based on communication tower clusters as described in claim 7, characterized in that, The processing steps of the intelligent spectrum analysis algorithm include: S1: Acquire the preprocessed electromagnetic spectrum data detected; S2: Extract the maximum, average, and standard deviation of signal strength from electromagnetic spectrum data; S3: Select a suitable machine learning model in the intelligent analytics engine; S4: Train the selected machine learning model using historical monitoring data, adjust the model parameters through cross-validation, and evaluate the machine learning model; S5: Update the model periodically using real-time acquired electromagnetic spectrum data; S6: Use the updated model to predict the spectrum occupancy at a specified time and location.

Images