Multi-dimensional characteristic carrier real-time intelligent management and control system based on RFID signals

Abstract

The invention relates to the technical field of Internet of Things and intelligent management, in particular to a multi-dimensional characteristic carrier real-time intelligent management and control system based on an RFID signal, and the system comprises an intelligent RFID read-write module, a signal characteristic analysis unit, a carrier state diagnosis engine, a dynamic power management module and a cloud management platform. The intelligent RFID read-write module supports dual frequency bands of UHF (Ultra High Frequency) RFID (Radio Frequency Identification Device) (860MHz-960MHz) and HF (High Frequency) RFID (Radio Frequency Identification Device) (13.56 MHz); in the scheme, through close cooperative work of the intelligent RFID read-write module, the signal feature analysis unit, the carrier state diagnosis engine, the dynamic power management module and the cloud management platform, high-precision positioning, real-time state monitoring and automatic management of the carrier are realized; advanced technologies such as multi-dimensional signal fusion, self-adaptive power control and AI state diagnosis are adopted by the system, many problems of a traditional RFID system in a complex environment are effectively solved, and the operation efficiency and reliability in the fields of logistics, storage, medical treatment, book and archive management and the like are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things and intelligent management, and particularly relates to a real-time intelligent control system for multi-dimensional feature carriers based on RFID signals. Background Art

[0002] At present, with the booming development of Internet of Things technology, the efficient management of various physical carriers (such as logistics parcels, warehouse goods, books and archives, medical equipment, etc.) has become the key for many fields to improve operation efficiency and ensure stable operation of business.

[0003] In traditional RFID systems in densely deployed scenarios (such as logistics warehouses), due to the superposition of multi-tag signals and the attenuation of radio frequency signals by metal / liquid carriers, the reading rate decreases. Moreover, existing solutions can only obtain the carrier ID and basic attributes, and cannot perceive the physical state of the carrier in real time. There may be problems such as position deviation, temperature and humidity changes, and damage risks. The fixed-power reader needs to be frequently adjusted in long-distance or dynamic environments, resulting in high energy consumption and response delay.

[0004] Therefore, the present invention proposes a carrier management system based on multi-dimensional feature analysis of RFID (radio frequency identification) signals, which is used to track, monitor the status and intelligently manage various physical carriers (such as logistics parcels, warehouse goods, books and archives, medical equipment, etc.), and solve the problems of signal interference, multi-tag conflict and poor adaptability to dynamic environments in traditional RFID systems in complex scenarios. Summary of the Invention

[0005] Technical problems to be solved: The problems of signal interference, multi-tag conflict and poor adaptability to dynamic environments in traditional RFID systems in complex scenarios.

[0006] In view of the deficiencies of the prior art, the present invention provides a real-time intelligent control system for multi-dimensional feature carriers based on RFID signals, thereby solving the technical problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] A real-time intelligent control system for multi-dimensional feature carriers based on RFID signals, the system includes an intelligent RFID reading and writing module, a signal feature analysis unit, a carrier status diagnosis engine, a dynamic power management module and a cloud management platform;

[0009] The intelligent RFID reading and writing module supports dual bands of UHF RFID (860MHz - 960MHz) and HF RFID (13.56MHz), is built-in with a 0.1mW - 2W adjustable power amplifier, integrates a four-polarization antenna array and adopts beamforming technology;

[0010] The signal feature analysis unit is used to collect the RSSI, phase difference, Doppler frequency shift and backscatter modulation features of RFID signals in real time, and process the signals using the Kalman filtering algorithm;

[0011] The carrier status diagnosis engine constructs a lightweight graph neural network to classify and judge the carrier status, and uses a federated learning framework to co-train the model;

[0012] The dynamic power management module dynamically optimizes the transmitter power of the reader / writer using reinforcement learning;

[0013] The cloud management platform is used for data storage, display and interaction with external systems.

[0014] In a possible implementation, the intelligent RFID reader / writer module initializes the operating frequency band and transmitter power according to the preset configuration at system startup, adjusts the power according to the instructions of the dynamic power management module during the scanning task, and uses beamforming technology to direct the signal to cover the target area. The received RFID signal is transmitted to the signal feature analysis unit.

[0015] In a possible implementation, the signal feature analysis unit obtains the raw signal from the intelligent RFID reader / writer module, performs feature extraction and Kalman filtering processing, calculates the carrier motion trajectory, stability index and motion acceleration, transmits the processed information to the carrier status diagnosis engine, and at the same time feeds back the information to the dynamic power management module.

[0016] In a possible implementation, the carrier status diagnosis engine receives the signal feature matrix of the signal feature analysis unit, judges the carrier status through the lightweight graph neural network model, transmits the judgment result to the cloud management platform, triggers an alarm mechanism according to the abnormal status, and sends control instructions to the dynamic power management module and automation equipment, and performs model parameter interaction and update with other nodes in the federated learning.

[0017] In a possible implementation, the dynamic power management module obtains information such as the carrier distance, motion state, and signal attenuation degree from the signal feature analysis unit and the carrier status diagnosis engine, calculates the optimal transmitter power through the DQN algorithm and sends instructions to the intelligent RFID reader / writer module, and at the same time feeds back the power adjustment record and relevant data to the cloud management platform.

[0018] Beneficial effects compared with the prior art:

[0019] 1. In this solution, through the close collaborative work of the intelligent RFID reading and writing module, signal feature analysis unit, carrier status diagnosis engine, dynamic power management module, and cloud management platform, high-precision positioning, real-time status monitoring, and automated management of the carrier are achieved. The advanced technologies such as multi-dimensional signal fusion, adaptive power control, and AI status diagnosis adopted by the system effectively solve many problems of traditional RFID systems in complex environments, significantly improve the operation efficiency and reliability in the fields of logistics, warehousing, medical care, library and archives management, etc., and have broad application prospects and promotion value. In actual applications, system parameters and functions can be further optimized according to different scenario requirements to better meet user needs.

[0020] 2. In this solution, through the intelligent RFID reading and writing module that supports dual-band, adjustable power, and four-polarized antenna array and beamforming technology, combined with the multi-feature acquisition and Kalman filtering algorithm of the signal feature analysis unit, high-precision identification and stable signal transmission of the carrier label in complex environments are achieved. For example, in a logistics warehouse with a dense metal environment, it can effectively overcome signal interference and multipath effects, with a reading rate ≥ 99.5%, a multi-tag processing speed ≥ 500 tags / s, accurately obtain information such as the position and motion state of the carrier, and provide a reliable data basis for subsequent management decisions.

[0021] 3. In this solution, with the help of the lightweight GNN model and federated learning framework of the carrier status diagnosis engine, and its collaborative work with other modules, accurate monitoring and intelligent diagnosis of the physical state of the carrier are achieved. For example, in warehousing logistics, abnormal states such as the inclination, fall, and moisture of goods can be detected in a timely manner, the detection sensitivity of the fall event is ≥ 95%, the warning response time for abnormal temperature and humidity is ≤ 1 s, and the generalization ability of the model can be improved through multi-node collaborative training to ensure the safety of the carrier and reduce cargo losses.

[0022] 4. In this solution, based on the deep Q-network (DQN) reinforcement learning algorithm of the dynamic power management module, in collaboration with the intelligent RFID reading and writing module, etc., a significant reduction in system energy consumption and optimization of reading performance are achieved. While meeting the carrier identification requirements in different scenarios, the energy consumption is reduced by ≥ 40%, and the reading rate is still > 90% in the low-power mode. For example, in a logistics warehouse, the power of the reader is dynamically adjusted according to the distance of the goods and signal attenuation, which not only saves energy but also ensures the data acquisition efficiency and reduces operating costs. Description of the Drawings

[0023] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly and implement it according to the content of the specification, the following takes the preferred embodiments of the present invention and combines the drawings to describe in detail as follows.

[0024] Figure 1 It is a schematic diagram of the system framework of the present invention. Detailed implementation manners

[0025] Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention can be implemented in various different forms, and thus the present invention is not limited to the embodiments described below;

[0026] The technical solutions in the embodiments of the present application are to solve the problems in the above-mentioned background technology, and the general idea is as follows:

[0027] Embodiment:

[0028] Please refer to Figure 1 As shown, this embodiment introduces a real-time intelligent management and control system for multi-dimensional feature carriers based on RFID signals. The system mainly consists of five core parts: an intelligent RFID reading and writing module, a signal feature analysis unit, a carrier status diagnosis engine, a dynamic power management module, and a cloud management platform. Each part works in coordination to achieve all-round management and control of the carrier, specifically as follows:

[0029] 1. Intelligent RFID reading and writing module

[0030] 1.1 Technical means

[0031] This module supports dual bands of UHF RFID (860 MHz - 960 MHz) and HF RFID (13.56 MHz), and is built-in with a 0.1 mW - 2 W adjustable power amplifier, integrating a four-polarized antenna array. With the help of beamforming technology, the radiation direction is controlled in real time through FPGA programming; in different scenarios, the frequency band can be flexibly switched. For example, the UHF band is used for long-distance identification of a large number of goods in a logistics warehouse, and the HF band is selected for close-range management of medical equipment; by adjusting the power amplifier, the transmission power is accurately controlled according to the tag distance and signal strength, reducing energy waste and signal interference; the four-polarized antenna array can receive and transmit signals in multiple polarization directions, and combined with beamforming technology, it can achieve full coverage of the target area and effectively improve the signal reception ability;

[0032] 1.2 Management and coordination

[0033] When the system starts, the working frequency band and transmission power are initialized according to the preset configuration; after the scanning task is issued, the transmission power is adjusted according to the instructions of the dynamic power management module, and the signal is directionally covered in the target area according to the beamforming technology; after receiving the RFID signal, it is transmitted to the signal feature analysis unit in a timely manner; it maintains real-time communication with other modules and adjusts the working state according to the overall requirements of the system;

[0034] 1.3 Effects

[0035] 1.3.1 Effects of frequency band switching and power adjustment

[0036] By supporting both UHF and HF frequency bands, the system can flexibly select the operating frequency band according to the actual scenario; in scenarios such as logistics warehouses where a large number of goods need to be identified over a long distance, the UHF band, with its longer identification distance and higher transmission rate, enables rapid batch reading of goods tags; while in scenarios such as medical equipment, books, and archives where close-range identification and management are required, the HF band has good identification effect within a short distance and certain resistance to interference from metals and liquids, ensuring stable signal transmission and accurate tag identification; at the same time, the built-in adjustable power amplifier (0.1mW - 2W) adjusts the transmission power in real time according to the tag distance and signal strength; when the tag is close to the reader, the power is reduced to reduce energy consumption and signal interference; when the tag is far away or the signal is weak, the power is increased to ensure effective signal transmission;

[0037] For example, in the scenario of warehouse shelves, for the goods tags close to the edge of the shelf, the reader automatically reduces the transmission power to 0.3mW, which not only ensures the reading success rate but also avoids interfering with tags in other areas; while for the goods tags deep in the shelf, the power is increased to 1W to achieve stable reading, thus significantly improving the reading success rate of tags and the energy utilization efficiency of the system;

[0038] 1.3.2, Antenna Array and Beamforming Effect

[0039] The integrated four-polarization antenna array combined with beamforming technology controls the radiation direction in real time through FPGA programming; the four-polarization antenna can receive and transmit signals in multiple polarization directions, greatly expanding the signal coverage range compared with traditional single-polarization antennas; the beamforming technology enables the reader to concentrate the signal energy on the target area and enhance the signal strength; in a complex warehouse environment with numerous shelves, goods, and other obstacles, traditional antennas are prone to signal blind spots, but the four-polarization antenna array of this system combined with beamforming technology can accurately radiate the signal to the area where the target shelf or goods are located, reducing the influence of signal occlusion and reflection, effectively reducing signal blind spots, and enhancing the signal reception ability of the system to ensure stable and efficient communication with tags in a complex environment;

[0040] 2. Signal Feature Analysis Unit

[0041] 2.1 Technical Means

[0042] This unit effectively eliminates the interference of environmental noise and multipath effects by collecting the RSSI, phase difference, Doppler frequency shift, and backscatter modulation characteristics of RFID signals in real time and applying the Kalman filter algorithm. Through in-depth analysis of the processed signals, it accurately calculates the carrier motion trajectory, stability index, and motion acceleration; for example, it judges the carrier position offset according to the change of the phase difference, analyzes the change of the distance between the carrier and the reader based on the RSSI fluctuation, and determines the carrier motion speed in combination with the Doppler frequency shift;

[0043] 2.2. Management and Collaboration

[0044] After obtaining the original signal from the intelligent RFID reading and writing module, feature extraction and processing are carried out quickly; information such as the processed carrier motion trajectory and stability index is transmitted to the carrier status diagnosis engine in a timely manner, providing key data support for carrier status judgment. At the same time, information such as the calculated carrier motion acceleration is fed back to the dynamic power management module to assist it in adjusting the transmitter power of the reader-writer;

[0045] 2.3. Effects

[0046] 2.3.1. Multi-feature Acquisition and Processing Effects

[0047] The RSSI, phase difference, Doppler frequency shift, and backscatter modulation features of the RFID signal are collected in real time, providing a rich data basis for comprehensively understanding the carrier status; RSSI reflects the signal strength, and by analyzing its changes, the distance change between the carrier and the reader-writer and whether the signal is interfered can be judged; the phase difference can accurately detect the position offset of the carrier; the Doppler frequency shift is used to determine the motion speed of the carrier; the backscatter modulation feature contains specific information of the carrier tag;

[0048] For example, during the logistics transportation process, by continuously monitoring the fluctuation of RSSI, if it is found that the RSSI suddenly drops, combined with the change of the phase difference, it can be judged that the goods may have moved in position or been blocked by other objects; at the same time, the motion speed of the goods is calculated according to the Doppler frequency shift to grasp the transportation status of the goods in real time. These multi-dimensional features complement each other, providing strong support for accurately judging the carrier status subsequently;

[0049] 2.3.2. Application Effects of Kalman Filter

[0050] The Kalman filter algorithm is used to eliminate the interference of environmental noise and multipath effects, significantly improving the signal quality; multipath effects are common in the actual environment, and the signal will propagate to the reader-writer through different paths, resulting in signal distortion and interference; the Kalman filter algorithm uses the state estimate of the previous moment and the observed value of the current moment, and through two steps of prediction and update, optimizes the signal; in the prediction step, the current state is predicted according to the system model; in the update step, the predicted state is corrected by combining the actual observed value;

[0051] For example, in a metal shelf warehouse, the multipath effect seriously affects the signal quality. After being processed by the Kalman filter, the noise of the signal is effectively suppressed, making the parameters such as the carrier motion trajectory, stability index, and motion acceleration calculated based on these signals more accurate and reliable, laying a solid foundation for the accurate judgment of the carrier status;

[0052] 3. Carrier Status Diagnosis Engine

[0053] 3.1. Technical Means

[0054] Construct a lightweight Graph Neural Network (GNN) with the signal feature matrix (time × frequency band × parameter) output by the signal feature parsing unit as the input; through deep learning and analysis of complex signal features, the GNN accurately outputs the classification results of the carrier state, such as normal, tilted, dropped, abnormal temperature and humidity, etc.; introduce the federated learning framework to support multi-node collaborative training of the model; each node uses local data to train the model, and aggregates and updates the model parameters through the federated learning algorithm without transmitting the original data, improving the generalization ability of the model while protecting data privacy;

[0055] 3.2. Management and Collaboration

[0056] The carrier state diagnosis engine constructs a lightweight Graph Neural Network (GNN) and continuously receives the signal feature matrix (time × frequency band × parameter) transmitted by the signal feature parsing unit; the GNN model deeply analyzes these complex signal features, such as monitoring the states of the carrier's tilt, drop, abnormal temperature and humidity, etc.; taking the goods warehousing scenario as an example, when the RSSI attenuation rate of the goods is parsed to be abnormal, combined with the alarm of the temperature and humidity sensor, and the standard deviation of the phase difference is greater than 5° and lasts for 3 seconds, the GNN model determines that the goods may be tilted and damp, and determines that the carrier state is abnormal;

[0057] Once the carrier state diagnosis engine identifies an abnormal carrier state, it will generate corresponding control instructions according to the preset logic; these instructions contain key information such as the type of abnormality, location information, and the corrective actions to be performed; for example, when it is determined that the goods are tilted, the instructions will specify the angle and direction that the robotic arm needs to adjust to restore the goods to the normal position; if there is an abnormal temperature and humidity, it may instruct the AGV to transfer the goods to an area with a suitable environment;

[0058] The carrier state diagnosis engine sends the generated control instructions to the corresponding automation equipment through the internal communication network of the system; in a modern industrial network environment, data transmission is usually carried out in a wired or wireless manner, such as communication protocols like industrial Ethernet, Wi-Fi, or Zigbee, to ensure that the instructions can be quickly and accurately delivered to the target equipment; taking the robotic arm as an example, it is equipped with a dedicated controller that receives the instructions from the carrier state diagnosis engine and parses and processes the instructions;

[0059] After receiving the control instruction, the automated equipment performs corresponding rectification actions according to the instruction content; the robotic arm precisely adjusts its joint movements according to the angle and position information in the instruction, grabs and straightens the tilted goods; the AGV moves to the designated position according to the path planned by the instruction and transports the goods in the area with abnormal temperature and humidity to a suitable environment; during the execution process, the automated equipment will real-time feedback the execution status to the carrier status diagnosis engine for subsequent monitoring and evaluation;

[0060] After the automated equipment completes the rectification operation, it will send a feedback signal to the carrier status diagnosis engine; after receiving the feedback, the carrier status diagnosis engine obtains the relevant signal characteristics of the carrier through the signal feature analysis unit again to verify the rectification effect; if it is verified that the carrier status returns to normal, then this rectification operation is completed; if there are still abnormalities, the instruction may be adjusted again to continue controlling the automated equipment for further rectification operations until the carrier status returns to normal;

[0061] 3.3. Effect

[0062] 3.3.1. GNN Model Analysis Effect

[0063] The constructed lightweight graph neural network (GNN) takes the signal feature matrix output by the signal feature analysis unit as input, and through in-depth learning and analysis of complex signal features, realizes the accurate classification of the carrier status; GNN can capture the complex relationships and patterns between signal features; for example, when the carrier drops, multiple features of its RFID signal (such as the mutation of RSSI, the abnormal change of phase difference, etc.) will change simultaneously, and GNN can learn the association between these features to accurately judge that the carrier is in the dropped state; compared with the traditional state judgment methods based on single features or simple models, GNN can comprehensively consider the mutual influence of multiple features, greatly improving the accuracy and reliability of state recognition; in the medical device management scenario, through continuous monitoring of the device RFID signal features and GNN model analysis, it is possible to timely detect the abnormal state of the device, such as whether the device is tilted due to misoperation or whether it has been collided, providing an effective guarantee for the safe use of medical devices;

[0064] 3.3.2. Federated Learning Collaboration Effect

[0065] The introduced federated learning framework supports multi-node collaborative model training; on different nodes (such as different warehouses, different hospital departments, etc.), local data is used for model training; since the data of different nodes has different characteristics and distributions, through federated learning, each node aggregates and updates the model parameters without transmitting the original data;

[0066] For example, in multiple logistics warehouses, the types of goods, storage environments, and logistics processes in each warehouse are different. Through federated learning, the systems of each warehouse can share the results of model training, improving the generalization ability of the model in different scenarios. At the same time, this method protects data privacy, avoids the leakage of sensitive data, enables the system to better adapt to the complex and changing actual environment, and improves the accuracy and generality of carrier state diagnosis.

[0067] 4. Dynamic Power Management Module

[0068] 4.1 Technical Means

[0069] Adopt reinforcement learning (DQN algorithm) to dynamically optimize the transmitter power of the reader according to information such as the carrier distance, signal attenuation degree, and carrier motion state. During the operation of the system, continuously collect environmental information and reader working state data, learn the optimal power adjustment strategy through the DQN algorithm, and achieve the goal of reducing energy consumption by ≥40%. For example, when the carrier is close to the reader and the signal strength is good, reduce the transmitter power; when the carrier is far away or the signal is severely interfered, appropriately increase the transmitter power.

[0070] 4.2 Management and Collaboration

[0071] Obtain information such as carrier distance and motion state from the signal feature analysis unit, and obtain signal attenuation degree and carrier state information from the carrier state diagnosis engine. According to this information, calculate the optimal transmitter power through the DQN algorithm, and send the power adjustment instruction to the intelligent RFID reading and writing module. At the same time, feedback the power adjustment record and relevant data to the cloud management platform for energy consumption analysis and system optimization.

[0072] 4.3 Effects

[0073] 4.3.1 Reinforcement Learning Decision-making Effects

[0074] Using reinforcement learning (DQN algorithm), the transmission power of the reader is dynamically optimized based on information such as the carrier distance, signal attenuation degree, and carrier motion state. The DQN algorithm learns the optimal power adjustment strategy by continuously trying different power adjustment actions and according to the rewards feedback from the environment. When the carrier is close to the reader and the signal strength is good, the algorithm will choose the action of reducing the transmission power, and a positive reward is given at this time because it not only saves energy but also does not affect tag reading. When the carrier is far away or the signal is severely interfered, the algorithm will try to increase the transmission power. If the tag is successfully read, a positive reward is given, otherwise a negative reward is given. Through continuous training and learning, the DQN algorithm can make optimal power adjustment decisions in various complex situations, achieving the goal of reducing energy consumption by ≥40%. In the actual warehousing scenario, after a period of training, the system can dynamically adjust the reader power according to the position of the goods on the shelf, effectively reducing the system energy consumption while ensuring the reading rate.

[0075] 4.3.2. Coordinated energy-saving effect of modules

[0076] The dynamic power management module closely cooperates with the intelligent RFID reading and writing module, the signal feature analysis unit, and the carrier status diagnosis engine to achieve a balance between energy saving and efficient identification. Information such as the distance and motion state of the carrier is obtained from the signal feature analysis unit, and the signal attenuation degree and carrier status information are obtained from the carrier status diagnosis engine. These information provide a decision-making basis for the dynamic power management module.

[0077] For example, when the carrier status diagnosis engine detects that a certain cargo is in an abnormal state and needs to be monitored keyly, the dynamic power management module will appropriately increase the transmission power of the reader for the label of this cargo according to the distance information provided by the signal feature analysis unit to ensure that the signal can be stably obtained without excessive energy consumption. Through the coordinated work between these modules, the system can still operate efficiently in the low-power mode with the tag reading rate greater than 90%, effectively improving the energy utilization efficiency and reducing the operating cost.

[0078] 5. Cloud management platform

[0079] 5.1 Technical means

[0080] Using visualization technology, display the carrier distribution heat map, status alarms, and historical trajectories according to the requirements of the handover documents. Through docking the API of the enterprise ERP / WMS system, realize data interaction and sharing. Store, manage, and analyze various types of data generated by the system to provide support for enterprise decision-making. For example, analyze the historical trajectories and status data of the carrier to optimize the warehouse layout and adjust the logistics distribution strategy, etc.

[0081] 5.2 Management and coordination

[0082] Receive carrier status information, alarm information transmitted by the carrier status diagnosis engine, and data such as power adjustment records fed back by the dynamic power management module; store and process this data and then display it to the management personnel through a visualization interface; send control instructions to other modules according to the operation instructions of the management personnel to achieve remote management and control of the entire system; conduct data interaction with the enterprise ERP / WMS system, synchronize the carrier management data to the enterprise business management system, and achieve business process collaboration;

[0083] 5.3. Effects

[0084] 5.3.1 Visualization display effects

[0085] Use visualization technology to display the carrier distribution heat map, status alarms, and historical trajectories, providing an intuitive and convenient management interface for management personnel; the carrier distribution heat map visually presents the distribution density of carriers in different regions through different colors or brightness, helping management personnel quickly understand the storage locations and quantity distributions of goods, facilitating the rational planning of warehouse layouts and logistics paths; status alarms prominently prompt management personnel of the abnormal status of carriers, such as equipment failures, missing goods, etc., and take corresponding measures in a timely manner; the historical trajectory function records the movement history of carriers, and by analyzing these trajectories, the logistics process can be optimized and the resource utilization efficiency can be improved;

[0086] For example, in the logistics warehouse management, management personnel can quickly find the areas where goods are concentrated by viewing the heat map and reasonably arrange handling equipment; when receiving status alarms, deal with abnormal goods in a timely manner to reduce losses; by analyzing historical trajectories, discover the flow patterns of certain goods, optimize storage locations, and improve the efficiency of goods inbound and outbound;

[0087] 5.3.2 System integration and data interaction effects

[0088] Realize data interaction and sharing by docking with the API of the enterprise ERP / WMS system; the cloud management platform synchronizes the carrier management data to the enterprise's business management system, enabling each department of the enterprise to obtain the latest information of the carrier in real time and achieving business process collaboration;

[0089] For example, in the enterprise's supply chain management, the sales department can adjust sales strategies in a timely manner according to the inventory information provided by the cloud management platform; the purchasing department can replenish goods in a timely manner according to the consumption of goods; the production department can reasonably arrange production plans according to the inventory status of raw materials; at the same time, the enterprise ERP / WMS system can also send instructions to the cloud management platform, such as adjusting the storage location of goods, querying specific goods information, etc., to achieve remote control and management of the carrier management system, improving the overall informatization management level of the enterprise, enhancing the operation efficiency and the scientific nature of management decisions.

[0090] Finally, it should be noted that: Obviously, the above embodiments are merely examples given to clearly illustrate the present invention, rather than limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation manners here. And the obvious changes or modifications derived therefrom still fall within the protection scope of the present invention.

Claims

1. A real-time intelligent control system for multi-dimensional feature carriers based on RFID signals, characterized in that The system includes an intelligent RFID reading and writing module, a signal feature analysis unit, a carrier state diagnosis engine, a dynamic power management module, and a cloud management platform; The intelligent RFID reading and writing module supports dual bands of UHF RFID (860 MHz - 960 MHz) and HF RFID (13.56 MHz), has a built-in adjustable power amplifier of 0.1 mW - 2 W, integrates a four-polarization antenna array and uses beamforming technology; The signal feature analysis unit is used to collect the RSSI, phase difference, Doppler frequency shift, and backscatter modulation features of RFID signals in real time, and processes the signals using the Kalman filter algorithm; The carrier state diagnosis engine constructs a lightweight graph neural network to classify and judge the carrier state, and uses a federated learning framework to co-train the model; The dynamic power management module uses reinforcement learning to dynamically optimize the transmitter power of the reader; The cloud management platform is used for data storage, display, and interaction with external systems.

2. The real-time intelligent control system for multi-dimensional feature carriers based on RFID signals according to claim 1, characterized in that, When the system starts, the intelligent RFID reading and writing module initializes the working frequency band and transmission power according to the preset configuration, adjusts the power according to the instructions of the dynamic power management module during the scanning task, and uses beamforming technology to direct the signal to cover the target area. The received RFID signal is transmitted to the signal feature analysis unit.

3. The real-time intelligent control system for multi-dimensional feature carriers based on RFID signals according to claim 1, wherein The signal feature analysis unit obtains the original signal from the intelligent RFID reading and writing module, performs feature extraction and Kalman filter processing, calculates the carrier movement trajectory, stability index, and movement acceleration, transmits the processed information to the carrier state diagnosis engine, and at the same time feeds back the information to the dynamic power management module.

4. The real-time intelligent control system for multi-dimensional feature carriers based on RFID signals according to claim 1, wherein The carrier state diagnosis engine receives the signal feature matrix from the signal feature analysis unit, judges the carrier state through the lightweight graph neural network model, transmits the judgment result to the cloud management platform, triggers an alarm mechanism according to the abnormal state, and sends control instructions to the dynamic power management module and automation equipment, and conducts model parameter interaction and update with other nodes in the federated learning.

5. The real-time intelligent control system for multi-dimensional feature carriers based on RFID signals according to claim 1, wherein The dynamic power management module obtains information such as the carrier distance, movement state, and signal attenuation degree from the signal feature analysis unit and the carrier state diagnosis engine, calculates the optimal transmission power through the DQN algorithm and sends instructions to the intelligent RFID reading and writing module, and at the same time feeds back the power adjustment record and related data to the cloud management platform.

Images