Anti-interference multi-band communication system and implementation method thereof

Abstract

The application discloses an anti-interference multi-frequency band communication system and an implementation method thereof, and belongs to the field of communication technology.The anti-interference multi-frequency band communication system comprises a perception layer, a decision layer and an execution layer.The perception layer comprises a service classification engine embedded in a protocol stack, which is used for dynamically marking a service type and a priority of a user PDU session according to QoS parameters, and an edge computing unit disposed on a side of a radio frequency unit, which is used for rapidly extracting spectrum characteristics by using an FPGA.The decision layer comprises a decision middleware, a cooperative control module and an anti-interference module, the decision middleware is used for synchronizing multi-source data by fusing spectrum data of the anti-interference module and device parameters of the cooperative control module and aligning time stamps, the decision middleware is used for structurally integrating the synchronized data, generating a fusion decision input package, generating a frequency band decision result according to a double-layer decision algorithm and encapsulating the frequency band decision result into a frequency band allocation instruction.The execution layer is used for adjusting radio frequency parameters according to the decision result and supporting dynamic bandwidth guarantee by using a radio frequency unit resource management.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to an anti-interference multi-band communication system and its implementation method. Background Technology

[0002] With the large-scale commercialization of 5G networks in the Sub-6GHz and millimeter-wave bands, the increased density of base stations has led to a sharp increase in the probability of co-channel / adjacent-channel interference. The diverse needs of various business scenarios for latency, reliability, and throughput have further exacerbated spectrum allocation conflicts. Traditional static frequency band allocation schemes cannot dynamically detect sudden interference (such as radar pulses or unauthorized equipment occupancy) and fluctuations in service load, resulting in the deterioration of QoS for critical services. Single-band anti-interference technology is limited by finite spectrum freedom and isolation, making it difficult to balance interference avoidance and capacity improvement in complex electromagnetic environments. Therefore, it is urgent to use a multi-band collaborative communication mechanism to aggregate discrete spectrum resources to form a virtual broadband channel. By combining real-time interference detection and multi-site collaborative decision-making, carrier frequency points, bandwidth, and spatial reuse modes can be dynamically reconstructed. This will enable precise isolation of interference sources, adaptation to service requirements, and maximization of spectrum efficiency in the three-dimensional space of space-frequency-time, thereby improving the overall network resilience while ensuring the reliability of high-priority services.

[0003] In existing technologies, when anti-interference and collaborative control are started in parallel during joint control, the data interaction between the two (such as spectrum data and frequency band parameters) lacks a clear priority or fusion mechanism, which may lead to a lengthy decision-making process, affect the real-time interference avoidance effect, and cause resource conflicts. Therefore, an anti-interference multi-band communication system and its implementation method are proposed. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an anti-interference multi-band communication system and its implementation method.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: An anti-interference multi-band communication system, comprising: Perception layer: The service type and priority are dynamically marked by parsing the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack. At the same time, the FPGA is used to quickly extract spectrum features through the edge computing unit deployed on the radio frequency unit side. Decision-making layer: It includes a decision-making middleware, a collaborative control module, and an anti-interference module. Through the decision-making middleware, the spectrum data of the anti-interference module and the device parameters of the collaborative control module are fused and time-stamped aligned for multi-source data synchronization. The decision-making middleware structurally integrates the synchronized data to generate a fused decision input package, generates a frequency band decision result according to the double-layer decision algorithm, and then the decision-making middleware encapsulates the frequency band decision result into a frequency band allocation instruction. The collaborative control module adopts a hierarchical multicast architecture to elect regional proxy nodes to aggregate the information within the group and synchronize the summary. The collaborative control module combines the incremental update protocol and the adaptive heartbeat mechanism. The anti-interference module performs emergency interference response and geographic association interference positioning through the pre-decision mechanism, data fusion, and FPGA acceleration to avoid misjudgment; Execution layer: Adjust the radio frequency parameters according to the decision result through radio frequency unit resource management and support dynamic bandwidth guarantee. The protocol stack adaptation dynamically adjusts the SSB / PDCCH configuration through the carrier reconfiguration sub-module, controls the enabling / disabling of broadcasting through the cell broadcast management sub-module, and provides a reconfiguration window through the user access query sub-module.

[0006] The above technical solution further includes: Preferably, the specific steps for the sensing layer to perform three-dimensional sensing of "service type - priority - spectrum status" through the collaboration of the service classification engine and spectrum feature extraction are as follows: Step 1: PDU session parsing and 5G QoS identification extraction: The service classification engine in the protocol stack first parses the PDU session establishment request initiated by the user equipment and extracts the 5G QoS identification parameters therein; Step 2: Mapping of 5G QoS identification to service type: According to the correspondence between the pre-defined 5G QoS identification and service type in the 3GPP standard, convert the 5G QoS identification parameters into specific service types; Step 3: Priority dynamic marking: Assign priorities according to the service type; Step 4: Acquisition of original radio frequency signals: The edge computing unit on the RU side receives wireless signals through the radio frequency front end and converts them into digital baseband signals ; Step 5: FFT transformation and spectrum analysis: Utilize the parallel computing ability of the FPGA to perform fast Fourier transform, converting the time-domain signal into a frequency-domain signal , expressed as , where k represents the frequency point index, 0 ≤ k < N, and N is the number of FFT points; Step 6: Calculation of signal power and interference intensity: Calculate the signal power of each frequency point , and statistically calculate the total power of the target frequency band , the frequency point power calculation formula is expressed as , the frequency band total power calculation formula is expressed as ,in, Indicates the frequency band range; Step 7: Interference Threshold Detection: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] With preset interference threshold In comparison, if > If so, it is determined that there is interference in the frequency band, and an interference event is triggered; Step 8: Extraction of interference feature parameters: If an interference event is triggered, further extract the interference features, including: interference start / stop frequency, interference intensity, and interference duration.

[0007] Preferably, the specific steps for the collaborative control module to perform collaborative control of the equipment are as follows: Regional agent node election and multicast architecture initialization: The collaborative control module first constructs a hierarchical multicast architecture and defines parameters, including device PCI coefficients and device computing capabilities. All devices broadcast their own PCI coefficients and computing capabilities to the multicast address. Each device calculates its own weight W according to the weighting formula and elects the device with the smallest weight as the regional agent node. The regional agent node establishes a multicast group, and devices in the group register with the agent. The agent maintains the device list. Intra-group information aggregation and summary synchronization: The regional agent node periodically performs the following operations: Frequency band parameter collection: The agent sends parameter request messages via multicast, and devices within the group respond with parameter reporting messages; Information aggregation: The agent statistically analyzes the frequency band occupancy within the group and generates a summary information table, which includes: frequency band occupancy rate, distribution of interfering frequency bands, and distribution of service types; Summary synchronization: The agent sends the summary information table to all devices via multicast, and the devices update their local frequency band status view; Incremental update protocol: The device sends a multicast update when the following conditions are met: Frequency band parameters change beyond a preset threshold, resulting in a change in interference level; Message format: Incremental update messages include: version number, changed field flag, and specific changed value; Adaptive Heartbeat Mechanism: Survival Status Detection and On-Demand Triggering: The collaborative control module employs a two-layer heartbeat detection mechanism. Basic heartbeat cycle: The device sends a heartbeat message every cycle T, and the cycle is dynamically adjusted according to the group size. Where K is the base period, α≈0.5 is a constant, and N is the number of devices in the group; Agent on-demand detection: If the agent does not receive a heartbeat from a device within 1.5T, it will actively send a status query message to check; if there is still no response within 2T, the agent will mark the device as "offline" and remove it from the device list, and trigger a multicast notification at the same time. Status recovery: After the device comes back online, it sends a parameter request message to the agent, which updates the device list and synchronizes the latest summary information.

[0008] Preferably, the specific steps for the anti-interference module to perform emergency interference response and interference localization are as follows: Real-time spectrum scanning and FPGA pre-filtering: The edge computing unit on the RF unit side performs high-spectrum-efficiency interference detection through FPGA, and only encapsulates the data that meets the emergency interference conditions into an emergency interference alarm message, which is then reported to the anti-interference module of the baseband processing unit through the internal bus. Data Fusion: Geographic Tag and Device Location Matching: After receiving an emergency interference event reported by the radio frequency unit, the anti-interference module performs the following operations: Geographic tag addition: Add geographic tags to interference events to record the physical location where the interference occurred; Device location association: Query the device locations recorded in the protocol stack and calculate the geographical distance between the interference location and each device; Interference correlation determination: If the distance between the device and the interference location is less than the threshold, the device is determined to be affected by interference; otherwise, it is marked as "remote interference". Pre-decision mechanism: Local reconfiguration triggering and execution: If an interference event is determined to be urgent and affects local equipment, the anti-interference module immediately initiates the pre-decision process without waiting for the global decision of the collaborative control module. Based on the interference frequency band and the equipment service type, it selects a carrier configuration to avoid interference from the pre-stored anti-interference frequency band configuration list, generates a carrier reconfiguration instruction message, and adjusts the radio frequency parameters of the affected equipment through the radio frequency unit resource management module. Reconfiguration result verification: After reconfiguration is completed, cell broadcast is enabled, and user access status is monitored through the user access query submodule; if interference still exists, the above process is repeated until the interference is resolved. Conflict resolution between collaborative control and anti-interference: If the collaborative control module initiates frequency band adjustment requests simultaneously, the anti-interference module ensures priority through the following rules: If the anti-interference pre-decision has triggered local reconfiguration, the decision of the collaborative control module must wait for the anti-interference process to be completed before it is executed. The anti-interference module synchronizes the reconfigured frequency band information to the collaborative control module through the decision middleware.

[0009] Preferably, the specific steps for the decision middleware to synchronize multi-source data by fusing the spectrum data of the anti-interference module with the device parameters of the collaborative control module and aligning the timestamps are as follows: Multi-source data acquisition and timestamp marking: The decision middleware first receives raw data streams from two core modules: spectrum data from the anti-interference module and device parameters from the collaborative control module; Timestamp alignment and data synchronization: Calculate the time difference between the spectrum data timestamp and the device parameter timestamp. If the absolute value of the time difference is greater than the threshold, it is determined that there is a synchronization deviation in the data. If the spectrum data is delayed, the spectrum data is cached and waits for the device parameters to be updated to the same time point. If the device parameters are delayed, the state of the device parameters at the time of the spectrum data is predicted by interpolation based on historical data. The aligned data packets are uniformly marked as decision timestamps. Data association and structured fusion: The aligned data needs to be associated based on the device identifier to generate a structured input package. Input packet validation and exception handling: Integrity verification: Checks whether the data packet contains required fields; missing fields are filled in using historical data or retransmission is triggered. Consistency check: Verify whether the device's current carrier configuration overlaps with spectrum interference bands; Time validity verification: If the difference between the data packet timestamp and the current system time exceeds the threshold, the data is discarded and re-collection is triggered; Integrate data inputs into a two-level decision algorithm: The structured fusion decision input packet is fed into a two-layer decision algorithm to generate the final frequency band decision result; Decision result feedback and status update: After the decision result is sent to the execution layer through the decision middleware, the new carrier configuration is synchronized to the device management information table of the collaborative control module, the frequency band occupancy status of each device is updated, and the correspondence between the decision timestamp, input data, and decision result is saved.

[0010] Preferably, the specific steps for generating frequency band decision results based on the two-level decision algorithm are as follows: Algorithm execution: Taking the fused decision input package as input, perform the following operations: First-level decision: Group devices according to preset rules, and arrange the devices into group queues in descending order based on the "business type and priority" field; Second-level decision-making: Select an allocation strategy based on group characteristics and adjust frequency band allocation in conjunction with interference information in spectrum data. If multiple devices need to be adjusted to the same frequency band, allocate resources according to priority and generate the final frequency band decision result. Output: The frequency band decision results generated by the two-layer decision algorithm are encapsulated into frequency band allocation instructions by the decision middleware. The decision middleware is responsible for the version management and conflict verification of the instructions in this process, and finally sends the instructions to the radio frequency unit resource management module and protocol stack adaptation module of the execution layer. Feedback loop: After the execution layer completes carrier reconfiguration, it feeds back the reconfiguration result to the perception layer through the decision middleware. During this process, the decision middleware performs result aggregation and anomaly detection.

[0011] Preferably, the execution layer utilizes a collaborative mechanism of hard adjustment of radio frequency unit parameters, soft adaptation of protocol stack configuration, and flexible management of user access to perform specific steps for carrier reconfiguration response and zero interruption of user services: Command parsing and parameter extraction: The RF unit resource management module receives frequency band allocation commands from the decision middleware, parses the key fields in the commands, and generates a local reconfiguration parameter table. RF parameter adjustment and bandwidth allocation: According to the parameters in the instruction, the RF unit adjusts the carrier center frequency to the target frequency through the RF front end, and adjusts the filter bandwidth according to the target bandwidth; Parameter activation and status feedback: After the adjustment is completed, a carrier reconfiguration confirmation message is reported to the baseband processing unit via the internal bus, and the reconfiguration status is marked as complete; SSB / PDCCH Dynamic Adjustment: After the protocol stack adaptation module receives the reconfiguration status feedback from the radio frequency unit, the carrier reconfiguration submodule adjusts the air interface parameters of the protocol stack according to the SSB / PDCCH configuration in the instruction. Community broadcast management submodule: Broadcast enable / disable control: Disable broadcast before reconfiguration: Upon receiving the reconfiguration command, immediately send a cell broadcast disable message to the radio frequency unit to stop cell broadcasting and prevent new users from accessing the network; Enable broadcast after reconfiguration: After the radio frequency unit completes parameter adjustment, the protocol stack sends a cell broadcast enable message to restore cell broadcast and allow user access; User access query submodule: Reconfiguration window management: Reconfiguration when no user access: If the user access query submodule detects "no user access", the reconfiguration process will be triggered immediately; Periodic monitoring when a user accesses the device: If a user accesses the device, a reconfiguration will be triggered according to a preset period, and the user device will be notified to suspend services before the reconfiguration is performed via a user access query message. Closed-loop verification: Reconfiguration result feedback: The protocol stack reports the new carrier configuration to the decision middleware through the carrier reconfiguration confirmation message, and updates the global frequency band view; Interference rescan verification: The edge computing unit on the radio frequency unit side performs a spectrum scan on the new carrier frequency band to confirm whether the interference has been eliminated; User access recovery monitoring: The user access query submodule counts the number of users accessing the network after reconfiguration. If the access rate is lower than the threshold, the collaborative control module is triggered to readjust the frequency band allocation.

[0012] A method for implementing an anti-interference multi-band communication system includes the following steps; Step 1: The perception layer parses the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack, dynamically marks the service type and priority for service perception, and at the same time, the edge computing unit deployed on the RU side uses FPGA to quickly extract spectrum features and reports the device parameters and spectrum data with service tags to the decision layer. Step 2: The decision middleware of the decision layer integrates the spectrum data of the anti-interference module and the equipment parameters of the collaborative control module, completes the synchronization of multi-source data through timestamp alignment, performs structured integration of the synchronized data to generate a fusion decision input package, generates frequency band decision results by combining the two-layer decision algorithm, and then encapsulates the results into frequency band allocation instructions; Step 3: The execution layer adjusts the radio frequency parameters based on the decision results through RU resource management, and dynamically adjusts the SSB / PDCCH configuration through the carrier reconfiguration submodule adapted to the protocol stack, controls the broadcast enable / disable through the cell broadcast management submodule, and provides a reconfiguration window through the user access query submodule to ensure user service continuity. At the same time, the execution layer feeds back the carrier status and user access status to the perception layer.

[0013] The present invention has the following beneficial effects: In this invention, anti-interference spectrum scanning and collaborative parameter collection are separated into independent pipeline stages. A decision middleware acts as a unified data hub for dynamic priority arbitration. The middleware first receives and timestamps the spectrum scanning results reported by the anti-interference module, and asynchronously integrates the device frequency band parameters transmitted by the collaborative module. Then, it performs real-time alignment and spatial correlation of heterogeneous data based on a preset conflict strategy. When an emergency interference event is detected, a local reconfiguration command is immediately triggered to bypass the collaborative process and take effect directly. Regular decisions are generated by integrating spectrum availability, service priority, and bandwidth requirements to create a global optimization scheme, which is finally distributed to each execution node through a standard interface. This not only eliminates the resource conflicts that occur when anti-interference and collaborative control are started in parallel in the joint control of existing technologies, but also compresses the response time of high-priority interference to the millisecond level through a hybrid triggering mode of event-driven and periodic scheduling, while ensuring resource utilization in low-frequency band adjustment scenarios. This achieves a synergistic leap in interference avoidance success rate and multi-service fairness in complex wireless environments. Attached Figure Description

[0014] Figure 1 This is a system architecture diagram of an anti-interference multi-band communication system proposed in this invention; Figure 2 This is a flowchart illustrating the implementation method of an anti-interference multi-band communication system proposed in this invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] like Figures 1-2 As shown, an anti-interference multi-band communication system includes: The perception layer: It achieves fine-grained service perception by parsing the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack and dynamically marking the service type and priority. At the same time, it uses the FPGA to quickly extract spectrum features through the edge computing unit deployed on the radio frequency unit (RU) side to support sub-millisecond interference event triggering and improve the real-time performance of spectrum data acquisition. The decision layer comprises a decision middleware, a collaborative control module, and an anti-interference module. The decision middleware fuses the spectrum data from the anti-interference module with the device parameters from the collaborative control module, aligning them with timestamps for multi-source data synchronization. The middleware then structurally integrates the synchronized data, generating a fused decision input package. Based on a two-layer decision algorithm, it generates frequency band decision results, which are then encapsulated into frequency band allocation instructions. The collaborative control module employs a hierarchical multicast architecture to elect regional agent nodes weighted by PCI coefficient and computing power, aggregating group information and synchronizing summaries. The collaborative control module also incorporates an incremental update protocol (only updating when frequency band parameter changes exceed a threshold or interference occurs). When the level changes, a multicast message containing version number and change field markers is sent. An adaptive heartbeat mechanism (survival detection period T=K*(group size)^α, α≈0.5, K is the baseline value, and the agent triggers on-demand detection) reduces multicast overhead and improves the scalability of large-scale networking and the real-time synchronization capability of device status. The anti-interference module uses a pre-decision mechanism (when the emergency interference intensity is >90%, local reconfiguration is triggered immediately without waiting for coordination), data fusion (geographical tags are added to spectrum data and intelligently matched with device location), and FPGA acceleration (the edge computing unit on the RU side pre-filters interference data to reduce BBU load) to perform emergency interference response and geographically associated interference location to avoid misjudgment. Execution layer: Through radio frequency unit (RU) resource management, radio frequency parameters (such as carrier frequency and bandwidth) are adjusted according to the decision results, and dynamic bandwidth guarantee is supported (a 5MHz dedicated frequency band pool is reserved for mMTC, and URLLC can preempt eMBB bandwidth but retains the minimum threshold). Protocol stack adaptation dynamically adjusts SSB / PDCCH configuration through the carrier reconfiguration submodule, controls broadcast enable / disable through the cell broadcast management submodule, and provides a reconfiguration window through the user access query submodule to ensure user service continuity.

[0017] Each layer forms a closed-loop optimization link through data interaction: the perception layer reports equipment parameters and spectrum data with service tags to the decision layer; the decision layer issues frequency band allocation instructions to the execution layer; and the execution layer feeds back carrier status and user access status to the perception layer.

[0018] In one embodiment, the specific steps of the perception layer in performing three-dimensional perception of "service type-priority-spectrum status" through the collaboration of a service classification engine and spectrum feature extraction are as follows: Step 1: PDU Session Parsing and 5G QoS Identifier (5QI) Extraction: The service classification engine in the protocol stack first parses the PDU session establishment request initiated by the user equipment (UE) and extracts the 5G QoS Identifier (5QI) parameter. 5QI is a QoS feature index defined in the 3GPP standard, which is directly related to the core indicators of the service such as latency, packet loss rate, and priority. Step 2: Mapping 5QI to service type: Based on the predefined correspondence between 5G QoS identifiers (5QI) and service types in the 3GPP standard, convert 5QI to specific service types (eMBB / URLLC / mMTC). 5QI=1, 2 / 3: corresponds to URLLC (Ultra-Reliable Low-Latency Communication), latency <1ms, reliability >99.999%; 5QI=4 / 5 / 6: corresponds to eMBB (enhanced mobile broadband), with peak rates >1Gbps; 5QI = 7 / 8 / 9: corresponds to mMTC (massive machine-type communications), with a connection density > 10^6 devices / km. 2 .

[0019] Step 3: Dynamic Priority Tagging: Assign priorities based on service type (configurable). The default rule is: URLLC > mMTC > eMBB. For example, the priority weight of URLLC is set to 10, mMTC to 5, and eMBB to 1. Example: When a UE initiates a PDU session, it carries 5QI=1. After parsing by the service classification engine, it is marked as "URLLC" with a priority weight of 10, and a structured data packet (containing the UE's PCI, 5QI, service type, and priority) is generated and reported to the decision middleware via the PROTOCOL_STACK_INFO message.

[0020] Step 4: Raw RF signal acquisition: The edge computing unit on the RU side receives wireless signals through the RF front end (such as an antenna or mixer) and converts them into digital baseband signals. (The sampling rate is typically on the order of tens of MHz); Step 5: FFT Transformation and Spectrum Analysis: Utilizing the parallel computing capabilities of the FPGA, the FFT transformation and spectrum analysis are performed. Perform a Fast Fourier Transform (FFT) to convert the time-domain signal into a frequency-domain signal , denoted as , where k represents the frequency index, 0 ≤ k < N, and N is the number of FFT points; Step 6: Signal power and interference intensity calculation: Calculate the signal power at each frequency point , and count the total power in the target frequency band (such as the n78 band from 3.4 - 3.6 GHz) , the formula for calculating the frequency point power is denoted as , the formula for calculating the total power of the frequency band is denoted as , where represents the frequency band range; Step 7: Interference threshold detection: Compare with the preset interference threshold (such as -80 dBm). If > , it is determined that there is interference in this frequency band, and an interference event is triggered; Step 8: Interference characteristic parameter extraction: If an interference event is triggered, further extract interference characteristics, including: interference start / stop frequency points, interference intensity, and interference duration.

[0021] Example: The edge computing unit on the RU side scans the n78 band and detects a total power = -75 dBm (exceeding the threshold -80 dBm), starting frequency point 3.5 GHz, ending frequency point 3.55 GHz, interference intensity -75 dBm, and duration 2 ms. The FPGA immediately generates a SPECTRUM_SCAN_RESULT message (including the frequency band range, interference intensity, and timestamp), reports it to the anti-interference module through the internal bus, and triggers the pre-decision process.

[0022] In one embodiment, the specific steps for the collaborative control module to perform collaborative control on the device are as follows: Regional proxy node election and multicast architecture initialization: The collaborative control module first constructs a hierarchical multicast architecture and defines parameters, including weighting both the device PCI coefficient (unique identifier, range 1~N, N is the number of devices in the group) and the device computing power (quantified as the number of CPU cores, e.g., device A has 4 cores and device B has 8 cores): (α and β are configurable weights, with default values ​​of α=0.7 and β=0.3). All devices broadcast their PCI coefficient and computing power to the multicast address. Each device calculates its own weight W according to the weighting formula and elects the device with the smallest weight as the regional proxy node (because the PCI coefficient is dominant, it ensures uniqueness; computing power assists in ensuring proxy processing performance). The regional proxy node establishes a multicast group, and devices within the group register with the proxy. The proxy maintains a device list (including PCI, IP address, and liveness status). Intra-group information aggregation and summary synchronization: The regional agent node periodically performs the following operations: Frequency band parameter collection: The agent sends a parameter request message (PARAM_REQUEST) via multicast, and the devices in the group respond with a parameter reporting message (PARAM_REPORT) (including the current carrier frequency, bandwidth, interference frequency band, and service type); Information aggregation: The agent statistically analyzes the frequency band occupancy within the group and generates a summary information table (AGENT_SUMMARY), which includes: frequency band occupancy rate, distribution of interfering frequency bands, and distribution of service types; Summary synchronization: The agent sends the summary information table (AGENT_SUMMARY) to all devices via multicast, and the devices update their local frequency band status view, avoiding the transmission of all parameters; Incremental update protocol: The device sends a multicast update when the following conditions are met: Frequency band parameter changes (frequency point shift, bandwidth change) exceed preset thresholds, interference level changes (interference intensity change, interference state switch (no interference → interference or vice versa)). Message format: The incremental update message (MULTICAST_DELTA_INFO) includes: version number (V, an incrementing integer to ensure message order), change field flags (such as FREQ_POINT_CHANGED, BANDWIDTH_CHANGED, INTERFERENCE_LEVEL_CHANGED), and specific change values ​​(such as new frequency 3.55GHz, new bandwidth 20MHz). Adaptive Heartbeat Mechanism: Survival Status Detection and On-Demand Triggering: The collaborative control module employs a two-layer heartbeat detection mechanism. Basic heartbeat cycle: The device sends a heartbeat message (HEARTBEAT) every cycle T, and the cycle is dynamically adjusted according to the group size. Where K is the base period, α≈0.5 is a constant, and N is the number of devices in the group; Agent on-demand detection: If the agent does not receive a heartbeat from a device within 1.5T, it will actively send a status query message (QUERY_STATUS) to query; if there is still no response within 2T, the agent will mark the device as "offline" and remove it from the device list, and trigger a multicast notification (DEVICE_OFFLINE). Status recovery: After the device comes back online, it sends a parameter request message (REGISTER_REQUEST) to the agent, and the agent updates the device list and synchronizes the latest summary information.

[0023] Example: Taking a 5-device network scenario (PCI=1~5, agent is PCI=1) as an example, the complete process is as follows: Initialization: Agent election (PCI=1 elected), device registration, agent sends the first PARAM_REQUEST, and collects initial frequency band parameters.

[0024] Normal operation: Device 2 detects interference at frequency 3.5GHz (intensity -75dBm), adjusts to 3.6GHz (bandwidth 15MHz), and triggers incremental update (MULTICAST_DELTA_INFO (incremental update message), V=1).

[0025] The agent aggregates the summary (band occupancy: 3.6GHz occupancy 20%) and multicasts it to all devices.

[0026] Heartbeat detection: Device 3 did not send a heartbeat due to a fault. The agent triggered QUERY_STATUS (status query message) after 316ms × 1.5 = 474ms, but did not respond. After 828ms, device 3 was marked as offline and multicast DEVICE_OFFLINE (device offline notification).

[0027] Status recovery: After device 3 is repaired, it sends a REGISTER_REQUEST (parameter request message), and the agent updates the list and synchronizes the latest summary (including the frequency band configuration after device 3 comes back online).

[0028] In one embodiment, the specific steps of the anti-interference module in performing emergency interference response and interference localization are as follows: Real-time spectrum scanning and FPGA pre-filtering (RF unit (RU) side): The edge computing unit on the RF unit (RU) side performs high-spectrum-efficiency interference detection through the FPGA, and only encapsulates data that meets the emergency interference conditions into an emergency interference alarm (URGENT_INTERFERENCE_ALERT) message (including frequency band range, interference intensity, and timestamp), and reports it to the anti-interference module of the baseband processing unit (BBU) through the internal bus to filter low-priority interference and reduce the load on the baseband processing unit (BBU); Data Fusion: Geographic Tag and Device Location Matching (Baseband Processing Unit (BBU) Side): After receiving an emergency interference event reported by the Radio Frequency Unit (RU), the anti-interference module performs the following operations: Geographic tag addition: Add geographic tags to interference events to record the physical location where the interference occurred; Device location association: Query the device location recorded in the protocol stack (via UE positioning or base station triangulation) and calculate the geographical distance between the interference location and each device; Interference correlation determination: If the distance between the device and the interference location is less than the threshold, the device is determined to be affected by interference; otherwise, it is marked as "far-end interference" to avoid false judgment. Pre-decision mechanism: Local reconfiguration triggering and execution: If an interference event is determined to be urgent and affects local equipment, the anti-interference module immediately initiates the pre-decision process without waiting for the global decision of the collaborative control module. Based on the interference frequency band (e.g., 3.5-3.55GHz) and the equipment service type (e.g., URLLC), it selects a carrier configuration to avoid interference from the pre-stored anti-interference frequency band configuration list (e.g., adjust to 3.6GHz, bandwidth 20MHz), generates a carrier reconfiguration instruction (CARRIER_RECONFIG_CMD) message (including new frequency, bandwidth, and SSB configuration), adjusts the radio frequency parameters of the affected equipment through the radio frequency unit (RU) resource management module, and disables cell broadcast (CELL_BROADCAST_DISABLE) to avoid user access interruption. Reconfiguration result verification: After reconfiguration is completed, cell broadcast is enabled (CELL_BROADCAST_ENABLE), and user access status is monitored through the user access query submodule; if interference still exists (e.g., residual interference is detected by spectrum scanning), the above process is repeated until the interference is eliminated; Conflict resolution between collaborative control and anti-interference: If the collaborative control module initiates frequency band adjustment requests simultaneously, the anti-interference module ensures priority through the following rules: If the anti-interference pre-decision has triggered local reconfiguration, the decision of the collaborative control module must wait for the anti-interference process to be completed before it is executed. The anti-interference module synchronizes the reconfigured frequency band information to the collaborative control module through the decision middleware to ensure the consistency of the global frequency band view.

[0029] In one embodiment, the specific steps for the decision middleware to fuse the spectrum data of the anti-interference module and the device parameters of the collaborative control module and align the timestamps for multi-source data synchronization are as follows: Multi-source data acquisition and timestamp marking: The decision middleware first receives raw data streams from two core modules: spectrum data from the anti-interference module and device parameters from the collaborative control module; Timestamp alignment and data synchronization: Calculate the time difference between the spectrum data timestamp and the device parameter timestamp. If the absolute value of the time difference is greater than the threshold, it is determined that there is a synchronization deviation in the data. If the spectrum data is delayed, the spectrum data is cached and waits for the device parameters to be updated to the same time point. If the device parameters are delayed, the state of the device parameters at the time of the spectrum data is predicted based on historical data interpolation (such as linear interpolation). The aligned data packets are uniformly marked as decision timestamps. Example: The spectrum data timestamp is 12:00:00.123456, and the device parameter timestamp is 12:00:00.120000. Δt = 3.456ms > 1ms. The middleware caches the spectrum data until device 1 reports new parameters (carrier frequency adjusted to 3.6GHz) at 12:00:00.123000. At this time, Δt = 0.456ms < 1ms, and alignment is complete.

[0030] Data association and structured fusion: The aligned data needs to be associated based on the device identifier (PCI) to generate a structured input packet. Device-Spectrum Mapping: Using the device's PCI as a key, interference frequency bands and intensities in the spectrum data are matched to specific devices. For example, if device 1 has a PCI of 1, and its 3.5-3.55 GHz frequency band is marked as interference in its spectrum data, then a mapping relationship is established: Device 1 → Interference frequency band: 3.5-3.55GHz, intensity: -75dBm Business type and priority integration: By combining the service type (such as URLLC) in the device parameters and the preset priority weights (URLLC=10, mMTC=5, eMBB=1), a device priority label is generated.

[0031] Generation of fusion decision input packages: The final generated structured data packet (FUSED_DECISION_INPUT) contains the following fields: Device PCI; Current carrier configuration (frequency, bandwidth); Service type and priority; Associated interference frequency bands and intensity; Decision timestamp; Input packet validation and exception handling: Integrity verification: Check whether the data packet contains required fields such as PCI, carrier configuration, and interference information. Missing fields are filled in by historical data or retransmission is triggered. Consistency check: Verify whether the current carrier configuration of the device overlaps with the spectrum interference band (e.g., if the carrier frequency of device 1 is 3.6GHz and the interference band is 3.5-3.55GHz, it passes if there is no overlap; if the carrier frequency is 3.5GHz and it overlaps with the interference band, it is marked as "conflict state"). Time validity verification: If the difference between the data packet timestamp and the current system time exceeds the threshold, the data is discarded and re-collection is triggered; Integrate data inputs into a two-level decision algorithm: The structured fusion decision input packet (FUSED_DECISION_INPUT) is input into the two-layer decision algorithm to generate the final frequency band decision result (including carrier configuration and SSB parameters of each device). Decision result feedback and status update: After the decision result is sent to the execution layer through the decision middleware, the new carrier configuration is synchronized to the device management information table of the collaborative control module, the frequency band occupancy status of each device is updated, and the correspondence between the decision timestamp, input data and decision result is saved for subsequent performance backtracking and algorithm optimization.

[0032] In one embodiment, the specific steps for generating frequency band decision results based on the two-level decision algorithm are as follows: Algorithm execution: Taking the fused decision input package as input, perform the following operations: First-level decision (business priority grouping): Group devices according to preset rules (URLLC > mMTC > eMBB), and arrange devices in descending order into group queues based on the "business type and priority" field to ensure that high-priority businesses (such as URLLC) are allocated resources first; The second-level decision-making (dynamic allocation within groups) selects an allocation strategy based on group characteristics (e.g., the URLLC group uses "minimum continuous frequency band priority" to ensure low latency, while other groups use "optimal bandwidth utilization" to improve capacity), and adjusts frequency band allocation based on interference information in the spectrum data (e.g., avoiding emergency interference frequency bands). The algorithm relies on structured data provided by the middleware (e.g., a list of devices with associated service tags and a spectrum diagram of marked interference areas) to avoid the overhead of raw data parsing and improve decision-making efficiency. If multiple devices need to be adjusted to the same frequency band, resources are allocated according to priority (URLLC>mMTC>eMBB) to generate the final frequency band decision result. Output: The frequency band decision results generated by the two-layer decision algorithm (including carrier frequency, bandwidth, and SSB configuration of each device) are encapsulated into frequency band allocation instructions (FREQUENCY_ALLOCATION_CMD) by the decision middleware. During this process, the decision middleware is responsible for the version management of the instructions (such as adding decision timestamps and algorithm version numbers) and conflict verification (such as detecting frequency band overlap and triggering reallocation). Finally, the instructions are sent to the radio frequency unit (RU) resource management module and protocol stack adaptation module of the execution layer. Feedback loop: After the execution layer completes carrier reconfiguration, it feeds back the reconfiguration results (such as carrier status and user access status) to the perception layer through the decision middleware. During this process, the decision middleware performs result aggregation (such as statistical analysis of user access rates for each service type) and anomaly detection (such as marking devices that failed to reconfigure), providing historical data reference for the next round of decision-making.

[0033] In one embodiment, the execution layer performs the following specific steps for responding to carrier reconfiguration and ensuring zero interruption of user services through a collaborative mechanism of hard tuning of radio frequency unit (RU) parameters, soft adaptation of protocol stack configuration, and elastic management of user access: Command parsing and parameter extraction: The Radio Unit (RU) resource management module receives the frequency band allocation command (FREQUENCY_ALLOCATION_CMD) issued by the decision middleware, parses the key fields in the command, and generates a local reconfiguration parameter table.

[0034] Radio frequency parameter adjustment and bandwidth allocation: According to the parameters in the instruction, the radio frequency unit (RU) adjusts the carrier center frequency to the target frequency (e.g., 3.6GHz) through the radio frequency front end (e.g., mixer, oscillator), and adjusts the filter bandwidth according to the target bandwidth (20MHz) to ensure that the signal does not overflow into the interference band (e.g., 3.5-3.55GHz). Dynamic bandwidth guarantee: mMTC Dedicated Frequency Band Pool: If the instruction is marked as "mMTC Dedicated", the RU will reserve 5MHz of bandwidth (e.g., 3.65-3.70GHz) and only allow mMTC devices to access it.

[0035] URLLC preemption mechanism: If a URLLC device needs to access the network and the eMBB bandwidth exceeds the minimum threshold (e.g., 10MHz), the RU can forcibly reclaim the eMBB bandwidth to the minimum threshold (e.g., reserve 10MHz) and allocate the remaining bandwidth (e.g., 10MHz) to the URLLC.

[0036] Example: eMBB original bandwidth is 20MHz, minimum threshold is 10MHz, URLLC requires 15MHz, then URLLC allocates 10MHz (min(20-10,15)), and eMBB reserves 10MHz.

[0037] Parameter activation and status feedback: After the adjustment is completed, a carrier reconfiguration confirmation message (CARRIER_RECONFIG_ACK) (including new frequency, bandwidth, and SSB configuration) is reported to the baseband processing unit (BBU) via the internal bus, marking the reconfiguration status as complete; SSB / PDCCH Dynamic Adjustment: After receiving the reconfiguration status feedback from the Radio Unit (RU), the carrier reconfiguration submodule adjusts the air interface parameters according to the SSB / PDCCH configuration in the instruction. SSB Configuration: Set the SSB's time and frequency position (e.g., send once every 5ms, subcarrier interval 30kHz) to ensure that the UE can correctly receive the synchronization signal.

[0038] PDCCH configuration: Adjust the aggregation level of the control channel (e.g., CCE aggregation level = 8) to improve the reliability of control signaling; Community broadcast management submodule: Broadcast enable / disable control: Disable broadcast before reconfiguration: Upon receiving the reconfiguration command, immediately send a cell broadcast disable message (CELL_BROADCAST_DISABLE) to the radio frequency unit (RU) to stop cell broadcasting and prevent new users from accessing the network; Enable broadcast after reconfiguration: After the radio frequency unit (RU) completes parameter adjustment, the protocol stack sends a cell broadcast enable message (CELL_BROADCAST_ENABLE) to restore cell broadcast and allow user access; User access query submodule: Reconfiguration window management: Reconfiguration when no user access: If the user access query submodule detects "no user access" (e.g., user count = 0 via air interface), the reconfiguration process is triggered immediately without waiting for the business window; Periodic monitoring when a user accesses the network: If a user accesses the network, a reconfiguration will be triggered at a preset period (e.g., every 10 minutes). Before the reconfiguration, the user equipment (UE) will be notified to suspend services (e.g., switch to another frequency band or enter a buffer state) through a user access query message (USER_ACCESS_QUERY). Example: The user access query submodule detects that 5 users are using the service. The protocol stack sends a USER_ACCESS_QUERY message to the UE, notifying it that "service will be suspended for frequency band adjustment in 10 seconds". After the UE responds, it enters a buffer state. After the RU completes the reconfiguration, the protocol stack restores the service and the UE reconnects.

[0039] Closed-loop verification: Reconfiguration result feedback: The protocol stack reports the new carrier configuration (frequency point, bandwidth, SSB / PDCCH parameters) to the decision middleware through the carrier reconfiguration confirmation message (CARRIER_RECONFIG_ACK) to update the global frequency band view; Interference rescan verification: The edge computing unit on the radio frequency unit (RU) side performs a spectrum scan on the new carrier frequency band to confirm whether the interference has been eliminated; User access recovery monitoring: The user access query submodule counts the number of users accessing the network after reconfiguration. If the access rate is lower than the threshold (e.g., 95%), the collaborative control module is triggered to readjust the frequency band allocation.

[0040] Example: Taking device 1 (URLLC service, original carrier 3.5GHz) as an example of interference from 3.5-3.55GHz, the complete execution layer process is as follows: Decision command issuance: The decision middleware generates FREQUENCY_ALLOCATION_CMD (frequency 3.6GHz, bandwidth 20MHz, SSB subcarrier spacing 30kHz, dynamic marking: URLLC preemption).

[0041] RU parameter adjustment: Adjust the RU frequency to 3.6GHz, bandwidth 20MHz, reserve 10MHz for eMBB (original bandwidth 20MHz, minimum threshold 10MHz), and allocate 10MHz for URLLC.

[0042] Protocol stack adaptation: The carrier reconfiguration submodule adjusts the SSB / PDCCH configuration (subcarrier spacing 30kHz, aggregation level 8).

[0043] The community broadcast management submodule disables broadcasting and enables it after reconfiguration.

[0044] The user access query submodule detects 3 users and notifies them to suspend service for 10 seconds, after which it will resume after reconfiguration.

[0045] Result verification: RU spectrum scanning confirmed no interference in the 3.6 GHz band.

[0046] User access rate has recovered to 100%, and business continuity has been successfully ensured.

[0047] A method for implementing an anti-interference multi-band communication system includes the following steps; Step 1: The perception layer parses the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack, dynamically marks the service type and priority for service perception, and at the same time, the edge computing unit deployed on the RU side uses FPGA to quickly extract spectrum features, supports sub-millisecond interference event triggering to improve the real-time performance of spectrum data acquisition, and reports the device parameters and spectrum data with service tags to the decision layer. Step 2: The decision middleware of the decision layer integrates the spectrum data of the anti-interference module and the equipment parameters of the collaborative control module, completes the synchronization of multi-source data through timestamp alignment, performs structured integration of the synchronized data to generate a fusion decision input package, generates frequency band decision results by combining the two-layer decision algorithm, and then encapsulates the results into frequency band allocation instructions; Step 3: The execution layer adjusts the radio frequency parameters based on the decision results through RU resource management, and dynamically adjusts the SSB / PDCCH configuration through the carrier reconfiguration submodule adapted to the protocol stack, controls the broadcast enable / disable through the cell broadcast management submodule, and provides a reconfiguration window through the user access query submodule to ensure user service continuity. At the same time, the execution layer feeds back the carrier status and user access status to the perception layer.

[0048] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An anti-interference multi-band communication system, characterized in that, include: Perception layer: The service type and priority are dynamically marked by parsing the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack. At the same time, the FPGA is used to quickly extract spectrum features through the edge computing unit deployed on the radio frequency unit side. The decision layer comprises a decision middleware, a collaborative control module, and an anti-interference module. The decision middleware fuses the spectrum data from the anti-interference module with the device parameters from the collaborative control module and aligns them with timestamps for multi-source data synchronization. The decision middleware then performs structured integration of the synchronized data to generate a fused decision input package. Based on a two-layer decision algorithm, it generates frequency band decision results and encapsulates these results into frequency band allocation instructions. The collaborative control module uses a hierarchical multicast architecture to elect regional agent nodes, aggregate information within the group, and synchronize summaries. The collaborative control module combines an incremental update protocol and an adaptive heartbeat mechanism. The anti-interference module uses a pre-decision mechanism, data fusion, and FPGA acceleration to perform emergency interference response and geographically correlated interference location to avoid misjudgment. Execution layer: Through radio frequency unit resource management, radio frequency parameters are adjusted according to the decision results and dynamic bandwidth guarantee is supported. Protocol stack adaptation dynamically adjusts SSB / PDCCH configuration through carrier reconfiguration submodule, controls broadcast enable / disable through cell broadcast management submodule, and provides reconfiguration window through user access query submodule.

2. The anti-interference multi-band communication system according to claim 1, characterized in that, The specific steps of the perception layer in performing three-dimensional perception of "service type-priority-spectrum status" through the collaboration of the service classification engine and spectrum feature extraction are as follows: Step 1: PDU Session Parsing and 5G QoS Identifier Extraction: The service classification engine in the protocol stack first parses the PDU session establishment request initiated by the user equipment and extracts the 5G QoS identifier parameters from it; Step 2: Mapping 5G QoS Identifiers to Service Types: Based on the predefined correspondence between 5G QoS identifiers and service types in the 3GPP standard, convert the 5G QoS identifier parameters into specific service types; Step 3: Dynamic Priority Tagging: Assign priorities based on business type; Step 4: Raw RF signal acquisition: The edge computing unit on the RU side receives wireless signals through the RF front end and converts them into digital baseband signals. ; Step 5: FFT Transformation and Spectrum Analysis: Utilize the parallel computing ability of the FPGA to perform a fast Fourier transform on to convert the time-domain signal into a frequency-domain signal , which is expressed as , where k represents the frequency-point index, 0 ≤ k < N, and N is the number of FFT points; Step 6: Signal Power and Interference Intensity Calculation: Calculate the signal power at each frequency point. And count the total power of the target frequency band. The formula for calculating frequency power is expressed as follows: The formula for calculating the total power of a frequency band is expressed as follows: ,in, Indicates the frequency band range; Step 7: Interference Threshold Detection: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] With preset interference threshold In comparison, if > If so, it is determined that there is interference in the frequency band, and an interference event is triggered; Step 8: Extraction of interference feature parameters: If an interference event is triggered, further extract the interference features, including: interference start / stop frequency, interference intensity, and interference duration.

3. The anti-interference multi-band communication system according to claim 1, characterized in that, The specific steps for the collaborative control module to perform collaborative control of the equipment are as follows: Regional agent node election and multicast architecture initialization: The collaborative control module first constructs a hierarchical multicast architecture and defines parameters, including device PCI coefficients and device computing capabilities. All devices broadcast their own PCI coefficients and computing capabilities to the multicast address. Each device calculates its own weight W according to the weighting formula and elects the device with the smallest weight as the regional agent node. The regional agent node establishes a multicast group, and devices in the group register with the agent. The agent maintains the device list. Intra-group information aggregation and summary synchronization: The regional agent node periodically performs the following operations: Frequency band parameter collection: The agent sends parameter request messages via multicast, and devices within the group respond with parameter reporting messages; Information aggregation: The agent statistically analyzes the frequency band occupancy within the group and generates a summary information table, which includes: frequency band occupancy rate, distribution of interfering frequency bands, and distribution of service types; Summary synchronization: The agent sends the summary information table to all devices via multicast, and the devices update their local frequency band status view; Incremental update protocol: The device sends a multicast update when the following conditions are met: Frequency band parameters change beyond a preset threshold, resulting in a change in interference level; Message format: Incremental update messages include: version number, changed field flag, and specific changed value; Adaptive Heartbeat Mechanism: Survival Status Detection and On-Demand Triggering: The collaborative control module employs a two-layer heartbeat detection mechanism. Basic heartbeat cycle: The device sends a heartbeat message every cycle T, and the cycle is dynamically adjusted according to the group size. Where K is the base period, α≈0.5 is a constant, and N is the number of devices in the group; Agent on-demand detection: If the agent does not receive a heartbeat from a device within 1.5T, it will actively send a status query message to check; if there is still no response within 2T, the agent will mark the device as "offline" and remove it from the device list, and trigger a multicast notification at the same time. Status recovery: After the device comes back online, it sends a parameter request message to the agent, which updates the device list and synchronizes the latest summary information.

4. The anti-interference multi-band communication system according to claim 2, characterized in that, The specific steps for the anti-interference module to perform emergency interference response and interference localization are as follows: Real-time spectrum scanning and FPGA pre-filtering: The edge computing unit on the RF unit side performs high-spectrum-efficiency interference detection through FPGA, and only encapsulates the data that meets the emergency interference conditions into an emergency interference alarm message, which is then reported to the anti-interference module of the baseband processing unit through the internal bus. Data Fusion: Geographic Tag and Device Location Matching: After receiving an emergency interference event reported by the radio frequency unit, the anti-interference module performs the following operations: Geographic tag addition: Add geographic tags to interference events to record the physical location where the interference occurred; Device location association: Query the device locations recorded in the protocol stack and calculate the geographical distance between the interference location and each device; Interference correlation determination: If the distance between the device and the interference location is less than the threshold, the device is determined to be affected by interference; otherwise, it is marked as "far-end interference". Pre-decision mechanism: Local reconfiguration triggering and execution: If an interference event is determined to be urgent and affects local equipment, the anti-interference module immediately initiates the pre-decision process without waiting for the global decision of the collaborative control module. Based on the interference frequency band and the equipment service type, it selects a carrier configuration to avoid interference from the pre-stored anti-interference frequency band configuration list, generates a carrier reconfiguration instruction message, and adjusts the radio frequency parameters of the affected equipment through the radio frequency unit resource management module. Reconfiguration result verification: After reconfiguration is completed, cell broadcast is enabled, and user access status is monitored through the user access query submodule; if interference still exists, the above process is repeated until the interference is resolved. Conflict resolution between collaborative control and anti-interference: If the collaborative control module initiates frequency band adjustment requests simultaneously, the anti-interference module ensures priority through the following rules: If the anti-interference pre-decision has triggered local reconfiguration, the decision of the collaborative control module must wait for the anti-interference process to be completed before it is executed. The anti-interference module synchronizes the reconfigured frequency band information to the collaborative control module through the decision middleware.

5. The anti-interference multi-band communication system according to claim 1, characterized in that, The specific steps for the decision middleware to synchronize multi-source data by fusing the spectrum data from the anti-interference module and the device parameters from the collaborative control module and aligning them with timestamps are as follows: Multi-source data acquisition and timestamp marking: The decision middleware first receives raw data streams from two core modules: spectrum data from the anti-interference module and device parameters from the collaborative control module; Timestamp alignment and data synchronization: Calculate the time difference between the spectrum data timestamp and the device parameter timestamp. If the absolute value of the time difference is greater than the threshold, it is determined that there is a synchronization deviation in the data. If the spectrum data is delayed, the spectrum data is cached and waits for the device parameters to be updated to the same time point. If the device parameters are delayed, the state of the device parameters at the time of the spectrum data is predicted by interpolation based on historical data. The aligned data packets are uniformly marked as decision timestamps. Data association and structured fusion: The aligned data needs to be associated based on the device identifier to generate a structured input package. Input packet validation and exception handling: Integrity verification: Checks whether the data packet contains required fields; missing fields are filled in using historical data or retransmission is triggered. Consistency check: Verify whether the device's current carrier configuration overlaps with spectrum interference bands; Time validity verification: If the difference between the data packet timestamp and the current system time exceeds the threshold, the data is discarded and re-collection is triggered; Integrate data inputs into a two-level decision algorithm: The structured fusion decision input packet is fed into a two-layer decision algorithm to generate the final frequency band decision result; Decision result feedback and status update: After the decision result is sent to the execution layer through the decision middleware, the new carrier configuration is synchronized to the device management information table of the collaborative control module, the frequency band occupancy status of each device is updated, and the correspondence between the decision timestamp, input data, and decision result is saved.

6. The anti-interference multi-band communication system according to claim 5, characterized in that, The specific steps for generating frequency band decision results based on the two-level decision algorithm are as follows: Algorithm execution: Taking the fused decision input package as input, perform the following operations: First-level decision: Group devices according to preset rules, and arrange the devices into group queues in descending order based on the "business type and priority" field; Second-level decision-making: Select an allocation strategy based on group characteristics and adjust frequency band allocation in conjunction with interference information in spectrum data. If multiple devices need to be adjusted to the same frequency band, allocate resources according to priority and generate the final frequency band decision result. Output: The frequency band decision results generated by the two-layer decision algorithm are encapsulated into frequency band allocation instructions by the decision middleware. The decision middleware is responsible for the version management and conflict verification of the instructions in this process, and finally sends the instructions to the radio frequency unit resource management module and protocol stack adaptation module of the execution layer. Feedback loop: After the execution layer completes carrier reconfiguration, it feeds back the reconfiguration result to the perception layer through the decision middleware. During this process, the decision middleware performs result aggregation and anomaly detection.

7. The anti-interference multi-band communication system according to claim 1, characterized in that, The execution layer employs a collaborative mechanism involving hard tuning of radio frequency unit parameters, soft adaptation of protocol stack configuration, and flexible management of user access to implement specific steps for carrier reconfiguration response and zero-interruption of user services: Command parsing and parameter extraction: The RF unit resource management module receives frequency band allocation commands from the decision middleware, parses the key fields in the commands, and generates a local reconfiguration parameter table. RF parameter adjustment and bandwidth allocation: According to the parameters in the instruction, the RF unit adjusts the carrier center frequency to the target frequency through the RF front end, and adjusts the filter bandwidth according to the target bandwidth; Parameter activation and status feedback: After the adjustment is completed, a carrier reconfiguration confirmation message is reported to the baseband processing unit via the internal bus, and the reconfiguration status is marked as complete; SSB / PDCCH Dynamic Adjustment: After the protocol stack adaptation module receives the reconfiguration status feedback from the radio frequency unit, the carrier reconfiguration submodule adjusts the air interface parameters of the protocol stack according to the SSB / PDCCH configuration in the instruction. Community Broadcast Management Submodule: Broadcast Enable / Disable Control: Disable broadcast before reconfiguration: Upon receiving the reconfiguration command, immediately send a cell broadcast disable message to the radio frequency unit to stop cell broadcasting and prevent new users from accessing the network; Enable broadcast after reconfiguration: After the radio frequency unit completes parameter adjustment, the protocol stack sends a cell broadcast enable message to restore cell broadcast and allow user access; User access query submodule: Reconfiguration window management: Reconfiguration when no user access: If the user access query submodule detects "no user access", the reconfiguration process will be triggered immediately; Periodic monitoring when a user accesses the device: If a user accesses the device, a reconfiguration will be triggered according to a preset period, and the user device will be notified to suspend services before the reconfiguration is performed via a user access query message. Closed-loop verification: Reconfiguration result feedback: The protocol stack reports the new carrier configuration to the decision middleware through the carrier reconfiguration confirmation message, and updates the global frequency band view; Interference rescan verification: The edge computing unit on the radio frequency unit side performs a spectrum scan on the new carrier frequency band to confirm whether the interference has been eliminated; User access recovery monitoring: The user access query submodule counts the number of users accessing the network after reconfiguration. If the access rate is lower than the threshold, the collaborative control module is triggered to readjust the frequency band allocation.

8. A method for implementing an anti-interference multi-band communication system, using the anti-interference multi-band communication system as described in claim 1, characterized in that, Includes the following steps; Step 1: The perception layer parses the QoS parameters of the user PDU session through the service classification engine embedded in the protocol stack, dynamically marks the service type and priority for service perception, and at the same time, the edge computing unit deployed on the RU side uses FPGA to quickly extract spectrum features and reports the device parameters and spectrum data with service tags to the decision layer. Step 2: The decision middleware of the decision layer integrates the spectrum data of the anti-interference module and the equipment parameters of the collaborative control module, completes the synchronization of multi-source data through timestamp alignment, performs structured integration of the synchronized data to generate a fusion decision input package, generates frequency band decision results by combining the two-layer decision algorithm, and then encapsulates the results into frequency band allocation instructions; Step 3: The execution layer adjusts the radio frequency parameters based on the decision results through RU resource management, and dynamically adjusts the SSB / PDCCH configuration through the carrier reconfiguration submodule adapted to the protocol stack, controls the broadcast enable / disable through the cell broadcast management submodule, and provides a reconfiguration window through the user access query submodule to ensure user service continuity. At the same time, the execution layer feeds back the carrier status and user access status to the perception layer.

Images