A cloud collaborative networked unmanned aerial vehicle countermeasure system and method

Abstract

This invention discloses a cloud-based collaborative networked UAV countermeasure system and method, specifically relating to the field of UAV technology. The system includes a radio frequency data acquisition unit, a data modeling and processing unit, a sequence deduction and prediction unit, a scheduling command generation unit, and an edge execution collaboration unit. By acquiring the radio frequency signals of the target UAV's communication link, the system extracts carrier frequency, frequency hopping sequence, pulse arrival time information, and bandwidth parameters. Graph structure mapping processing is performed in the cloud, mapping discrete frequency values ​​to graph nodes and frequency hopping relationships to weighted directed edges. The calculation results construct a frequency hopping topology graph, predicting the frequency dwell sequence and corresponding time parameters within future time windows. Structured control command data is generated and sent to edge devices for synchronous scheduling. This invention improves the identification and prediction accuracy of complex frequency hopping communication links, enhancing cloud-edge collaborative scheduling efficiency, execution accuracy, response speed, and system stability.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicles (UAVs), and in particular to a cloud-based collaborative networked UAV countermeasure system and method. Background Technology

[0002] With the rapid development of drone technology, unmanned aerial vehicles (UAVs) are increasingly widely used in civilian and commercial fields such as aerial photography, inspection, and logistics. A typical UAV system consists of an aircraft platform, flight control system, navigation and positioning system, power transmission system, and mission payload system. It primarily relies on radio data links for ground remote control or autonomous flight according to preset programs. However, the surge in the number of UAVs and the lowering of application barriers have also brought serious challenges to low-altitude airspace security. Unauthorized intrusion incidents are frequent, posing a significant threat to the protection of important facilities, security for large-scale events, and public safety. Therefore, detection and countermeasures technologies for low-altitude UAVs have gradually become a research hotspot in related fields, and building an effective UAV management system has become an important guarantee for maintaining low-altitude security.

[0003] Currently, existing drone defense networks typically employ a cloud-based, networked countermeasure architecture. These systems rely on cloud servers and distributed edge detection devices to operate, primarily covering radio frequency band detection, identity verification, and radio frequency interference handling for low-altitude intruding drones. In terms of specific operation, the system uses radio frequency receiving antennas deployed at different geographical nodes to receive radio frequency signals in real time from common drone communication frequency bands such as 2.4GHz or 5.8GHz, extracting time-frequency waterfall plot features and frequency hopping sequences. Subsequently, the feature data is transmitted to the cloud server via the network, where classifiers such as convolutional neural networks are used for model comparison and attribute identification. After target confirmation, the system uses a multi-base station time-of-arrival (TOA) positioning method to calculate the target drone's three-dimensional spatial coordinates and further issues scheduling commands via the network link to control ground-based directional transmitting antennas to continuously radiate broadband noise interference electromagnetic waves or send spoofed satellite navigation messages towards the target location, thereby taking over or forcing the target drone to land.

[0004] In existing technologies, countermeasures against UAV communication links mostly employ local single-point monitoring, fixed-frequency scanning, or passive response methods based on real-time detection results. These methods lack the ability to continuously model and predict the frequency hopping patterns of the target communication link, making it difficult to adapt to complex and dynamically changing frequency hopping communication scenarios. At the same time, existing solutions typically focus on local processing of front-end devices, lacking a unified data organization method and high-precision collaborative scheduling mechanism between cloud analysis and edge execution. This results in lag in the identification of the target link, insufficient prediction accuracy, poor consistency in multi-device collaborative control, and susceptibility to network jitter, clock deviation, and rapid switching of target frequencies. Consequently, it is difficult to balance reaction speed, execution accuracy, and overall system stability. Summary of the Invention

[0005] The main objective of this invention is to provide a cloud-based collaborative networked UAV countermeasure system and method to address the problems in the prior art, such as insufficient identification of frequency hopping behavior in the target UAV communication link, inadequate sequence prediction capabilities, difficulty in directly supporting collaborative execution of edge devices with cloud analysis results, and low accuracy of networked scheduling.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A cloud-based, networked drone countermeasure system includes: The radio frequency data acquisition unit is used to acquire radio frequency signals from the communication link of the target UAV and extract discrete feature data from the radio frequency signals, including at least carrier frequency, frequency jump sequence, pulse arrival time information and bandwidth parameters. The data modeling and processing unit, located on a cloud server, is used to perform graph structure mapping processing on the discrete feature data, mapping discrete frequency values ​​to graph nodes, mapping frequency hopping relationships to weighted directed edges, and constructing a frequency hopping topology graph based on the state transition probability calculation results. The sequence deduction and prediction unit is used to predict the frequency dwell sequence and corresponding time parameters of the target UAV communication link in the future time window based on the frequency hopping topology map, through probability weight analysis and path deduction algorithm. The scheduling instruction generation unit is used to convert the frequency dwell sequence and the corresponding time parameters into structured control instruction data, wherein the structured control instruction data includes at least frequency parameters, time parameters and execution priority information; The edge execution coordination unit is used to receive the structured control command data and perform synchronous scheduling according to the time parameters to generate and output the corresponding edge execution control signal.

[0007] Preferably, the data modeling and processing unit adopts a probability calculation method based on the Markov state transition model to quantify the transition probability between adjacent frequency nodes. Specifically, this includes: counting the state transition frequency according to the continuously observed frequency jump sequence; normalizing each state transition frequency to obtain a transition probability matrix; introducing a time decay factor to weight and update the historical data to obtain an updated transition probability matrix; and mapping the updated transition probability matrix to the weight values ​​of the weighted directed edges.

[0008] Preferably, the frequency hopping topology graph is a multidimensional graph structure containing a time dimension, wherein each graph node includes at least a frequency value attribute and a timestamp attribute, each directed edge includes at least a transition probability attribute and a time interval attribute, the arrival time interval of adjacent pulses determined by the pulse arrival time information is encoded as a time difference parameter between nodes and embedded in the multidimensional graph structure to form a temporal constraint, and the multidimensional graph structure is stored and retrieved using a gridded coordinate method or a time series indexing method.

[0009] Preferably, the sequence deduction and prediction unit includes: a loop identification module, used to identify closed paths in the frequency hopping topology map through a graph traversal algorithm and calculate the cycle period characteristics of each closed path; a parameter back-inference module, used to estimate the pseudo-random sequence generation parameters based on the node sequence of the closed path using a polynomial fitting model or a linear feedback shift register model; and a path expansion module, used to recursively calculate the frequency hopping sequence for multiple future cycles based on the pseudo-random sequence generation parameters, so as to establish a structured model of the frequency hopping sequence generation mechanism of the target UAV communication link.

[0010] Preferably, the sequence inference and prediction unit adopts an optimal path selection strategy based on path weight accumulation, specifically: accumulating or averaging the directed edge weights of candidate paths; introducing a confidence evaluation index to quantify the stability of candidate paths; sorting and selecting among multiple candidate paths, outputting the most confident frequency-resident sequence; and dynamically updating and rolling correcting the prediction results.

[0011] Preferably, the scheduling instruction generation unit encapsulates the structured control instruction data using a unified data format. The frequency field is used to characterize the center frequency and frequency band range information of the target frequency point corresponding to the frequency dwell sequence, the time field is used to characterize the absolute timestamp and relative time offset, the execution priority field is used to characterize the scheduling order of different control instructions, and the structured control instruction data supports batch generation, queued management, and network transmission and parsing through standardized interface protocols.

[0012] Preferably, the edge execution coordination unit includes: a time synchronization module for achieving system time unification based on a network time protocol or a hardware clock signal; a scheduling cache module for preloading and caching the received structured control instruction data; and a trigger control module for triggering the execution of the corresponding control instruction when the time parameter in the structured control instruction data is reached, so as to achieve microsecond-level time precision scheduling control and continuous instruction switching.

[0013] Preferably, the system adopts a cloud-edge collaborative architecture, wherein the cloud server is used to perform frequency hopping topology map construction, probability calculation and sequence prediction tasks, and the edge device carrying the edge execution collaboration unit is used to perform control command parsing, caching and real-time scheduling control tasks based on time parameters. The cloud and the edge interact with each other through network links and support task distribution and result feedback.

[0014] A cloud-based, networked drone countermeasure method includes the following steps: The radio frequency signal of the communication link of the target UAV is collected, and discrete feature data is extracted from the radio frequency signal. The discrete feature data includes at least carrier frequency, frequency jump sequence, pulse arrival time information and bandwidth parameters. The discrete feature data is uploaded to a cloud server, and graph structure mapping processing is performed on the discrete feature data. The discrete frequency values ​​are mapped to graph nodes, and the frequency hopping relationship is mapped to weighted directed edges. A frequency hopping topology graph is constructed based on the state transition probability calculation results. Based on the frequency hopping topology map, the frequency dwell sequence and corresponding time parameters of the target UAV communication link in the future time window are predicted through probability weight analysis and path deduction. Structured control instruction data is generated based on the frequency dwell sequence and the corresponding time parameters. The structured control instruction data includes at least frequency parameters, time parameters, and execution priority information. The structured control command data is sent to the edge execution device; The edge execution device synchronizes and schedules the structured control command data according to the time parameters, generates and outputs corresponding edge execution control signals, so as to realize networked collaborative countermeasures against the target UAV's communication link.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention, through the coordinated efforts of radio frequency data acquisition, cloud-based graph structure modeling, state transition probability calculation, and sequence extrapolation prediction, can extract the frequency hopping patterns of the target UAV communication link from continuous observation results and make advance judgments on the frequency dwell sequence and time parameters within future time windows. This improves the identification capability and prediction accuracy of complex frequency hopping communication links, reduces response lag, misjudgment, and resource waste, enhances the system's adaptability and continuous tracking capability to dynamically changing targets, and can be rolled over by combining historical observation results to improve the overall stability, practicality, and reliability of the model.

[0016] 2. This invention further converts the predicted frequency dwell sequence and corresponding time parameters into structured control command data, which is then preloaded, queued, and triggered for execution by the edge execution coordination unit according to a unified time base. This enables cloud analysis results to directly support real-time scheduling and control of edge devices. Compared to methods that rely solely on local real-time decision-making, this invention improves the data connection efficiency between the cloud and the edge, the collaborative control capabilities under networked deployment conditions, and the consistency of continuous switching between multiple devices. It also reduces the impact of network jitter and clock skew on execution performance, thereby enhancing the overall system performance, including scheduling accuracy, response efficiency, scalability, and engineering application value. Attached Figure Description

[0017] Figure 1 This is a structural block diagram of a cloud-based collaborative networked drone countermeasure system according to one embodiment of the present invention. Detailed Implementation

[0018] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the linguistic context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0019] It should be understood that the terms "system," "device," "unit," and / or "module" as used in this specification are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0020] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0021] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0022] The cloud-based collaborative networked drone countermeasure system and method provided in the embodiments of this specification will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 This is an exemplary flowchart of a cloud-collaborative networked drone countermeasure system and method according to some embodiments of this specification. In some embodiments, the cloud-collaborative networked drone countermeasure system and method can be executed by processing logic, which may include hardware (e.g., circuits, dedicated logic, programmable logic, microcode, etc.), software (running on a processing device to execute hardware-simulated instructions), and any combination thereof. In some embodiments, Figure 1 One or more operations in the flowchart of the cloud-collaborative networked drone countermeasure system and method shown can be implemented by a processing device and / or a terminal device. For example, the cloud-collaborative networked drone countermeasure system and method can be stored in a storage device in the form of computer programs and / or instructions, and invoked and / or executed by the processing device and / or the terminal device.

[0024] This embodiment provides a cloud-collaborative networked UAV countermeasure system. The system adopts a cloud-edge collaborative architecture, including a radio frequency data acquisition unit, a data modeling and processing unit, a sequence deduction and prediction unit, a scheduling instruction generation unit, and an edge execution collaboration unit. The cloud server performs tasks such as frequency hopping topology map construction, probability calculation, and sequence prediction. Edge devices perform control instruction parsing, caching, and real-time scheduling control tasks based on time parameters. Data interaction between the cloud and edge is achieved through a network link, supporting task distribution and result feedback.

[0025] The radio frequency (RF) data acquisition unit is used to acquire RF signals from the target UAV's communication link and extract discrete feature data from the RF signals. The discrete feature data includes at least the carrier frequency, frequency transition sequence, pulse arrival time information, and bandwidth parameters. Specifically, the RF data acquisition unit can continuously monitor the target UAV's communication link, recording the frequency points of the target communication signal, the switching relationships between adjacent frequency points, the arrival time information of each pulse, and the corresponding bandwidth information in chronological order during continuous observation, thereby forming a set of discrete feature data for subsequent processing.

[0026] The data modeling and processing unit, located on a cloud server, performs graph structure mapping processing on the discrete feature data. This graph structure mapping processing includes: mapping discrete frequency values ​​to graph nodes, mapping frequency hopping relationships to weighted directed edges, and constructing a frequency hopping topology graph based on the state transition probability calculation results. Through this method, the original discrete feature data can be converted into graph structure data suitable for unified management and continuous updating on the cloud server, facilitating subsequent sequence inference and prediction.

[0027] In this embodiment, the data modeling and processing unit employs a probability calculation method based on a Markov state transition model to quantify the transition probabilities between adjacent frequency nodes. Specifically, this includes: statistically analyzing the state transition frequencies based on continuously observed frequency hopping sequences; normalizing each state transition frequency to obtain a transition probability matrix; introducing a time decay factor to weight and update historical data to obtain an updated transition probability matrix; and mapping the updated transition probability matrix to the weight values ​​of the weighted directed edges. This processing method allows recent observation results to be more fully reflected in the frequency hopping topology map, while retaining the reference value of historical data for identifying overall sequence patterns.

[0028] In this embodiment, the frequency-hopping topology graph is a multidimensional graph structure containing a time dimension. Each graph node includes at least a frequency value attribute and a timestamp attribute, and each directed edge includes at least a transition probability attribute and a time interval attribute. The arrival time interval between adjacent pulses, determined by the pulse arrival time information, is encoded as a time difference parameter between nodes and embedded in the multidimensional graph structure to form a temporal constraint. The multidimensional graph structure can be stored and retrieved using a gridded coordinate method or a time series indexing method, thereby facilitating the cloud server to query, update, and compare graph nodes and directed edges within different time windows.

[0029] The sequence extrapolation and prediction unit is used to predict the frequency dwell sequence and corresponding time parameters of the target UAV communication link within a future time window based on the frequency hopping topology map, through probability weight analysis and path extrapolation algorithms. The time parameters may include one or more of the following: start time, end time, duration, absolute timestamp, and relative time offset for each dwell frequency point. Preferably, the sequence extrapolation and prediction unit is deployed on the same cloud server as the data modeling and processing unit to reduce map access latency and facilitate rolling model updates.

[0030] In this embodiment, the sequence deduction and prediction unit includes a loop identification module, a parameter inversion module, and a path expansion module. The loop identification module is used to identify closed paths in the frequency hopping topology map using a graph traversal algorithm and calculate the cyclic periodic characteristics of each closed path. By identifying the closed paths, the periodic repetition part can be extracted from the frequency hopping relationship obtained from continuous observation, providing a basis for subsequent parameter inversion and path expansion.

[0031] The parameter back-calculation module is used to estimate the pseudo-random sequence generation parameters based on the node sequence of the closed path, using a multinomial fitting model or a linear feedback shift register model. Specifically, when the node sequence of the closed path has a clear periodic pattern, its generation mechanism can be fitted based on the temporal variation of the node sequence; when the closed path exhibits pseudo-random state transition characteristics, the pseudo-random sequence generation parameters can be estimated using a linear feedback shift register model. Thus, key parameters describing the frequency hopping sequence generation mechanism of the target UAV communication link can be obtained.

[0032] The path expansion module is used to recursively calculate the frequency hopping sequence for multiple future periods based on the pseudo-random sequence generation parameters, in order to establish a structured model of the frequency hopping sequence generation mechanism of the target UAV communication link. This structured model can be used for predicting the frequency dwell sequence within the current time window, or it can be used as a basis for rolling corrections and updates after new observation data arrives.

[0033] In this embodiment, the sequence deduction and prediction unit adopts an optimal path selection strategy based on path weight accumulation. Specifically, it performs cumulative calculation or weighted average calculation of the directed edge weights of candidate paths; introduces a confidence evaluation index to quantify the stability of candidate paths; sorts and selects among multiple candidate paths, outputting the frequency-resident sequence with the highest confidence; and dynamically updates and rolls the prediction results. Through this processing method, prediction results more suitable for actual scheduling can be selected from multiple possible frequency hopping paths.

[0034] The scheduling instruction generation unit converts the frequency dwell sequence and corresponding time parameters into structured control instruction data. The structured control instruction data includes at least frequency parameters, time parameters, and execution priority information. Preferably, the scheduling instruction generation unit encapsulates the structured control instruction data using a unified data format, wherein the frequency field characterizes the target frequency point's center frequency and frequency band range information, the time field characterizes the absolute timestamp and relative time offset, and the execution priority field characterizes the scheduling order of different control instructions. Furthermore, the structured control instruction data supports batch generation, queued management, and network transmission and parsing via standardized interface protocols.

[0035] The edge execution coordination unit receives the structured control instruction data and performs synchronous scheduling according to the time parameters to generate and output corresponding edge execution control signals. In this embodiment, the edge execution coordination unit includes a time synchronization module, a scheduling cache module, and a trigger control module. The time synchronization module achieves system time unification based on a network time protocol or hardware clock signal; the scheduling cache module preloads and queues the received structured control instruction data; and the trigger control module triggers the execution of the corresponding control instruction when the time parameters are reached, achieving microsecond-level time precision scheduling control and continuous instruction switching. Because the edge device completes local caching and clock unification before execution, even if there is network jitter between the cloud and the edge, real-time scheduling control can still be performed according to the predetermined time parameters.

[0036] In another embodiment, a cloud-based collaborative networked UAV countermeasure method is provided. The method, executed by the aforementioned system, includes the following steps: acquiring radio frequency (RF) signals from the target UAV's communication link, and extracting discrete feature data from the RF signals, including at least carrier frequency, frequency hopping sequence, pulse arrival time information, and bandwidth parameters; uploading the discrete feature data to a cloud server, and having the data modeling processing unit perform graph structure mapping processing on the discrete feature data, mapping discrete frequency values ​​to graph nodes, mapping frequency hopping relationships to weighted directed edges, and constructing a frequency hopping topology graph based on state transition probability calculation results; having the sequence deduction and prediction unit predict the frequency dwell sequence and corresponding time parameters of the target UAV's communication link within a future time window based on the frequency hopping topology graph, using probability weight analysis and path deduction algorithms; having the scheduling instruction generation unit convert the frequency dwell sequence and corresponding time parameters into structured control instruction data; sending the structured control instruction data to an edge execution coordination unit; and having the edge execution coordination unit perform synchronous scheduling based on the time parameters to generate and output corresponding edge execution control signals.

[0037] To facilitate understanding of the technical solution of the present invention, a complete embodiment will be described below.

[0038] In this embodiment, the system includes a data modeling and processing unit located on a cloud server, a sequence inference and prediction unit, and a scheduling instruction generation unit. It also includes a radio frequency (RF) data acquisition unit deployed at the front end and an edge execution coordination unit deployed in an edge device. The RF data acquisition unit continuously monitors the communication link of the target UAV and acquires a set of continuously changing discrete feature data over multiple consecutive sampling periods. The discrete feature data includes multiple different carrier frequencies, a frequency jump sequence formed in chronological order, pulse arrival time information characterizing the occurrence time of each pulse, and corresponding bandwidth parameters.

[0039] In this embodiment, the radio frequency data acquisition unit uploads the aforementioned discrete feature data to the cloud server in real time. The data modeling and processing unit preprocesses the discrete feature data, mapping different discrete frequency values ​​to multiple graph nodes, mapping frequency jump relationships in adjacent sampling periods to weighted directed edges, and statistically analyzing state transition frequencies based on continuously observed frequency jump sequences. Subsequently, the state transition frequencies are normalized to obtain a transition probability matrix; a time decay factor is then introduced to weight and update the historical data, resulting in an updated transition probability matrix, which is then mapped to the weight values ​​of each weighted directed edge. Simultaneously, the data modeling and processing unit encodes the adjacent pulse arrival time intervals determined by the pulse arrival time information as time difference parameters between nodes and embeds them into the multidimensional graph structure to form temporal constraints, thereby obtaining a frequency hopping topology graph containing a time dimension. Each graph node in this frequency hopping topology graph includes a frequency value attribute and a timestamp attribute, and each directed edge includes a transition probability attribute and a time interval attribute, all stored and retrieved using a time-series indexing method.

[0040] In this embodiment, after reading the frequency hopping topology map, the sequence inference and prediction unit first traverses the directed edge connections in the map through the loop identification module to identify at least one closed path and calculate the cycle period characteristics of the closed path. Then, the parameter back-inference module estimates the pseudo-random sequence generation parameters based on the node sequence of the closed path using a multinomial fitting model or a linear feedback shift register model. The path expansion module recursively calculates the frequency hopping sequences for multiple future cycles based on the pseudo-random sequence generation parameters to establish a structured model of the frequency hopping sequence generation mechanism of the target UAV communication link. Further, the sequence inference and prediction unit adopts an optimal path selection strategy based on path weight accumulation, accumulating or weighted averaging the directed edge weights of multiple candidate paths, and introducing a confidence evaluation index to quantify the stability of the candidate paths. It then ranks and selects among the multiple candidate paths, outputting the frequency dwell sequence with the highest confidence, and simultaneously providing the corresponding time parameters for the frequency dwell sequence. As new discrete feature data continues to arrive, the sequence inference and prediction unit also dynamically updates and rolls over the prediction results.

[0041] In this embodiment, the scheduling instruction generation unit generates structured control instruction data based on the frequency dwell sequence and corresponding time parameters. The scheduling instruction generation unit encapsulates the structured control instruction data using a unified data format. The frequency field characterizes the target frequency point's center frequency and frequency band range information, the time field characterizes the absolute timestamp and relative time offset, and the execution priority field characterizes the scheduling order of different control instructions. Multiple structured control instruction data are arranged into an instruction queue according to their execution order, and batch generation, queued management, and network transmission and parsing via standardized interface protocols are supported. The structured control instruction data includes at least frequency parameters, time parameters, and execution priority information, and may also include auxiliary information such as device identifiers, verification identifiers, and status return identifiers as needed.

[0042] In this embodiment, after receiving the structured control command data, the edge execution coordination unit uses a time synchronization module to unify the system time based on a network time protocol or a hardware clock signal. Then, a scheduling cache module preloads and queues the received structured control command data. When the corresponding time parameter is reached, the trigger control module executes the corresponding control command to generate and output the corresponding edge execution control signal. Since each structured control command data contains a clear absolute timestamp or relative time offset, the edge execution coordination unit can perform synchronous scheduling based on the time parameters to achieve continuous command switching. In a preferred embodiment, scheduling control with microsecond-level time accuracy can be achieved through local caching and unified clock control.

[0043] In the complete operation of this embodiment, the cloud server performs tasks such as frequency hopping topology map construction, probability calculation, and sequence prediction, while the edge device performs tasks such as control command parsing, caching, and real-time scheduling control based on time parameters. The cloud and edge interact with each other via a network link, supporting task distribution and result feedback. This completes the entire processing flow from acquiring radio frequency signals from the target UAV's communication link, extracting discrete feature data, processing graph structure mapping, calculating state transition probabilities, constructing the frequency hopping topology map, predicting frequency dwell sequences and corresponding time parameters within future time windows, generating structured control command data, to synchronously scheduling and outputting edge execution control signals by the edge execution coordination unit.

[0044] Those skilled in the art will understand that, without departing from the basic concept of this invention, the deployment form of the radio frequency data acquisition unit, the specific update strategy of the data modeling and processing unit, the candidate path selection method of the sequence deduction and prediction unit, the encapsulation format of the scheduling instruction generation unit, and the synchronization method of the edge execution coordination unit in the above embodiments can all be adjusted according to the actual application scenario. As long as the technical solution still uses graph structure mapping processing of the discrete feature data of the target UAV communication link, constructs a frequency hopping topology map based on the state transition probability calculation results, predicts the frequency dwell sequence and corresponding time parameters within the future time window through probability weight analysis and path deduction algorithms, and then converts it into structured control instruction data and performs synchronous scheduling by the edge execution coordination unit according to the time parameters, it should all be considered part of the disclosure of this invention.

[0045] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A cloud-based collaborative networked drone countermeasure system, characterized in that, The system includes: The radio frequency data acquisition unit is used to acquire radio frequency signals from the communication link of the target UAV and extract discrete feature data from the radio frequency signals, including at least carrier frequency, frequency jump sequence, pulse arrival time information and bandwidth parameters. The data modeling and processing unit, located on a cloud server, is used to perform graph structure mapping processing on the discrete feature data, mapping discrete frequency values ​​to graph nodes, mapping frequency hopping relationships to weighted directed edges, and constructing a frequency hopping topology graph based on the state transition probability calculation results. The sequence deduction and prediction unit is used to predict the frequency dwell sequence and corresponding time parameters of the target UAV communication link in the future time window based on the frequency hopping topology map, through probability weight analysis and path deduction algorithm. The scheduling instruction generation unit is used to convert the frequency dwell sequence and the corresponding time parameters into structured control instruction data, wherein the structured control instruction data includes at least frequency parameters, time parameters and execution priority information; The edge execution coordination unit is used to receive the structured control command data and perform synchronous scheduling according to the time parameters to generate and output the corresponding edge execution control signal.

2. The system according to claim 1, characterized in that, The data modeling and processing unit adopts a probability calculation method based on the Markov state transition model to quantify the transition probability between adjacent frequency nodes. Specifically, it includes: counting the state transition frequency according to the continuously observed frequency jump sequence; normalizing each state transition frequency to obtain a transition probability matrix; introducing a time decay factor to weight and update the historical data to obtain an updated transition probability matrix; and mapping the updated transition probability matrix to the weight values ​​of the weighted directed edges.

3. The system according to claim 1, characterized in that, The frequency hopping topology graph is a multidimensional graph structure containing a time dimension. Each graph node includes at least a frequency value attribute and a timestamp attribute, and each directed edge includes at least a transition probability attribute and a time interval attribute. The arrival time interval of adjacent pulses determined by the pulse arrival time information is encoded as a time difference parameter between nodes and embedded in the multidimensional graph structure to form a temporal constraint. The multidimensional graph structure is stored and retrieved using a gridded coordinate method or a time series indexing method.

4. The system according to claim 1, characterized in that, The sequence deduction and prediction unit includes: a loop identification module, used to identify closed paths in the frequency hopping topology map using a graph traversal algorithm, and calculate the cycle period characteristics of each closed path; a parameter back-inference module, used to estimate the pseudo-random sequence generation parameters based on the node sequence of the closed path using a polynomial fitting model or a linear feedback shift register model; and a path expansion module, used to recursively calculate the frequency hopping sequence for multiple future cycles based on the pseudo-random sequence generation parameters, so as to establish a structured model of the frequency hopping sequence generation mechanism of the target UAV communication link.

5. The system according to claim 1, characterized in that, The sequence inference and prediction unit adopts an optimal path selection strategy based on path weight accumulation, specifically: the directed edge weights of candidate paths are accumulated or calculated by weighted average; a confidence evaluation index is introduced to quantify the stability of candidate paths; multiple candidate paths are sorted and selected, and the most frequent resident sequence with the highest confidence is output; and the prediction results are dynamically updated and rolled for correction.

6. The system according to claim 1, characterized in that, The scheduling instruction generation unit encapsulates the structured control instruction data using a unified data format. The frequency field is used to characterize the center frequency and frequency band range information of the target frequency point corresponding to the frequency dwell sequence. The time field is used to characterize the absolute timestamp and relative time offset. The execution priority field is used to characterize the scheduling order of different control instructions. The structured control instruction data supports batch generation, queued management, and network transmission and parsing through standardized interface protocols.

7. The system according to claim 1, characterized in that, The edge execution coordination unit includes: a time synchronization module for achieving system time unification based on network time protocols or hardware clock signals; a scheduling cache module for preloading and caching the received structured control instruction data; and a trigger control module for triggering the execution of corresponding control instructions when the time parameters in the structured control instruction data are reached, so as to achieve scheduling control and continuous instruction switching with microsecond-level time precision.

8. The system according to claim 1, characterized in that, The system adopts a cloud-edge collaborative architecture, in which the cloud server is used to perform frequency hopping topology map construction, probability calculation and sequence prediction tasks, and the edge device carrying the edge execution collaboration unit is used to perform control command parsing, caching and real-time scheduling control tasks based on time parameters. The cloud and the edge interact with each other through network links and support task distribution and result feedback.

9. A cloud-based collaborative networked drone countermeasure method, executed by the method described in any one of claims 1-8, characterized in that, Includes the following steps: The radio frequency signal of the communication link of the target UAV is collected, and discrete feature data is extracted from the radio frequency signal. The discrete feature data includes at least carrier frequency, frequency jump sequence, pulse arrival time information and bandwidth parameters. The discrete feature data is uploaded to a cloud server, and graph structure mapping processing is performed on the discrete feature data. The discrete frequency values ​​are mapped to graph nodes, and the frequency hopping relationship is mapped to weighted directed edges. A frequency hopping topology graph is constructed based on the state transition probability calculation results. Based on the frequency hopping topology map, the frequency dwell sequence and corresponding time parameters of the target UAV communication link in the future time window are predicted through probability weight analysis and path deduction. Structured control instruction data is generated based on the frequency dwell sequence and the corresponding time parameters. The structured control instruction data includes at least frequency parameters, time parameters, and execution priority information. The structured control command data is sent to the edge execution device; The edge execution device synchronizes and schedules the structured control command data according to the time parameters, generates and outputs corresponding edge execution control signals, so as to realize networked collaborative countermeasures against the target UAV's communication link.

Images