Satellite-ground cooperative anti-interference system and method based on sensing integrated intelligent cognition

Abstract

The invention discloses a satellite-ground cooperative anti-interference system based on communication and sensing integrated intelligent cognition, and the system comprises a satellite-borne communication equipment module which comprises a first communication receiving and processing module, an interference detection module, an interference recognition module, an interference parameter estimation module, an anti-interference strategy generation module, and a first communication transmitting and processing module; an electromagnetic environment signal is received through a first communication receiving and processing module, an electromagnetic interference result and a signal to interference plus noise ratio are obtained through an interference detection module, an interference type is obtained through an interference identification module, an interference parameter is obtained through an interference parameter estimation module, and an anti-interference strategy is obtained through an anti-interference strategy generation module. An anti-interference strategy is sent through a first communication emission processing module; and the satellite ground station module comprises a second communication receiving and processing module, a second communication transmitting and processing module and an anti-interference strategy synchronization module, the anti-interference strategy is synchronized to the anti-interference strategy synchronization module, and the sent data is processed by the anti-interference strategy.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication networks, and in particular relates to a satellite-ground collaborative anti-interference system and method based on synaesthesia-integrated intelligent cognition. Background Art

[0002] Satellite communication anti-interference technology is a key technology to ensure efficient and reliable information transmission and the security of satellite constellations. It is related to the stable operation of satellite communications in complex electromagnetic environments, and is of great significance for resisting intentional or unintentional radio frequency signal interference, preventing information leakage and network paralysis. It is a necessary means to deal with dynamically changing interference situations.

[0003] Satellite communications are widely used in many fields, but the links are affected by complex electromagnetic environments, are susceptible to interference, and face security challenges. Anti-interference technology is of great significance, and there are many types of technologies, such as antennas and spread spectrum. Spread spectrum and frequency hopping are widely used, and there are multi-mode systems that can switch systems to resist interference, but at this stage, the selection of anti-interference systems depends on ground settings, has limited parameters, cannot be adaptively adjusted on board, lacks autonomous decision-making capabilities, and is difficult to deal with dynamic interference. Summary of the invention

[0004] The present invention proposes a satellite-ground collaborative anti-interference system and method based on synaesthesia-integrated intelligent cognition.

[0005] The inter-sensory data frame structure realizes the processing of different tasks by dividing the sensing time slot, synchronization time slot, and communication time slot to realize the organic combination of the three functions of sensing, synchronization, and communication in the satellite communication system, providing a framework for the anti-interference strategy. By detecting and identifying interference in real time and combining a two-layer decision-making mechanism, an effective anti-interference strategy can be adaptively generated, which not only improves the system's anti-interference ability and communication efficiency, but also ensures the stability and reliability of data transmission.

[0006] The first aspect of the present invention provides a satellite-ground collaborative anti-interference system based on synaesthesia and intelligent cognition, comprising: The satellite communication equipment module includes a first communication receiving processing module, an interference detection module, an interference identification module, an interference parameter estimation module, an anti-interference strategy generation module, and a first communication transmitting processing module; it is used to receive the electromagnetic environment signal and the data sent by the satellite ground station through the first communication receiving processing module, obtain the result of electromagnetic interference and the signal to noise ratio based on the electromagnetic environment signal through the interference detection module, obtain the interference type based on the electromagnetic environment signal through the interference identification module, obtain the interference parameter based on the electromagnetic environment signal through the interference parameter estimation module, obtain the anti-interference strategy based on the interference type, the signal to noise ratio and the interference parameter through the anti-interference strategy generation module, send the anti-interference strategy through the first communication transmitting processing module, and send the anti-interference strategy through the first communication transmitting processing module.

[0007] The satellite ground station module includes a second communication receiving and processing module, a second communication transmitting and processing module, and an anti-interference strategy synchronization module, which is used to receive the anti-interference strategy through the second communication receiving and processing module and synchronize to the anti-interference strategy synchronization module, and send data processed by the anti-interference strategy through the second communication transmitting and processing module.

[0008] The second aspect of the present invention provides a satellite-ground collaborative anti-interference method based on synaesthesia integrated intelligent cognition, which is applied to the satellite-ground collaborative anti-interference system described above. The satellite-ground collaborative anti-interference system includes a satellite-borne communication equipment module and a satellite ground station module, and includes the following steps: The preset periodic tasks include, in order of time, a perception periodic task, a synchronization periodic task, and a communication periodic task, and the periodic tasks are executed simultaneously on the onboard communication device and the satellite ground station; The sensing period task: the satellite-borne communication device collects electromagnetic environment signals and obtains an anti-interference strategy based on the electromagnetic environment signals; The synchronous periodic task: the satellite-borne communication device transmits the anti-interference strategy to the satellite ground station, and the satellite ground station updates the anti-interference strategy; The communication cycle task: the satellite ground station adopts the anti-interference strategy to transmit data information to the satellite-borne communication equipment, and the satellite-borne communication equipment demodulates to obtain the data information.

[0009] Preferably, the step of obtaining the anti-interference strategy based on the electromagnetic environment signal specifically includes: Detecting the signal-to-interference-noise ratio and electromagnetic interference results based on the electromagnetic environment signal by an energy method; Obtaining interference type and interference parameters based on the electromagnetic environment signal and the electromagnetic interference result; The anti-interference strategy is obtained based on the interference type, the signal to interference noise ratio and the interference parameter through deep reinforcement learning DQN.

[0010] Preferably, the step of obtaining the signal-to-interference-noise ratio and the electromagnetic interference result based on the electromagnetic environment signal by the energy method specifically includes: Preset a first threshold range and signal power; The interference power is obtained based on the electromagnetic environment signal by detecting by the energy method, and the noise power is obtained based on the electromagnetic environment signal by collecting; The signal to interference plus noise ratio is obtained based on the signal power, the interference power, and the noise power. The calculation expression is: Wherein Ps is the signal power, PJ is the interference power, and PN is the noise power; The result of electromagnetic interference is obtained by comparing the signal to interference noise ratio with the first threshold range.

[0011] Preferably, the step of obtaining the interference type and interference parameters based on the electromagnetic environment signal and the result of the electromagnetic interference specifically includes: If the result of judging the electromagnetic interference is that interference exists, the interference type is obtained based on the electromagnetic environment signal by the interference identification module of the satellite communication equipment, and the interference parameter is obtained based on the electromagnetic environment signal by the interference parameter estimation module of the satellite communication equipment.

[0012] Preferably, the step of obtaining the anti-interference strategy based on the interference type, the signal to interference noise ratio and the interference parameter through the deep reinforcement learning DQN specifically includes: Pre-constructing a modulation mode set, a frequency point set, an information rate set, a channel coding rate set, and a transmission power set, wherein the modulation mode set includes: BPSK, DSSS, and FHSS; The modulation mode set, the frequency point set, the information rate set, the channel coding rate set, and the transmit power set are combined to construct an anti-interference strategy set; Obtaining the modulation mode based on the signal to interference noise ratio and the interference type through a rule-based decision tree; The anti-interference strategy is obtained through the deep reinforcement learning DQN based on the modulation mode, the anti-interference strategy set, the interference type and the interference parameter.

[0013] Preferably, the step of obtaining the modulation mode based on the signal to interference noise ratio and the interference type through the decision tree of the rule specifically includes: Presetting a second threshold range; If the interference type is any one of single-tone interference, narrowband interference, and comb interference, the modulation mode remains unchanged; if the interference type is any one of broadband noise interference and swept frequency interference, it is further determined whether the signal to interference plus noise ratio is greater than the upper limit value of the second threshold range; if so, the modulation mode is updated to BPSK; if the signal to interference plus noise ratio hits the second threshold range, the modulation mode is updated to DSSS; if the signal to interference plus noise ratio is less than the lower limit value of the second threshold range, the modulation mode is updated to FHSS.

[0014] Preferably, the step of obtaining the anti-interference strategy based on the modulation mode, the anti-interference strategy set, the interference type and the interference parameter through the deep reinforcement learning DQN specifically includes: Timing interference information is constructed based on the interference parameters and interference types, and the calculation expression is: Where f j is the center frequency of the interference parameter, B J is the interference signal bandwidth of the interference parameter, A J is the interference amplitude of the interference parameter, C J is the interference type, P J is the interference power of the interference parameter; Based on the anti-interference strategy set, different four-dimensional tuples are constructed according to the modulation mode dimension, and the calculation expression is: Where P is the transmit power value of the anti-interference strategy, f s is the frequency value of the anti-interference strategy, R is the information rate value of the anti-interference strategy, and η is the channel coding rate of the anti-interference strategy; Based on the four-dimensional tuple and the timing interference information, a signal to interference plus noise ratio reward value, an information rate reward value, and a spectrum utilization reward value are calculated; The deep reinforcement learning DQN obtains the maximum Q value based on the signal to noise ratio reward value, the information rate reward value, and the spectrum utilization reward value to obtain an adaptive anti-interference strategy, and the calculation expression is: In the formula, reward SINR 、reward R 、reward η are the signal to interference and noise ratio reward value, the information rate reward value, and the spectrum utilization reward value, respectively, t+1 is an anti-interference strategy, θ and θ' are the strategy network weight and the target network weight respectively, and the value of the strategy network weight is updated to the target network weight by training the strategy network.

[0015] Preferably, the step of updating the value of the policy network weight to the target network weight by training the policy network specifically includes: Pre-built samples containing states, rewards, actions, and target actions; Based on the sample training network strategy, the loss function is obtained to update the value of the strategy network weight to the target network weight. The calculation expression is: .

[0016] Preferably, the step of calculating the signal to interference and noise ratio reward value, the information rate reward value, and the spectrum utilization reward value based on the four-dimensional tuple and the timing interference information specifically includes: Preset signal power P s , the first information rate R l , the second information rate R m , the third information rate R h , the first information rate result V R,l , the second information rate result V R,m , the third information rate result V R,h , the first spectrum utilization η l , the second spectrum utilization η m , the third spectrum utilization η h , the first spectrum utilization result V η,l , the second spectrum utilization result V η,m , the third spectrum utilization result V η,h ; Construct the signal to noise ratio reward value reward SINR , the information rate reward value reward R , the spectrum utilization reward value reward η The calculation formula is: .

[0017] Due to the adoption of the above technical solution, the present invention has the following advantages and positive effects compared with the prior art: The integrated data frame structure realizes the processing of different tasks by dividing the perception time slot, synchronization time slot and communication time slot, and realizes the organic combination of the three major functions of perception, synchronization and communication in the satellite communication system, providing a framework foundation for the anti-interference strategy.

[0018] By detecting and identifying interference in real time and combining it with a two-layer decision-making mechanism, effective anti-interference strategies can be adaptively generated, which not only improves the system's anti-interference capability and communication efficiency, but also ensures the stability and reliability of data transmission. Specifically: Real-time detection and identification of interference: Satellite-borne communication equipment collects electromagnetic signals in the environment, and obtains several elements of the following signals based on electromagnetic signal processing: the result of electromagnetic interference, the type of interference, the signal-to-noise ratio and the interference parameters; Adaptive generation of anti-interference strategies based on the elements through two-layer decision-making, the first layer of decision-making adopts a rule-based decision tree method, based on expert experience, to build a rule-based anti-interference intelligent decision-making engine, and automatically make decisions and operations according to a set of pre-defined rules. The rule engine can make automatic anti-interference decisions based on input data and expert anti-interference libraries, output parameter control actions, and also support subsequent model evolution incremental learning. The second-level decision-making through deep reinforcement learning: In order to solve the parameter selection problem after the system change, a data-based deep reinforcement learning method (DQN) is adopted. The perceived interference situation is taken as the state, and the signal-to-interference-noise ratio, information rate and spectrum utilization are taken as the joint optimization goals. An adjustable parameter strategy model is formed, and the selection weights of various multi-dimensional parameter selection actions facing the State are saved. Finally, the parameter decision model of power P, bit rate and information rate is obtained through the DQN method. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings, wherein: Figure 1 It is a schematic diagram of the framework of the satellite-ground collaborative anti-interference system based on synaesthesia integrated intelligent cognition of the present invention; Figure 2 It is a task module diagram of the satellite-ground collaborative anti-interference method based on synaesthesia integrated intelligent cognition of the present invention; Figure 3 It is a flow chart of the satellite-ground collaborative anti-interference method based on synaesthesia integrated intelligent cognition of the present invention; Figure 4 It is a schematic diagram of a double-layer anti-interference strategy generation method for reinforcement learning multi-dimensional parameter regulation in the satellite-ground collaborative anti-interference method based on synaesthesia integrated intelligent cognition of the present invention; Figure 5 This is a performance comparison diagram of the satellite-ground collaborative anti-interference method based on synaesthesia integrated intelligent cognition of the present invention and that without an anti-interference strategy; DETAILED DESCRIPTION

[0020] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise ratios, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0021] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0022] Embodiment 1 See also Figure 1 The first aspect of the present invention provides a satellite-ground collaborative anti-interference system based on synaesthesia and intelligent cognition, comprising: The satellite communication equipment module includes a first communication receiving processing module, an interference detection module, an interference identification module, an interference parameter estimation module, an anti-interference strategy generation module, and a first communication transmitting processing module; used to receive electromagnetic environment signals and data sent by a satellite ground station through the first communication receiving processing module, obtain electromagnetic interference results and a signal to noise ratio based on the electromagnetic environment signal through the interference detection module, obtain interference types based on the electromagnetic environment signal through the interference identification module, obtain interference parameters based on the electromagnetic environment signal through the interference parameter estimation module, obtain anti-interference strategies based on the interference types, signal to noise ratios, and interference parameters through the anti-interference strategy generation module, send anti-interference strategies through the first communication transmitting processing module, and send anti-interference strategies through the first communication transmitting processing module; The satellite ground station module includes a second communication receiving and processing module, a second communication transmitting and processing module, and an anti-interference strategy synchronization module. It is used to receive the anti-interference strategy through the second communication receiving and processing module and synchronize it to the anti-interference strategy synchronization module, and send data processed by the anti-interference strategy through the second communication transmitting and processing module.

[0023] In the perception cycle, the first communication transmission processing module collects electromagnetic environment signals through the first communication reception processing module in the perception cycle, obtains the result of electromagnetic interference based on the electromagnetic environment signal through the interference detection module, obtains the interference type based on the electromagnetic environment signal through the interference identification module, obtains the interference parameter based on the electromagnetic environment signal through the interference parameter estimation module, and obtains the anti-interference strategy based on the interference type, the signal-to-interference-noise ratio and the interference parameter through the anti-interference strategy generation module. In the synchronization cycle, the second communication reception processing module receives the anti-interference strategy and updates the anti-interference strategy synchronization module to the latest value, ensuring that the anti-interference measures of the entire communication link are consistent, ensuring the real-time consistency of the anti-interference strategies of the satellite communication equipment and the satellite ground station, and ensuring the consistency of the communication mechanism. In the communication cycle, the second communication transmission processing module encodes and modulates the data information to be sent according to the synchronized anti-interference strategy, and then transmits the data to the first communication reception processing module through the uplink. After receiving the data, the first communication reception processing module performs demodulation and decoding operations to obtain the original data information. In the perception, synchronization, and communication cycle tasks, the satellite communication equipment and the satellite ground station ensure collaborative work to ensure the timely transmission and processing of information, thereby improving the response speed and flexibility of the system.

[0024] Embodiment 2 See also Figure 1 , Figure 2 and Figure 3 The second aspect of the present invention provides a satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition, which is applied to the above-mentioned satellite-ground collaborative anti-interference system. The satellite-ground collaborative anti-interference system includes a satellite-borne communication equipment module and a satellite ground station module, and includes the following steps: The preset periodic tasks include: perception periodic tasks, synchronization periodic tasks, and communication periodic tasks in order of time. The periodic tasks are executed simultaneously on the onboard communication equipment and the satellite ground station; Perception cycle tasks: The satellite-borne communication equipment collects electromagnetic environment signals and obtains anti-interference strategies based on the electromagnetic environment signals; Synchronous periodic tasks: The satellite-borne communication equipment transmits the anti-interference strategy to the satellite ground station, and the satellite ground station updates the anti-interference strategy; Communication cycle tasks: The satellite ground station uses an anti-interference strategy to transmit data information to the onboard communication equipment, and the onboard communication equipment demodulates to obtain data information.

[0025] See also Figure 1 Preferably, the step of obtaining an anti-interference strategy based on an electromagnetic environment signal specifically includes: The signal-to-interference-to-noise ratio and electromagnetic interference results are obtained based on the electromagnetic environment signal through energy method detection; Obtain interference type and interference parameters based on the results of electromagnetic environment signals and electromagnetic interference; Through deep reinforcement learning DQN, the anti-interference strategy is obtained based on interference type, signal-to-interference-noise ratio and interference parameters.

[0026] The anti-interference strategy generated or updated by the satellite communication equipment in the perception cycle is transmitted to the satellite ground station. In the synchronization cycle, the satellite ground station receives the anti-interference strategy and updates it to synchronously respond to the interference conditions in the current environment, ensuring that the anti-interference measures of the entire communication link are consistent, ensuring the real-time consistency of the anti-interference strategies of the satellite communication equipment and the satellite ground station, and ensuring the consistency of the communication mechanism. In the communication cycle, the satellite ground station encodes and modulates the data information to be sent according to the synchronized anti-interference strategy, and then transmits the data to the satellite communication equipment through the uplink. After receiving the data, the satellite communication equipment performs demodulation and decoding operations to obtain the original data information. In the perception, synchronization, and communication cycle tasks, the satellite communication equipment and the satellite ground station ensure collaborative work to ensure the timely transmission and processing of information, thereby improving the response speed and flexibility of the system. There is no restriction on the duration of the perception cycle tasks, synchronization cycle tasks, and communication cycle tasks in this embodiment. Figure 5 It is demonstrated that the anti-interference performance of the present invention is superior compared with that without anti-interference strategy.

[0027] See also Figure 1 Preferably, the step of obtaining the signal-to-interference-noise ratio and the electromagnetic interference result based on the electromagnetic environment signal by the energy method specifically includes: Preset a first threshold range and signal power; The interference power is obtained by detecting the electromagnetic environment signal through the energy method, and the noise power is obtained by collecting the electromagnetic environment signal; The signal-to-interference-to-noise ratio is obtained based on signal power, interference power, and noise power. The calculation expression is: Where Ps is the signal power, PJ is the interference power, and PN is the noise power.

[0028] The result of electromagnetic interference is obtained by comparing the signal to interference noise ratio with the first threshold range.

[0029] The first threshold range quantifies whether there is electromagnetic interference and the degree of interference, which is set by performance requirements, experience and relevant standards, and is not limited in this embodiment. Signal power is used as an input parameter for calculating the signal-to-interference-to-noise ratio, and the interference power is obtained by extracting the electromagnetic environment signal. The noise power is used to evaluate the degree of influence on the communication signal. The calculated signal to interference noise ratio is compared with a preset first threshold range: if the signal to interference noise ratio exceeds the threshold range, it indicates that electromagnetic interference exists and further analysis and corresponding anti-interference measures are required.

[0030] See also Figure 1 Preferably, the step of obtaining interference type and interference parameter based on the electromagnetic environment signal and the electromagnetic interference result specifically includes: If the result of judging the electromagnetic interference is that interference exists, the interference type is obtained based on the electromagnetic environment signal by the interference identification module of the satellite communication equipment, and the interference parameter is obtained based on the electromagnetic environment signal by the interference parameter estimation module of the satellite communication equipment.

[0031] The interference identification module and interference parameter estimation module of the satellite communication equipment can accurately obtain the interference type and interference parameters. Identifying the interference type and estimating the interference parameters enable the satellite-ground collaborative anti-interference system to flexibly respond to various complex electromagnetic scenarios, ensure communication stability, and reduce communication interruptions and bit error rates caused by interference.

[0032] Preferably, the step of obtaining an anti-interference strategy based on interference type, signal-to-interference-noise ratio and interference parameter through deep reinforcement learning DQN specifically includes: Pre-constructed modulation mode set, frequency point set, information rate set, channel coding rate set, and transmit power set, where the modulation mode set includes: BPSK, DSSS, and FHSS; The modulation mode set, frequency point set, information rate set, channel coding rate set, and transmission power set are combined to construct an anti-interference strategy set; The modulation mode is obtained based on the signal-to-interference-noise ratio and the interference type through a rule-based decision tree; The anti-interference strategy is obtained through deep reinforcement learning DQN based on the modulation mode, anti-interference strategy set, interference type and interference parameters.

[0033] A pre-built modulation mode set includes modulation modes such as BPSK, DSSS, and FHSS. These modulation modes have different anti-interference characteristics and are suitable for different electromagnetic environments.

[0034] At the same time, a frequency point set, an information rate set, a channel coding rate set and a transmission power set are constructed. This embodiment does not restrict the set of frequency point sets, information rate sets, channel coding rate sets and transmission power sets. The frequency point set covers the optional communication frequency points; the information rate set determines the optional range of data transmission rate; the channel coding rate set includes different coding rate options for error correction and improving transmission reliability; the transmission power set specifies the possible value range of signal transmission power. The modulation mode set, frequency point set, information rate set, channel coding rate set and transmission power set are traversed and combined to construct a comprehensive anti-interference strategy set. Each combination represents a possible anti-interference strategy configuration. The first-level decision obtains the modulation mode based on the signal-to-noise ratio and interference type through a rule-based decision tree. The decision tree selects a modulation mode that matches it according to the preset rules. The second-level decision is continuously learned and trained by DQN, and the strategy that can maximize the long-term reward is selected from the anti-interference strategy set as the output according to the input modulation mode, interference type and interference parameter state information.

[0035] See also Figure 4 Preferably, the step of obtaining the modulation mode based on the signal to interference noise ratio and the interference type through a rule-based decision tree specifically includes: Presetting a second threshold range; If the interference type is any one of single-tone interference, narrowband interference, and comb interference, the modulation mode remains unchanged. If the interference type is any one of broadband noise interference and swept frequency interference, it is further determined whether the signal to interference and noise ratio is greater than the upper limit value of the second threshold range. If so, the modulation mode is updated to BPSK. If the signal to interference and noise ratio hits the second threshold range, the modulation mode is updated to DSSS. If the signal to interference and noise ratio is less than the lower limit value of the second threshold range, the modulation mode is updated to FHSS.

[0036] The interference type is determined to determine whether it is single-tone interference, narrowband interference or comb interference. If it is one of the above types, the current modulation mode is maintained unchanged. If the interference type is one of broadband noise interference or swept frequency interference, it is necessary to further determine the relationship between the signal to noise ratio and the second threshold range. If the signal to noise ratio is greater than the upper limit of the second threshold range, the modulation mode is updated to BPSK. The BPSK modulation mode has a higher spectrum efficiency under certain signal to noise ratio conditions. If the signal to noise ratio hits the second threshold range, that is, the signal to noise ratio is between the upper limit and the lower limit, the modulation mode is updated to DSSS. The DSSS modulation mode uses pseudo-random coding for spread spectrum communication and has a strong anti-interference ability under medium signal to noise ratio conditions. It is suitable for dealing with such situations with certain interference but not excellent channel conditions. If the signal to noise ratio is less than the lower limit of the second threshold range, it indicates that the current channel is subject to large interference and low signal to noise ratio. At this time, the modulation mode is updated to FHSS. The FHSS modulation mode avoids interference by fast frequency hopping, which can effectively improve the reliability of communication and reduce the impact of interference on communication in a low signal to noise ratio environment.

[0037] See also Figure 4 Preferably, the step of obtaining the anti-interference strategy based on the modulation mode, the anti-interference strategy set, the interference type and the interference parameter through deep reinforcement learning DQN specifically includes: Timing interference information is constructed based on interference parameters and interference types, and the calculation expression is: Where f j is the center frequency of the interference parameter, B J is the interference signal bandwidth of the interference parameter, A J is the interference amplitude of the interference parameter, C J is the interference type, P J is the interference power of the interference parameter; Based on the anti-interference strategy set, different four-dimensional tuples are constructed according to the modulation mode dimension. The calculation expression is: Where P is the transmit power value of the anti-interference strategy, f s is the frequency value of the anti-interference strategy, R is the information rate value of the anti-interference strategy, and η is the channel coding rate of the anti-interference strategy; Based on the four-dimensional tuple and the timing interference information, the signal-to-interference-noise ratio reward value, the information rate reward value, and the spectrum utilization reward value are calculated; Through deep reinforcement learning DQN, the maximum Q value is obtained based on the signal-to-noise ratio reward value, information rate reward value, and spectrum utilization reward value to obtain an adaptive anti-interference strategy. The calculation expression is: In the formula, rewardSINR 、reward R 、reward η are the signal-to-interference-noise ratio reward value, information rate reward value, and spectrum utilization reward value, respectively. t+1 For the anti-interference strategy, θ and θ' are the policy network weight and the target network weight respectively. The value of the policy network weight is updated to the target network weight by training the policy network.

[0038] The signal-to-interference-to-noise ratio reward value can reflect the effect of the strategy on improving the signal-to-interference-to-noise ratio under different interference conditions, as one of the quantitative strategy values. The information rate reward value quantifies the quantitative performance of the information transmission rate in the anti-interference strategy. The spectrum utilization reward value quantifies the efficiency of the strategy in utilizing spectrum resources, prompting the system to choose a strategy that can both effectively resist interference and reasonably utilize the spectrum. The quantification of the above values ​​helps to achieve a balance between resource utilization and performance in the communication system. Using deep reinforcement learning DQN to update the policy network weights based on multiple reward values ​​enables the system to dynamically adapt to the changing interference environment and communication needs. As the training progresses, the policy network weights are continuously updated to the target network weights, so that the system can adaptively find the optimal anti-interference strategy. This dynamic learning and optimization mechanism ensures that the communication system can always select the most suitable anti-interference strategy under different electromagnetic environments, different interference types and parameter changes.

[0039] See also Figure 4 Preferably, the step of updating the value of the policy network weight to the target network weight by training the policy network specifically includes: Pre-built samples containing states, rewards, actions, and target actions; Based on the sample training network strategy, the loss function is obtained to update the value of the strategy network weight to the target network weight. The calculation expression is: .

[0040] By pre-building samples containing states, rewards, actions, and target actions, rich data support is provided for deep reinforcement learning algorithms. These samples cover various states of the communication system under different interference scenarios, as well as the corresponding anti-interference actions and rewards obtained, so that the policy network can fully learn the characteristics and patterns of the optimal strategy in different situations, thereby achieving more accurate policy decisions. The loss function obtained by training the network strategy based on the samples helps to accelerate the convergence process of the policy network. By calculating the loss function and back-propagating the error, the parameters of the policy network can be accurately adjusted so that the network can learn an effective anti-interference strategy. At the same time, this method based on the target network weight update can increase the stability of training, avoid errors in the weight update process, and ensure that the system gradually tends to the optimal strategy selection during the training process. As the training progresses, the policy network weights are continuously updated to the target network weights, so that the system can dynamically adapt to the changing electromagnetic environment and interference conditions. In different time steps or scenarios, the system can adjust the strategy in time according to the new sample data to ensure that good anti-interference performance is always maintained in a complex and changing communication environment, effectively improving the reliability and stability of the communication system.

[0041] See also Figure 4 Preferably, the steps of calculating the signal to interference and noise ratio reward value, the information rate reward value, and the spectrum utilization reward value based on the four-dimensional tuple and the timing interference information specifically include: Preset signal power P s , the first information rate R l , the second information rate R m , the third information rate R h , the first information rate result V R,l , the second information rate result V R,m , the third information rate result V R,h , the first spectrum utilization η l , the second spectrum utilization η m , the third spectrum utilization η h , the first spectrum utilization result V η,l , the second spectrum utilization result V η,m , the third spectrum utilization result V η,h ; Construct the signal-to-interference-noise ratio reward SINR 、Information rate reward R , spectrum utilization reward η The calculation formula is: .

[0042] The first information rate R l , the second information rate R m , the third information rate R h , the first information rate result V R,l , the second information rate result V R,m , the third information rate result V R,h , the first spectrum utilization η l , the second spectrum utilization η m , the third spectrum utilization η h , the first spectrum utilization result V η,l , the second spectrum utilization result V η,m , the third spectrum utilization result V η,h Formulas for constructing signal-to-interference-and-noise ratio reward values, information rate reward values, and spectrum utilization reward values, the above variables are not limited in this embodiment.

[0043] In the description of this application, it should be noted that the terms "inside", "outside", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the application is usually placed when in use, which is only for the convenience of describing this application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0044] It should also be noted that, unless otherwise clearly specified and limited, the terms "disposed" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0045] Those skilled in the art can clearly understand that, for the sake of convenience and brevity in description, the identification content specifically executed by the above-described system and device can refer to the corresponding process in the aforementioned method embodiment.

[0046] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the protection scope of the present invention.

Claims

1. A satellite-ground collaborative anti-interference system based on synaesthesia and intelligent cognition, characterized in that: include: The satellite communication equipment module includes a first communication receiving processing module, an interference detection module, an interference identification module, an interference parameter estimation module, an anti-interference strategy generation module, and a first communication transmitting processing module; used to receive the electromagnetic environment signal and the data sent by the satellite ground station through the first communication receiving processing module, obtain the result of electromagnetic interference and the signal to noise ratio based on the electromagnetic environment signal through the interference detection module, obtain the interference type based on the electromagnetic environment signal through the interference identification module, obtain the interference parameter based on the electromagnetic environment signal through the interference parameter estimation module, obtain the anti-interference strategy based on the interference type, the signal to noise ratio and the interference parameter through the anti-interference strategy generation module, send the anti-interference strategy through the first communication transmitting processing module, and send the anti-interference strategy through the first communication transmitting processing module; The satellite ground station module includes a second communication receiving and processing module, a second communication transmitting and processing module, and an anti-interference strategy synchronization module, which is used to receive the anti-interference strategy through the second communication receiving and processing module and synchronize to the anti-interference strategy synchronization module, and send data processed by the anti-interference strategy through the second communication transmitting and processing module.

2. A satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition, applied to the satellite-ground collaborative anti-interference system according to claim 1, wherein the satellite-ground collaborative anti-interference system comprises a satellite-borne communication equipment module and a satellite ground station module, characterized in that: The steps include: The preset periodic tasks include, in order of time, a perception periodic task, a synchronization periodic task, and a communication periodic task, and the periodic tasks are executed simultaneously on the onboard communication device and the satellite ground station; The sensing period task: the satellite-borne communication device collects electromagnetic environment signals and obtains an anti-interference strategy based on the electromagnetic environment signals; The synchronous periodic task: the satellite-borne communication device transmits the anti-interference strategy to the satellite ground station, and the satellite ground station updates the anti-interference strategy; The communication cycle task: the satellite ground station adopts the anti-interference strategy to transmit data information to the satellite-borne communication equipment, and the satellite-borne communication equipment demodulates to obtain the data information.

3. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 2 is characterized in that: The step of obtaining the anti-interference strategy based on the electromagnetic environment signal specifically includes: Detecting the signal-to-interference-noise ratio and electromagnetic interference results based on the electromagnetic environment signal by an energy method; Obtaining interference type and interference parameters based on the electromagnetic environment signal and the electromagnetic interference result; The anti-interference strategy is obtained based on the interference type, the signal to interference noise ratio and the interference parameter through deep reinforcement learning DQN.

4. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 3 is characterized in that: The step of obtaining the signal-to-interference-noise ratio and the electromagnetic interference result based on the electromagnetic environment signal by the energy method specifically includes: Preset a first threshold range and signal power; The interference power is obtained based on the electromagnetic environment signal by detecting by the energy method, and the noise power is obtained based on the electromagnetic environment signal by collecting; The signal to interference plus noise ratio is obtained based on the signal power, the interference power, and the noise power. The calculation expression is: Wherein Ps is the signal power, PJ is the interference power, and PN is the noise power; The result of electromagnetic interference is obtained by comparing the signal to interference noise ratio with the first threshold range.

5. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 3 is characterized in that: The step of obtaining interference type and interference parameter based on the electromagnetic environment signal and the result of the electromagnetic interference specifically includes: If the result of judging the electromagnetic interference is that interference exists, the interference type is obtained based on the electromagnetic environment signal by the interference identification module of the satellite communication equipment, and the interference parameter is obtained based on the electromagnetic environment signal by the interference parameter estimation module of the satellite communication equipment.

6. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 3 is characterized in that: The step of obtaining the anti-interference strategy based on the interference type, the signal to interference noise ratio and the interference parameter through the deep reinforcement learning DQN specifically includes: Pre-constructing a modulation mode set, a frequency point set, an information rate set, a channel coding rate set, and a transmission power set, wherein the modulation mode set includes: BPSK, DSSS, and FHSS; The modulation mode set, the frequency point set, the information rate set, the channel coding rate set, and the transmit power set are combined to construct an anti-interference strategy set; Obtaining the modulation mode based on the signal to interference noise ratio and the interference type through a rule-based decision tree; The anti-interference strategy is obtained through the deep reinforcement learning DQN based on the modulation mode, the anti-interference strategy set, the interference type and the interference parameter.

7. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 6 is characterized in that: The step of obtaining the modulation mode based on the signal to interference noise ratio and the interference type through the decision tree of the rule specifically includes: Presetting a second threshold range; If the interference type is any one of single-tone interference, narrowband interference, and comb interference, the modulation mode remains unchanged; if the interference type is any one of broadband noise interference and swept frequency interference, it is further determined whether the signal to interference plus noise ratio is greater than the upper limit value of the second threshold range; if so, the modulation mode is updated to BPSK; if the signal to interference plus noise ratio hits the second threshold range, the modulation mode is updated to DSSS; if the signal to interference plus noise ratio is less than the lower limit value of the second threshold range, the modulation mode is updated to FHSS.

8. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 6 is characterized in that: The step of obtaining the anti-interference strategy based on the modulation mode, the anti-interference strategy set, the interference type and the interference parameter through the deep reinforcement learning DQN specifically includes: Timing interference information is constructed based on the interference parameters and interference types, and the calculation expression is: Where f j is the center frequency of the interference parameter, B J is the interference signal bandwidth of the interference parameter, A J is the interference amplitude of the interference parameter, C J is the interference type, P J is the interference power of the interference parameter; Based on the anti-interference strategy set, different four-dimensional tuples are constructed according to the modulation mode dimension, and the calculation expression is: Where P is the transmit power value of the anti-interference strategy, f s is the frequency value of the anti-interference strategy, R is the information rate value of the anti-interference strategy, and η is the channel coding rate of the anti-interference strategy; Based on the four-dimensional tuple and the timing interference information, a signal to interference plus noise ratio reward value, an information rate reward value, and a spectrum utilization reward value are calculated; The deep reinforcement learning DQN obtains the maximum Q value based on the signal to noise ratio reward value, the information rate reward value, and the spectrum utilization reward value to obtain an adaptive anti-interference strategy, and the calculation expression is: In the formula, reward SINR 、reward R 、reward η are the signal to interference and noise ratio reward value, the information rate reward value, and the spectrum utilization reward value, respectively, t+1 is an anti-interference strategy, θ and θ' are the strategy network weight and the target network weight respectively, and the value of the strategy network weight is updated to the target network weight by training the strategy network.

9. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 8 is characterized in that: The step of updating the value of the strategy network weight to the target network weight by training the strategy network specifically includes: Pre-built samples containing states, rewards, actions, and target actions; Based on the sample training network strategy, the loss function is obtained to update the value of the strategy network weight to the target network weight. The calculation expression is: 。 10. The satellite-ground collaborative anti-interference method based on synaesthesia and intelligent cognition according to claim 8 is characterized in that: The step of calculating the signal to interference plus noise ratio bonus value, the information rate bonus value, and the spectrum utilization bonus value based on the four-dimensional tuple and the timing interference information specifically includes: Preset signal power P s , the first information rate R l , the second information rate R m , the third information rate R h , the first information rate result V R,l , the second information rate result V R,m , the third information rate result V R,h , the first spectrum utilization η l , the second spectrum utilization η m , the third spectrum utilization η h , the first spectrum utilization result V η,l , the second spectrum utilization result V η,m , the third spectrum utilization result V η,h ; Construct the signal to noise ratio reward value reward SINR , the information rate reward value reward R , the spectrum utilization reward value reward η The calculation formula is: 。

Images