An anti-jamming navigation signal deception waveform design optimization method and system

Abstract

The present application relates to the technical field of anti-interference navigation signal deception, in particular to an anti-interference navigation signal deception waveform design optimization method and system; the system extracts space-time features by listening to the internal communication signals of the target network; the core is to use the extended Kalman filter to iteratively solve the network topology state in reverse, and combine the virtual constellation parameters to construct a systematic deception quality index; the optimal deception waveform is determined by optimizing the parameters to maximize the index; the system also corrects the topology state and strategy weight based on the deception effect error closed loop; the present application realizes the transition from open-loop blind deception to dynamic passive sensing, and solves the problem of being unable to perceive the target topology.

Description

Technical Field

[0001] This invention relates to the field of anti-interference navigation signal deception technology, specifically to an anti-interference navigation signal deception waveform design optimization method and system. Background Technology

[0002] In modern navigation countermeasures scenarios, distributed receiving networks widely employ spatiotemporal collaborative verification mechanisms to identify spoofing signals; traditional spoofing waveform generation methods are mostly open-loop systems, and their strategies are usually statically preset.

[0003] While such open-loop methods may focus on the fidelity of single-point signals, the signals they generate are physically derived from a single point and lack spatiotemporal asynchrony, making them easily detected by networked collaborative verification. Furthermore, traditional methods cannot passively perceive the topological state of the target network and lack a closed-loop mechanism for real-time feedback correction based on the deception effect, making it difficult to cope with the dynamic changes of the target network.

[0004] Therefore, how to design a deception waveform generation method that can dynamically sense the target network topology, balance signal fidelity and network confusion, and achieve adaptive closed-loop optimization has become an urgent technical problem to be solved. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for designing and optimizing anti-interference navigation signal deception waveforms. Specifically, the technical solution of this invention is as follows:

[0006] Listen to the internal communication signals of the target's receiving network, and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset;

[0007] Based on the spatiotemporal feature dataset, the network topology state vector is determined by iterative inverse calculation using an extended Kalman filter.

[0008] By combining the network topology state vector and the preset virtual constellation parameters, the deception feature matrix is ​​calculated;

[0009] Based on the deception feature matrix, a systematic deception quality index is constructed;

[0010] By adjusting the virtual constellation parameters to maximize the systematic deception quality index, the optimized deception feature matrix is ​​determined.

[0011] Based on the optimized deception feature matrix, a radio frequency output waveform is synthesized;

[0012] Transmit the RF output waveform and evaluate the spoofing effect, calculate the topology estimation error and spoofing failure indication error;

[0013] The network topology state vector is corrected based on the topology estimation error, and the preset deception strategy weight factor is dynamically adjusted based on the deception failure indication error.

[0014] Preferably, the network topology state vector includes: multiple estimated spatial location vectors of network nodes and clock offsets.

[0015] Preferably, solving the deception feature matrix includes:

[0016] For each receiving node in the network topology state vector, calculate the signal propagation delay from each virtual satellite;

[0017] For each receiving node in the network topology state vector, calculate the Doppler shift from each virtual satellite;

[0018] The calculated signal propagation delay and Doppler shift are combined to form a deception feature matrix.

[0019] Preferably, a systematic deception quality index is constructed, including:

[0020] Calculate the average fidelity of the nodes;

[0021] Calculate network confusion;

[0022] The average fidelity of nodes and the network confusion are weighted and summed using deception strategy weighting factors to obtain a systematic deception quality index.

[0023] Preferably, the calculation of the node average fidelity includes:

[0024] Obtain the normalized correlation peak quality of the synthesized waveform and the local desired signal reproduction at N receiving nodes;

[0025] Calculate the average value of the normalized correlation peak quality of N receiving nodes, and determine it as the node average fidelity.

[0026] Preferably, calculating network confusion includes:

[0027] For the i-th node among N nodes, calculate the spatiotemporal characteristic differences among M virtual satellites and determine the asynchronous characteristic index;

[0028] The average of the asynchronous characteristic indices of all N nodes is calculated and determined as the network confusion.

[0029] Preferably, the asynchronous characteristic index is based on the signal propagation delay variance and Doppler frequency shift variance calculated along the virtual satellite index axis, and is normalized by weighting coefficients and reference variance benchmarks.

[0030] Preferably, dynamically adjusting the preset deception strategy weight factors includes:

[0031] Determine whether the deception has failed based on whether the target network broadcasts global alarm information, and set the deception failure indication error accordingly;

[0032] The deception strategy weighting factor is adjusted based on the deception failure indication error and the preset learning rate.

[0033] An anti-interference navigation signal deception waveform design and optimization system, comprising:

[0034] The listening module is used to listen to the internal communication signals of the target receiving network and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset.

[0035] The state calculation module is used to determine the network topology state vector by performing iterative inverse calculation using an extended Kalman filter based on a spatiotemporal feature dataset.

[0036] The feature calculation module is used to calculate the deception feature matrix by combining the network topology state vector and the preset virtual constellation parameters;

[0037] The index construction module is used to construct a systematic deception quality index based on the deception feature matrix;

[0038] The optimization module is used to maximize the systematic deception quality index by adjusting the virtual constellation parameters and determine the optimized deception feature matrix.

[0039] The waveform synthesis module is used to synthesize radio frequency output waveforms based on the optimized deception feature matrix.

[0040] The feedback module is used to transmit the RF output waveform and evaluate the spoofing effect, and calculate the topology estimation error and spoofing failure indication error.

[0041] The correction module is used to correct the network topology state vector based on the topology estimation error and dynamically adjust the preset deception strategy weight factor based on the deception failure indication error.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. This invention achieves dynamic and passive perception of the target network topology by listening to the internal communication signals of the target receiving network and using an extended Kalman filter for iterative reverse calculation, thus solving the problem that traditional open-loop methods cannot perceive the target network topology.

[0044] 2. This invention constructs a systematic deception quality index that integrates node average fidelity and network confusion, and uses this as an optimization target. It can intelligently balance the two conflicting objectives of signal fidelity and network confusion, thus solving the problem that traditional methods are static and difficult to balance.

[0045] 3. This invention solves the deception feature matrix containing precise signal propagation delay and Doppler frequency shift, and quantifies the network confusion, thereby synthesizing a deception waveform with spatiotemporal asynchronous characteristics. This effectively counters the spatiotemporal collaborative verification mechanism of distributed networks and solves the defect that traditional single-point source signals are easily detected.

[0046] 4. This invention constructs a dual closed-loop feedback correction mechanism by evaluating the deception effect, calculating the topology estimation error and the deception failure indication error; it can correct the estimation of the network topology state and dynamically adjust the weight factor of the deception strategy, thereby achieving the ability to adapt to dynamic adversarial environments and overcoming the difficulty of traditional methods in dealing with dynamic changes in the target network. Attached Figure Description

[0047] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0048] Figure 1 This is a flowchart of the method of the present invention.

[0049] Figure 2 This is a structural diagram of the system of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0051] Example 1:

[0052] Please see Figure 1 An anti-interference navigation signal deception waveform design optimization method, comprising:

[0053] Listen to the internal communication signals of the target's receiving network, and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset;

[0054] Based on the spatiotemporal feature dataset, the network topology state vector is determined by iterative inverse calculation using an extended Kalman filter.

[0055] By combining the network topology state vector and the preset virtual constellation parameters, the deception feature matrix is ​​calculated;

[0056] Based on the deception feature matrix, a systematic deception quality index is constructed;

[0057] By adjusting the virtual constellation parameters to maximize the systematic deception quality index, the optimized deception feature matrix is ​​determined.

[0058] Based on the optimized deception feature matrix, a radio frequency output waveform is synthesized;

[0059] Transmit the RF output waveform and evaluate the spoofing effect, calculate the topology estimation error and spoofing failure indication error;

[0060] The network topology state vector is corrected based on the topology estimation error, and the preset deception strategy weight factor is dynamically adjusted based on the deception failure indication error.

[0061] This invention provides a method for designing and optimizing anti-interference navigation signal deception waveforms; this method constitutes a complete and self-consistent technical loop.

[0062] The system performs the task of listening to the internal communication signals of the target receiving network; its purpose is to passively perceive the state of the target network; in this embodiment, the system continuously listens to the navigation signal frequency band in the target area through a broadband radio frequency front-end, focusing on capturing and demodulating the internal communication signals exchanged between nodes of the distributed intelligent receiving network for spatiotemporal collaborative verification.

[0063] The Time Difference of Arrival (TDOA) and Frequency Difference of Arrival (FDOA) were extracted from the internal communication signals. These key spatiotemporal feature parameters were extracted using a standard signal processing procedure that performed multi-point localization on the captured internal communication signals. The extracted TDOA and FDOA datasets were used to construct a spatiotemporal feature dataset.

[0064] Based on the spatiotemporal feature dataset, the system performs iterative inverse computation using an Extended Kalman Filter (EKF). The purpose is to iteratively inversely compute the physical states of each node in the target receiving network, thereby determining the network topology state vector, denoted as... The vector It forms the basis for all subsequent deception strategies, providing dynamic target parameters for subsequent steps;

[0065] In obtaining the target state Then, the system combines the network topology state vector Given preset virtual constellation parameters, such as the initial orbital parameters of M virtual satellites, calculate the deception feature matrix, denoted as... The matrix It is a direct design blueprint for generating collaborative deception waveforms;

[0066] Based on the deception feature matrix Construct a systematic deception quality index, denoted as Systematic Deception Quality Index This refers to an optimization objective function used to unify the inherently conflicting goals of high fidelity and resistance to networked recognition.

[0067] To obtain the optimal deception parameters, the system adjusts the virtual constellation parameters to maximize the systematic deception quality index. This optimization process enables the system to find the optimal policy balance point and determine the optimized deception feature matrix. ;

[0068] Based on the optimized deception feature matrix System generated Baseband navigation signal of a virtual satellite Then, through a multi-channel signal synthesizer and precoding technology, and after up-conversion, a radio frequency output waveform is synthesized, denoted as... ;

[0069] The system enters the closed-loop feedback correction phase; the system transmits radio frequency output waveforms. The evaluation method involves returning to the first step, continuously monitoring the internal communication of the target's receiving network, and calculating the topology estimation error by comparing the data changes before and after the deception. And deception failure indication error ;

[0070] The system is based on topology estimation error. Correcting the network topology state vector Specifically, This information is fed back to the extended Kalman filter in the first step to correct the topological state vector at the next time step. ; and based on the error of deception failure indication Dynamically adjust the preset deception strategy weight factors, denoted as ;

[0071] The method in this embodiment overcomes the shortcomings of traditional open-loop deception methods in dealing with dynamic changes in the target network by constructing a complete closed loop of perception-decision-execution-feedback; it can not only accurately perceive the target network topology. Furthermore, it can be based on a quantifiable quality index that integrates fidelity and confusion. To intelligently design deception waveforms More importantly, it utilizes the feedback of deception. , To simultaneously correct the perception of the target. and its own strategy This enables the system to maintain optimal deception effectiveness and high robustness in dynamic confrontations.

[0072] Example 2:

[0073] The network topology state vector includes: the spatial location vectors of multiple estimated network nodes and clock offsets.

[0074] Based on Example 1, the network topology state vector is denoted as... Its purpose is to provide a target receiving network in A precise mathematical description of the instantaneous physical state; this vector is used in subsequent calculations of the deception feature matrix. The basis of dynamic parameters;

[0075] In this embodiment, Specifically, this includes: spatial location vectors of multiple estimated network nodes. velocity vector and clock bias Its mathematical form is defined as follows:

[0076] ;

[0077] in, represent The network topology state vector at any given time; This represents the estimated number of network nodes, derived from the calculation of internal communication signals; Represents the node index, from 1 to N; represent The estimated time of the first The node spatial location vector is a three-dimensional coordinate vector, and its source is obtained from the EKF iterative solution in the first step. This represents the velocity vector of the i-th node estimated at time t. Its data type is a three-dimensional coordinate vector, and its source is obtained by the EKF iterative solution in the first step. represent The estimated time of the first The node clock bias, whose data type is a scalar and whose unit is seconds, is obtained from the EKF iteration solution in the first step;

[0078] By using network topology state vectors Explicitly defined as a spatial location containing N nodes and clock bias By providing a set of data, this invention ensures that the description of the target network state is high-dimensional and accurate; this refined state description is essential for subsequent calculations of the precise signal propagation delay between the virtual constellation and each node. and Doppler shift This is a necessary prerequisite, and it greatly enhances the deception feature matrix. The physical reality.

[0079] Example 3:

[0080] Solving the deception feature matrix includes:

[0081] For each receiving node in the network topology state vector, calculate the signal propagation delay from each virtual satellite;

[0082] For each receiving node in the network topology state vector, calculate the Doppler shift from each virtual satellite;

[0083] The calculated signal propagation delay and Doppler shift are combined to form a deception feature matrix.

[0084] Based on Example 1, the deception feature matrix is ​​calculated. The steps are intended to construct a A 2D matrix, which defines M different spatiotemporal features that need to be accurately represented at N different spatial nodes, is the synthesized RF output waveform. Direct design blueprints; this step specifically includes:

[0085] For network topology state vectors Each receiving node in Calculate from each virtual satellite The signal propagation delay is denoted as . Signal propagation delay This refers to simulated virtual satellites. The signal propagates through space and reaches the receiving node. The time elapsed, taking into account the clock difference between the two and atmospheric delay along the signal propagation path; its calculation is based on the speed of light propagation model and superimposed with the standard atmospheric model, and the calculation method is as follows:

[0086] ;

[0087] in, The signal propagation delay is measured in seconds. The speed of light is a physical constant. For the first The node spatial location vector originates from the network topology state vector. supply; for Time of the first The position vectors of the virtual satellites are adjustable parameters that need to be optimized in this step; For the first Node clock bias, which originates from the network topology state vector. supply; for Time of the first Clock offset of a virtual satellite, and These are all adjustable parameters that need to be optimized in this step; and These represent calculations based on the standard atmospheric model and data from virtual satellites. To the node The ionospheric and tropospheric delays are functions of time, node position, and virtual satellite position;

[0088] For each receiving node in the network topology state vector Calculate from each virtual satellite The Doppler frequency shift, denoted as Doppler frequency shift This refers to the fact that due to the nodes With virtual satellite The change in signal carrier frequency caused by the relative motion between them; its calculation is based on the fundamental formula of the Doppler effect, and the calculation method is as follows:

[0089] ;

[0090] in, This is the Doppler frequency shift, and its dimension is Hertz; The navigation signal carrier frequency is a known system parameter. Let be the velocity vector of the i-th node at time t, which originates from the network topology state vector. Provided directly; for Time of the first The velocity vector of the virtual satellite is an adjustable parameter that needs to be optimized in this step; the remaining parameters... , , Same as above; ;

[0091] All calculated signal propagation delay and Doppler shift Combine to form a deception feature matrix ;

[0092] By applying the physical laws of light propagation model and the Doppler effect, and utilizing precise, real-time calculations of the target state... , To calculate separately and This invention ensures Each element in the matrix possesses a high degree of physical realism; this design ensures that the final synthesized waveform... The ability to accurately reproduce M sets of virtual signals with spatiotemporal asynchronous characteristics at N target node locations in space is a key prerequisite for deceiving a networked collaborative verification system.

[0093] Example 4:

[0094] Construct a systematic deception quality index, including:

[0095] Calculate the average fidelity of the nodes;

[0096] Calculate network confusion;

[0097] The average fidelity of nodes and the network confusion are weighted and summed using deception strategy weighting factors to obtain a systematic deception quality index.

[0098] Based on Example 1, a systematic deception quality index is constructed. The core purpose of this step is to provide a quantifiable objective function for the optimization algorithm, which creatively resolves the inherent conflict between the two objectives of high fidelity and resistance to network-based recognition. This step specifically includes:

[0099] Calculate the average fidelity of nodes Node average fidelity This refers to the quantification of deception signals. Average signal level similarity across all N receiving nodes; The higher the value, the more easily a single node is fooled;

[0100] Calculate network confusion Network confusion This refers to the ability to quantify deceptive signals against network collaborative verification; The higher the value, the stronger the asynchronicity of the constellation simulated by the system, and the more difficult it is for the target network to identify the single-point sound emission physical characteristics of the deception source through collaborative verification;

[0101] Average fidelity of nodes Network confusion By deceiving the strategy weight factors By performing a weighted summation, a systematic deception quality index is obtained. The index This is the core optimization objective of the present invention, and its construction method is as follows:

[0102] ;

[0103] in, To systematize the deception quality index is the optimization objective of step 1f in the optimization module; To deceive the strategy weight factor, its data type is a scalar between 0 and 1, and its source is a dynamically adjustable strategy parameter; The average fidelity of the nodes; Network obfuscation level;

[0104] The logic of this formula lies in, through Values ​​are used to balance the two sub-objectives: when When the value is close to 1, the system prioritizes optimization. fidelity; when When the value is close to 0, the system prioritizes optimization. Anti-identification; the system adjusts virtual constellation parameters Seeking The maximum value;

[0105] By constructing this unified, weighted harmonic quality index This invention enables high-fidelity targeting of two conflicting tactical objectives. and high confusion It is integrated into a computable and optimizable mathematical framework. This makes deception strategies no longer a binary choice, but rather something that can be adjusted... This allows for continuous and intelligent switching, enabling the system to consistently find the optimal strategic balance point in dynamic confrontations.

[0106] Example 5:

[0107] The average fidelity of the nodes is calculated, including:

[0108] Obtain the normalized correlation peak quality of the synthesized waveform and the local desired signal reproduction at N receiving nodes;

[0109] Calculate the average value of the normalized correlation peak quality of N receiving nodes, and determine it as the node average fidelity.

[0110] Based on Example 4, the average node fidelity is calculated. The steps, the purpose of which is to The high-fidelity term in the index is quantified; this step specifically includes:

[0111] During the optimization calculation phase, it is predicted that there are N receiving nodes. place, by The composite waveform formed by superimposing virtual signals, and the node The quality of the normalized correlation peak expected to be generated when performing correlation processing on the local desired signal is denoted as . ;Should It is a scalar value between 0 and 1, and the closer its value is to 1, the higher the fidelity of the system simulation; this predicted value can be calculated through the relevant processing of the simulated target receiver; specifically, It can be simulated Synthetic waveform at each node The expected signal reproduced locally at this node Perform relevant calculations to make predictions; This can be modeled as the ratio of the normalized main peak energy to the total received energy from the correlation operation output; normalized correlation peak quality. This refers to a scalar between 0 and 1, where the closer the value is to 1, the stronger the node. Received composite waveform The higher the fidelity;

[0112] Calculate the normalized correlation peak quality of N receiving nodes The average value is determined as the node average fidelity. The calculation method is as follows:

[0113] ;

[0114] in, The average fidelity of the nodes; The number of network nodes, whose source is the same Definition; For node indexing; For the first The quality of the normalized correlation peak of each receiving node;

[0115] By increasing fidelity Defined as N nodes The arithmetic mean of the present invention ensures the optimization objective. The fidelity term in the algorithm is applied to the entire network, rather than optimizing just a single node; this averaging process forces the optimization algorithm to seek a value that ensures all target nodes achieve a high fidelity. The solution provides a value, thus ensuring the deception waveform is correct. It has universal high fidelity for all target nodes, avoiding the risk of some nodes recognizing the deception due to neglecting some.

[0116] Example 6:

[0117] Calculating network confusion includes:

[0118] For the i-th node among N nodes, calculate the spatiotemporal characteristic differences among M virtual satellites and determine the asynchronous characteristic index;

[0119] The average of the asynchronous characteristic indices of all N nodes is calculated and determined as the network confusion.

[0120] The asynchronous characteristic index is based on the signal propagation delay variance and Doppler frequency shift variance calculated along the virtual satellite index axis, and is normalized using weighting coefficients and a reference variance benchmark.

[0121] Based on Example 4, calculate network confusion. The steps, the purpose of which is to The resistance to network identification in the index is quantified; the innovation of this invention lies in defining resistance to identification as the natural asynchronicity in simulated spacetime. The higher the value, the more spatially dispersed and asynchronous the simulated constellation; this step specifically includes:

[0122] For the i-th node among N nodes, calculate the spatiotemporal characteristic differences among M virtual satellites, determine the asynchronous characteristic index, and denot it as... ;

[0123] The asynchronous characteristic index The goal is to quantify at a single node. Spatiotemporal characteristics of the M virtual satellites observed at the location and The degree of dispersion; its construction is based on the virtual satellite index. Axial calculation of signal propagation delay variance and Doppler frequency shift variance And through weighting coefficients and reference variance benchmark Normalization is performed; the calculation method is as follows:

[0124] ;

[0125] in, For the first The asynchronous characteristic index of each node; Indicates the virtual satellite index Axial direction, i.e. Calculated variance; and The signal propagation delay and Doppler shift originate from the deception feature matrix. supply; and These are weighting coefficients for delay and frequency variance, derived from preset values, used to balance the importance of delay and frequency differences; for example, in a target network where TDOA information is the primary basis for collaborative verification, these can be set. Conversely, when FDOA is the primary basis, it can be set... , or settings Taking a balanced approach; and The reference variance benchmark used for normalization is derived from a preset value, such as the statistical variance of the actual navigation constellation in this scenario, so that... It has a clear physical meaning; the reference variance benchmark and The specific calibration method includes: collecting TDOA and FDOA data of real navigation constellations in typical scenarios, calculating their statistical variance, and using the statistical variance or a specific multiple thereof as the benchmark value; the weighting coefficient and The calibration method includes: testing the deception effect of different weight combinations on a specific type of collaborative verification network through a simulation platform, and selecting the optimal weight coefficient combination; to ensure the numerical stability of the calculation, and All should be greater than a certain minimum positive value. The value is used to prevent division by zero errors;

[0126] After calculating all N nodes Then, the average of the asynchronous characteristic indices of all N nodes is calculated and determined as the network confusion. The calculation method is as follows:

[0127] ;

[0128] in, The overall network confusion level; The number of nodes; For the first The asynchronous characteristic index of each node is derived from the calculation in the previous step;

[0129] It provides a sophisticated and computable method for quantifying resistance to identification; by adjusting the degree of confusion... Defined as spatiotemporal characteristic variance The average value of this invention transforms an abstract tactical concept of anti-cooperative verification into a concrete mathematical indicator. The higher the value, the more dispersed the deception signal is in terms of spatiotemporal characteristics. The more difficult it is for the target network to identify the physical characteristics that all signals come from a single sound source by comparing the data of N nodes, thus greatly improving the systematic countermeasure capability and success rate of deception.

[0130] Example 7:

[0131] Dynamically adjust the preset deception strategy weight factors, including:

[0132] Determine whether the deception has failed based on whether the target network broadcasts global alarm information, and set the deception failure indication error accordingly;

[0133] The deception strategy weighting factor is adjusted based on the deception failure indication error and the preset learning rate.

[0134] Based on Example 1, the preset deception strategy weight factor is dynamically adjusted. The following steps are crucial for achieving closed-loop correction and policy adaptation across the entire system; these steps include:

[0135] After transmitting the waveform, the system evaluates the effect through the feedback module as described in step 1h of Example 1. Based on observable behaviors such as whether the target network broadcasts global alarm information, it determines whether the spoofing has failed and sets a spoofing failure indication error. Deception failure indicates error This refers to an error signal used to indicate whether a deception strategy has succeeded or failed. In this embodiment, the rule is: if the network collaborative verification concludes that deception has occurred, then the deception is considered to have failed. It is set to a large positive value; if the internal network communication characteristics change significantly, indicating that it has switched operating modes or rejected signals, it is judged as a failed deception even without broadcast alarms. Set to a positive value; if the nodes become chaotic and cannot reach a consensus, then... It is set to a negative value; for example, to ensure the feasibility of the algorithm, a baseline error value can be set. When the deception fails, set When the network is judged to be in chaos, set... ; or can The value is normalized to Within the range, to ensure the learning rate Stable convergence;

[0136] Error based on deception failure indication and preset learning rate Adjust the weighting factors of the deception strategy For use in the next moment of Exponential calculation; this update rule forms a feedback loop, the strategy weight factor It has a preset initial value. The initial value It can be determined based on prior knowledge or simulation test results; the calculation method is as follows:

[0137] ;

[0138] in, The updated policy weights will be used for Moment calculate; The current weight; A preset learning rate, whose data type is a positive scalar, is used to control the adjustment step size and sensitivity; A larger value will lead to The adjustments are faster to adapt to rapidly changing adversarial scenarios, but this may cause strategic turmoil. Smaller values ​​result in smoother adjustments and more stable strategies, but slower responses. Specific calibration methods include simulating multiple adversarial processes in a simulation environment and testing different... Consider the impact of the value on the convergence speed and stability of the strategy, and choose a value that strikes a balance between convergence and stability; The error in the deception failure indication is derived from the evaluation of the target network's behavior; To limit the results to A function of intervals, to ensure Always keep within the effective range;

[0139] The logical fusion effect of this update rule is as follows:

[0140] When the deception fails ;at this time , It will decrease; according to , Decrease means Increase the value, and the system will automatically make the strategy more resistant to identification, i.e., increase the value. The weights;

[0141] When the deception succeeds ;at this time , This will increase; the system will be more inclined towards fidelity. In order to ensure success while maximizing signal quality;

[0142] By introducing based of The present invention employs a dynamic adjustment mechanism to implement deception strategies. The system has evolved from a static, manually preset value to a dynamically evolving, real-time driven intelligent parameter based on adversarial effects; this allows the system to automatically adjust fidelity. and confusion Finding the optimal strategy balance point in the current adversarial scenario greatly improves the system's adaptability and robustness in long-term, continuous adversarial processes.

[0143] Example 8:

[0144] Please see Figure 2 An anti-interference navigation signal deception waveform design and optimization system, comprising:

[0145] The listening module is used to listen to the internal communication signals of the target receiving network and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset.

[0146] The state calculation module is used to determine the network topology state vector by performing iterative inverse calculation using an extended Kalman filter based on a spatiotemporal feature dataset.

[0147] The feature calculation module is used to calculate the deception feature matrix by combining the network topology state vector and the preset virtual constellation parameters;

[0148] The index construction module is used to construct a systematic deception quality index based on the deception feature matrix;

[0149] The optimization module is used to maximize the systematic deception quality index by adjusting the virtual constellation parameters and determine the optimized deception feature matrix.

[0150] The waveform synthesis module is used to synthesize radio frequency output waveforms based on the optimized deception feature matrix.

[0151] The feedback module is used to transmit the RF output waveform and evaluate the spoofing effect, and calculate the topology estimation error and spoofing failure indication error.

[0152] The correction module is used to correct the network topology state vector based on the topology estimation error and dynamically adjust the preset deception strategy weight factor based on the deception failure indication error.

[0153] This invention also provides an anti-interference navigation signal deception waveform design and optimization system, which is a hardware or hardware + software functional entity specifically designed for executing the method; such as Figure 2 As shown, the system includes:

[0154] A listening module is designed to passively sense a target network. In this embodiment, the module is configured to listen to the internal communication signals of the target receiving network and extract the time difference of arrival (TDOA) and frequency difference of arrival (FDOA) from the internal communication signals to construct a spatiotemporal feature dataset.

[0155] A state calculation module, whose purpose is to estimate the instantaneous state of the target, is connected to the listening module and configured to perform iterative back-calculation using an Extended Kalman Filter (EKF) based on a spatiotemporal feature dataset to determine the network topology state vector. Its definition is as shown in Example 2;

[0156] A feature resolution module, designed to construct a deception blueprint, is connected to a state resolution module and configured to incorporate network topology state vectors. Using preset virtual constellation parameters, calculate the deception feature matrix. Its definition is as shown in Example 3, including and ;

[0157] An index building module, whose purpose is to establish the optimization objective, is connected to the feature solving module and is configured to be based on the deceptive feature matrix. Construct a systematic deception quality index Its definition is as shown in Examples 4-7, fusion and ;

[0158] An optimization module, designed to solve for the optimal deception parameters, connects the index construction module and the feature calculation module. This module is configured to maximize the systematic deception quality index by adjusting the virtual constellation parameters. And finally determine the optimized deception feature matrix. ;

[0159] A waveform synthesis module is designed to generate actual physical signals; this module is connected to an optimization module and is configured to use an optimized deception feature matrix. Synthesized RF output waveform ;

[0160] A feedback module is included to evaluate the actual effectiveness of the deception; this module outputs the system's transmitted radio frequency waveform. Subsequently, it was configured to evaluate the deception effect and solve for the topology estimation error. And deception failure indication error ;

[0161] A correction module, designed to implement dual closed-loop correction of the system, is connected to the feedback module and configured to perform two correction actions: one based on the topology estimation error. Correcting the network topology state vector For example, This information is fed back to the EKF in the state resolution module; and based on the deception failure indication error... Dynamically adjust the preset deception strategy weight factors And will update Feedback is sent to the index construction module;

[0162] The aforementioned modules work closely together to form a fully functional, logically closed-loop deception waveform optimization system; the listening and state calculation module enables accurate target perception; and the feature calculation, exponent construction, and optimization module implements... and The intelligent decision-making of the balancing strategy is handled by the waveform synthesis module, while the feedback and correction module implements the sensing. and decision The system employs a dual closed-loop correction mechanism, thus possessing dynamic adversarial capabilities, adaptability, and a high deception success rate that are not found in traditional open-loop deception systems.

[0163] The above embodiments only illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the appended embodiments.

[0164] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for designing and optimizing anti-interference navigation signal deception waveforms, characterized in that, include: Listen to the internal communication signals of the target's receiving network, and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset; Based on the spatiotemporal feature dataset, the network topology state vector is determined by iterative inverse calculation using an extended Kalman filter. By combining the network topology state vector and the preset virtual constellation parameters, the deception feature matrix is ​​calculated; Based on the deception feature matrix, a systematic deception quality index is constructed; By adjusting the virtual constellation parameters to maximize the systematic deception quality index, the optimized deception feature matrix is ​​determined. Based on the optimized deception feature matrix, a radio frequency output waveform is synthesized; Transmit the RF output waveform and evaluate the spoofing effect, calculate the topology estimation error and spoofing failure indication error; The network topology state vector is corrected based on the topology estimation error, and the preset deception strategy weight factor is dynamically adjusted based on the deception failure indication error. Solving the deception feature matrix includes: For each receiving node in the network topology state vector, calculate the signal propagation delay from each virtual satellite; For each receiving node in the network topology state vector, calculate the Doppler shift from each virtual satellite; The calculated signal propagation delay and Doppler shift are combined to form a deception feature matrix.

2. The anti-interference navigation signal deception waveform design and optimization method according to claim 1, characterized in that, The network topology state vector includes: the spatial location vectors of multiple estimated network nodes and clock offsets.

3. The anti-interference navigation signal deception waveform design and optimization method according to claim 1, characterized in that, Construct a systematic deception quality index, including: Calculate the average fidelity of the nodes; Calculate network confusion; The average fidelity of nodes and the network confusion are weighted and summed using deception strategy weighting factors to obtain a systematic deception quality index.

4. The anti-interference navigation signal deception waveform design and optimization method according to claim 3, characterized in that, The average fidelity of the nodes is calculated, including: Obtain the normalized correlation peak quality of the synthesized waveform and the local desired signal reproduction at N receiving nodes; Calculate the average value of the normalized correlation peak quality of N receiving nodes, and determine it as the node average fidelity.

5. The anti-interference navigation signal deception waveform design and optimization method according to claim 3, characterized in that, Calculating network confusion includes: For the i-th node among N nodes, calculate the spatiotemporal characteristic differences among M virtual satellites and determine the asynchronous characteristic index; The average of the asynchronous characteristic indices of all N nodes is calculated and determined as the network confusion.

6. The anti-interference navigation signal deception waveform design and optimization method according to claim 5, characterized in that, The asynchronous characteristic index is based on the signal propagation delay variance and Doppler frequency shift variance calculated along the virtual satellite index axis, and is normalized using weighting coefficients and a reference variance benchmark.

7. The anti-interference navigation signal deception waveform design and optimization method according to claim 1, characterized in that, Dynamically adjust the preset deception strategy weight factors, including: Determine whether the deception has failed based on whether the target network broadcasts global alarm information, and set the deception failure indication error accordingly; The deception strategy weighting factor is adjusted based on the deception failure indication error and the preset learning rate.

8. An anti-interference navigation signal deception waveform design and optimization system, based on the anti-interference navigation signal deception waveform design and optimization method according to any one of claims 1-7, characterized in that, include: The listening module is used to listen to the internal communication signals of the target receiving network and extract the signal arrival time difference and signal arrival frequency difference from the internal communication signals to construct a spatiotemporal feature dataset. The state calculation module is used to determine the network topology state vector by performing iterative inverse calculation using an extended Kalman filter based on a spatiotemporal feature dataset. The feature calculation module is used to calculate the deception feature matrix by combining the network topology state vector and the preset virtual constellation parameters; The index construction module is used to construct a systematic deception quality index based on the deception feature matrix; The optimization module is used to maximize the systematic deception quality index by adjusting the virtual constellation parameters and determine the optimized deception feature matrix. The waveform synthesis module is used to synthesize radio frequency output waveforms based on the optimized deception feature matrix. The feedback module is used to transmit the RF output waveform and evaluate the spoofing effect, and calculate the topology estimation error and spoofing failure indication error. The correction module is used to correct the network topology state vector based on the topology estimation error and dynamically adjust the preset deception strategy weight factor based on the deception failure indication error.

Images