Distributed intelligent resource allocation method and system based on SA-PSO algorithm

Abstract

The application discloses a kind of distributed interference intelligent resource allocation method and system based on SA-PSO algorithm, mainly solve the problem that the effect of existing technology is poor under the condition of limited interference resource multiple part distributed jammer to airborne phased array radar space-time adaptive processing interference.It is as follows: the scene of distributed interference resource allocation is established, and the situation parameter set of radar, target and jammer is obtained;Interference resource decision vector containing interference power, unified interference style, forwarding times and sampling pulse width is constructed;Resource constraint is established and is mapped as equivalent interference covariance matrix, and it is used as the input of minimum space-time adaptive processing output target signal-to-clutter-noise ratio, and the target function value is calculated;The optimal resource allocation scheme is obtained by searching according to the function value.The application can reduce the target detection performance of airborne phased array radar after space-time adaptive processing under the condition of limited interference resource, improve the interference effect on airborne phased array radar, and can be used for intelligent optimization configuration of target radar interference resource.

Description

Technical Field

[0001] This invention belongs to the field of radar countermeasures technology, and further relates to a distributed jamming intelligent resource allocation method and system. It can be used for intelligent optimization and configuration of jamming resources for target radars, improving the cooperative jamming effectiveness against target radars. It can also effectively jam non-cooperative phased array radars. Background Technology

[0002] Airborne radar is susceptible to widespread ground clutter, and the movement of the aircraft further expands the clutter spectrum, posing significant challenges to its clutter suppression capabilities. The development of active phased array (APA) technology has not only enhanced radar performance but also expanded the operational modes of combat aircraft. Compared to other airborne radars, APA-equipped airborne phased array radars offer advantages such as the ability to engage multiple targets, beam scanning unaffected by inertia, fast response speed, and strong anti-jamming capabilities. For phased array radars, the space-time adaptive processing (STAP) technology allows for joint processing of both space and time dimensions, creating notches in the clutter spectrum to filter out clutter and significantly reduce its impact. Furthermore, STAP can also create interference notches in the direction of interference incidence, making it difficult for traditional sidelobe active jamming to effectively interfere with it. The superior ability of technology to suppress clutter and interference makes it difficult for traditional jamming methods to effectively jam airborne phased array radars. Distributed jamming is an effective measure to jam this type of radar. Under limited resource conditions, the allocation and scheduling of jamming resources from multiple jammers is particularly important. If allocation is done manually, it not only consumes a lot of manpower but may also result in insufficient jamming coverage density, leading to poor jamming effect. Therefore, there is an urgent need to propose a feasible method for jamming resource allocation.

[0003] Patent document CN121333347A discloses a method for allocating frequency-hopping jamming resources under reconnaissance error conditions. It utilizes an asynchronous deep reinforcement learning algorithm to solve the frequency-hopping jamming resource allocation problem. The method trains a policy network through multiple jamming tasks, obtaining multiple policy gradients corresponding to these tasks. A gradient conflict resolution mechanism is then used to correct the policy gradients, resulting in an optimized allocation policy network. However, when the number of distributed jamming devices is large, the training model becomes complex, leading to high implementation complexity in practical applications.

[0004] In his graduation thesis, "Research on Distributed Interference of STAP Technology for Airborne Early Warning Radar," Xi Xin proposed a method for optimizing distributed interference resource allocation using the firefly algorithm. This method solves the objective function problem by leveraging the attraction between fireflies based on their luminosity and evaluates the merits of the current allocation scheme using a fitness function. While this method can perform optimization without relying on gradient information of the objective function, it suffers from several drawbacks. The use of high-luminosity individuals can prematurely dominate the population, leading to local optima. Furthermore, with a large population size, the numerous comparisons between individuals significantly increase computational complexity.

[0005] In his graduation thesis, "Distributed Interference Decision and Evaluation Technology," Zhang Lianwei proposed a method using a genetic algorithm to optimize the allocation of interference resources in distributed interference. This method encodes the solutions to the problem using chromosomes, constructs a fitness function to calculate the fitness value of each individual, and outputs the optimization result through a genetic operation process. However, this method, due to the need for selection, crossover, and mutation in each generation during the genetic operation, is prone to producing individuals that do not meet the constraints. This necessitates additional repair mechanisms or penalty functions, resulting in slow convergence and requiring more generations to iterate and obtain the desired result. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the prior art by proposing a distributed jamming intelligent resource allocation method and system based on the SA-PSO algorithm, so as to improve the convergence speed in the jamming resource allocation process and enhance the jamming effect on airborne phased array radar.

[0007] The technical approach to achieving this invention is as follows: By combining the optimization process of the PSO algorithm, the particle swarm can accept inferior solutions in a probabilistic manner while evolving along the historical optimal direction, significantly reducing the required average number of iterations and improving the convergence speed in the optimization process; by introducing the iteration rules of the particle swarm and the simulated annealing criterion, the optimal result is obtained by comparing the individual optimal value with the global optimal value, avoiding the algorithm from converging prematurely and getting trapped in local optima, and finally obtaining the optimal resource allocation scheme, thereby improving the interference effect on airborne phased array radar.

[0008] Based on the above ideas, the technical solution of the present invention includes:

[0009] 1. A distributed interference intelligent resource allocation method based on SA-PSO algorithm

[0010] 1. A distributed interference intelligent resource allocation method based on the SA-PSO algorithm, characterized in that it includes:

[0011] (1) Establish a distributed jamming resource allocation scenario for space-time adaptive processing of airborne phased array radar, and obtain the parameter sets of airborne phased array radar, target to be detected and multiple distributed jammers;

[0012] (2) Based on the situation parameter set and the type of interference resources to be allocated, construct a distributed interference resource decision vector;

[0013] (3) Establish resource constraints based on the interference resource decision vector;

[0014] (4) The interference resource decision vector that satisfies the constraints is mapped to the equivalent interference covariance matrix of the input end of the airborne phased array radar space-time adaptive processing, and a distributed interference resource allocation optimization model is constructed with the goal of reducing the target signal-to-clutter ratio at the output end of the space-time adaptive processing.

[0015] (5) Encode the interference resource decision vector into the particle position in the particle swarm algorithm, and initialize the particle swarm algorithm parameters and simulated annealing algorithm parameters;

[0016] (6) Update the particle positions according to the particle swarm update rule, calculate the target signal-to-noise ratio by constraining the updated particle positions, and update the individual optimal position and the global optimal position using the simulated annealing acceptance criterion.

[0017] (7) Repeat step (6) until the termination condition is met, decode the global optimal position into the optimal interference resource allocation scheme of each distributed jammer and output it.

[0018] Furthermore, the implementation of the distributed interference resource allocation optimization model in (4) includes:

[0019] 4a) Based on the set resource constraints, the interference power is non-negatively truncated and normalized, and the uniform interference pattern, forwarding times and sampling pulse width are projected onto the corresponding feasible sets to extract the interference decision vector that meets the resource constraints.

[0020] 4b) Based on the interference resource decision vector and the Space-time steering vector of the distributed jammer Calculate the first The equivalent interference contribution of each distributed jammer to the airborne phased array radar's space-time processing input is calculated, and the equivalent interference contributions of all distributed jammers are superimposed to obtain the result from the jamming resource decision vector. Determined equivalent disturbance covariance matrix :

[0021] 4d) Using the equivalent interference covariance matrix constructed in 4b), define a method to minimize the target signal-to-clutter ratio at the output of the airborne phased array radar's space-time adaptive processing. .

[0022] Furthermore, the constraint calculation of the target signal-to-noise ratio for the updated particle positions in step (6) includes the following implementation:

[0023] 6a) in the In the iteration, the particle's velocity and position are updated respectively, resulting in the... After the nth iteration The speed of each particle and location :

[0024] 6b) Based on resource constraints After applying constraints, the first allocation is calculated based on the optimal allocation model. After the nth iteration The target signal-to-noise ratio for each particle .

[0025] Furthermore, the process of updating the individual optimal position and the global optimal position using the simulated annealing acceptance criterion in step (6) includes the following implementation:

[0026] 6c) The constrained first The candidate interference resource allocation scheme corresponding to the particle position in the next iteration is denoted as... Let the resource allocation scheme corresponding to the current individual's optimal position be denoted as Define the objective function difference :

[0027] 6d) Based on the difference of the objective function Given the range of values, select an acceptable resource allocation scheme and update the current optimal position of the individual based on the accepted resource allocation scheme:

[0028] when Then accept Find the resource allocation scheme corresponding to the current individual's optimal position, and update the current individual's optimal position. ;

[0029] when When, then according to probability Choose whether to accept :

[0030] If random number satisfy Then accept Develop the current resource allocation plan and update the optimal position of the current individual. ;

[0031] like If so, the current resource allocation scheme will be retained. ;

[0032] 6e) Based on the magnitude of the objective function value, select the position that accepts the global optimum:

[0033] If the global optimal position corresponds to the objective function value Greater than the objective function value corresponding to the current individual's optimal position Then retain the current globally optimal position. ;

[0034] If the global optimal position corresponds to the objective function value Less than the objective function value corresponding to the current individual's optimal position Then update the current global optimal position to ;

[0035] 6f) After completing 6a) to 6e), update the temperature. ;

[0036] 2. A distributed interference intelligent resource allocation system based on the SA-PSO algorithm, characterized in that it comprises:

[0037] The scenario establishment module is used to establish a distributed jamming resource allocation scenario for airborne phased array radar space-time adaptive processing, and to obtain the parameter sets of airborne phased array radar, the target to be detected, and multiple distributed jammers;

[0038] The parameter mapping module is used to transform the acquired parameter set into a distributed interference resource decision vector, and to construct resource constraints and collaborative constraints based on the distributed interference resource decision vector;

[0039] The vector encoding module is used to construct a distributed interference resource allocation optimization model, encode the distributed interference decision vector into the initialization parameters of the SA-PSO algorithm, and correct the particle swarm parameters after iteration according to the constraint rules of the parameter mapping module.

[0040] The update iteration module is used to iteratively update parameters such as particle swarm position and velocity, and to update the individual optimal position and the global optimal position using the simulated annealing acceptance criterion.

[0041] The location decoding module is used to output the optimal interference resource allocation scheme.

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

[0043] Firstly, this invention constructs a disturbance resource decision vector, maps the disturbance resource decision vector to the basic parameters of the SA-PSO algorithm, clarifies the iterative relationship of particle swarm parameters, and introduces a simulated annealing algorithm to update the individual optimal position and the global optimal position, thereby maintaining the directional convergence of the particle swarm and improving the convergence speed in the process of disturbance resource allocation.

[0044] Secondly, this invention constructs a distributed interference resource allocation optimization model, using the defined minimum target signal-to-clutter ratio (SNR) at the output of the airborne phased array radar's space-time adaptive processing as the objective function. Guided by the objective function, the model iterative relationship is established, and the optimal interference resource decision is selected to obtain the globally minimum SNR, thereby improving the interference effect on airborne phased array radar. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the implementation of the distributed interference intelligent resource allocation method based on the SA-PSO algorithm of this invention.

[0046] Figure 2 This is a distributed interference resource optimization scenario diagram constructed in this invention;

[0047] Figure 3 This is a block diagram of the distributed interference intelligent resource allocation system based on the SA-PSO algorithm of this invention.

[0048] Figure 4 These are images showing the noise convolution interference effects after resource allocation using the present invention and existing genetic algorithms, respectively.

[0049] Figure 5 These are images showing the effects of dense false target interference after resource allocation using the present invention and existing genetic algorithms, respectively.

[0050] Figure 6 These are diagrams showing the intermittent sampling frequency shifting and forwarding interference effects after resource allocation using the present invention and existing genetic algorithms, respectively. Detailed Implementation

[0051] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should all fall within the protection scope of the present invention. It should be noted that the step numbers in the specification and claims of the present invention are only for clear description of the embodiments of the present invention and for ease of understanding, and their order is not limited.

[0052] Example 1: A Distributed Interference Intelligent Resource Allocation Method Based on the SA-PSO Algorithm

[0053] Reference Figure 1 The implementation steps of this example include the following:

[0054] Step 1: Establish a distributed interference scenario and obtain relevant parameters.

[0055] Reference Figure 2 The distributed jamming resource optimization scenario established in this step includes an airborne phased array radar, the target being detected, and multiple distributed jammers. The airborne phased array radar uses space-time adaptive processing technology to jointly suppress clutter and interference; the multiple distributed jammers carry out coordinated jamming on the airborne phased array radar from different directions; after receiving the jamming signal, the airborne phased array radar obtains the space-time power spectrum through space-time adaptive processing, and raises its suppression notch at the location of the jammer.

[0056] The relevant parameters include the operating parameters of the airborne phased array radar, target parameters, and jammer parameters, among which:

[0057] The operating parameters of an airborne phased array radar include: carrier frequency, pulse repetition period, number of array elements and number of coherently processed pulses, clutter covariance matrix, noise power and target spacetime steering vector;

[0058] Target parameters include: relative distance to the target, relative azimuth, and target echo power;

[0059] The jammer parameters include the distance and azimuth information of the distributed jammer relative to the airborne phased array radar.

[0060] The obtained parameters are used to calculate the value of the objective function in the subsequent construction.

[0061] Step 2: Construct a distributed interference resource decision vector.

[0062] Distributed jamming resource allocation, under conditions of limited power and time resources, determines the combination of jamming parameters for each jammer to produce the strongest possible suppression effect on the airborne phased array radar's space-time processing output. The equivalent jamming contribution of a distributed jammer to the radar receiver is mainly determined by its transmit power, jamming pattern, jamming repeater count, and sampling pulse width. Therefore, these parameters can be used as the basic decision vector for constructing the distributed jamming resource decision vector, the implementation of which includes:

[0063] 2.1) Set the number of distributed jammers to [number]. , No. The resource variables corresponding to the distributed jammer include normalized jamming power. Unified interference style Number of times of interference and forwarding and sampling pulse width ;

[0064] 2.2) Combine the set distributed jamming machine parameters into a vector to form a jamming resource decision vector. :

[0065] .

[0066] This interference resource decision vector can unify power resources, interference pattern resources, and time resources into the same optimization variable, providing a parameter basis for subsequent resource constraint establishment, equivalent interference covariance matrix construction, and objective function solution.

[0067] Step 3: Establish resource constraints.

[0068] In distributed cooperative jamming, the total transmit power available to the jammers is limited. Therefore, it is necessary to rationally allocate the jamming power of each distributed jammer under the condition of limited total power. The implementation includes:

[0069] 3.1) The single-station power constraint is set as follows:

[0070] , ,

[0071] in, For the number of distributed jammers, For the first The resource variables corresponding to the distributed jamming machine include normalized jamming power;

[0072] 3.2) In distributed cooperative jamming, multiple jammers act on the same target radar. If the jamming patterns are inconsistent, it can easily lead to inconsistencies in time-frequency structure and forwarding characteristics, weakening the cooperative jamming effect. Therefore, it is necessary to set a unified jamming pattern constraint: , To ensure that the jamming patterns of each jammer are identical,

[0073] in, The interference patterns are selected from the set of interference patterns that can be chosen by the distributed jammers. This is a set of interference patterns that can be selected by distributed jammers. To standardize interference patterns;

[0074] 3.3) The forwarding count constraint and sampling pulse width constraint are set as follows:

[0075] , , ,

[0076] in, To unify the set of possible forwarding counts corresponding to interference pattern s, To unify the set of optional sampling pulse widths corresponding to interference pattern s, For the first Number of times the distributed jammer interferes with forwarding. For the first The distributed jammer samples the pulse width.

[0077] 3.4) Distributed coordinated jamming typically needs to be completed within a limited radar pulse repetition period. Based on the constraints of the number of relays and the sampling pulse width, the time resource constraints are expressed as follows:

[0078] ,

[0079] in, This indicates the upper limit of time resources available for coordinated interference;

[0080] By defining the feasible range of distributed interference resource allocation through the above resource constraints, a clear feasible domain can be provided for subsequent resource optimization solutions, making the obtained resource allocation scheme engineering-executable. Specifically, power constraints ensure that the total interference resources do not exceed the limit, uniform interference pattern constraints ensure the consistency of the coordinated action of multiple jammers, forwarding count and sampling pulse width constraints ensure that the interference parameters are achievable, and time resource constraints ensure that the time occupied by each jammer does not exceed the system's allowable range.

[0081] Step 4: Construct a distributed interference resource allocation optimization model.

[0082] To construct the objective function of the interference resource allocation optimization model, it is necessary to consider the interference resource decision vector. Based on the parameter characteristics and resource constraints, the decision vector based on interference resources is analyzed. Constraint repair is performed; based on the spatiotemporal characteristics of airborne phased array radar spatiotemporal processing, a decision vector for interference resources is constructed. The target signal-to-clutter ratio (SNR) at the output of the airborne phased array radar is minimized based on the input. Its implementation includes:

[0083] 4.1) Based on the set resource constraints, the interference power is non-negatively truncated and normalized, and the unified interference pattern, forwarding count, and sampling pulse width are projected onto the corresponding feasible sets respectively. Interference decision vectors that meet the resource constraints are extracted. By uniformly mapping continuous resource variables and discrete interference parameters to the feasible solution space, it can be ensured that the interference decision vectors participating in subsequent optimization calculations all meet the constraints of power, pattern, forwarding count, sampling pulse width, and time resources, thus avoiding the particle swarm algorithm from searching in the infeasible region.

[0084] 4.2) Based on the interference resource decision vector and the Space-time steering vector of the distributed jammer Calculate the first The equivalent interference contribution of each distributed jammer to the airborne phased array radar's space-time processing input is calculated, and the equivalent interference contributions of all distributed jammers are superimposed to obtain the result from the jamming resource decision vector. Determined equivalent disturbance covariance matrix :

[0085] ,

[0086] in, The interference effectiveness coefficient is determined by the forwarding gain and the sampling pulse width. For the number of reposts The corresponding forwarding gain function, The radar pulse repetition period; For the first The resource variables corresponding to the distributed jamming machine include normalized jamming power; For a unified jamming pattern of distributed jammers, For the first Number of times the distributed jammer interferes with forwarding. For the first The distributed jammer samples the pulse width. This refers to the number of distributed jammers.

[0087] Disturbance resource decision vector It not only determines the power allocation of each distributed jammer, but also influences the equivalent jamming covariance matrix through jamming pattern, relay count, and sampling pulse width. The covariance matrix can characterize the comprehensive impact of distributed jamming signals on the space-time processing input of airborne phased array radar, and is the basis for subsequent evaluation of jamming effect and construction of objective function;

[0088] 4.3) Utilizing the aforementioned equivalent interference covariance matrix Calculate the decision vector for interfering resources The target signal-to-noise ratio at the output of the space-time adaptive processing under the action :

[0089] ,

[0090] in, For the target echo power, When the weight vector is empty, adaptive processing is performed. Here is the clutter covariance matrix. The noise covariance matrix is... The target spacetime guidance vector;

[0091] 4.4) Based on the target signal-to-noise ratio Define the target signal-to-clutter ratio at the output of the space-time adaptive processing of an airborne phased array radar. :

[0092] ,

[0093] in, Penalties for violating resource constraints This is the penalty coefficient.

[0094] This step defines the minimum target signal-to-clutter ratio at the output of the airborne phased array radar's space-time adaptive processing. It can be used to measure the impact of different jamming resource configurations on radar target detection performance. When the jamming covariance matrix has stronger energy and more obvious coupling with the target space-time steering vector, the target signal-to-clutter ratio at the radar output is usually lower, indicating that the jamming resource configuration is more effective.

[0095] Step 5: Encode the interference resource decision vector and initialize the algorithm parameters.

[0096] The interference resource decision vector, as the particle position in the particle swarm optimization algorithm, needs to be encoded so that each particle corresponds to a candidate distributed interference resource allocation scheme. This distributed interference resource allocation scheme includes parameters such as power allocation for each distributed jammer, uniform interference pattern, number of interference forwardings, and sampling pulse width. The particle position can completely represent the result of an executable interference resource configuration. Its implementation includes:

[0097] 5.1) The first The position of each particle The encoding is a set of candidate interference resource allocation schemes, represented as follows:

[0098] ,

[0099] in, For the first The first particle The amount of interference power allocated to the distributed jammer. For the first The first particle The number of times the distributed jammer forwards the interference. For the first The first particle The sampling pulse width of the distributed jammer;

[0100] 5.2) Based on the attribute that particle velocity can be used to represent the adjustment direction and adjustment magnitude of candidate resource allocation schemes in the search space, the position and velocity of the particles are updated to gradually approach a better interference resource allocation scheme:

[0101] Existing methods for updating particle position and velocity include integrated learning particle swarm optimization (PSO), artificial bee swarm optimization (ASO), firefly optimization, and PSO. This example uses, but is not limited to, the PSO algorithm to update particle position and velocity, and the specific implementation is as follows:

[0102] 5.2.1) Therefore, the first... The speed of each particle Let the change of the candidate interference resource allocation scheme in the optimization space be set as follows:

[0103] ,

[0104] in, For the first The first particle The velocity component corresponding to the interference power allocation of the distributed jammer. For the velocity components corresponding to the same disturbance pattern, For the first The first particle The velocity component corresponding to the number of jamming relays by a distributed jammer. For the first The first particle The velocity component corresponding to the sampling pulse width of the distributed jammer;

[0105] 5.2.2) Initialize the particle swarm optimization (PSO) algorithm parameters according to the composition of the interference resource decision vector:

[0106] Set the number of particles to Used to generate One candidate interference resource allocation scheme;

[0107] Set the maximum number of iterations. , used to limit the search termination condition of the algorithm;

[0108] Set the particle dimension to Two numbers are used to characterize the two numbers that need to be optimized in each candidate solution, since each interference resource decision vector contains One power allocation variable, one uniform interference pattern variable. The number of forwarding variables and The sampling pulse width variable, therefore the particle dimension satisfies: , The number of distributed jammers;

[0109] Set the first Initial position of each particle , used to represent the initial interference resource allocation scheme corresponding to the particle;

[0110] Set the first Initial velocity of each particle This is used to represent the initial adjustment direction and adjustment magnitude of the particle in the resource search space;

[0111] 5.3) Initialize the simulated annealing algorithm parameters based on the structure of the system particle position solutions and the optimization objective:

[0112] Set initial temperature This is used to control the probability of the algorithm accepting a non-optimal resource allocation scheme in the early stage, so that the search process has a certain ability to escape local optima;

[0113] Set the current temperature to , used to characterize the Annealing state after the next iteration;

[0114] Set the cooling coefficient to This is used to control the current temperature to gradually decrease with each iteration.

[0115] along with As the value decreases, the probability of the algorithm accepting a poor solution gradually decreases, causing the search process to gradually shift from global exploration in the early stage to local optimization in the later stage.

[0116] Step 6: Particle position update and optimal solution selection.

[0117] In the During the iteration, it is necessary to adjust the position of the first particle based on its current position, particle velocity, individual optimal position, and global optimal position. The search state of each particle is updated. Through velocity updates, the particle retains its original search trend while simultaneously referencing both the individual optimal resource allocation scheme and the global optimal resource allocation scheme, thereby adjusting its movement direction and amplitude in the interference resource search space. The simulated annealing algorithm is used to select and update the individual optimal position and the global optimal position, the implementation of which includes:

[0118] 6.1) Based on the iterative relationship of the particle swarm optimization algorithm, the th... After the nth iteration The speed of each particle and location :

[0119] ,

[0120] ,

[0121] in, The inertial weight is used to control the degree to which a particle inherits the search state from the previous iteration. and For two learning factors with different values, and Given two random numbers in the interval [0,1], Indicates the first The position corresponding to the historical optimal resource allocation scheme of each particle. This indicates the position corresponding to the current globally optimal resource allocation scheme. For the first After the nth iteration The velocity components corresponding to each particle For the first After the nth iteration The position components corresponding to each particle;

[0122] 6.2) Based on resource constraints Constraints are applied to ensure that the particle positions meet requirements such as power, forwarding pattern, number of forwardings, sampling pulse width, and time resources, thus obtaining the constrained particle positions. And calculate the first Particle position after the next iteration The target signal-to-noise ratio at the output of the space-time adaptive processing under the action :

[0123] ,

[0124] in, For the target echo power, When the weight vector is empty, adaptive processing is performed. Here is the clutter covariance matrix. The noise covariance matrix is... The target spacetime guidance vector;

[0125] 6.3) Calculate the first [unit / item] based on the optimal allocation model. After the nth iteration The objective function corresponding to each particle :

[0126] ,

[0127] in, This is a penalty term for violating resource constraints, used to characterize the degree to which the resource allocation scheme satisfies the constraints. This is the penalty coefficient;

[0128] 6.3) The first one after resource constraint processing The particle position in the next iteration is used as a new candidate interference resource allocation scheme, denoted as... The resource allocation scheme corresponding to the current optimal position of this particle is denoted as... Define the difference in objective functions between the two. :

[0129] ,

[0130] Since the objective function aims to minimize the target signal-to-noise ratio at the output, therefore The smaller the value, the more beneficial the new resource allocation scheme is to reducing the target signal-to-clutter ratio at the output of the airborne phased array radar.

[0131] 6.4) Based on the difference of the objective function The range of values ​​for is used to select an acceptable resource allocation scheme using the simulated annealing acceptance criterion, and the optimal position of the current individual is updated based on the accepted resource allocation scheme:

[0132] when This indicates that the new candidate interference resource allocation scheme is no greater than the resource allocation corresponding to the current individual's optimal position.

[0133] The proposal will be accepted. Find the resource allocation scheme corresponding to the current individual's optimal position, and update the current individual's optimal position. ;

[0134] when If this means that the resource allocation scheme corresponding to the current individual's optimal position is better than the new candidate interference resource allocation scheme, then according to probability... Choose whether to accept If random numbers satisfy Then accept Develop the current resource allocation plan and update the optimal position of the current individual. ;like If so, the current resource allocation scheme will be retained. ;

[0135] This step allows the algorithm to accept poor solutions with a certain probability, thereby enhancing its global search capability and avoiding excessive particle swarm optimization.

[0136] It gets trapped in a local optimum early on.

[0137] 6.5) After updating the individual optimal positions of each particle, update the global optimal position based on the magnitude of the objective function value:

[0138] If the global optimal position corresponds to the objective function value Greater than the objective function value corresponding to the current individual's optimal position Then retain the current globally optimal position. ;

[0139] If the global optimal position corresponds to the objective function value Less than the objective function value corresponding to the current individual's optimal position Then update the current global optimal position to ;

[0140] 6.6) After completing particle position updates, objective function calculations, individual optimal position updates, and global optimal position updates, the simulated annealing temperature is updated to obtain the current updated temperature: ,in, The cooling coefficient, Represented as the first The current temperature after the next iteration.

[0141] As the iterations proceed, the temperature gradually decreases, and the probability of the algorithm accepting a poor solution gradually decreases, causing the search process to gradually shift from the early global exploration to the later local fine-tuning optimization.

[0142] Step 7: Iteration terminates and the optimal solution is output.

[0143] Repeat step 6 to iteratively update the particle position, particle velocity, objective function value, individual optimal position, and global optimal position until the maximum number of iterations is reached. Output the optimal interference resource decision vector:

[0144] ,

[0145] in, Indicates the first The globally optimal position obtained in the next iteration. This represents the optimal resource allocation scheme obtained from particle swarm optimization. The scheme is a resource allocation table arranged by the distributed jammer number, and the resource allocation in the table includes: [The table contains information about the first resource allocation scheme]. Optimal normalized jamming power of distributed jammer Optimal unified interference pattern Optimal number of interference forwardings and optimal sampling pulse width It realizes the distributed collaborative interference resource allocation under the condition of limited interference resources.

[0146] Through the above steps, the distributed jamming resource allocation problem can be transformed into a position search problem in particle swarm optimization, achieving joint optimization of power allocation, jamming patterns, forwarding times, and sampling pulse width. During the iteration process, particles continuously update candidate resource allocation schemes based on their individual optimal positions and the global optimal position, and evaluate their suppression effect on the target signal-to-clutter ratio at the radar output end by combining resource constraints and the objective function. Simultaneously, a simulated annealing acceptance criterion is introduced to improve the algorithm's ability to escape local optima. When the number of iterations reaches its maximum value, the optimal jamming resource configuration scheme for each distributed jammer is obtained, thus providing executable parameter basis for distributed cooperative jamming implementation.

[0147] It should be noted that the step numbers in this specification and claims are only for the purpose of clearly describing the embodiments of the present invention and facilitating understanding, and their order is not limited.

[0148] Example 2: Distributed Interference Intelligent Resource Allocation System Based on SA-PSO Algorithm.

[0149] Reference Figure 3 This example includes: a scene establishment module 1, a parameter mapping module 2, a vector encoding module 3, an update and iteration module 4, and a location decoding module 5. The vector encoding module 3 includes: a resource mapping submodule 31 and a vector correction submodule 32. The update and iteration module 4 includes: an optimization evaluation submodule 41 and an optimization scheme selection submodule.

[0150] The working principle of the entire system is as follows:

[0151] The scenario establishment module 1 is used to establish a distributed jamming resource allocation scenario for space-time adaptive processing of airborne phased array radar, obtain parameter sets of airborne phased array radar, target to be detected and multiple distributed jammers, and use the obtained parameter sets as input to parameter mapping module 2.

[0152] The parameter mapping module 2 is used to convert the parameter set obtained by the scenario establishment module 1 into a distributed interference resource decision vector, and to establish resource constraints including power resource constraints, forwarding count constraints, sampling pulse width constraints and time resource constraints according to the established distributed interference resource scenario, and transmit the converted distributed interference resource decision vector to the vector encoding module 3.

[0153] The vector encoding module 3 is used to construct a distributed interference resource allocation optimization model using the output of the parameter mapping module 2 as basic parameters. The resource mapping submodule 31 is used to combine and map the distributed interference resource decision variables output by the parameter mapping module 2, including normalized interference power, unified interference pattern, interference forwarding count, and sampling pulse width, into particle position vectors in the SA-PSO algorithm according to the parameters of each distributed jammer, in a preset order. Each particle position corresponds to a candidate distributed interference resource allocation scheme, and the mapped particle positions are then passed to the vector correction submodule 32. The vector correction submodule 32 is used to perform feasibility correction on the particle position vectors based on the resource constraints established by the parameter mapping module 2 and the particle position information output by the resource mapping module 31. This includes: normalizing the normalized interference power to ensure that the power allocation of each distributed jammer meets the total power constraint; discretizing the interference pattern variables to unify the interference patterns of each distributed jammer; truncating or projecting the interference forwarding count and sampling pulse width, and constraining the combination of forwarding count and sampling pulse width according to the time resource constraint, and then passing the corrected particle position vectors to the update iteration module 4.

[0154] The update iteration module 4 is used to iteratively update the particle velocity and particle position according to the position update rules of the particle swarm algorithm, and calculate the objective function value of the distributed interference resource allocation optimization model constructed in the vector encoding module 3 based on the updated particle position. The optimization evaluation submodule 41 is used to evaluate the performance of the iterative candidate interference resource allocation schemes. Specifically, it decodes the particle positions to obtain information about each distributed jammer, including normalized interference power, uniform interference pattern, interference forwarding count, and sampling pulse width. Using this information, it calculates the equivalent interference contribution of each distributed jammer to the airborne phased array radar's space-time processing input, and superimposes these contributions to obtain an equivalent interference covariance matrix. This matrix is ​​then substituted into the target signal-to-clutter-to-noise ratio expression at the space-time adaptive processing output to calculate the objective function value corresponding to the current candidate resource allocation scheme. This objective function value is then passed to the optimization scheme selection submodule 42. The optimization scheme selection submodule 42, based on the objective function value output by the optimization evaluation submodule 41, completes the selection of candidate schemes and updates the optimal position. Specifically, it compares the objective function value of the current candidate scheme with the objective function value of the particle's historical best scheme. If the current candidate scheme is superior to the historical best scheme, the individual's optimal position is directly updated. If the current candidate scheme is inferior to the historical best scheme, the simulated annealing acceptance criterion is used to accept the candidate scheme, thereby enhancing the algorithm's ability to escape local optima. After updating the individual optimal position, the individual optimal positions of each particle are further compared, and the resource allocation scheme with the smallest objective function value is selected as the current global optimal position. The selection and update of candidate interference resource allocation schemes are completed, and the updated global optimal position is input to the position decoding module 5.

[0155] The location decoding module 5 is used to decode the global optimal location output by the update iteration module 4 into the optimal interference resource allocation scheme for each distributed jammer after the iteration number reaches the maximum iteration number, and output the normalized interference power, unified interference pattern, interference forwarding number and sampling pulse width corresponding to each distributed jammer.

[0156] Through the coordinated connection between the aforementioned modules, this embodiment can transform the scene parameters of the airborne phased array radar, the target to be detected, and the distributed jammers into optimizable jamming resource decision variables. Under constraints such as power, jamming pattern, number of forwards, sampling pulse width, and time resources, the SA-PSO algorithm is used to iteratively optimize the distributed jamming resources. Finally, the globally optimal position obtained from the search is decoded into the optimal resource configuration scheme for each distributed jammer, achieving joint optimization of normalized jamming power, unified jamming pattern, number of jamming forwards, and sampling pulse width, thereby improving the rationality and feasibility of distributed cooperative jamming resource allocation.

[0157] The effects of this invention can be further illustrated by the following simulation results:

[0158] I. Simulation Parameters

[0159]

[0160] II. Simulation Content

[0161] Simulation 1: Under the above simulation parameters, simulations were performed using the present invention and existing genetic algorithms for the noise convolution interference pattern, and the resource allocation results are shown in Table 1 and Table 2.

[0162] Table 1 shows the optimal interference resource allocation table after resource allocation using the existing genetic algorithm.

[0163]

[0164] As shown in Table 1, under the noise convolutional interference pattern, the optimal interference resource allocation for the first jammer is: normalized jammer power 0.367, interference forwarding times 16, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.105, jamming forwarding times 12, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.159, jamming forwarding times 16, and sampling pulse width... The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.369, jamming forwarding times 12, and sampling pulse width... ;

[0165] Table 2 shows the optimal interference resource allocation table after resource allocation using the method of the present invention.

[0166]

[0167] As shown in Table 2, under the noise convolutional interference pattern, the optimal interference resource allocation for the first jammer is: normalized jammer power 0.130, interference forwarding times 1, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.064, jamming forwarding times 1, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.477, jamming forwarding times 1, and sampling pulse width... The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.329, jamming forwarding times 1, and sampling pulse width... .

[0168] Simulation 2: Under the above simulation parameters, the optimal interference resource allocation table obtained under the noise convolution interference pattern was simulated using the present invention and existing genetic algorithms, respectively. The results are as follows: Figure 4 ,in:

[0169] 4(a) shows the two-dimensional power spectrum of the space-time after resource allocation by the existing genetic algorithm;

[0170] 4(b) is a cross-sectional view of the spatiotemporal power spectrum after resource allocation using the existing genetic algorithm;

[0171] 4(c) is a two-dimensional space-time power spectrum after resource allocation by the method of the present invention;

[0172] 4(d) is a cross-sectional view of the space-time power spectrum after resource allocation by the method of the present invention.

[0173] from Figure 4 (a) It can be seen that after resource allocation using the existing genetic algorithm, the distributed noise convolution interference energy is still adaptively suppressed by STAP processing on the spatiotemporal two-dimensional plane, and obvious notches are formed at multiple interference locations.

[0174] from Figure 4 (b) As can be seen, after resource allocation using the existing genetic algorithm, the suppression notch depths at the jamming azimuths of the jammer at angles [17°, 23°, 41°, 58°] are -47.092dB, -36.154dB, -47.932dB, and -29.714dB, respectively. Compared with the suppression notch depths of -50.083dB, -49.055dB, -49.542dB, and -32.185dB without jamming resource allocation, the existing genetic algorithm reduces the average suppression effect of STAP on distributed noise convolutional jamming by about 5dB, indicating that the genetic algorithm can affect the jamming effect of airborne phased array radar to a certain extent.

[0175] from Figure 4 (c) It can be seen that after using the method of the present invention to allocate interference resources, the residual response of distributed noise convolution interference in the space-time two-dimensional power spectrum is enhanced, and the notch depth at multiple interference azimuths is significantly reduced.

[0176] from Figure 4(d) As can be seen, after resource allocation using the method of this invention, the suppression notch depths at the interference locations [17°, 23°, 41°, 58°] are -44.773dB, -31.204dB, -34.960dB, and -27.668dB, respectively. Compared with no interference resource allocation, this invention reduces the average suppression effect of STAP on distributed noise convolution interference by approximately 11dB; compared with the interference resource optimization allocation results of existing genetic algorithms, this invention can further reduce the average suppression effect of STAP on interference by approximately 6dB.

[0177] Simulation results show that, under noisy convolutional interference scenarios, this invention can allocate a better interference resource scheme, significantly raising the STAP suppression notch at the interference azimuth, thereby reducing the target radar's spatiotemporal adaptive suppression capability against distributed interference. Compared with existing genetic algorithms, this invention has stronger interference resource optimization capabilities and interference effect improvement capabilities.

[0178] Simulation 3: Under the above simulation parameters, simulations were performed using the present invention and existing genetic algorithms for dense false target interference patterns. The resource allocation results are shown in Tables 3 and 4, where:

[0179] Table 3. Optimal Disturbance Resource Allocation Table after Resource Allocation by Existing Genetic Algorithm

[0180]

[0181] As shown in Table 3, under the dense decoy jamming pattern, the optimal jamming resource allocation for the first jammer is: normalized jammer power 0.323, jamming forwarding times 6, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.236, jamming forwarding times 8, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.265, jamming forwarding times 10, and sampling pulse width... The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.176, jamming forwarding times 14, and sampling pulse width... ;

[0182] Table 4. Optimal Interference Resource Allocation Table After Resource Allocation by the Method of the Invention

[0183]

[0184] As shown in Table 4, under the dense decoy jamming pattern, the optimal jamming resource allocation for the first jammer is: normalized jammer power 0.304, jamming forwarding times 12, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.392, jamming forwarding times 14, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.080, jamming forwarding times 10, and sampling pulse width... The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.224, jamming forwarding times 14, and sampling pulse width... .

[0185] Simulation 4: Under the above simulation parameters, the optimal interference resource allocation tables obtained under the dense false target interference pattern were simulated using the present invention and existing genetic algorithms, respectively. The results are as follows: Figure 5 ,in:

[0186] 5(a) shows the two-dimensional power spectrum of the space-time after resource allocation by the existing genetic algorithm;

[0187] 5(b) is a cross-sectional view of the spatiotemporal power spectrum after resource allocation using the existing genetic algorithm;

[0188] 5(c) is a two-dimensional space-time power spectrum after resource allocation by the method of the present invention;

[0189] 5(d) is a cross-sectional view of the space-time power spectrum after resource allocation by the method of the present invention.

[0190] from Figure 5 (a) It can be seen that after resource allocation using the existing genetic algorithm, the energy of distributed dense false target interference is still subject to adaptive suppression by STAP processing on the two-dimensional space-time plane, and obvious notches are formed at multiple interference azimuths.

[0191] from Figure 5 (b) As can be seen, after resource allocation using the existing genetic algorithm, the suppression notch depths at the jamming azimuths of the jammer at angles [17°, 23°, 41°, 58°] are -51.204dB, -47.213dB, -41.296dB, and -36.331dB, respectively. Compared with the suppression notch depths of -53.230dB, -51.682dB, -50.456dB, and -41.462dB without jamming resource allocation, the existing genetic algorithm reduces the average suppression effect of STAP against distributed dense false target jamming by approximately 5.2dB, indicating that the genetic algorithm can affect the jamming effect of airborne phased array radar to a certain extent.

[0192] from Figure 5(c) It can be seen that after using the method of the present invention to allocate interference resources, the residual response of distributed dense false target interference in the space-time two-dimensional power spectrum is enhanced, and the notch depth at multiple interference azimuths is significantly reduced.

[0193] from Figure 5 (d) As can be seen, after using the method of this invention to allocate interference resources, the suppression notch depths at the interference azimuths of the jammer at angles [17°, 23°, 41°, 58°] are -45.955 dB, -36.667 dB, -34.232 dB, and -36.196 dB, respectively. Compared with no interference resource allocation, this invention reduces the average suppression effect of STAP on distributed dense false target interference by 10.9 dB; compared with the interference resource optimization allocation results of existing genetic algorithms, this invention can further reduce the average suppression effect of STAP on interference by approximately 5.7 dB.

[0194] Simulation results show that, in dense decoy interference scenarios, this invention can allocate better interference resources, significantly raising the STAP suppression notch at the interference azimuth, thereby reducing the target radar's spatiotemporal adaptive suppression capability against distributed interference. Compared to existing genetic algorithms, this invention has stronger interference resource optimization capabilities and interference effect improvement capabilities.

[0195] Simulation 5: Under the above simulation parameters, simulations were performed using the present invention and existing genetic algorithms for intermittent sampling frequency shifting and forwarding interference patterns, respectively. The resource allocation results are shown in Tables 5 and 6.

[0196] Table 5. Optimal Disturbance Resource Allocation Table after Resource Allocation by Existing Genetic Algorithm

[0197]

[0198] As shown in Table 5, under the intermittent sampling frequency shift and forwarding jamming pattern, the optimal jamming resource allocation for the first jammer is: normalized jammer power 0.325, jamming forwarding times 12, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.315, jamming forwarding times 14, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.279, jamming forwarding times 10, and sampling pulse width... The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.081, jamming forwarding times 12, and sampling pulse width... ;

[0199] Table 6. Optimal Interference Resource Allocation Table After Resource Allocation by the Method of the Invention

[0200]

[0201] As shown in Table 6, under the intermittent sampling frequency shift and forwarding jamming pattern, the optimal jamming resource allocation result for the first jammer is: normalized jammer power 0.134, jamming forwarding times 8, and sampling pulse width... The optimal jamming resource allocation for the second jammer is: normalized jammer power 0.752, jamming forwarding times 10, and sampling pulse width... The optimal jamming resource allocation for the third jammer is: normalized jammer power 0.093, jamming forwarding times 6, and sampling pulse width. The optimal jamming resource allocation for the fourth jammer is: normalized jammer power 0.021, jamming forwarding times 14, and sampling pulse width... .

[0202] Simulation 6: Under the above simulation parameters, the optimal interference resource allocation tables obtained under the intermittent sampling frequency shifting and forwarding interference pattern were simulated using the present invention and existing genetic algorithms, respectively. The results are as follows: Figure 6 ,in:

[0203] 6(a) shows the two-dimensional space-time power spectrum after resource allocation using the existing genetic algorithm;

[0204] 6(b) is a cross-sectional view of the spatiotemporal power spectrum after resource allocation using the existing genetic algorithm;

[0205] 6(c) is a two-dimensional space-time power spectrum after resource allocation by the method of the present invention;

[0206] 6(d) is a cross-sectional view of the space-time power spectrum after resource allocation by the method of the present invention.

[0207] from Figure 6 (a) It can be seen that after resource allocation using the existing genetic algorithm, the distributed intermittent sampling frequency shift forwarding interference energy is still adaptively suppressed by STAP processing in the two-dimensional space-time plane, and obvious notches are formed at multiple interference locations.

[0208] from Figure 6(b) As can be seen, after resource allocation using the existing genetic algorithm, the suppression notch depths at the jamming azimuths of the jammer at angles [17°, 23°, 41°, 58°] are -49.370dB, -35.528dB, -40.459dB, and -39.953dB, respectively. Compared with the suppression notch depths of -53.755dB, -45.920dB, -48.519dB, and -40.011dB without jamming resource allocation, the existing genetic algorithm reduces the average suppression effect of STAP on distributed intermittent sampling frequency shifting and forwarding jamming by approximately 5.7dB, indicating that the genetic algorithm can affect the jamming effect of airborne phased array radar to a certain extent.

[0209] from Figure 6 (c) It can be seen that after using the method of the present invention to allocate interference resources, the residual response of distributed intermittent sampling frequency shift forwarding interference in the space-time two-dimensional power spectrum is enhanced, and the notch depth at multiple interference azimuths is significantly reduced.

[0210] from Figure 6 (d) As can be seen, after using the method of this invention to allocate interference resources, the suppression notch depths at the interference azimuths of the jammer at angles [17°, 23°, 41°, 58°] are -35.405dB, -31.962dB, -37.238dB, and -33.024dB, respectively. Compared with no interference resource allocation, this invention reduces the average suppression effect of STAP on distributed intermittent sampling frequency shifting and forwarding interference by approximately 12.5dB; compared with the interference resource optimization allocation results of existing genetic algorithms, this invention can further reduce the average suppression effect of STAP on interference by approximately 6.8dB.

[0211] Simulation results show that, in the intermittent sampling frequency shifting and forwarding jamming scenario, this invention can allocate a better jamming resource scheme, significantly raising the STAP suppression notch at the jamming azimuth, thereby reducing the target radar's spatiotemporal adaptive suppression capability against distributed jamming. Compared with existing genetic algorithms, this invention has stronger jamming resource optimization capabilities and jamming effect improvement capabilities.

Claims

1. A distributed interference intelligent resource allocation method based on the SA-PSO algorithm, characterized in that, include: (1) Establish a distributed jamming resource allocation scenario for space-time adaptive processing of airborne phased array radar, and obtain the parameter sets of airborne phased array radar, target to be detected and multiple distributed jammers; (2) Based on the situation parameter set and the type of interference resources to be allocated, construct a distributed interference resource decision vector; (3) Establish resource constraints based on the interference resource decision vector; (4) The interference resource decision vector that satisfies the constraints is mapped to the equivalent interference covariance matrix of the input end of the airborne phased array radar space-time adaptive processing, and a distributed interference resource allocation optimization model is constructed with the goal of reducing the target signal-to-clutter ratio at the output end of the space-time adaptive processing. (5) Encode the interference resource decision vector into particle positions in the algorithm and initialize the algorithm parameters; (6) Update the particle positions according to the particle swarm update rule, calculate the target signal-to-noise ratio by constraining the updated particle positions, and update the individual optimal position and the global optimal position using the simulated annealing acceptance criterion. (7) Repeat step (6) until the termination condition is met, decode the global optimal position into the optimal interference resource allocation scheme of each distributed jammer and output it.

2. The method according to claim 1, characterized in that: The (1) establishes a distributed jamming resource allocation scenario for space-time adaptive processing of airborne phased array radar, which includes airborne phased array radar, the target being detected and multiple distributed jammers; the airborne phased array radar uses space-time adaptive processing technology to jointly suppress clutter and interference, and the distributed jammers carry out coordinated jamming of the airborne phased array radar from different directions. The parameter set obtained in (1) for the airborne phased array radar, the target to be detected, and multiple distributed jammers includes: 1a) Obtain radar operating parameters including carrier frequency and pulse repetition period. Number of array elements and number of coherent processing pulses; 1b) Obtaining space-time processing parameters, including clutter covariance matrix. Noise power and target spacetime guidance vector The target parameters include the target relative distance, relative azimuth, and target echo power. 1c) Obtaining jammer parameters includes the first The range, azimuth, and set of selectable jamming patterns of the distributed jammer relative to the airborne phased array radar.

3. The method according to claim 1, characterized in that, In step (2), a distributed interference resource decision vector is constructed based on the situation parameter set and the type of interference resource to be allocated. Its implementation includes: 2a) Let the number of distributed jammers be... , No. The resource variables corresponding to the distributed jammer include normalized jamming power. Unified interference pattern s, number of interference forwardings and sampling pulse width ; 2b) Combine the distributed jammer parameters set in 2a) into a vector to form a jamming resource decision vector. : 。 4. The method according to claim 1, characterized in that, The establishment of resource constraints based on the interference resource decision vector in (3) includes the following implementation: 3a) The single-station power constraint is set as follows: ， ， in, For the number of distributed jammers, For the first The resource variables corresponding to the distributed jamming machine include normalized jamming power; 3b) Set uniform interference style constraints: , , in, The interference patterns are selected from the set of interference patterns that can be chosen by the distributed jammers. This is a set of interference patterns that can be selected by distributed jammers. To standardize interference patterns; 3c) The forwarding count constraint and the sampling pulse width constraint are set as follows: ， ， ， in, To unify the set of possible forwarding counts corresponding to interference pattern s, To unify the set of selectable sampling pulse widths corresponding to interference pattern s, For the first Number of times the distributed jammer interferes with forwarding. For the first The distributed jammer samples the pulse width. 3d) Based on the forwarding count constraint and the sampling pulse width constraint, the time resource constraint is set as follows: ， in, This indicates the upper limit of time resources available for coordinated interference.

5. The method according to claim 1, characterized in that, The implementation of the distributed interference resource allocation optimization model in (4) includes: 4a) Based on the set resource constraints, the interference power is non-negatively truncated and normalized, and the uniform interference pattern, forwarding times and sampling pulse width are projected onto the corresponding feasible sets to extract the interference decision vector that meets the resource constraints. 4b) Based on the interference resource decision vector and the Space-time steering vector of the distributed jammer Calculate the first The equivalent interference contribution of each distributed jammer to the airborne phased array radar's space-time processing input is calculated, and the equivalent interference contributions of all distributed jammers are superimposed to obtain the result from the jamming resource decision vector. Determined equivalent disturbance covariance matrix : ， in, The interference effectiveness coefficient is determined by the forwarding gain and the sampling pulse width. For the number of reposts The corresponding forwarding gain function, The radar pulse repetition period; For the first The resource variables corresponding to the distributed jamming machine include normalized jamming power; 's' indicates a uniform interference pattern. For the first Number of times the distributed jammer interferes with forwarding. For the first The distributed jammer samples the pulse width. This refers to the number of distributed jammers. 4d) Using the equivalent interference covariance matrix constructed in 4b), define a method to minimize the target signal-to-clutter ratio at the output of the airborne phased array radar's space-time adaptive processing. : ， in, To interfere with resource decision vectors The target signal-to-noise ratio at the output of the space-time adaptive processing under the action of [the system] Penalties for violating resource constraints The penalty coefficient is... For the target echo power, When the weight vector is empty, adaptive processing is performed. Here is the clutter covariance matrix. The noise covariance matrix is... The target spacetime guiding vector.

6. The method according to claim 1, characterized in that, In step (5), the interference resource decision vector is encoded as the particle position in the algorithm, and the algorithm parameters are initialized. The implementation of this step includes: 5a) The first The position of each particle The encoding is a set of candidate interference resource allocation schemes, represented as follows: ， in, For the first The first particle The amount of interference power allocated to the distributed jammer. For the first The first particle The number of times the distributed jammer forwards the interference. For the first The first particle The sampling pulse width of the distributed jammer; 5b) The first The speed of each particle Let the change of the candidate interference resource allocation scheme in the optimization space be set as follows: ， in, For the first The first particle The velocity component corresponding to the interference power allocation of the distributed jammer. For the velocity components corresponding to the same disturbance pattern, For the first The first particle The velocity component corresponding to the number of jamming relays by a distributed jammer. For the first The first particle The velocity component corresponding to the sampling pulse width of the distributed jammer; 5c) Initialize the number of particles in the particle swarm optimization algorithm Maximum number of iterations Particle dimension initial position of the particle and the initial velocity of the particles Where the particle dimension satisfies: , The number of distributed jammers; 5d) The cooling coefficient of the initial simulated annealing algorithm is: The current temperature is The initial temperature is .

7. The method according to claim 1, characterized in that, The process of constraining the updated particle positions to calculate the target signal-to-noise ratio in step (6) includes the following: 6a) in the In the iteration, the particle's velocity and position are updated respectively, resulting in the... After the nth iteration The speed of each particle and location : ， ， in, For inertial weights, and For two learning factors with different values, and Given two random numbers in the interval [0,1], Indicates the first The position corresponding to the historical optimal resource allocation scheme of each particle. This indicates the position corresponding to the current globally optimal resource allocation scheme. For the first After the nth iteration The velocity components corresponding to each particle For the first After the nth iteration The position components corresponding to each particle; 6b) Based on resource constraints After applying constraints, the first allocation is calculated based on the optimal allocation model. After the nth iteration The target signal-to-noise ratio corresponding to each particle : ， in, To interfere with resource decision vectors The target signal-to-noise ratio at the output of the space-time adaptive processing under the action of [the system] Penalties for violating resource constraints The penalty coefficient is... For the target echo power, When the weight vector is empty, adaptive processing is performed. Here is the clutter covariance matrix. The noise covariance matrix is... The target spacetime guiding vector.

8. The method according to claim 1, characterized in that, The method for updating the individual optimal position and the global optimal position using the simulated annealing acceptance criterion in (6) includes the following implementation: 6c) The constrained first The candidate interference resource allocation scheme corresponding to the particle position in the next iteration is denoted as... Let the resource allocation scheme corresponding to the current individual's optimal position be denoted as Define the objective function difference : ， 6d) Based on the difference of the objective function Given the range of values, select an acceptable resource allocation scheme and update the current optimal position of the individual based on the accepted resource allocation scheme: when Then accept Find the resource allocation scheme corresponding to the current individual's optimal position, and update the current individual's optimal position. ; when When, then according to probability Choose whether to accept : If random number satisfy Then accept Develop the current resource allocation plan and update the optimal position of the current individual. ; like If so, the current resource allocation scheme will be retained. ; 6e) Based on the magnitude of the objective function value, select the position that accepts the global optimum: If the global optimal position corresponds to the objective function value Greater than the objective function value corresponding to the current individual's optimal position Then retain the current globally optimal position. ; If the global optimal position corresponds to the objective function value Less than the objective function value corresponding to the current individual's optimal position Then update the current global optimal position to ; 6f) After completing 6a)~6e), update the temperature according to the following formula: ； in, The cooling coefficient, Represented as the first The current temperature after the next iteration.

9. A distributed interference intelligent resource allocation system based on the SA-PSO algorithm, characterized in that, include: The scenario establishment module is used to establish a distributed jamming resource allocation scenario for airborne phased array radar space-time adaptive processing, and to obtain the parameter sets of airborne phased array radar, the target to be detected, and multiple distributed jammers; The parameter mapping module is used to transform the acquired parameter set into a distributed interference resource decision vector, and to construct resource constraints and collaborative constraints based on the distributed interference resource decision vector. The vector encoding module is used to construct a distributed interference resource allocation optimization model, encode the distributed interference decision vector into the initialization parameters of the SA-PSO algorithm, and correct the particle swarm parameters after iteration according to the constraint rules of the parameter mapping module. The update iteration module is used to iteratively update parameters such as particle swarm position and velocity, and to update the individual optimal position and the global optimal position using the simulated annealing acceptance criterion. The location decoding module is used to output the optimal interference resource allocation scheme.

10. The system according to claim 9, characterized in that: The vector encoding module includes: The resource mapping submodule is used to map the existing allocation scheme to information such as the position and velocity of the particle swarm in the SA-PSO algorithm, and to clarify the initialization parameters and particle swarm iteration criteria of the algorithm. The vector correction submodule is used to adjust the vector parameter values ​​according to the current resource constraints, including the set of distributed jammer jamming patterns, the set of forwarding jamming counts, and the set of sampling pulse lengths of forwarding jamming, until they can be mapped to the set. The update and iteration module includes: The optimization evaluation submodule is used to establish a signal-to-noise ratio function with minimization as the optimal evaluation criterion, mathematically model and quantify the interference contribution of the distributed jammer to the radar system, and transform the interference resource allocation problem into a constrained optimization problem. The optimization scheme selection submodule is used to update the resource optimization allocation scheme corresponding to the current best individual position and the resource optimization allocation scheme corresponding to the current best global position based on the iterative resource optimization allocation scheme and the simulated annealing acceptance criterion.

Images