MPC-based radar radiation power control method, device and equipment

Abstract

The invention provides a radar radiation power control method, device and equipment based on MPC. The method comprises: establishing a target model; the target model comprises a target radiation model and a target state model, and the process of the target radiation signal follows the Poisson process; under a radar networking architecture, constructing a measurement model through the target state model; and at a preset decision moment, measuring tracking precision through a measurement model, a model prediction control MPC algorithm and a posterior Cramer-Rao bound PCRLB, judging whether a central station in the radar networking architecture needs active tracking or not, and obtaining a radar radiation control result. According to the invention, by combining the target radiation model, the target state model and the measurement model, the tracking precision and the anti-interference capability are effectively improved, on the basis, the MPC algorithm and the PCRLB are adopted to measure the tracking precision, the optimization of the active tracking decision process is realized, and the resource utilization rate and the tracking effect are improved.

Description

Technical Field

[0001] The present invention relates to the field of radar signal processing technology, and in particular to a radar radiation power control method, device and equipment based on MPC (Model Predictive Control). Background Art

[0002] In the field of radar signal processing, target tracking is a crucial technology, which is widely used in military reconnaissance, civil aviation monitoring, traffic management and other aspects. According to the different radar systems, target tracking technology is mainly divided into two categories: passive tracking and active tracking. Active tracking technology has performed well in many application scenarios due to its high precision, all-weather working ability and real-time positioning ability. However, its significant radiation signal makes the radar easily become the target of enemy attack, which seriously threatens the survivability of the radar. In contrast, passive tracking technology effectively improves the stealth and survivability of the radar by only using the radiation signal of the target for detection, while reducing the cost, but its lack of tracking accuracy limits its scope of application. Therefore, how to combine the advantages of active tracking and passive tracking to achieve efficient and concealed target tracking has become a technical problem that needs to be solved urgently.

[0003] In order to overcome the above technical difficulties, researchers have proposed a strategy to build a passive / active radar network and perform radiation control. This strategy aims to balance the accuracy requirements of active tracking and the concealment requirements of passive tracking by optimizing the radar's radiation pattern. For example, a literature has proposed a radar / electronic support measures (ECM) collaborative tracking method under radar radiation control. By controlling the radar to perform intermittent radiation, it not only maintains the continuity of tracking but also reduces the risk of being detected by the enemy. Another literature models the tracking problem as a partially observable Markov decision process (POMDP), using the posterior Cramer-Rao bound and hidden Markov model filter to predict future target tracking accuracy and radar radiation cost, so as to achieve more intelligent resource allocation. The introduction of these methods marks initial progress in the collaborative optimization of radar radiation control and target tracking.

[0004] Although the above methods have improved the comprehensive performance of radar networks to a certain extent, there are still many challenges. First, the decision-making process of existing methods is relatively simple and difficult to adapt to complex and changeable battlefield environments and target behavior patterns. Secondly, the cost functions of these methods often rely on the real state of the target for comparative evaluation. In practical applications, the real state of the target is often unknown, which limits the practicality and accuracy of the methods. In addition, some algorithms attempt to generate virtual states by simulating and extrapolating the radiation model and motion model of the target, and calculate the tracking accuracy based on this. Although it solves the problem of unknown real state to a certain extent, in the case of model mismatch, the performance of the algorithm may drop sharply, resulting in poor tracking effect. Therefore, developing a target tracking and radar radiation control method that can adapt to complex scenarios, does not rely on the real state of the target, and is robust to model mismatch has become an important direction of current research. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a radar radiation power control method, device and equipment based on MPC.

[0006] The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0007] In a first aspect, the present invention provides a radar radiation power control method based on MPC, comprising:

[0008] Establishing a target model; the target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys a Poisson process, and the target state model is described by the approximate uniform linear motion of the target;

[0009] Under the radar networking architecture, a measurement model is constructed through a target state model; the measurement model includes: a target-based measurement model and a central station-based measurement model;

[0010] At the preset decision moment, the tracking accuracy is measured through the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB to determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

[0011] Optionally, the target radiation model is expressed as:

[0012] t=[t1,t2,…,t K ];

[0013] Among them, t represents the time set of the target radiation signal, K represents the total number of target radiation signals, t K Indicates the time when the target radiates the signal for the Kth time;

[0014] The target state model is expressed as:

[0015]

[0016] in, Indicates that the target is at t k The state at time t k represents the time when the target radiates the signal for the kth time, Indicates that the target is at t k-1 The state of the moment, Indicates that the target is at t k-1 The state transfer matrix at time t k-1 It indicates the moment when the target radiates the signal for the k-1th time. Indicates that the target is at t k-1 The process noise at the moment.

[0017] Optionally, the radar networking architecture includes: a central station and multiple auxiliary stations;

[0018] The central station is used to radiate signals to the outside and receive signals radiated by the target, and the auxiliary station is used to receive signals radiated by the target and the central station; the radar networking architecture uses an extended Kalman filter to track the target.

[0019] Optionally, the target-based measurement model is expressed as:

[0020]

[0021] in, Indicates that the nth auxiliary station is at the target t k The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the target t k The zero-mean measurement noise obtained when the signal is radiated at time t k It indicates the time when the target is at the kth radiation signal. Indicates that the target is at t k The state of the moment;

[0022] The measurement model based on the central station is expressed as:

[0023]

[0024] in, Indicates that the nth auxiliary station is at the central station T m The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the central station T m The zero-mean measurement noise obtained when the signal is radiated at time instant, Indicates T m The state of the target at time T m Indicates the time when the central station radiates the signal for the mth time.

[0025] Optionally, at a preset decision moment, the tracking accuracy is measured by a measurement model, a model predictive control MPC algorithm, and a posterior Cramer-Rao bound PCRLB to determine whether the central station in the radar networking architecture needs to perform active tracking and obtain radar radiation control results, including:

[0026] Build a target prediction model based on the target's historical external radiation data;

[0027] The cost function is constructed based on the measurement model, target prediction model and the posterior Cramer-Rao bound PCRLB;

[0028] The cost function and model predictive control MPC algorithm are used to determine whether the current tracking accuracy meets the preset accuracy threshold, and based on the judgment result, it is determined whether the central station needs to perform active tracking to obtain the radar radiation control result.

[0029] Alternatively, the cost function is expressed as:

[0030]

[0031] in, represents the value of the cost function, Indicates that the target is at the preset final time t end Status information when represents the prior information, G D1 Indicates passive measurement information, G D2 Represents active measurement information; where G D1 The target prediction model and the target-based measurement model are used to construct G D2 It is constructed using a target prediction model and a measurement model based on a central station.

[0032] Optionally, using a cost function and a model predictive control MPC algorithm, determining whether the current tracking accuracy meets a preset accuracy threshold, and determining whether the central station needs to perform active tracking based on the judgment result to obtain a radar radiation control result, including:

[0033] The decision coefficients in the active measurement information in the control cost function are all 0, and the cost function is processed by preset processing to obtain an error value, and the error value is used as the current tracking accuracy; the preset processing is to sequentially perform inversion, trace and square root processing on the cost function;

[0034] Determine whether the error value is less than a preset accuracy threshold;

[0035] When the error value is less than the preset accuracy threshold, it is determined that the central station does not need to perform active tracking, and the absence of active tracking is taken as the radar radiation control result;

[0036] When the error value is greater than a preset accuracy threshold, it is determined that the central station needs to perform active tracking, and the need for active tracking is taken as a radar radiation control result.

[0037] Optionally, when the error value is greater than a preset accuracy threshold, it is determined that the central station needs to perform active tracking, and the need for active tracking is used as a radar radiation control result, including:

[0038] When the error value is greater than the preset accuracy threshold, the Euclidean distance between the central station and the estimated target position at the preset decision time is calculated according to the position of the central station and the estimated target position, and the decision distance value is obtained;

[0039] Sort the decision distance values ​​from small to large to obtain the distance sorting result;

[0040] The decision coefficients in the active measurement information corresponding to the distance sorting results are set to 1, and the error value is recalculated to obtain an updated error value;

[0041] The preset decision time corresponding to when the update error value is greater than the preset accuracy threshold is used as the start time of active tracking of the central station;

[0042] The start time of active tracking is taken as the radar radiation control result.

[0043] In a second aspect, the present invention provides a radar radiation power control device based on MPC, and the radar radiation power control device based on MPC includes: a model building unit and a decision judgment unit;

[0044] The model building unit is used to: establish a target model; the target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys the Poisson process, and the target state model is described by the approximate uniform linear motion of the target;

[0045] The model building unit is also used to: build a measurement model through a target state model under a radar networking architecture; the measurement model includes: a target-based measurement model and a central station-based measurement model;

[0046] The decision judgment unit is used to: at the preset decision moment, measure the tracking accuracy through the measurement model, model predictive control MPC algorithm and posterior Cramer-Rao bound PCRLB, determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

[0047] In a third aspect, the present invention provides an MPC-based radar radiation power control device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the MPC-based radar radiation power control device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the MPC-based radar radiation power control method in the first aspect as described above.

[0048] The present invention provides a radar radiation power control method, device and equipment based on MPC. Among them, a radar radiation power control method based on MPC includes: establishing a target model; the target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys a Poisson process, and the target state model is described by the approximate uniform linear motion of the target; under the radar networking architecture, a measurement model is constructed through the target state model; the measurement model includes: a target-based measurement model and a central station-based measurement model; at a preset decision moment, the tracking accuracy is measured through the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB, and it is determined whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result. In the present invention, by combining the target radiation model, the target state model and the measurement model, accurate tracking of the target is achieved, and the tracking accuracy and anti-interference ability are effectively improved. On this basis, the MPC algorithm and PCRLB are used to measure the tracking accuracy, and the optimization of the active tracking decision process is achieved, and the resource utilization and tracking effect are improved.

[0049] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic diagram of a flow chart of a radar radiation power control method based on MPC provided in an embodiment of the present invention;

[0051] Figure 2 The spatial position relationship diagram of the target away from the radar network is shown exemplarily;

[0052] Figure 3 The spatial position relationship diagram of the target close to the radar network is shown exemplarily;

[0053] Figure 4 The spatial position relationship diagram of a target approaching and then moving away from the radar network is shown as an example;

[0054] Figure 5 The comparison result diagram of passive tracking and decision tracking when the target is far away from the radar network is shown as an example;

[0055] Figure 6The comparison result diagram of passive tracking and decision tracking when the target is close to the radar network is shown as an example;

[0056] Figure 7 The comparison diagram of the results of passive tracking and decision tracking when the target first approaches and then moves away from the radar network is shown as an example;

[0057] Figure 8 A schematic diagram of the structure of a radar radiation power control device based on MPC provided by an embodiment of the present invention;

[0058] Fig. 9 A schematic structural diagram of a radar radiation power control device based on MPC provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0060] In order to optimize the active tracking decision process and improve resource utilization and tracking effect, an embodiment of the present invention provides a radar radiation power control method based on MPC. Figure 1 The present invention provides a flow chart of a radar radiation power control method based on MPC. Figure 1 As shown, the method includes:

[0061] S101. Establish a target model, where the target model includes a target radiation model and a target state model.

[0062] The process of the target radiation signal obeys the Poisson process, and the target state model is described by the target's approximate uniform linear motion.

[0063] Optionally, in this embodiment, the initial state of the target can be expressed as Among them, x0, y0, z0 represent the horizontal coordinate, vertical coordinate and vertical coordinate of the target in the initial state respectively. Respectively represent the lateral velocity, longitudinal velocity and vertical velocity of the target in the initial state, and T represents the transpose of the matrix. In this embodiment, an approximate uniform linear motion model is used to describe the target motion to obtain a target state model. Since the process of the target radiation signal obeys the Poisson process, the time set t of the target radiation signal is expressed as:

[0064] t=[t1,t2,…,t K ];

[0065] Among them, t represents the time set of the target radiation signal, K represents the total number of target radiation signals, t K Indicates the time when the target radiates the signal for the Kth time;

[0066] The target state model is expressed as:

[0067]

[0068] in, Indicates that the target is at t k The state at time t k represents the time when the target radiates the signal for the kth time, Indicates that the target is at t k-1 The state of the moment, Indicates that the target is at t k-1 The state transfer matrix at time t k-1 It indicates the moment when the target radiates the signal for the k-1th time. Indicates that the target is at t k-1 The process noise at the moment.

[0069] S102. Under the radar networking architecture, a measurement model is constructed through a target state model.

[0070] The measurement models include: target-based measurement model and central station-based measurement model.

[0071] It should be noted that the radar networking architecture includes: a central station and multiple auxiliary stations;

[0072] The central station is used to radiate signals to the outside and receive signals radiated by the target, and the auxiliary station is used to receive signals radiated by the target and the central station; the radar networking architecture uses an extended Kalman filter to track the target.

[0073] In this embodiment, the position of the central station can be expressed as s = [x, y, z] T , the position of the nth auxiliary station can be expressed as s n =[x n ,y n ,z n ] T ,n∈[1,…,N](N is known). Where x,y,z represent the horizontal coordinate, vertical coordinate and vertical coordinate of the central station respectively. n ,y n ,z n They respectively represent the horizontal coordinate, vertical coordinate and ordinate of the nth auxiliary station, and N represents the total number of auxiliary stations.

[0074] Assume that the target is at t k The signal is transmitted at all times, and the central station is set as the reference node. The measurement obtained by the nth auxiliary station based on the target radiation signal, that is, the target-based measurement model can be expressed as:

[0075]

[0076] in, Indicates that the nth auxiliary station is at the target t k The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the target t k The zero-mean measurement noise obtained when the signal is radiated at time t k It indicates the time when the target is at the kth radiation signal. Indicates that the target is at t k The state of the moment. It can be calculated through fuzzy function estimation theory.

[0077] in, It represents the difference between the Euclidean distance from the target to the central station and the Euclidean distance from the target to the nth auxiliary station divided by the signal propagation speed. Indicates that the target measured by the nth auxiliary station is at t k The arrival frequency difference of the radiation signal at time , can be calculated using the Doppler frequency shift.

[0078] The measurement model based on the central station is expressed as:

[0079]

[0080] in, Indicates that the nth auxiliary station is at the central station T m The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the central station T m The zero-mean measurement noise obtained when the signal is radiated at time instant, Indicates T m The state of the target at time T m Indicates the time when the central station radiates the signal for the mth time.

[0081] in, It represents the sum of the Euclidean distances from the target to the central station and from the target to the nth auxiliary station measured by the nth auxiliary station. It indicates the elevation angle of the central station to the target measured by the nth auxiliary station. It can be obtained using the maximum likelihood estimation method.

[0082] S103. At the preset decision moment, the tracking accuracy is measured by the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB to determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

[0083] The embodiment of the present invention provides a radar radiation power control method based on MPC, which realizes accurate tracking of the target by combining the target radiation model, the target state model and the measurement model, effectively improves the tracking accuracy and anti-interference capability, and on this basis adopts the MPC algorithm and PCRLB to measure the tracking accuracy, realizes the optimization of the active tracking decision process, and improves the resource utilization and tracking effect.

[0084] Optionally, S103 may specifically include:

[0085] Build a target prediction model based on the target's historical external radiation data;

[0086] The cost function is constructed based on the measurement model, target prediction model and the posterior Cramer-Rao bound PCRLB;

[0087] The cost function and model predictive control MPC algorithm are used to determine whether the current tracking accuracy meets the preset accuracy threshold, and based on the judgment result, it is determined whether the central station needs to perform active tracking to obtain the radar radiation control result.

[0088] In this embodiment, it is assumed that the time T m , m∈[1,…,M] is the decision time, the central station needs to predict the radiation model and state model of the target. The central station will count the number of external radiations of the target in the past Δt time And use this to estimate the target at time T m Assuming that the target is to radiate signals according to the Poisson distribution process, and the maximum likelihood estimate of the Poisson distribution parameter is its average value, then at time T m About the target Poisson process parameters The maximum likelihood estimate of is:

[0089]

[0090] According to the Poisson process parameters and the final decision time T ω It can be estimated that the target is at T ω The number of radiation times within a certain period is Np, and the estimated target radiation time set can be expressed as Indicates the estimated time when the target radiates signals outward for the Npth time.

[0091] While estimating the parameters of the target radiation Poisson distribution, the central station also estimates the target at T m Status of the moment

[0092]

[0093] in, Indicates that the central station is based on the target T m Passive tracking results obtained based on the radiation information at each moment.

[0094] according to and the estimated target radiation time set It can be estimated that the target is at T ω The state set when the signal is radiated in time and the state information set at the decision time Indicates that the estimated target is Status information when Indicates that the estimated target is at T M Status information at all times.

[0095] In practical applications, since the real state of the target cannot be known, the embodiment of the present invention adopts the posterior Cramer-Rao bound (PCRLB) to measure the tracking accuracy, and the PCRLB of the k-th radiation signal of the target is estimated to be:

[0096]

[0097] in, Indicates the estimated time when the target radiates the signal for the kth time The central station estimates the target state based on the measurement and The mathematical expectation of the difference between the target state at each moment. express The target measurement information set obtained by the central station at all times, Indicates that the radar Measurements obtained at all times The estimated target is The state of the moment.

[0098] Represents the estimated target Status of the moment The Bayesian information matrix of can be calculated iteratively as:

[0099]

[0100] in, and Respectively The Fisher information matrix of the central station’s prior information and measurement information about the target at each moment.

[0101] According to the above formula, PCRLB needs to be updated once each time a new measurement information is obtained, and an inverse operation needs to be performed each time the update is performed. ω The target radiates Np signals in total within a certain period of time, and multiple inversion operations are required, which greatly affects the operation speed. The multiple inversion processes are approximated and directly from T m Always get T ω The PCRLB at time t, approximates the multiple inversion processes to a single inversion and multiple accumulations, greatly reducing the complexity of the algorithm. The final cost function after approximate inversion is expressed as:

[0102]

[0103] in, represents the value of the cost function, Indicates that the target is at the preset final time t end Status information when represents the prior information, G D1 Indicates passive measurement information, G D2 Represents active measurement information; where G D1 The target prediction model and the target-based measurement model are used to construct G D2 It is constructed using a target prediction model and a measurement model based on a central station.

[0104] In an embodiment of the present invention, The acquisition process of can refer to the relevant content of formula 3-12 on page 30 of the paper "Research on Resource Allocation Algorithm in Cognitive Radar". It can be expressed as:

[0105]

[0106] in, Indicates that from T m Time to t end The process error matrix at time , Indicates that the target is at T m The state transfer matrix at time , Indicates T m The value of the cost function at time .

[0107] Item 2G D1 is the passive measurement information, indicating that at the final moment T ω The internal target radiates Np signals outward in total. Each time the target radiates a signal, the radar performs passive tracking, which is specifically expressed as:

[0108]

[0109] express The Jacobian matrix of express exist The value of the moment. for exist The value of the moment, Indicates that the prediction target is The state information at time j indicates the estimated jth outward radiation signal of the target.

[0110] The third term of the cost function is the information of active measurement, but not every decision moment will choose to turn on the power, so it is necessary to determine which decision moment needs to turn on the power.

[0111] Set the decision variable u for the mth decision m for:

[0112]

[0113] Among them, L m Indicates that the mth extrapolation simulation contains L m A startup decision point, u i =1 indicates whether the i-th decision point in the m-th decision is powered on. D2 It can be expressed as:

[0114]

[0115] in, express The Jacobian matrix of express In T m The value of the moment, In T m The value of the moment, Indicates that the prediction target is in T i The state information at the moment, i represents the i-th decision point, T i represents the time corresponding to the i-th decision point.

[0116] Optionally, using a cost function and a model predictive control MPC algorithm, determining whether the current tracking accuracy meets a preset accuracy threshold, and determining whether the central station needs to perform active tracking based on the judgment result to obtain a radar radiation control result, including:

[0117] The decision coefficients in the active measurement information in the control cost function are all 0, and the cost function is processed by preset processing to obtain an error value, and the error value is used as the current tracking accuracy; the preset processing is to sequentially perform inversion, trace and square root processing on the cost function;

[0118] Determine whether the error value is less than a preset accuracy threshold;

[0119] When the error value is less than the preset accuracy threshold, it is determined that the central station does not need to perform active tracking, and the absence of active tracking is taken as the radar radiation control result;

[0120] When the error value is greater than a preset accuracy threshold, it is determined that the central station needs to perform active tracking, and the need for active tracking is taken as a radar radiation control result.

[0121] Optionally, when the error value is greater than a preset accuracy threshold, it is determined that the central station needs to perform active tracking, and the need for active tracking is used as a radar radiation control result, including:

[0122] When the error value is greater than the preset accuracy threshold, the Euclidean distance between the central station and the estimated target position at the preset decision time is calculated according to the position of the central station and the estimated target position, and the decision distance value is obtained;

[0123] Sort the decision distance values ​​from small to large to obtain the distance sorting result;

[0124] The decision coefficients in the active measurement information corresponding to the distance sorting results are set to 1, and the error value is recalculated to obtain an updated error value;

[0125] The preset decision time corresponding to when the update error value is greater than the preset accuracy threshold is used as the start time of active tracking of the central station;

[0126] The start time of active tracking is taken as the radar radiation control result.

[0127] In the embodiment of the present invention, the error value is expressed as At the beginning, all u i =0, set the tracking accuracy requirement η, judge Is it true? If so, it means that there is no need to turn on the machine, and the tracking accuracy requirement can be achieved by using only passive measurement. If not, according to the central station position s and the estimated target position Calculate the Euclidean distance between the central station and the estimated target location at the preset decision time Arrange them from near to far, and let the corresponding u i =1, and recalculate judge Is it true? Until Established or all ui =1, output the decision variable u at this time m If u1=1, it means that the machine needs to be powered on at the current moment, and the corresponding power optimization needs to be performed. The power optimization method adopts the binary method. i =0 means that the central station does not need to be turned on at the current moment.

[0128] In summary, the radar radiation power control method based on MPC provided by the embodiment of the present invention has the following main ideas: determining the source / active radar networking; simulating the target radiation and motion behavior; performing passive tracking according to the target radiation and passive radar at the non-decision-making moment, making a decision on the active radar at the decision-making moment, extrapolating the simulation according to the target radiation frequency and the passive tracking result, generating a virtual position and a virtual measurement, calculating the posterior Cramer-Rao bound, judging whether to start up in turn according to the distance from the radar networking position, and optimizing the decision-making strategy; outputting the decision result, and the decision-making strategy is radar power radiation control based on MPC.

[0129] Combining the present invention with the prior art, the effects of the method of the present invention are mainly as follows:

[0130] 1. Adapt to complex scenarios and dynamic target behaviors:

[0131] As mentioned in the background technology, the decision-making process of the existing methods is simple and difficult to adapt to the complex and changeable battlefield environment and target behavior patterns. The present invention can more accurately predict the future state of the target by simulating the target radiation and motion behavior, thereby formulating a more reasonable radar radiation power control strategy. This prediction capability enables the radar system to better adapt to dynamically changing scenes and target behaviors, and improve the accuracy and robustness of tracking.

[0132] 2. No need to rely on the actual state of the target:

[0133] Existing methods rely on the real state of the target when calculating the cost function, which is unrealistic in practical applications. The present invention generates virtual positions and virtual measurements through simulation and extrapolation, and uses these virtual data to calculate the posterior Cramer-Rao bound (PCRLB) as a basis for evaluating tracking accuracy and making decisions. This method avoids dependence on the real state of the target and improves the practicality and accuracy of the algorithm.

[0134] 3. Optimize radar startup times and radiation power:

[0135] The present invention uses passive radar for passive tracking at non-decision-making moments, reduces dependence on active radar, and reduces the number of times the radar is turned on and the radiation power. At the decision-making moment, the optimal radar radiation power control strategy is formulated by comprehensively considering factors such as target radiation frequency, passive tracking results, and radar networking location. This strategy minimizes the number of times the radar is turned on and the radiation power while ensuring tracking accuracy, thereby improving the survivability and concealment of the radar.

[0136] 4. Balance the advantages of passive and active tracking:

[0137] The present invention realizes the complementary advantages of passive tracking and active tracking by constructing a passive / active radar network and combining the MPC method for radiation power control. In the passive tracking stage, the target radiation signal is used for perception, which reduces the radiation power of the radar and the risk of being discovered; in the active tracking stage, high-precision tracking is achieved by optimizing the radiation power and startup strategy of the radar. This balancing strategy enables the radar system to achieve optimal performance in different scenarios.

[0138] 5. Optimize decision-making strategies:

[0139] The MPC method used in the present invention is an optimization control algorithm that can formulate the best decision-making strategy based on the target state, radar resources and environmental constraints in the future. This strategy not only takes into account the current tracking needs, but also takes into account possible future situations, thereby achieving a global optimal control effect.

[0140] In order to verify the effectiveness of the radar radiation power control method based on MPC proposed in the embodiment of the present invention, a simulation experiment is also carried out as follows:

[0141] 1. Experimental scenario:

[0142] Set the simulation scenario and simulation parameters. For the simulation scenario and parameters, refer to Table 1.

[0143] Table 1 Radar networking and target simulation parameter settings

[0144]

[0145] 2. Simulation content:

[0146] In a set simulation scenario, the tracking errors of the tracking method of the present invention and the passive tracking method alone are analyzed. In the subsequent description, the tracking method of the present invention is referred to as decision tracking.

[0147] The simulation scenarios can be divided into three categories:

[0148] 1. When the target is far away from the radar network. Correspondingly, Figure 2The spatial position relationship diagram of the target away from the radar network is shown as an example. Figure 2 As shown in the figure, the radar network includes: central station, auxiliary station and target. Figure 2 The five-pointed star in the figure represents the initial position of the target, and the dotted line represents the actual trajectory of the target, and the actual trajectory of the target gradually moves away from the radar network.

[0149] 2. When the target is close to the radar network. Correspondingly, Figure 3 The spatial position relationship diagram of the target close to the radar network is shown as an example. Figure 3 As shown in the figure, the radar network includes: central station, auxiliary station and target. Figure 3 The five-pointed star in the figure represents the initial position of the target, and the dotted line represents the actual trajectory of the target, and the actual trajectory of the target gradually approaches the radar network.

[0150] 3. When the target approaches and then moves away from the radar network. Correspondingly, Figure 4 The spatial position relationship diagram of a target approaching and then moving away from the radar network is shown as an example. Figure 4 As shown in the figure, the radar network includes: central station, auxiliary station and target. Figure 4 The five-pointed star in the figure represents the initial position of the target, and the dotted line represents the actual trajectory of the target. The actual trajectory of the target is first approaching the radar network and then gradually moving away from the radar network.

[0151] 3. Analysis of simulation results:

[0152] Figure 5 The following figure shows the comparison results of passive tracking and decision tracking when the target is far away from the radar network. Figure 5 The results show that when the target is far away from the radar network and the active radar (center radar) is not turned on, the decision tracking result is exactly the same as the passive tracking result; when the active radar is turned on, the tracking error is greatly reduced. Although the tracking error will slowly increase afterwards, it is still lower than the pure passive tracking error. The power-on point in the figure is represented by a circle, and the power-on power is represented by the size of the circle. A total of 4 power-on decision points are set at 10s, 20s, 30s, and 40s. Since the target has been moving away from the radar network, the power-on is chosen at all four power-on decision points; and at 40s, since it has been turned on three times, the power-on power at 40s has decreased, reducing the probability of being discovered by the target.

[0153] Figure 6 The following figure shows the comparison results of passive tracking and decision tracking when the target is close to the radar network. Figure 6The results show that when the target is close to the radar network and the active radar is not turned on, the decision tracking result is exactly the same as the passive tracking result; when the active radar is turned on, the tracking error is greatly reduced. Although the tracking error will slowly increase afterwards, it is also lower than the pure passive tracking error. The power-on point in the figure is represented by a circle, and the power-on power is represented by the size of the circle. A total of 4 power-on decision points are set at 10s, 20s, 30s, and 40s. When the decision is made at the 10th second, the target is far away from the radar network, so the decision is to turn on; because it is turned on at the 10th second and the target is always close to the radar network, the decision is not to turn on at the 20th and 30th seconds; because it is not turned on at 20s and 30s, the prediction accuracy becomes higher at the 40th second, so the decision is to turn on.

[0154] Figure 7 The following is an example of a comparison of the results of passive tracking and decision tracking when the target first approaches and then moves away from the radar network. Figure 7 The results show that when the target approaches the radar network first and then moves away, and the active radar is not turned on, the decision tracking result is exactly the same as the passive tracking result; when the active radar is turned on, the tracking error is greatly reduced. Although the tracking error will slowly increase afterwards, it is also lower than the pure passive tracking error. The power-on point in the figure is represented by a circle, and the power-on power is represented by the size of the circle. A total of 4 power-on decision points are set at 10s, 20s, 30s, and 40s. When making the decision at the 10th second, since the target is closer to the radar network at 20s and 30s, it is judged that turning on at 20s and 30s can meet the accuracy requirements, so the decision is not to turn on at 10s; since it is not turned on at 10s, the decision is turned on at 20s and 30s; at 40s, the target is far away from the radar network, so the decision is turned on; and at 40s, since it has been turned on twice, the power at 40s is reduced, which reduces the probability of being discovered by the target.

[0155] The method provided in the embodiment of the present invention can be applied to an electronic device. Specifically, the electronic device can be: a desktop computer, a portable computer, an intelligent mobile terminal, a server, etc., which is not limited in the embodiment of the present invention.

[0156] Based on the same inventive concept, an embodiment of the present invention also provides a radar radiation power control device based on MPC. Figure 8 The present invention provides a schematic diagram of the structure of a radar radiation power control device based on MPC. Figure 8 As shown, it includes: a model building unit 801 and a decision judging unit 802;

[0157] The model building unit 801 is used to: establish a target model; the target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys the Poisson process, and the target state model is described by the approximate uniform linear motion of the target;

[0158] The model building unit 801 is also used to: build a measurement model through the target state model under the radar networking architecture; the measurement model includes: a target-based measurement model and a central station-based measurement model;

[0159] The decision judgment unit 802 is used to: at a preset decision moment, measure the tracking accuracy through the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB, determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

[0160] Fig. 9 A schematic diagram of the structure of a radar radiation power control device based on MPC provided by an embodiment of the present invention includes: a processor 910, a storage medium 920 and a bus 930, wherein the storage medium 920 stores machine-readable instructions executable by the processor 910, and when the radar radiation power control device based on MPC is running, the processor 910 communicates with the storage medium 920 through the bus 930, and the processor 910 executes the machine-readable instructions to execute the steps of the above method embodiment. The specific implementation method and technical effect are similar and will not be repeated here.

[0161] The storage medium may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk storage. Optionally, the storage medium may also be at least one storage device located away from the aforementioned processor.

[0162] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0163] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification.

[0164] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art can understand and implement other changes of the above disclosed embodiments by viewing the drawings and the disclosed content. In the description of the present invention, the term "comprising" does not exclude other components or steps, "one" or "an" does not exclude multiple situations, and the meaning of "multiple" is two or more, unless otherwise clearly and specifically limited. In addition, certain measures are recorded in different embodiments, but this does not mean that these measures cannot be combined to produce good results.

[0165] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A radar radiation power control method based on MPC, characterized in that: include: Build a target model; The target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys a Poisson process, and the target state model is described by the approximate uniform linear motion of the target; Under the radar networking architecture, a measurement model is constructed through the target state model; The measurement model includes: a target-based measurement model and a central station-based measurement model; At the preset decision moment, the tracking accuracy is measured by the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB to determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

2. The radar radiation power control method based on MPC according to claim 1, characterized in that: The target radiation model is expressed as: t=[t1,t2,…,t K ]; Among them, t represents the time set of the target radiation signal, K represents the total number of target radiation signals, t K Indicates the time when the target radiates the signal for the Kth time; The target state model is expressed as: in, Indicates that the target is at t k The state at time t k represents the time when the target radiates the signal for the kth time, Indicates that the target is at t k-1 The state of the moment, Indicates that the target is at t k-1 The state transfer matrix at time t k-1 It indicates the moment when the target radiates the signal for the k-1th time. Indicates that the target is at t k-1 The process noise at the moment.

3. The radar radiation power control method based on MPC according to claim 1, characterized in that: The radar networking architecture includes: a central station and multiple auxiliary stations; The central station is used to radiate signals externally and receive signals radiated by the target, and the auxiliary station is used to receive signals radiated by the target and the central station; the radar networking architecture uses an extended Kalman filter method to track the target.

4. The radar radiation power control method based on MPC according to claim 1, characterized in that: The target-based measurement model is expressed as: in, Indicates that the nth auxiliary station is at the target t k The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the target t k The zero-mean measurement noise obtained when the signal is radiated at time t k It indicates the time when the target is at the kth radiation signal. Indicates that the target is at t k The state of the moment; The central station-based measurement model is expressed as: in, Indicates that the nth auxiliary station is at the central station T m The measurement obtained when the signal is radiated at any moment, Indicates that the nth auxiliary station is at the target The actual measurement of the state, Indicates that the nth auxiliary station is at the central station T m The zero-mean measurement noise obtained when the signal is radiated at time instant, Indicates T m The state of the target at time T m Indicates the time when the central station radiates the signal for the mth time.

5. The radar radiation power control method based on MPC according to claim 1, characterized in that: At the preset decision moment, the tracking accuracy is measured by the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB, and it is determined whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result, including: Build a target prediction model based on the target's historical external radiation data; A cost function is constructed according to the measurement model, the target prediction model and the posterior Cramer-Rao bound PCRLB; The cost function and the model predictive control MPC algorithm are used to determine whether the current tracking accuracy meets the preset accuracy threshold, and based on the result of the determination, it is determined whether the central station needs to perform active tracking to obtain the radar radiation control result.

6. The radar radiation power control method based on MPC according to claim 5, characterized in that: The cost function is expressed as: in, represents the value of the cost function, Indicates that the target is at the preset final time t end Status information when represents the prior information, G D1 Indicates passive measurement information, G D2 Represents active measurement information; where G D1 The target prediction model and the target-based measurement model are used to construct G D2 It is constructed using a target prediction model and a measurement model based on a central station.

7. The radar radiation power control method based on MPC according to claim 5, characterized in that: The using of the cost function and the model predictive control MPC algorithm to judge whether the current tracking accuracy meets a preset accuracy threshold, and determining whether the central station needs to perform active tracking according to the judgment result to obtain the radar radiation control result, includes: Controlling the decision coefficients in the active measurement information in the cost function to be 0, performing preset processing on the cost function to obtain an error value, and using the error value as the current tracking accuracy; the preset processing is to perform inversion, trace and square root processing on the cost function in sequence; Determining whether the error value is less than the preset accuracy threshold; When the error value is less than the preset accuracy threshold, it is determined that the central station does not need to perform active tracking, and the absence of active tracking is taken as the radar radiation control result; When the error value is greater than the preset accuracy threshold, it is determined that the central station needs to perform active tracking, and the need for active tracking is taken as the radar radiation control result.

8. The radar radiation power control method based on MPC according to claim 7, characterized in that: When the error value is greater than the preset accuracy threshold, determining that the central station needs to perform active tracking, and taking the need for active tracking as the radar radiation control result, includes: When the error value is greater than the preset accuracy threshold, the Euclidean distance between the central station and the estimated target position at the preset decision time is calculated according to the position of the central station and the estimated target position to obtain a decision distance value; Sort the decision distance values ​​from small to large to obtain a distance sorting result; The decision coefficients in the active measurement information corresponding to the distance sorting result are set to 1, and the error value is recalculated to obtain an updated error value; The preset decision time corresponding to when the update error value is greater than the preset accuracy threshold is used as the start time of active tracking of the central station; The start time of the active tracking is used as the radar radiation control result.

9. A radar radiation power control device based on MPC, characterized in that: The MPC-based radar radiation power control device includes: a model building unit and a decision-making unit; The model building unit is used to: establish a target model; the target model includes a target radiation model and a target state model, wherein the process of the target radiation signal obeys a Poisson process, and the target state model is described by the approximate uniform linear motion of the target; The model building unit is also used to: build a measurement model through the target state model under the radar networking architecture; the measurement model includes: a target-based measurement model and a central station-based measurement model; The decision judgment unit is used to: at a preset decision moment, measure the tracking accuracy through the measurement model, the model predictive control MPC algorithm and the posterior Cramer-Rao bound PCRLB, determine whether the central station in the radar networking architecture needs to perform active tracking and obtain the radar radiation control result.

10. A radar radiation power control device based on MPC, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the MPC-based radar radiation power control device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the MPC-based radar radiation power control method as described in any one of claims 1 to 8.

Images