Clustered deployment method for cluster airborne MIMO radar nodes under phase reference constraint

Abstract

The application discloses a cluster airborne MIMO radar node clustering deployment method under coherent constraints, which is applied to the field of radar signal processing and aims at the problem that the prior art does not consider the coherent constraint condition in the echo signal accumulation process, thereby failing to accurately reflect the monitoring problem of a MIMO radar to a target area in a real situation. The application firstly performs gridding processing on a monitoring area and a station arrangement area, then sets a clustering queue of cluster unmanned aerial vehicles to calculate coherent constraint conditions of nodes in each cluster, then calculates a detection probability information matrix of the gridding processed monitoring area according to the station arrangement area and positions of each radar node, subsequently constructs an optimization problem with radar effective coverage rate as an index based on the detection probability information matrix, and finally solves the optimization problem with constraints based on a cuckoo particle swarm optimization algorithm, so that the cluster airborne distributed MIMO radar node clustering optimization deployment under coherent constraints is realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of radar signal processing, and particularly relates to a radar node clustering deployment technology. BACKGROUND

[0002] Distributed MIMO (Multiple-Input Multiple-Output) radar has become one of the important directions of the next generation of radar development due to its flexible reconfigurability, mutual enhancement of node capabilities, large space coverage, strong anti-interference capability and other characteristics. At the same time, compared with traditional land / sea-based radars, airborne distributed MIMO radars have unique advantages such as wide detection range, strong anti-destroying capability and high platform flexibility, and will certainly play a very important role in future wars. According to the literature “A.M. Haimovich, R.S. Blum and L.J. Cimini, “MIMO radar with widely separated antennas,” IEEE Signal Process. Mag., vol. 25, no. 1, pp. 116-129, 2008.”, the detection coverage performance of a distributed MIMO radar system is affected by the radar node positions and the coherent constraints of the echoes between the nodes. Therefore, optimizing the positions of the distributed MIMO radar nodes under the constraint of the maximum echo accumulation coherent constraint can improve the detection performance of the radar and the effective coverage area of the system.

[0003] From the existing work, the research on the optimization deployment of cluster airborne distributed MIMO radar nodes under the coherent constraint is still less. Most researchers study the optimization deployment of unmanned aerial vehicle distributed MIMO radar nodes without considering the coherent constraint condition of radar echo. For example, the literature "Fast Optimal Antenna Placement for Distributed MIMO Radar with Surveillance Performance, IEEE Signal Processing Letters, 2015, pp.1955-1959." Some scholars consider the coherent constraint condition of radar echo, but due to the complexity of the model, only consider that all MIMO radar nodes work in non-coherent mode, and give the optimization deployment method of nodes, such as the literature "Target detection and localization using multi-frame information for noncoherent MIMO radar, IET Int. Radar Conf., pp.1-6, 2015." The above methods improve the monitoring performance and radar effective coverage rate of the cluster airborne radar system, but the effectiveness of such algorithms in the distributed MIMO system with coherent constraint and mixed accumulation is severely challenged. SUMMARY

[0004] To solve the above technical problems, the present application provides a cluster airborne MIMO radar node clustering deployment method under coherent constraint.

[0005] The technical scheme adopted by the present application is: a cluster airborne MIMO radar node clustering deployment method under coherent constraint, comprising:

[0006] S1. The monitoring area and the station area are subjected to grid processing, and the area to be monitored is evenly divided into a plurality of square grid units of the same size;

[0007] S2. According to the resource situation of the airborne MIMO radar node, the radar node is divided into different coherent clusters in advance, and the corresponding coherent constraint condition is calculated according to the number of radar nodes in the cluster group;

[0008] S3. The detection probability information matrix of the grid processing monitoring area is calculated according to the station area and the position of each radar node;

[0009] S4. An optimization model with radar effective coverage rate as an index is constructed based on the detection probability information matrix;

[0010] S5. The cuckoo particle swarm optimization algorithm is used to solve this constrained optimization problem.

[0011] The method of the present application uses the cuckoo particle swarm optimization algorithm to obtain the cluster airborne MIMO radar node clustering optimization deployment under the coherent accumulation constraint condition. The present application first performs grid processing on the monitoring area and the station area, then sets the cluster queue of the cluster unmanned aerial vehicle to calculate the coherent constraint condition of each cluster node, then calculates the detection probability information matrix of the grid processing monitoring area according to the station area and the position of each radar node, then constructs an optimization problem with radar effective coverage rate as an index based on the detection probability information matrix, and finally solves the optimization problem with constraints based on the cuckoo particle swarm optimization algorithm; so as to realize the cluster airborne MIMO radar node clustering optimization deployment under the coherent constraint condition. The present application has the advantages of realizing the optimization deployment of the cluster airborne MIMO radar node under the coherent constraint condition, the solving process is simple, and the calculation complexity is low. The present application can be applied to large-scale airborne MIMO radar cooperative detection, military application and other fields. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 The flow chart of the method of the present application is provided;

[0013] Figure 2 The schematic diagram of the cluster airborne MIMO radar node clustering optimization deployment under the coherent constraint condition adopted by the specific embodiment of the present application is provided;

[0014] Figure 3 The simulation effect diagram of the effective detection area of the target area in the case of single-cluster airborne MIMO radar of the present application is provided;

[0015] Figure 4 The iteration curve comparison diagram of the cuckoo particle swarm algorithm and the particle swarm algorithm and the cuckoo algorithm in the case of single-cluster airborne MIMO radar of the present application is provided;

[0016] Figure 5 The simulation effect diagram of the effective detection area of the target area in the case of multi-cluster airborne MIMO radar of the present application is provided;

[0017] Figure 6 The iteration curve comparison diagram of the cuckoo particle swarm algorithm and the particle swarm algorithm and the cuckoo algorithm in the case of multi-cluster airborne MIMO radar of the present application is provided. DETAILED DESCRIPTION

[0018] The present application is verified by the method of Matlab simulation experiment, and the correctness and effectiveness of the present application are verified on the scientific calculation software Matlab R2019a. The embodiments of the present application are further described below in combination with the drawings.

[0019] As Figure 1As shown, the application proposes a cluster airborne MIMO radar node clustering optimization deployment method under the condition of coherent accumulation based on the cuckoo particle swarm optimization algorithm, which is specifically realized through the following process:

[0020] S1. The monitoring area and the station area are rasterized, and the to-be-monitored area is evenly divided into a plurality of square grid units of the same size;

[0021] S2. The clustering queue of the cluster unmanned aerial vehicle is set, the coherent constraint condition of each node in the cluster is calculated, the radar nodes are preliminarily divided into different coherent clusters according to the resource condition of the airborne MIMO radar nodes, and the corresponding coherent constraint condition is calculated according to the number of radar nodes in the cluster;

[0022] S3. The detection probability information matrix of the rasterized monitoring area is calculated according to the station area and the position of each radar node;

[0023] S4. An optimization model taking the radar effective coverage rate as an index is constructed based on the detection probability information matrix;

[0024] S5. The cuckoo particle swarm optimization algorithm is used to solve the optimization problem with constraints.

[0025] The implementation process of the application is as follows:

[0026] Step 1: The monitoring area and the station area are rasterized, and the to-be-monitored area is evenly divided into a plurality of square grid units of the same size.

[0027] As Figure 2 shown is a cooperative detection schematic diagram of the cluster airborne MIMO radar under coherent constraint, and the to-be-monitored area of the radar is located on the plane z=0 and is evenly divided into L1*L2 grid units. Among them, is the real field of the matrix, represents the l1l 2th th grid unit in A, l1=1,...,L1, l2=1,...,L2. The station area is located on the same size area above A on z=H.

[0028] Step 2: The clustering queue of the cluster unmanned aerial vehicle is set, the radar nodes are preliminarily divided into different coherent clusters according to the resource condition of the airborne MIMO radar nodes, and the corresponding coherent constraint condition is calculated according to the number of radar nodes in the cluster, that is, the distance constraint condition that must be met between the nodes in each cluster for coherent processing. The specific calculation expression is:

[0029]

[0030] Wherein, λ is the wavelength of the radar signal, x tk , ytk Respectively represent the x-axis and y-axis coordinates of the transmitting node tk, x ti y ti Respectively represent the x-axis and y-axis coordinates of the transmitting node ti, x rl 、y rl Respectively represent the x-axis and y-axis coordinates of the receiving node rl, x rj 、y rj Respectively represent the x-axis and y-axis coordinates of the receiving node rj, θ tk ,θ ti are the two transmitting nodes in the cluster, θ rl ,θ rj are any two receiving nodes in the cluster, θ tk ,θ ti ,θ rl ,θ rj The coordinates are Similarly They represent the transmitting nodes θ tk The x-axis, y-axis, and z-axis coordinates of the grid are set to D according to the task and requirements. x 、D y , D x 、D y Take value according to demand, D x 、D y The value of can be understood as the concept of image resolution, << means much smaller than, X0 is the center coordinate of the grid unit. tk ,X0),d(θ ti ,X0),d(θ rl ,X0),d(θ rj ,X0) represent the Euclidean distance from the radar node to the grid unit, that is, the effective detection radius of the radar. When the detection power of a single radar node is constant, its size is related to the number of radar nodes. According to the radar far field assumption, it can be considered that d(θ tk ,X0)=d(θ ti ,X0)=d(θ rl ,X0)=d(θ rj ,X0)=d max At the same time, according to the radar equation, we can get Among them, M n represents the number of nodes in the cluster, R max It is the maximum detection range of a single radar node.

[0031] According to formula (3), all radar nodes are transceiver nodes. Transceiver work in a cluster is a working mode. Any four nodes in the same cluster meet the following conditions:

[0032] The difference between the derivative of the distance between the first transmitting node and the center coordinate of the grid unit and the derivative of the distance between the second transmitting node and the center coordinate of the grid unit is far less than the product of the derivative of the size of the grid unit along the x-axis and the wavelength of the radar signal.

[0033] The difference between the derivative of the distance between the first transmitting node and the center coordinate of the grid unit and the derivative of the distance between the second transmitting node and the center coordinate of the grid unit is far less than the product of the derivative of the size of the grid unit along the y-axis and the wavelength of the radar signal.

[0034] The difference between the derivative of the distance between the first receiving node and the center coordinate of the grid unit and the derivative of the distance between the second receiving node and the center coordinate of the grid unit is far less than the product of the derivative of the size of the grid unit along the x-axis and the wavelength of the radar signal.

[0035] The difference between the derivative of the distance between the first receiving node and the center coordinate of the grid unit and the derivative of the distance between the second receiving node and the center coordinate of the grid unit is far less than the product of the derivative of the size of the grid unit along the y-axis and the wavelength of the radar signal.

[0036] In the subsequent iteration solving process, the radar nodes in the same coherent cluster group need to satisfy formula (3). If a radar node does not satisfy formula (3), it does not belong to the same coherent cluster group, thereby optimizing the division of the coherent cluster group.

[0037] According to formula (3), the distance constraint condition of the radar nodes under coherent accumulation can be obtained as an iterative constraint for subsequent node position optimization solving.

[0038] Step 3: Calculate the detection probability information matrix of the monitoring area after grid processing according to the station arrangement area and the positions of the radar nodes.

[0039] Step 3.1: Calculate the detection probability information matrix of a single cluster,

[0040] Considering only the single-cluster radar system, the airborne MIMO radar system detects a target in a grid unit . ti The signal-to-noise ratio of the signal echo from the transmitting node θ rj to the receiving node θ

[0041]

[0042] where D0 is a radar monitoring factor, and σij is the bistatic radar RCS of the MIMO radar node to the target, σ is the radar RCS (radar cross section) of the monostatic radar node, R max is the maximum detection range of a single radar node. ti =d(θ ti ,X0),R rj =d(θ rj ,X0).

[0043] Based on the above calculation, for the n Single cluster radar system Θ of radar nodes n For grid cells The signal-to-noise ratio of the accumulated signal of the target is calculated as follows

[0044]

[0045] Among them, R ij is the node θ ti and θ rj The coherence constraint distance, Θ n The coherent constraint distances between all radar nodes are formed by the distance constraint matrix Get Θ n For the first After calculating the echo signal-to-noise ratio of the target, calculate Θ n right To monitor the probability information element, the process is as follows

[0046]

[0047] Where Q(·) is the Malkum function and (·)! is the factorial operation. T The minimum detection probability of the radar is set according to the mission requirements, and its value is related to the radar false alarm rate. So we search through all the elements in A and get all the monitoring probability information elements. Constructing the monitoring probability information matrix

[0048] Step 3.2 Calculate the multi-cluster detection probability information matrix

[0049] According to the clustered MIMO system detection model based on hybrid integrated detection, radars in any coherent cluster perform coherent processing, while radars perform incoherent signal processing between coherent clusters. Here, each cluster is regarded as a radar node that transmits and receives signals on its own. Therefore, for a multi-cluster radar system containing N clusters, the number of radars detected is For example, the signal-to-noise ratio of the radar system after accumulating all echoes is calculated as follows:

[0050]

[0051] where, L NCI is the non-coherent integration loss factor, which is calculated as

[0052]

[0053] N p is the number of non-coherent integration pulses, which is calculated as

[0054]

[0055] Therefore, the Θ vs. the first target echo signal-to-noise ratio can be obtained. Then, the Θ vs. monitoring probability information element is obtained by the following process

[0056]

[0057] Then, all the monitoring probability information elements are obtained by traversing all the elements in A, and a monitoring probability information matrix is formed

[0058] Step 4: Based on the detection probability information matrix, an optimization problem is constructed with radar effective coverage rate as an index

[0059] Step 4.1 Radar effective coverage rate calculation

[0060] The radar effective coverage area is defined as the set of grid cells that satisfy the system minimum detection probability γ Pd , and the calculation expression is as follows

[0061]

[0062] Further, the radar effective coverage rate is defined as the ratio of grid cells that satisfy the radar detection probability condition to the total grid cells, and the calculation is as follows

[0063]

[0064] Step 4.2 Construction of optimization model with radar effective coverage rate as index

[0065] The objective function expression of the optimization model is:

[0066]

[0067] where x min , x max , y min , y max are the boundary coordinates of the station area, and H is the height of the airborne node. Θ is a matrix composed of all node positions, and θ jrepresents any airborne MIMO radar node in the system, J represents the number of airborne MIMO radar nodes in the system. max(·) represents the maximum function, is the radar effective coverage rate.

[0068] Step 5: solving the optimization problem based on the cuckoo particle swarm optimization algorithm

[0069] The number of iterations is set to T, the number of particles is set to S, and s th The deployment scheme represented by the particle is I s (t), and the individual optimal solution is P s (t). The global optimal solution is P g (t). At t th The iteration step of the particle at iteration time is V s (t), and its update at t = t + 1 is

[0070] V s (t + 1) = ωV s (t) + c1ε1(P s (t) - I s (t)) + c2ε2(P g (t) - I s (t)) (14)

[0071] where c1 and c2 are inertia constants, and ε1 and ε2 are random numbers. ω is the inertia weight factor, and its calculation expression is

[0072] ω = ω max - (ω max - ω min ) * t / T (15)

[0073] where ω max and ω min are the upper and lower bounds of the inertia weight. At t = t + 1, the update of the deployment scheme represented by the s-th particle is

[0074] I s (t + 1) = I s (t) + V s (t + 1) (16)

[0075] The first individual optimal solution is updated, and the rules are as follows

[0076]

[0077] The second individual optimal solution is updated by random walk, and the rules are as follows

[0078]

[0079] where ζ and qs q is a random number obeying Gaussian distribution s Pa is the probability that the whole nest is discarded. Heaviside(·) is the Heaviside step function, is the element-wise multiplication. is any other nest solution. At time t = t + 1, the individual optimal solution of the deployment scheme represented by the s-th particle is updated as

[0080]

[0081] The update of the population optimal solution is

[0082]

[0083] Repeat (19)-(20) until the iteration terminates, i.e., the current iteration number reaches the preset maximum value T; the final optimized deployment scheme P is obtained g (T).

[0084] Figure 3 The simulation effect diagram of the effective detection area of the single-cluster airborne MIMO radar system on the target area is given, Figure 4 The iteration curve diagram of the cuckoo particle swarm algorithm for optimizing the station arrangement of the single-cluster airborne MIMO radar system is given. Figure 3 、 Figure 4 The corresponding parameters are shown in Table 1. As can be seen from the figure, the cuckoo particle swarm algorithm is used to solve the node deployment position of the single-cluster MIMO radar, which has a larger radar effective coverage area and a faster convergence speed compared with other algorithms. Figure 5 The simulation effect diagram of the effective detection area of the multi-cluster airborne MIMO radar system on the target area is given, Figure 6 The iteration curve diagram of the cuckoo particle swarm algorithm for optimizing the station arrangement of the multi-cluster airborne MIMO radar system is given. Figure 5 、 Figure 6 The corresponding parameters are shown in Table 2. As can be seen from the figure, the cuckoo particle swarm algorithm is used to solve the node deployment position of the clustered airborne MIMO radar, which has a larger radar effective coverage area and a faster convergence speed compared with other algorithms. Those skilled in the art should know that the higher the radar effective coverage, the higher the utilization rate of the radar node resources in the task; the faster the iteration curve converges, the better the performance of the algorithm.

[0085] Figure 3 UAV (Unmanned Aerial Vehicle) in the above represents a drone.

[0086] Those skilled in the art will appreciate that the embodiments described herein are presented for purposes of illustration and understanding of the principles of the application and are not intended to limit the scope of the application to the particular embodiments presented. Various modifications and changes can be made thereto by those skilled in the art without departing from the spirit and principles of the application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the application are intended to be included in the scope of the claims.

[0087] Table 1 Figure 3 , Figure 4 Corresponding parameters

[0088]

[0089] Table 2 Figure 5 , Figure 6 Corresponding parameters

[0090]

[0091] Those skilled in the art will appreciate that the embodiments described herein are presented for purposes of illustration and understanding of the principles of the application and are not intended to limit the scope of the application to the particular embodiments presented. Various modifications and changes can be made thereto by those skilled in the art without departing from the spirit and principles of the application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the application are intended to be included in the scope of the claims.

Claims

1. A clustering deployment method for clustered airborne MIMO radar nodes under coherence constraints, characterized by: include: S1. Gridding the monitoring area and the station area, dividing the monitored area into multiple square grid cells of equal size; S2. Based on the airborne MIMO radar node resource availability, pre-divide the radar nodes into different coherent clusters and calculate the corresponding coherent constraints based on the number of radar nodes in the clusters. S3. Calculate the detection probability information matrix of the monitoring area after raster processing based on the station area and the location of each radar node; S4. Construct an optimization model based on the detection probability information matrix and taking the radar effective coverage rate as an indicator; S5. Solve this constrained optimization problem based on the cuckoo particle swarm optimization algorithm.

2. The method for clustering and deploying airborne MIMO radar nodes under coherence constraints according to claim 1, characterized in that: Each pre-divided coherent cluster includes at least 4 radar nodes.

3. The method for clustering and deploying airborne MIMO radar nodes under coherence constraints according to claim 2, characterized in that: The specific coherence constraints are: Where λ is the wavelength of the radar signal, x tk 、y tk They represent the horizontal and vertical coordinates of the transmitting node tk in the coherent cluster, x ti y ti Respectively represent the horizontal and vertical coordinates of the transmitting node ti in the coherent cluster, x rl 、y rl Respectively represent the horizontal and vertical coordinates of the receiving node rl in the coherent cluster, x rj 、y rj Respectively represent the horizontal and vertical coordinates of the receiving node rj in the coherent cluster, θ tk ,θ ti are two transmitting nodes in a coherent cluster, θ rl ,θ rj are any two receiving nodes in the coherent cluster, θ tk ,θ ti ,θ rl ,θ rj The coordinates are D x 、D y are the length and width of the grid cell respectively; << means much less than, X0 is the center coordinate of the grid cell, d(θ tk ,X0),d(θ ti ,X0),d(θ rl ,X0),d(θ rj ,X0) represent the Euclidean distance from radar node to grid cell respectively.

4. The method for clustering and deploying airborne MIMO radar nodes under coherence constraints according to claim 3, characterized in that: Under the assumption of radar far field, max For d(θ tk ,X0),d(θ ti ,X0),d(θ rl ,X0),d(θ rj ,X0) to approximate the solution, that is, d(θ tk ,X0)=d(θ ti ,X0)=d(θ rl ,X0)=d(θ rj ,X0)=d max , Among them, M n represents the number of nodes in the coherent cluster, R max It is the maximum detection range of a single radar node.

5. The method for clustering and deploying airborne MIMO radar nodes under coherence constraints according to claim 4, characterized in that: Step S3 is specifically as follows: S31, calculating the detection probability information matrix of a single coherent cluster; Considering a single coherent cluster radar system, the airborne MIMO radar system detects grid cells When the target is in; through the transmitting node θ ti To the receiving node θ rj The signal-to-noise ratio of the signal echo is calculated as follows: in, Indicates the l1lth in the radar monitoring area A 2th Grid unit, the radar monitoring area A is evenly divided into L1×L2 grid units, l1=1,...,L1, l2=1,...,L2, D0 is the radar monitoring factor, σ ij is the bistatic radar cross section of the MIMO radar node to the target, σ is the radar cross section of the monostatic radar node, R max is the maximum detection range of a single radar node, R ti =d(θ ti ,X0),R rj =d(θ rj ,X0); For the inclusion of M n Single coherent cluster radar system of radar nodes Θ n For grid cells The signal-to-noise ratio of the accumulated signal of the target is calculated as follows Among them, R ij is the node θ ti and θ rj The coherence constraint distance, Θ n The coherent constraint distances between all radar nodes in the matrix are composed of the distance constraint matrix R n , we get Θ n For the first After calculating the echo signal-to-noise ratio of the target, calculate Θ n right To monitor the probability information element, the process is as follows Among them, Q(·) is the Malkum function, (·)! is the factorial operation, γ T is the minimum detection probability of the radar set according to mission requirements, represents the radar false alarm rate, m=0,1,2,…,M n ×M n -1; By traversing and searching all elements in A, all monitoring probability information elements are obtained Construct monitoring probability information matrix Pd n ; S32, calculating the multi-cluster detection probability information matrix; Each coherent cluster is regarded as a self-transmitting and self-receiving radar node; for a multi-coherent cluster radar system containing N coherent clusters, The signal-to-noise ratio of all the echoes accumulated by the radar system is calculated as follows: in, L NCI is the non-coherent accumulation loss factor, N p The number of pulses accumulated for non-coherent operation; Get Θ for the first After calculating the echo signal-to-noise ratio of the target, we can get Θ To monitor the probability information element, the process is as follows Traverse and search all elements in A to obtain all monitoring probability information elements to form the monitoring probability information matrix Pd.

6. The method for clustering and deploying airborne MIMO radar nodes under coherence constraints according to claim 5, characterized in that: Step S4 specifically includes the following sub-steps: S41, calculation of radar effective coverage; The effective coverage area of ​​the radar is defined as the area that satisfies the minimum detection probability of the system γ Pd The set of grid cells is calculated as follows The effective radar coverage is defined as the ratio of grid cells that meet the radar detection probability condition to the total grid cells, which is calculated as follows: S42. Construct an optimization model with radar effective coverage as an indicator; the objective function expression of the optimization model is: s.t.x min ≤x j ≤x max and min ≤y j ≤y max z j =H Among them, x min 、x max 、y min 、y max are the boundary coordinates of the station area, H is the height of the airborne node, Θ is the matrix composed of all node positions, θ j represents any airborne MIMO radar node in the system, J represents the number of airborne MIMO radar nodes in the system, max(·) represents the maximum function, is the effective coverage of the radar.

Images