Signal processing device, signal processing method and program

Abstract

The signal processing device (1) includes a plurality of observation systems that calculate smoothed values ​​and error covariance matrices of the smoothed values ​​of the position of the aircraft (100) using observation data of the aircraft (100), and an integration system that integrates the plurality of smoothed values ​​output from the plurality of observation systems using weights set to reduce the error covariance matrix, thereby estimating the position of the aircraft (100).

Description

[Technical Field]

[0001] The present disclosure relates to a signal processing device, a signal processing method, and a program. [Background technology]

[0002] As a signal processing device according to the present disclosure, for example, a signal processing device that estimates the position of a target from the time from when a radio wave is emitted to the target until the reflected wave of the radio wave is received and the direction of arrival of the reflected wave is known. Configuration examples of such a signal processing device (hereinafter also referred to as a "conventional device") are shown in Figures 16 to 18.

[0003] 16, the multiple sensors are radars 2 that observe the position of a target. Here, the sensors are radars, and the number of sensors is N, where N is an integer of 2 or more. The following describes a case where the target is an aircraft 100. The radar tracking filter 3 tracks the position of the aircraft 100 based on the observation values ​​transmitted by the radar 2, and obtains the smoothed position of the aircraft 100. The data integrator 10 integrates the data output from the N radar tracking filters 3.

[0004] As shown in Fig. 17, the radar tracking filter 3 includes a coordinate conversion unit 18, a Kalman gain calculation unit 19, a smoothed value calculation unit 20, and a predicted value calculation unit 21. The coordinate conversion unit 18 converts from a polar coordinate system to a Cartesian coordinate system. The Kalman gain calculation unit 19 calculates a Kalman gain used when calculating a smoothed value for determining the smoothed position of the target in the tracking process. The smoothed value calculation unit 20 calculates the smoothed value. The predicted value calculation unit 21 predicts the target position at the next sampling.

[0005] 18, the data integrator 10 includes a smoothed value memory circuit 22, a covariance memory circuit 23, and an estimated value calculator 24. The smoothed value memory circuit 22 stores the smoothed values ​​transmitted from the radar tracking filter 3. The covariance memory circuit 23 stores the error covariance matrix of the smoothed values ​​transmitted from the radar tracking filter 3. The estimated value calculator 24 integrates the smoothed values ​​based on the smoothed values ​​transmitted from the smoothed value memory circuit 22 and the error covariance matrix of the smoothed values ​​transmitted from the covariance memory circuit 23, and calculates an estimated value indicating the estimated position of the target.

[0006] In relation to the configuration of the conventional device, a configuration in which observation data of an aircraft 100 is observed by a plurality of radars 2 is described, for example, in Patent Document 1. Also, in relation to a configuration in which smoothed values ​​from radar tracking filters 3 are integrated, an algorithm for calculating the maximum likelihood estimate of a covariance matrix in a situation in which the covariance matrix is ​​unknown is described, for example, in Non-Patent Document 1.

[0007] Next, an example of the operation of the conventional device will be described. Note that, hereinafter, the subscript indicating each of the N radars 2 will be expressed as "-#n", where n is any integer from 1 to N (1≦n≦N).

[0008] The radar 2-#n detects the direction of the target aircraft 100 at an azimuth angle θ az,n and distance R d,n The radar 2-#n observes the direction θ of the aircraft 100 at the current time lT with a sampling interval of T. az,n,l and distance R d,n,l Observation data z consisting of n,l =(θ az,n,l ,R d,n,l ) to the radar tracking filter 3. The radar tracking filter 3 observes the observation data a predetermined number of sampling times L, and then transmits the estimated value x n,l For example, the observation data z at the current time lT transmitted from the radar 2-#n to the radar tracking filter 3 is n,lis transmitted to the coordinate transformation unit 18 and the Kalman gain calculation unit 19. The coordinate transformation unit 18 calculates the coordinate-transformed observation data z' according to the following equations (1) to (3). n,l Ask for. In the following formula (1), (x n,l ,y n,l ) are the coordinates of the observation position of the aircraft 100 after the coordinate transformation. Furthermore, in the following equations (2) and (3), the coordinates (x rd,n ,y rd,n ) represents the position coordinates of radar 2-#n. TIFF0007814627000001.tif13166 TIFF0007814627000002.tif11166 TIFF0007814627000003.tif11166

[0009] The observation data z'n,l is transmitted from the coordinate transformation unit 18 to the Kalman gain calculation unit 19 and the smoothed value calculation unit 20. In addition, the predicted value x'n,l calculated by the predicted value calculation unit 21 at the sampling time (l-1)T is n,l|l-1 The predicted value x is also transmitted from the predicted value calculation unit 21 to the Kalman gain calculation unit 19. n,l|l-1 is configured as a two-dimensional vector consisting of predicted values ​​of the x component and predicted values ​​of the y component. The Kalman gain calculation unit 19 calculates the Kalman gain K n,l Calculate. In the following formula (4), P l|l-1 is the prediction error covariance matrix calculated by the prediction value calculation unit 21 at the sampling time (l-1)T, and H is the observation matrix. In the following formula (6), σ AZ 2 is the error covariance of the azimuth angle observation, and σ R 2 is the error covariance of the distance observations. TIFF0007814627000004.tif13166 TIFF0007814627000005.tif15166 TIFF0007814627000006.tif16166 TIFF0007814627000007.tif15166

[0010] The Kalman gain K calculated by the Kalman gain calculation unit 19 n,l is transmitted to the smoothed value calculation unit 20. The smoothed value calculation unit 20 calculates the smoothed value x n,l|l In the following formula (8), z' is calculated. n,l is the coordinate-transformed observation data expressed by the above formula (1), and H is the observation matrix expressed by the above formula (5). TIFF0007814627000008.tif12166

[0011] The smoothed value calculation unit 20 also calculates the Kalman gain K n,l , the prediction error covariance matrix P l|l-1 Using the observation matrix H in the above equation (5), the smoothing error covariance matrix P is calculated according to the following equation (9). n,l|l The smoothed value x calculated by the smoothed value calculation unit 20 is calculated. n,l|l and the smoothing error covariance matrix P n,l|l is transmitted to the predicted value calculation unit 21. TIFF0007814627000009.tif11166

[0012] The predicted value calculation unit 21 calculates a predicted value x in the next sampling at the sampling time lT according to the following equations (10) and (11): n,l+1|l The calculated predicted value x n,l+1|l is transmitted to the smoothed value calculation unit 20. TIFF0007814627000010.tif12166 TIFF0007814627000011.tif24166

[0013] Furthermore, the predicted value calculation unit 21 uses the transition matrix Φ of the above equation (11) to calculate the prediction error covariance matrix P n,l+1|l Calculate the prediction error covariance matrix P n,l+1|l is transmitted from the predicted value calculation unit 21 to the Kalman gain calculation unit 19. In the following formula (13), σ x 2 is the driving noise power in the x-axis direction, and σ y 2 is the driving noise power in the y-axis direction. TIFF0007814627000012.tif11166 TIFF0007814627000013.tif46166

[0014] The radar tracking filter 3 calculates the smoothed value x output from the smoothed value calculation unit 20 at the sampling time LT. n,L|L The estimated value x of radar 2-#n n Output as a tilde. The radar tracking filter 3 also calculates the smoothing error covariance matrix P n,L|L is the error covariance matrix P n Output as a tilde. Estimated value x n tilde and its error covariance matrix P n The tilde is transmitted from the radar tracking filter 3 to the data integrator 10 .

[0015] The data integration unit 10 calculates the estimated value x1 tilde ~ x N The estimated value x tilde is calculated by integrating tilde. Non-Patent Document 1 describes that the estimated value x tilde is calculated according to the following equations (14) to (16). In the following equation (14), ω1, . . . ,ω N are the estimated values ​​x1 tilde, . . ., x N is the weight multiplied by the tilde, (0≦ω n ≦1,n=1,···,N), the sum is "1". Weights ω1,···,ω Nis set in advance and transmitted to the data integration unit 10. TIFF0007814627000014.tif18166 TIFF0007814627000015.tif19166 TIFF0007814627000016.tif18166

[0016] Estimated value x n The actual error covariance matrix P of tilde n As the error covariance matrix P n is the error covariance matrix P output from the radar tracking filter 3. n When it is guaranteed to be smaller than the tilde (P n Tilde-P n is a positive definite matrix), the estimate x1 tilde,···,x N Regardless of the correlation between , the estimated value x (ω1, , ω N ) actual error covariance matrix P cc (ω1, ,ω N ) is the error covariance matrix P(ω1, ,ω N ) is smaller than (P(ω1, ,ω N ) Tilde-P cc (ω1, ,ω N ) is guaranteed to be a positive definite matrix) as described in Non-Patent Document 1. Non-Patent Document 1 describes the case where the number of sensors N is "2", but the above points also apply to general numbers of sensors N. In addition, while the above configuration estimates the target position using only smoothed values ​​from N radars, when estimates are obtained sequentially for each sampling, it is also possible to configure the configuration such that the estimate from the previous sampling is used as one of the integrated data for the current sampling. [Prior art documents] [Patent documents]

[0017] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-45910 [Non-Patent Document 1] Simon J. Julier Jeffrey K. Uhlmann, “A Non-divergent Estimation Algorithm in the Presence of Unknown Correlations,” Proceedings of the American Control Conference Albuquerque, New Mexico June 1997 Summary of the Invention [Problem to be solved by the invention]

[0018] In conventional signal processing devices, weights ω1, ω N Therefore, the weights ω1, ω N However, there was a problem that the estimation accuracy of the target position decreased depending on the setting.

[0019] The present disclosure is intended to solve the above-mentioned problems, and has an object to provide a signal processing device that can improve the estimation accuracy of a target position compared to conventional devices. [Means for solving the problem]

[0020] The signal processing device according to the present disclosure calculates a smoothed value of a target position and The relevant Smoothed Value corresponds to Calculate the error covariance matrix and output Multiple observation systems, It corresponds to the estimated value of the target position obtained by integrating multiple smoothed values ​​output from multiple observation systems. error covariance matrix The diagonal sum of is minimum As it happens The weights assigned to each of the multiple smoothed values ​​are Set , set Using weights Complex and an integration system that integrates the smoothed values ​​of the number of the detected errors to estimate the target position. [Effects of the Invention]

[0021] According to the present disclosure, a signal processing device according to the present disclosure calculates a smoothed value of the target position and an error covariance matrix of the smoothed value using observation data of the target, and estimates the target position by integrating multiple smoothed values ​​output from multiple observation systems using weights set to reduce the error covariance matrix. This makes it possible for the signal processing device according to the present disclosure to improve the estimation accuracy of the target position compared to conventional methods. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a diagram illustrating an example of the configuration of a signal processing device according to a first embodiment. [Figure 2] 3 is a diagram illustrating an example of the configuration of a weight setting unit included in the signal processing device according to the first embodiment. FIG. [Figure 3] 3 is a flowchart showing a signal processing method according to the first embodiment. [Figure 4] FIG. 2 is a diagram illustrating a situation of a constrained optimization problem in the first embodiment. [Figure 5] 5A and 5B are diagrams illustrating an example of a hardware configuration for realizing the functions of the signal processing device according to the first embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of the configuration of a signal processing device according to a second embodiment. [Figure 7] FIG. 10 is a diagram illustrating an example of the configuration of a full search weight setting unit included in a signal processing device according to a second embodiment. [Figure 8] 10 is a flowchart showing a signal processing method according to the second embodiment. [Figure 9] FIG. 10 is a diagram illustrating an example of the configuration of a signal processing device according to a third embodiment. [Figure 10] FIG. 11 is a diagram illustrating an example of the configuration of a range-limited search-type weight setting unit included in a signal processing device according to a third embodiment. [Figure 11] 10 is a flowchart showing a signal processing method according to the third embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of the configuration of a signal processing device according to a fourth embodiment. [Figure 13] FIG. 10 is a diagram illustrating an example of the configuration of a sensor selection type weight setting unit included in a signal processing device according to a fourth embodiment. [Figure 14] 10 is a flowchart showing a signal processing method according to a fourth embodiment. [Figure 15] 13 is a flowchart showing a procedure for calculating weights in the fourth embodiment. [Figure 16] FIG. 1 is a diagram illustrating an example of the configuration of a conventional signal processing device. [Figure 17] 1 is a diagram illustrating an example of the configuration of a radar tracking filter in a conventional signal processing device and in the signal processing devices according to first to third embodiments. [Figure 18] 1 is a diagram illustrating an example of the configuration of a data integration unit in a conventional signal processing device and in the signal processing devices according to first to third embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0023] Embodiment 1 Fig. 1 is a diagram showing an example of the configuration of a signal processing device 1 according to embodiment 1. For example, as shown in Fig. 1, the signal processing device 1 includes a radar 2 which is a sensor, a radar tracking filter 3, a weight setting unit 9, and a data integrating unit 10. In the signal processing device 1, N radar tracking filters 3-#1 to 3-#N constitute a plurality of observation systems provided corresponding to the N radars 2-#1 to 2-#N, respectively. In addition, the weight setting unit 9 and the data integrating unit 10 configure an integrating system that integrates multiple smoothed values ​​output from multiple observation systems to estimate the target position, using weights that are set to reduce the error covariance matrix. The configuration examples of the radar 2, radar tracking filter 3, and data integrator 10 in the signal processing device 1 are the same as those of the conventional device described with reference to Figures 16 to 18, so repeated description will be omitted. Hereinafter, it is assumed that the target is an aircraft 100.

[0024] Fig. 2 is a diagram showing an example of the configuration of the weight setting unit 9. In Fig. 2, a memory circuit 4 stores observation data transmitted from the radar 2 and estimated values ​​transmitted from the radar tracking filter 3. A correlation coefficient calculation unit 5 calculates a correlation coefficient between the estimated values ​​output from the radar tracking filters 3 of different radars 2.

[0025] Furthermore, the error covariance matrix calculation unit 6 calculates an error covariance matrix based on the correlation coefficients transmitted from the correlation coefficient calculation unit 5. The evaluation function setting unit 7 calculates an evaluation function for integrating estimated values ​​by reducing the error covariance while guaranteeing an upper limit value of the error covariance matrix resulting from errors in the correlation coefficients. The optimal weight calculation unit 8 calculates optimal weights using the evaluation function transmitted from the evaluation function setting unit 7. The weight setting unit 9 is configured to include the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, the evaluation function setting unit 7, and the optimal weight calculation unit 8.

[0026] Next, an example of the operation of the signal processing device 1 according to the first embodiment will be described. First, the radar 2-#n observes the direction (azimuth angle) θaz,n and distance Rd,n of the target aircraft 100. After that, the signal processing device 1 operates in the same manner as the conventional device described above, and outputs the estimated value x 1,l Tilde, ,x N,l tilde and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the weight setting unit 9 and the data combining unit 10.

[0027] where the error covariance matrix P n,n The tilde indicates the estimated value x n,l The error covariance matrix P n,n Although the sampling number l is omitted in the tilde, it is actually updated for each sampling. Note that in the following formulas, the sampling number l is omitted to avoid complication of notation.

[0028] The memory circuit 4 of the weight setting unit 9 stores a predetermined number of samples l dWhile, observation z 1,l ,···,z n,l , and the estimated value x 1,l tilde,···,x N,l tilde is stored, and then each stored value is transmitted to the correlation coefficient calculation unit 5. As a result, the correlation coefficient calculation unit 5 calculates the correlation coefficient using observation data over several past samplings. The correlation coefficient calculation unit 5 calculates the correlation coefficient between the estimated values ​​of the radar tracking filters 3 of different radars 2. Specifically, the correlation coefficient calculation unit 5 calculates the correlation coefficient between the estimated values ​​of radar 2-#n and radar 2-#m (1≦n, m≦N).

[0029] For example, the correlation coefficient calculation unit 5 calculates the correlation coefficient k between the x-axis direction of the radar 2-#n and the x-axis direction of the radar 2-#m. xx,n,m tilde, the correlation coefficient k between the y-axis direction of radar 2-#n and the y-axis direction of radar 2-#m yy,n,m tilde, the correlation coefficient k between the x-axis direction of radar 2-#n and the y-axis direction of radar 2-#m xy,n,m tilde, the correlation coefficient k between the y-axis direction of radar 2-#n and the x-axis direction of radar 2-#m yx,n,m The tilde is calculated according to the following formulas (17) to (20). In the following formulas (17) to (20), x x,n,i The tilde is the estimate x n,i x coordinate of the tilde, x y,n,i The tilde is the estimate x n,y y coordinate of the tilde, z x,n,i is the observed value z n,i x coordinate, z coordinate y,n,i is the observed value z n,i and the y-coordinates of the TIFF0007814627000017.tif22166 TIFF0007814627000018.tif21166 TIFF0007814627000019.tif22166 TIFF0007814627000020.tif21166

[0030] Correlation coefficient k xx,n,m tilde, kyy,n,m tilde, k xy,n,m tilde, k yx,n,m tilde is transmitted from the correlation coefficient calculation unit 5 to the error covariance matrix calculation unit 6. The error covariance matrix calculation unit 6 first calculates the error covariance matrix P n,m is calculated using the following formulas (21) to (25). n,m(i,k) is the error covariance matrix P n,m represents the element in row i, column k of TIFF0007814627000021.tif13166 TIFF0007814627000022.tif14166 TIFF0007814627000023.tif14166 TIFF0007814627000024.tif14166 TIFF0007814627000025.tif13166

[0031] Inter-radar error covariance matrix P n,m (1≦n, m≦N) is transmitted from the error covariance matrix calculation unit 6 to the evaluation function setting unit 7. In addition, the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is also transmitted from each radar 2 to the evaluation function setting unit 7.

[0032] The evaluation function setting unit 7 calculates the evaluation function h according to the following equations (26) to (29): n (ω1, ,ω N ) is set as the evaluation function h n (ω1, ,ω N ) is transmitted from the evaluation function setting unit 7 to the optimum weight calculation unit 8. In the following equation (26), Λ represents a function for finding the diagonal sum of a matrix. TIFF0007814627000026.tif58166 TIFF0007814627000027.tif19166 TIFF0007814627000028.tif24166 TIFF0007814627000029.tif25166

[0033] The optimum weight calculation unit 8 calculates the weights that minimize the error covariance. 3 is a flowchart showing the signal processing method according to the first embodiment, and shows the processing procedure for calculating weights in the optimum weight calculation unit 8. This processing procedure is similar to the processing procedure of the Newton method.

[0034] First, in step ST1-1, the optimal weight calculation unit 8 calculates a vector ω=[ω1, ,ω N ] T The optimum weight calculation unit 8 sets the evaluation function h n (ω1, ,ω N The following equations (30) to (35) are used to calculate the evaluation function h n (ω1, ,ω N ) TIFF0007814627000030.tif112166 TIFF0007814627000031.tif21166 TIFF0007814627000032.tif26166 TIFF0007814627000033.tif86166 TIFF0007814627000034.tif31166 TIFF0007814627000035.tif25166

[0035] In step ST1-2, the optimal weight calculation unit 8 calculates the evaluation function h n (ω1, ,ω N ) is used as a component to calculate the matrix of the following equation (36). The optimal weight calculation unit 8 also calculates the evaluation function h n (ω1, ,ω N ) is calculated as the column vector of (37) below. TIFF0007814627000036.tif33166 TIFF0007814627000037.tif20166

[0036] In step ST1-3, the optimum weight calculation unit 8 calculates a correction vector Δ=φ when updating the weight vector ω by the Newton method. -1 In step ST1-4, the optimum weight calculation unit 8 sets the vector ω-Δ as a new weight vector ω. From the above formula (16), the weight ω n (n=1, ,N) is in the range 0≦ω n ≦1 and if the newly set weight vector ω has a negative component, the optimum weight calculation unit 8 corrects that component to zero in step ST1-5.

[0037] Furthermore, since the sum of the weights is "1" according to the above equation (16), in step ST1-6, the optimum weight calculation unit 8 corrects the weight vector ω to the following equation (38) so that the weighted sum becomes 1. TIFF0007814627000038.tif12166

[0038] In step ST1-7, the optimal weight calculation unit 8 determines whether or not the optimal weight has converged. The criteria for determining whether or not the optimal weight has converged include when the norm of the correction vector Δ becomes sufficiently small and when a predetermined number of updates have been performed. If it is determined that the optimal weight has converged, the optimal weight calculation unit 8 calculates the optimal weights ω1, . . . ,ω N If it is determined that convergence has not occurred, the process proceeds to step ST1-8.

[0039] In step ST1-8, the optimal weight calculation unit 8 calculates the evaluation function h n (ω1, ,ω N ) and its derivative. Evaluation function h n (ω1, ,ω N) and its derivative are transmitted to step ST1-2. After that, the process is repeated until it is determined in step ST1-7 that convergence has occurred. The optimal weights ω1, ,ω are output when it is determined that convergence has occurred. N are transmitted from the optimum weight calculation unit 8 to the data integrating unit 10. Thereafter, the signal processing device 1 operates in the same manner as the conventional device, and the smoothed values ​​transmitted from the radar tracking filter 3 are integrated by the data integrating unit 10.

[0040] Here, the evaluation function h n (ω1, ,ω N ) and the series of processes shown in Figure 3 are used to calculate the optimal weight. n (ω1, ,ω N ) is the weights ω1, ,ω such that the diagonal sum of the error covariance shown in the following equation (39) for the estimated value x(ω) tilde in the above equation (14) is minimized under the condition of the above equation (16). N The necessary conditions are expressed as follows. TIFF0007814627000039.tif8166

[0041] The error covariance shown in the above equation (39) is P cc (ω), then P cc (ω) can be expressed by the following formula (40) and formula (41). TIFF0007814627000040.tif24166 TIFF0007814627000041.tif19166

[0042] P cc (ω) diagonal sum Λ(P cc (ω)) by minimizing the weights ω1, ,ω N the constraint ω1+···+ω in the above equation (16). N By solving the constrained optimization problem under =1, the optimal weights ω1, ω N Here, the weight ω1 is changed to ω1+Δ1, and the weight ω n ω n +Δ nBy changing only this, we obtain the following equation (42). TIFF0007814627000042.tif84166

[0043] Figure 4 shows the situation of the constrained optimization problem at this time. Using the same idea as the Lagrange multiplier method, which is well known as a method for solving constrained optimization problems, Λ(P cc (ω1, ,ω N )) = constant weights ω1, ,ω N (solid line in the figure) and the constraint ω1+···+ω N The regions where .DELTA..times ... Therefore, in the vicinity of the optimal point, Λ(P cc (ω1, ,ω N )) = constant, the weight ω1 is changed by a small amount Δ1, and the weight ω n A small amount Δ n The slope Δ n / Δ1 is the constraint ω1+···+ω N = 1, the weight ω1 is set to a small amount Δ1, and the weight ω n A small amount Δ n The parts of the above equation (42) expressed by the following equation (43) on both sides are cancelled out, and Δ n / Δ1=1. TIFF0007814627000043.tif11166

[0044] Taking note of the above relationship, the above equation (42) can be expressed as the following equation (44). TIFF0007814627000044.tif51166

[0045] Therefore, the left side of the above equation (44) is expressed as the evaluation function h n (ω1, ,ω N ) to calculate the above equation (26). n (ω1, ,ω N)=0, the optimal weights ω1, ,ω N is a necessary condition for ω1+···+ω N =1 is also required. N =1 is always ω1+···+ω in the process shown in Figure 3. N = 1. In fact, by setting all elements in the first row of the matrix Φ to "1" and the value in the first row of the vector h to "0", the sum of each component of the change Δ in the weight vector ω is set to "0". Therefore, the initial value is ω1+···+ω N By setting the weight vector ω such that =1, ω1+ +ω N The weight vector ω is updated so that

[0046] The signal processing device 1 according to the first embodiment is configured as described above, and therefore can reduce the error covariance and calculate weights for integrating smoothed values, thereby improving the accuracy of the estimated value obtained by integrating the smoothed values ​​compared to conventional methods. An upper limit is guaranteed for the effect that the error in the correlation estimate between the smoothed values ​​of each sensor has on the error in the integrated estimate, and the signal processing device 1 can perform robust estimation with respect to the error in the correlation estimate. Furthermore, the optimum weight calculation unit 8 of the signal processing device 1 uses the Newton method, and can calculate the optimum weight at high speed. Here, an example in which the radar 2 is used as the sensor has been described, but the sensor is not limited to the radar 2 and other sensors can also be used. Furthermore, the signal processing device 1 can also be configured to use a system that uses time difference positioning and GPS positioning to estimate the position of the aircraft 100.

[0047] Next, with reference to Fig. 5, an example of a hardware configuration for realizing the functions of the signal processing device 1 according to the first embodiment will be described. The functions of the radar tracking filter 3 (coordinate conversion unit 18, Kalman gain calculation unit 19, smoothed value calculation unit 20, and predicted value calculation unit 21), the weight setting unit 9 (memory circuit 4, correlation coefficient calculation unit 5, error covariance matrix calculation unit 6, evaluation function setting unit 7, and optimal weight calculation unit 8), and the data integrating unit 10 (smoothed value memory circuit 22, covariance memory circuit 23, and estimated value calculation unit 24) included in the signal processing device 1 are realized by processing circuits. The processing circuit may be dedicated hardware as shown in Fig. 5A, or may be a CPU (also referred to as a central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor)) 52 that executes a program stored in a memory 53 as shown in Fig. 5B.

[0048] When the processing circuit is dedicated hardware, the processing circuit 51 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. The functions of the radar tracking filter 3, the weight setting unit 9, and the data integrating unit 10 may be realized by separate processing circuits 51, or all of the functions may be realized by a single processing circuit 51.

[0049] When the processing circuit is a CPU 52, the functions of the radar tracking filter 3, the weight setting unit 9, and the data integrating unit 10 are realized by software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the memory 53. The CPU 52 realizes the functions of each unit by reading and executing the programs recorded in the memory 53. That is, the signal processing device 1 includes a memory for storing a program that, when executed by the processing circuit, results in the execution of, for example, each step shown in FIG. 3. These programs also cause a computer to execute the procedures and methods of the radar tracking filter 3, the weight setting unit 9, and the data integrating unit 10. Here, examples of memory 53 include non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (Electrically EPROM), magnetic disk, flexible disk, optical disk, compact disk, mini disk, or DVD (Digital Versatile Disc).

[0050] It is also possible to realize some of the functions of the radar tracking filter 3, the weight setting unit 9, and the data integrating unit 10 with dedicated hardware and some with software or firmware. For example, the radar tracking filter 3 can be realized by a processing circuit as dedicated hardware, and the weight setting unit 9 and the data integrating unit 10 can be realized by the processing circuit reading and executing a program stored in the memory 53.

[0051] Thus, the processing circuitry can implement each of the above-described functions through hardware, software, firmware, or a combination thereof.

[0052] As described above, the signal processing device 1 according to the first embodiment includes a plurality of observation systems that calculate smoothed values ​​and error covariance matrices of the position of the aircraft 100 using observation data of the aircraft 100, and an integration system that integrates a plurality of smoothed values ​​output from the plurality of observation systems using weights set to reduce the error covariance matrix, to estimate the position of the aircraft 100. This makes it possible for the signal processing device 1 to improve the accuracy of estimating the position of the aircraft 100 compared to conventional systems.

[0053] In the signal processing device 1 according to the first embodiment, the N observation systems each include radar tracking filters 3-#1 to 3-#N that calculate smoothed values ​​using observation data including the direction of the aircraft 100 and the distance to the aircraft 100 acquired by each of the radars 2-#1 to 2-#N, and output the smoothed values ​​and an error covariance matrix. The integration system includes a weight setting unit 9 that sets weights for integrating the smoothed values ​​while suppressing an increase in the error covariance matrix output from the radar tracking filters 3-#1 to 3-#N, and a data integration unit 10 that integrates the smoothed values ​​using the set weights and estimates the position of the aircraft 100 based on the obtained estimated value. This enables the signal processing device 1 to improve the accuracy of estimating the position of the aircraft 100 compared to conventional systems.

[0054] In the signal processing device 1 according to the first embodiment, the radar tracking filter 3 includes a coordinate conversion unit 18 that converts the coordinate system of the observation data of the aircraft 100 from a polar coordinate system to a Cartesian coordinate system, a Kalman gain calculation unit 19 that calculates a Kalman gain used when calculating a smoothed value in the tracking process, a smoothed value calculation unit 20 that calculates a smoothed value using the observation data that has been coordinate-converted by the coordinate conversion unit 18 and the Kalman gain calculated by the Kalman gain calculation unit 19, and a predicted value calculation unit 21 that predicts the position of the aircraft 100 at the next sampling. This enables the signal processing device 1 to accurately calculate the smoothed value and the error covariance matrix of the smoothed value.

[0055] In the signal processing device 1 according to the first embodiment, the predicted value calculation unit 21 predicts the smoothed value and the error covariance matrix at the next sampling, the Kalman gain calculation unit 19 calculates the Kalman gain using the error covariance matrix at the next sampling predicted by the predicted value calculation unit 21, and the smoothed value calculation unit 20 calculates the error covariance matrix at the current sampling using the Kalman gain calculated by the Kalman gain calculation unit 19 and the error covariance matrix at the next sampling predicted by the predicted value calculation unit 21. This enables the signal processing device 1 to accurately calculate the smoothed value and the error covariance matrix of the smoothed value.

[0056] In the signal processing device 1 according to the first embodiment, the data integrator 10 includes a smoothed value memory circuit 22 that stores smoothed values ​​transmitted from the radar tracking filter 3, a covariance memory circuit 23 that stores error covariance matrices transmitted from the radar tracking filter 3, and an estimated value calculator 24 that integrates a plurality of smoothed values ​​to calculate an estimated value of the position of the aircraft 100 based on the smoothed values ​​stored in the smoothed value memory circuit 22 and the error covariance matrix stored in the covariance memory circuit 23. This enables the signal processing device 1 to calculate an estimated value with high accuracy.

[0057] In the signal processing device 1 according to the first embodiment, the weight setting unit 9 includes a memory circuit 4 that stores the observation data transmitted from the radar 2 and the smoothed values ​​transmitted from the radar tracking filters 3, a correlation coefficient calculation unit 5 that calculates a correlation coefficient between the smoothed values ​​output from the radar tracking filters 3 corresponding to different radars 2, an error covariance matrix calculation unit 6 that calculates an error covariance matrix based on the correlation coefficient transmitted from the correlation coefficient calculation unit 5, an evaluation function setting unit 7 that calculates an evaluation function for integrating the smoothed values ​​while suppressing an increase in the error covariance matrix and guaranteeing an upper limit value of the error covariance matrix caused by errors in the correlation coefficients, and an optimal weight calculation unit 8 that calculates weights using the evaluation function transmitted from the evaluation function setting unit 7. This enables the signal processing device 1 to accurately calculate the optimal weights.

[0058] In the signal processing device 1 according to the first embodiment, the optimal weight calculation unit 8 calculates the optimal weight by applying the Newton method using an evaluation function, thereby enabling the signal processing device 1 to calculate the optimal weight at high speed.

[0059] In the signal processing device 1 according to the first embodiment, the correlation coefficient calculation unit 5 calculates the correlation coefficient using observed data over several past samples, thereby enabling the signal processing device 1 to calculate the correlation coefficient with high accuracy.

[0060] In the signal processing device 1 according to the first embodiment, when calculating the optimal weights, if any weights have a negative sign, the optimal weight calculation unit 8 sets the weights to 0, and further adjusts the weights so that the total sum becomes 1. This allows the signal processing device 1 to efficiently calculate the optimal weights.

[0061] The signal processing method according to the first embodiment includes the steps of: a radar tracking filter 3 corresponding to each of the plurality of radars 2 calculating a smoothed value indicating the smoothed position of the aircraft 100 using observation data acquired by each radar 2, the smoothed value including the direction of the aircraft 100 and the distance to the target, and outputting the calculated smoothed value and the error covariance matrix of the smoothed value; a weight setting unit 9 setting an optimal weight for integrating the smoothed values ​​while suppressing an increase in the error covariance matrix of the output smoothed values; and a data integration unit 10 integrating the smoothed values ​​using the set optimal weight, and estimating the position of the aircraft 100 based on the obtained estimated value. By having the signal processing device 1 execute the above method, it is possible to improve the accuracy of estimating the position of the aircraft 100 compared to the prior art.

[0062] The program according to the first embodiment causes a computer to function as a plurality of radar tracking filters 3, each corresponding to a plurality of radars 2, which calculate a smoothed value indicating the smoothed position of the aircraft 100 using observation data acquired by each radar 2, including the direction of the aircraft 100 and the distance to the aircraft 100, and output the calculated smoothed value and the error covariance matrix of the smoothed value; a weight setting unit 9 which sets a weight for integrating the smoothed values ​​while suppressing an increase in the output error covariance matrix; and a data integration unit 10 which integrates the smoothed values ​​using the set weight and estimates the position of the aircraft 100 based on the obtained estimated value. By having a computer execute the above program, it is possible to provide a signal processing device 1 that has improved accuracy in estimating the position of the aircraft 100 compared to conventional devices.

[0063] Embodiment 2 In the first embodiment, a signal processing device was described that can reduce the error covariance, calculate weights for integrating smoothed values, and improve the accuracy of the estimated value obtained by integrating the smoothed values ​​compared to conventional methods. In the second embodiment, a signal processing device will be described that can prevent non-optimal weights such as local solutions from being calculated in addition to the above.

[0064] Fig. 6 is a diagram showing an example of the configuration of a signal processing device 1b according to embodiment 2. The signal processing device 1b differs from the signal processing device 1 shown in Fig. 1 in that the weight setting unit 9 is changed to a full search weight setting unit 13. Other configurations of the signal processing device 1b according to embodiment 2 are the same as those of the signal processing device 1 shown in Fig. 1, and therefore the same reference numerals are used and description thereof will be omitted.

[0065] Fig. 7 is a diagram showing a configuration example of the full search weight setting unit 13. In Fig. 7, the memory circuit 4, the correlation coefficient calculation unit 5, and the error covariance matrix calculation unit 6 are the same as the memory circuit 4, the correlation coefficient calculation unit 5, and the error covariance matrix calculation unit 6 included in the weight setting unit 9 in the first embodiment.

[0066] 7, a search-type evaluation function setting unit 11 searches to find an evaluation value for reducing the error covariance and integrating the estimated values. An exhaustive search-type optimal weight calculation unit 12 calculates optimal weights by exhaustive search using the evaluation function transmitted from the search-type evaluation function setting unit 11. The exhaustive search-type weight setting unit 13 includes a memory circuit 4, a correlation coefficient calculation unit 5, an error covariance matrix calculation unit 6, the search-type evaluation function setting unit 11, and the exhaustive search-type optimal weight calculation unit 12.

[0067] Next, an example of the operation of the signal processing device 1b will be described. The radar 2-#n detects the direction of the target aircraft 100 at an azimuth angle θ az,n and distance R d,n and observe. Thereafter, the signal processing device 1b operates in the same manner as the signal processing device 1, and outputs the estimated value x 1,l Tilde, ,x N,l tilde and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the data integration unit 10. Also, the observed value z from radar 2-#n 1,l ,···,z n,l However, the estimated value x from the radar tracking filter 3 1,l Tilde, ..., x N,l tilde, and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the exhaustive search weight setting unit 13 . Thereafter, the signal processing device 1b operates in the same manner as the signal processing device 1, and calculates the inter-radar error covariance matrix P n,m (1≦n, m≦N) is transmitted from the error covariance matrix calculation unit 6 to the search type evaluation function setting unit 11.

[0068] The search type evaluation function setting unit 11 searches for an evaluation value for reducing the error covariance and integrating the estimated values. Specifically, the search type evaluation function setting unit 11 calculates the evaluation function I(ω1, ,ω N ) is set. In the following equation (45), P cc (ω1, ,ωN ) tilde represents the weighted error covariance matrix defined by equation (15) above, and Q cc (ω1, ,ω N ) respectively represent the matrices defined by the above formula (27). Note that Λ represents a function for finding the diagonal sum of a matrix. TIFF0007814627000045.tif9166

[0069] The evaluation function I(ω1, ,ω N ) is transmitted from the search type evaluation function setting unit 11 to the exhaustive search type optimum weight calculation unit 12.

[0070] FIG. 8 is a flowchart showing a signal processing method according to the second embodiment, and illustrates the processing steps for calculating weights by the full search type optimum weight calculation unit 12. First, in step ST2-1, the exhaustive search type optimum weight calculation unit 12 sets various initial values. For example, the exhaustive search type optimum weight calculation unit 12 sets the number of divisions Nω within the range of 0 to 1 that the weight can take. Next, the exhaustive search type optimum weight calculation unit 12 sets the type of weight according to the following equations (46) and (47). In the following equation (46), ω k,nΔk (1≦k≦N) is the weight ω k This represents the type of weighting set for TIFF0007814627000046.tif32166 TIFF0007814627000047.tif13166

[0071] Next, in step ST2-2, the exhaustive search type optimum weight calculation unit 12 calculates the optimum weight with an additive value of 1(ω 1,nΔ1 +ω 2,nΔ2 +···+ω N,nΔN =1). If there are N such combinations, comb If there are n c (1≦n c ≦N combThe )th weight combination found can be expressed by the following equation (48). TIFF0007814627000048.tif13166

[0072] In step ST2-3, the exhaustive search type optimal weight calculation unit 12 calculates the initial value n c In step ST2-4, the exhaustive search type optimum weight calculation unit 12 calculates the evaluation function I(ω(n c )) is calculated. TIFF0007814627000049.tif11166 TIFF0007814627000050.tif11166

[0073] In step ST2-5, the exhaustive search type optimum weight calculation unit 12 calculates the number n c =N comb If the answer is YES, the process proceeds to step ST2-7. If not, the process proceeds to step ST2-6, and the exhaustive search type optimum weight calculation unit 12 determines whether the number n c n c As a +1, proceed to step ST2-4.

[0074] In step ST2-7, the exhaustive search type optimum weight calculation unit 12 calculates the number n that minimizes the evaluation function according to the following equation (51): c,min Check out. TIFF0007814627000051.tif14166

[0075] In step ST2-8, the exhaustive search type optimum weight calculation unit 12 calculates the evaluation function I(ω(n c )) is calculated as the weight (optimal weight). TIFF0007814627000052.tif20166

[0076] Optimal weights ω1, ,ω Nare transmitted from the exhaustive search optimal weight calculation unit 12 to the data integrating unit 10. Thereafter, the signal processing device 1b operates in the same manner as the signal processing device 1, and the smoothed values ​​transmitted from the radar tracking filter 3 are integrated in the data integrating unit 10.

[0077] The signal processing device 1b is configured as described above, and can calculate weights for integrating smoothed values ​​by reducing the error covariance, thereby improving the accuracy of the estimated value integrating the smoothed values ​​more than ever before. Furthermore, the signal processing device 1b calculates the optimal weights by exhaustive search, and can prevent calculation of non-optimal weights such as local solutions.

[0078] As described above, the signal processing device 1b according to the second embodiment includes, in place of the evaluation function setting unit 7, a search type evaluation function setting unit 11 that determines, by search, an evaluation function for integrating smoothed values ​​while suppressing an increase in error covariance, and, in place of the optimal weight calculation unit 8, an exhaustive search type optimal weight calculation unit 12 that calculates optimal weights by exhaustive search using the evaluation function transmitted from the search type evaluation function setting unit 11. This enables the signal processing device 1b to prevent non-optimal weights such as local solutions from being calculated.

[0079] Embodiment 3 In the second embodiment, a signal processing device that calculates optimal weights by exhaustive search has been described. In the third embodiment, a processing device that can limit the search range for optimal weights and reduce the processing load required for the search will be described.

[0080] Fig. 9 is a diagram showing an example of the configuration of a signal processing device 1c according to embodiment 3. The signal processing device 1c differs from the signal processing device 1b shown in Fig. 5 in that the full search type weight setting unit 13 is changed to a range limiting search type weight setting unit 15. Other configurations of the signal processing device 1c are the same as those of the signal processing device 1b shown in Fig. 5, and therefore the same reference numerals are used and description thereof will be omitted.

[0081] Fig. 10 is a diagram showing a configuration example of the range-limited search weight setting unit 15. In Fig. 10, the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, and the search type evaluation function setting unit 11 are the same as the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, and the search type evaluation function setting unit 11 provided in the full search weight setting unit 13 in the second embodiment.

[0082] 10, a range-limited search type optimal weight calculation unit 14 performs a range-limited search and calculates optimal weights using the evaluation function transmitted from the search type evaluation function setting unit 11. The range-limited search type weight setting unit 15 includes a memory circuit 4, a correlation coefficient calculation unit 5, an error covariance matrix calculation unit 6, the search type evaluation function setting unit 11, and the range-limited search type optimal weight calculation unit 14.

[0083] Next, an example of the operation of the signal processing device 1c will be described. The radar 2-#n detects the direction of the target aircraft 100 at an azimuth angle θ az,n and distance R d,n and observe. Thereafter, the signal processing device 1c operates in the same manner as the signal processing device 1, and outputs the estimated value x 1,l tilde,···,x N,l tilde and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the data integration unit 10. Also, the observed value z from radar 2-#n 1,l ,···,z n,l However, the estimated value x from the radar tracking filter 3 1,l tilde,..., x N,l tilde, and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the range-limited search type weight setting unit 15 . Thereafter, the signal processing device 1c operates in the same manner as the signal processing device 1b, and the evaluation function I(ω1, ,ω N ) is transmitted from the search type evaluation function setting unit 11 to the range limiting search type optimum weight calculation unit 14.

[0084] FIG. 11 is a flowchart showing a signal processing method according to the third embodiment, illustrating the procedure for calculating weights by the range-limited search-type optimum weight calculation unit 14. First, in step ST3-1, the range-limiting search type optimal weight calculation unit 14 sets various initial values. For example, like the full search type optimal weight calculation unit 12 in the second embodiment, the range-limiting search type optimal weight calculation unit 14 sets the number of divisions N in the range of 0 to 1 that the weight can take. ω Set.

[0085] Next, the range-limited search type optimum weight calculation unit 14 sets the type of weight according to the above formula (46) and formula (47).

[0086] Furthermore, the range-limited search type optimal weight calculation unit 14 calculates the weight at the center ω of the search range. n Tilde (1≦n≦N) and the width of the search range Δ ωn (1≦n≦N). Specifically, the range-limited search type optimum weight calculation unit 14 sets the center ω of the search range. n Tilde and search range width Δ ωn is set within a range where an optimal weight is expected to exist. For example, when tracking the aircraft 100, the optimal weight is calculated for each sampling, so the range-limited search type optimal weight calculation unit 14 can set the width of the search range to a range that can sufficiently reduce the processing load of the search, with the optimal weight calculated in the previous sampling being the center of the search range.

[0087] In step ST3-2, the range-limited search type optimum weight calculation unit 14 calculates the optimum weight with an additive value of 1(ω 1,nΔ1 +ω 2,nΔ2 +···+ω N,nΔN = 1) within the range of the following formula (53). TIFF0007814627000053.tif26166

[0088] Such a combination is N comb If there are n c (1≦n c ≦Ncomb The combination of weights found in the )th step is expressed by the above formula (48). The subsequent steps ST3-3 to ST3-8 are the same as steps ST2-3 to ST2-8 in the second embodiment. The optimal weights ω1, . . . ,ω N are transmitted from the range-limited search type optimum weight calculation unit 14 to the data integrating unit 10, and the smoothed values ​​transmitted from the radar tracking filter 3 are integrated in the data integrating unit 10.

[0089] The signal processing device 1c according to the third embodiment is configured as described above, and therefore can calculate weights for integrating smoothed values ​​by reducing the error covariance, thereby improving the accuracy of the estimated value integrating the smoothed values ​​compared to conventional devices. Furthermore, the signal processing device 1c calculates optimal weights by limiting the search range, which reduces the processing load compared to when optimal weights are calculated by a full search. Furthermore, the signal processing device 1c can prevent a local solution from being searched for and a non-optimal weight from being calculated.

[0090] As described above, the signal processing device 1c according to the third embodiment includes, in place of the optimal weight calculation unit 8, a range-limited search-type optimal weight calculation unit 14 that calculates optimal weights by performing a range-limited search using the evaluation function transmitted from the search-type evaluation function setting unit 11. This makes it possible for the signal processing device 1c to reduce the processing load required for the search in addition to the effects of the first embodiment.

[0091] In the signal processing device 1c according to the third embodiment, the range-limited search-type optimal weight calculation unit 14 searches for the optimal weight in the current sampling within a range limited to the vicinity of the optimal weight used in the previous sampling, thereby enabling the signal processing device 1c to reduce the processing load required for the search.

[0092] Embodiment 4 In the third embodiment, a processing device capable of limiting the search range for the optimal weight and reducing the processing load required for the search was described. In the fourth embodiment, a processing device will be described that can calculate the weight using only the smoothed values ​​of valid sensors whose weight sign is not negative, and can improve the accuracy of the estimated value obtained by integrating the smoothed values.

[0093] Fig. 12 is a diagram showing an example of the configuration of a signal processing device 1d according to embodiment 4. The signal processing device 1d differs from the signal processing device 1 shown in Fig. 1 in that the weight setting unit 9 is changed to a sensor selection type weight setting unit 17. Other configurations of the signal processing device 1d are the same as those of the signal processing device 1, so the same reference numerals are used and their description will be omitted.

[0094] Fig. 13 is a diagram showing a configuration example of the sensor selection type weight setting unit 17. In Fig. 13, the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, and the evaluation function setting unit 7 are the same as the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, and the evaluation function setting unit 7 provided in the weight setting unit 9 in the first embodiment.

[0095] 13, the sensor selection type optimal weight calculation unit 16 calculates the optimal weight excluding the sensor with a negative weight. The sensor selection type weight setting unit 17 includes the memory circuit 4, the correlation coefficient calculation unit 5, the error covariance matrix calculation unit 6, the evaluation function setting unit 7, and the sensor selection type optimal weight calculation unit 16.

[0096] Next, an example of the operation of the signal processing device 1d will be described. First, the radar 2-#n detects the direction (azimuth angle) θ of the target aircraft 100. az,n and distance R d,n and observe. Thereafter, the signal processing device 1d operates in the same manner as the signal processing device 1, and outputs the estimated value x 1,l tilde,···,x N,l tilde and the error covariance matrix P 1,1 Tilde,···,P N,N The tilde is transmitted to the data integration unit 10. Also, the observed value z from radar 2-#n 1,l ,···,z n,l However, the estimated value x from the radar tracking filter 3 1,l tilde,..., x N,l tilde, and the error covariance matrix P 1,1 Tilde,···,P N,NThe tilde is transmitted to the sensor selection type weight setting unit 17, respectively. The signal processing device 1d operates in the same manner as the signal processing device 1, and calculates the evaluation function h(ω1, ,ω N ) is transmitted from the evaluation function setting unit 7 to the sensor selection type optimal weight calculation unit 16.

[0097] FIG. 14 is a flowchart showing a signal processing method according to the fourth embodiment, illustrating the procedure for calculating weights by the sensor-selection-type optimum weight calculation unit 16. First, in step ST4-1, the sensor selection type optimal weight calculation unit 16 calculates a vector ω=[ω1 ω N ] T Next, the sensor selection type optimum weight calculation unit 16 sets the evaluation function h n (ω1, ,ω N The subsequent processes in steps ST4-2 to ST4-4 are the same as steps ST1-2 to ST1-4 in the first embodiment.

[0098] In step ST4-5, the sensor selection type optimum weight calculation unit 16 checks whether or not the weight ω has a component with a negative sign.

[0099] FIG. 15 is a flowchart showing the procedure for calculating weights in the fourth embodiment, and shows a series of processes executed between A and B in FIG. If there is a weight ω whose sign is negative, the sensor selection type optimum weight calculation unit 16 proceeds to step ST4-9 shown in FIG. In step ST4-9, the sensor selection type optimum weight calculation unit 16 excludes sensors corresponding to weights with a negative sign. In step ST4-10, the sensor selection type optimum weight calculation unit 16 selects only sensors corresponding to weights with a positive sign, and resets the type of sensor and the number of sensors N anew. In step ST4-11, the sensor selection type optimal weight calculation unit 16 calculates a vector ω=[ω1 ω N ] T Reset the setting. Furthermore, the sensor selection type optimal weight calculation unit 16 calculates the evaluation function h n (ω1, ,ω N ) is also derived. If there is no weight ω with a negative sign, the process proceeds to step ST4-6. The processes of steps ST4-6 to ST4-8 are the same as steps ST1-6 to ST1-8 in the first embodiment. After that, the sensor-selection type optimal weight calculation unit 16 operates in the same way as in the first embodiment, and calculates the optimal weights ω1, . . . ,ω N Output.

[0100] In this way, the sensor-selection type optimum weight calculation unit 16 calculates weights that minimize the error covariance. Figures 14 and 15 show a processing procedure for calculating weights similar to that of the Newton method.

[0101] Since the signal processing device 1d of embodiment 4 is configured as described above, it is possible to calculate weights using only the smoothed values ​​of valid sensors whose weight sign is not negative, thereby improving the accuracy of the estimated value obtained by integrating the smoothed values.

[0102] As described above, the signal processing device 1d according to the fourth embodiment includes the sensor-selection type optimal weight calculation unit 16 that calculates weights using only smoothed values ​​of valid sensors whose weights do not have negative signs, instead of the optimal weight calculation unit 8. This enables the signal processing device 1d to improve the accuracy of the estimated values ​​in addition to the effects of the first embodiment. [Industrial Applicability]

[0103] A signal processing device according to the present disclosure can be used in, for example, a radar device.

[0104] In addition, the present disclosure allows for free combination of the respective embodiments, modification of any of the components of the respective embodiments, or omission of any of the components of the respective embodiments. [Explanation of symbols]

[0105] 1, 1b, 1c, 1d signal processing device, 2 radar, 3 radar tracking filter, 4 memory circuit, 5 correlation coefficient calculation unit, 6 error covariance matrix calculation unit, 7 evaluation function setting unit, 8 optimal weight calculation unit, 9 weight setting unit, 10 data integration unit, 11 search type evaluation function setting unit, 12 full search type optimal weight calculation unit, 13 full search type weight setting unit, 14 range limited search type optimal weight calculation unit, 15 range limited search type weight setting unit, 16 sensor selection type optimal weight calculation unit, 17 sensor selection type weight setting unit, 18 coordinate transformation unit, 19 Kalman gain calculation unit, 20 smoothed value calculation unit, 21 predicted value calculation unit, 22 smoothed value memory circuit, 23 covariance memory circuit, 24 estimated value calculation unit, 51 processing circuit, 52 CPU, 53 memory, 100 aircraft.

Claims

1. a plurality of observation systems that calculate and output smoothed values ​​of the target position and error covariance matrices corresponding to the smoothed values ​​using observation data of the target; an integration system that sets weights to be assigned to the plurality of smoothed values ​​so that a diagonal sum of an error covariance matrix corresponding to the estimated value of the target position obtained by integrating the plurality of smoothed values ​​output from the plurality of observation systems is minimized, and that integrates the plurality of smoothed values ​​using the set weights to estimate the target position; A signal processing device comprising:

2. The plurality of observation systems include: a radar tracking filter that calculates and outputs the smoothed value and an error covariance matrix corresponding to the smoothed value using observation data including the direction of the target and the distance to the target, the observation data being acquired by each of a plurality of radars; The integrated system comprises: a weight setting unit that sets weights so that a diagonal sum of an error covariance matrix corresponding to the estimated value is minimized; a data integration unit that integrates the plurality of smoothed values ​​using a set weight to calculate the estimated value; 2. The signal processing device according to claim 1.

3. The radar tracking filter comprises: a coordinate conversion unit that converts a coordinate system of the observation data of the target from a polar coordinate system to a Cartesian coordinate system; a Kalman gain calculation unit that calculates a Kalman gain used when calculating the smoothed value in the tracking process; a smoothed value calculation unit that calculates the smoothed value using the observation data that has been coordinate-transformed by the coordinate transformation unit and the Kalman gain calculated by the Kalman gain calculation unit; a predicted value calculation unit that predicts the target position in the next sampling and an error covariance matrix corresponding to the target position; 3. The signal processing device according to claim 2.

4. The Kalman gain calculation unit calculates the Kalman gain using an error covariance matrix at the next sampling predicted by the predicted value calculation unit, The smoothed value calculation unit calculates an error covariance matrix at the current sampling using the Kalman gain calculated by the Kalman gain calculation unit and the error covariance matrix predicted by the predicted value calculation unit.

4. The signal processing device according to claim 3.

5. The data integration unit a smoothed value memory circuit for storing the smoothed value transmitted from the radar tracking filter; a covariance memory circuit that stores an error covariance matrix corresponding to the smoothed value transmitted from the radar tracking filter; an estimated value calculation unit that calculates the estimated value by integrating a plurality of the smoothed values ​​based on the smoothed values ​​stored in the smoothed value memory circuit and the error covariance matrix stored in the covariance memory circuit.

5. The signal processing device according to claim 2, wherein the first and second signals are input to the first and second input terminals.

6. The weight setting unit a memory circuit for storing the observation data transmitted from the radar and the smoothed value transmitted from the radar tracking filter; a correlation coefficient calculation unit that calculates a correlation coefficient between the smoothed values ​​output from the radar tracking filters corresponding to different radars; an error covariance matrix calculation unit that calculates an error covariance matrix between the radars based on the correlation coefficient transmitted from the correlation coefficient calculation unit; an evaluation function setting unit that calculates an evaluation function for integrating the smoothed values ​​while suppressing an increase in the error covariance matrix corresponding to the estimated value, while guaranteeing an upper limit value of the error covariance matrix between the radars due to errors in the correlation coefficient; an optimal weight calculation unit that calculates a weight using the evaluation function transmitted from the evaluation function setting unit; 5. The signal processing device according to claim 2, wherein the first and second signals are input to the first and second input terminals.

7. The optimal weight calculation unit calculates weights by applying the Newton method using the evaluation function.

7. The signal processing device according to claim 6.

8. The weight setting unit a memory circuit for storing the observation data transmitted from the radar and the smoothed value transmitted from the radar tracking filter; a correlation coefficient calculation unit that calculates a correlation coefficient between the smoothed values ​​output from the radar tracking filters corresponding to different radars; an error covariance matrix calculation unit that calculates an error covariance matrix between the radars based on the correlation coefficient transmitted from the correlation coefficient calculation unit; a search-type evaluation function setting unit that determines, by search, an evaluation function for integrating the smoothed values ​​while suppressing an increase in an error covariance matrix corresponding to the estimated values; an exhaustive search type optimum weight calculation unit that calculates weights by exhaustive search using the evaluation function transmitted from the search type evaluation function setting unit; 5. The signal processing device according to claim 2, wherein the first and second signals are input to the first and second input terminals.

9. The weight setting unit a memory circuit for storing the observation data transmitted from the radar and the smoothed value transmitted from the radar tracking filter; a correlation coefficient calculation unit that calculates a correlation coefficient between the smoothed values ​​output from the radar tracking filters corresponding to different radars; an error covariance matrix calculation unit that calculates an error covariance matrix between the radars based on the correlation coefficient transmitted from the correlation coefficient calculation unit; a search-type evaluation function setting unit that determines, by search, an evaluation function for integrating the smoothed values ​​while suppressing an increase in an error covariance matrix corresponding to the estimated values; a range-limited search-type optimal weight calculation unit that calculates weights by performing a range-limited search using the evaluation function transmitted from the search-type evaluation function setting unit; 5. The signal processing device according to claim 2, wherein the first and second signals are input to the first and second input terminals.

10. The range-limited search type optimum weight calculation unit searches for weights in the current sampling within the vicinity of the weights used in the previous sampling.

10. The signal processing device according to claim 9.

11. The correlation coefficient calculation unit calculates the correlation coefficient using observation data over several past samplings.

7. The signal processing device according to claim 6.

12. The optimum weight calculation unit sets a weight with a negative sign to 0 when calculating the weight, and further adjusts the weight so that the total weight becomes 1.

7. The signal processing device according to claim 6.

13. The weight setting unit a memory circuit for storing the observation data transmitted from the radar and the smoothed value transmitted from the radar tracking filter; a correlation coefficient calculation unit that calculates a correlation coefficient between the smoothed values ​​output from the radar tracking filters corresponding to different radars; an error covariance matrix calculation unit that calculates an error covariance matrix between the radars based on the correlation coefficient transmitted from the correlation coefficient calculation unit; an evaluation function setting unit that calculates an evaluation function for integrating the smoothed values ​​while suppressing an increase in the error covariance matrix corresponding to the estimated value, while guaranteeing an upper limit value of the error covariance matrix between the radars due to errors in the correlation coefficient; a sensor selection type optimum weight calculation unit that calculates weights using only the remaining sensors excluding sensors corresponding to weights with a negative sign when calculating weights using the evaluation function transmitted from the evaluation function setting unit.

5. The signal processing device according to claim 2, wherein the first and second signals are input to the first and second input terminals.

14. A signal processing method by a signal processing device, comprising: a step in which a radar tracking filter corresponding to each of the plurality of radars calculates a smoothed value of the target position using observation data acquired by each radar, the smoothed value including the direction of the target and the distance to the target, and outputs the calculated smoothed value and an error covariance matrix corresponding to the smoothed value; a weight setting unit setting weights so that a diagonal sum of an error covariance matrix corresponding to the estimated value of the target position obtained by integrating a plurality of the smoothed values ​​is minimized; a data integration unit integrating the plurality of smoothed values ​​using the weights set by the weight setting unit to estimate the target position; A signal processing method comprising:

15. Computer, a plurality of radar tracking filters, each corresponding to a plurality of radars, for calculating a smoothed value of a target position using observation data acquired by each radar, the smoothed value including a direction of a target and a distance to the target, and outputting the calculated smoothed value and an error covariance matrix corresponding to the smoothed value; a weight setting unit that sets weights so that a diagonal sum of an error covariance matrix corresponding to the estimated value of the target position obtained by integrating a plurality of the smoothed values ​​is minimized; and a data integration unit that integrates a plurality of the smoothed values ​​using the weights set by the weight setting unit to estimate the target position; A program to function as a

Images