Target object angle estimation method for performing high angle analysis by using large array antenna

Abstract

The invention provides a target object angle estimation method using a large-scale array antenna to carry out high angle analysis, which receives an input signal matrix through the array antenna, the input signal matrix is an emission signal or a reflection signal of at least one target object, and the method comprises the following steps: S1, inputting the input signal matrix to a first operation model, obtaining the number of target objects and roughly estimated target angles corresponding to the positions of the target objects; s2, inputting the number of the target objects and the input signal matrix into a second operation model for singular value decomposition, and obtaining a noise matrix; s3, obtaining an iteration angle range through the roughly estimated target angle obtained in the step S1; and S4, inputting the complex tracking matrix and the noise matrix corresponding to the iteration angle range into a third operation model, and performing iteration of the angle range, thereby obtaining a precise target angle.

Description

[0001] The present application is a divisional application, the original application's application number is CN202210466777.1, the application date is April 29, 2022, and the invention name is target angle estimation method for high angle resolution using large array antenna. TECHNICAL FIELD

[0002] The present application relates to an estimation method, in particular to a target angle estimation method for high angle resolution using large array antenna. BACKGROUND

[0003] Antenna modules are widely used in today's society, such as in the field of wireless communication, radar detection, etc. Antenna modules are used for transmitting and receiving wireless signals to perform signal transmission or position detection. In a multi-antenna signal source environment, the problem of mutual interference between signals, how to transmit the best signal path, and how to select the best signal source are issues that need to be addressed in the antenna field.

[0004] For example, the public uses mobile phones to connect to base stations through antennas to transmit network signals and telecommunications signals. Therefore, how to select a better base station for connection in a mobile phone in a multi-base station environment with different distances and different transmission paths to obtain better telecommunications signals and network signals is a problem that needs to be solved in the field of mobile phone communication. In another example, a car radar detection system is mainly used for detecting objects around the vehicle to confirm obstacles and assist driving to avoid collision hazards. Further, it can even actively drive automatically. The detection principle of the radar system of the car is to detect the distance between the objects around the vehicle and the vehicle through the radar antenna when the vehicle is running. The traditional radar solution is based on 2-3 transmitting antennas and 3-4 receiving antennas. Imaging radar uses multiple-input multiple-output (MIMO) antenna arrays to measure the surrounding environment with high resolution, providing high-precision image quality. It does not have the problem of Doppler ambiguity sensing under any external environmental conditions, which helps to achieve long-distance and wide-view decision-making to avoid collisions between vehicles and objects or vehicles, and prevent traffic accidents.

[0005] In the above example of the best signal search technology of the mobile phone and the base station, since the mobile phone does not urgently and dangerously search for the signal of the base station, the target signal detection method used by the mobile phone radar antenna can use a signal detection method with a longer calculation time to repeatedly search for the position of the base station and judge the signal strength of the base station to receive better signals. Moreover, the position of the base station is fixed, so the position of the base station is better determined.

[0006] However, the radar antenna of the automobile is to avoid collision of the vehicle when the vehicle is running, so the vehicle radar system needs to accurately detect the number and position of the object in a short time, and therefore a more accurate and rapid signal operation method is needed to achieve the above purpose. SUMMARY

[0007] The main purpose of the present application is to solve the problem that the target signal detection method of the prior art cannot accurately determine the position of the target object and the detection operation time is too long.

[0008] To achieve the above purpose, the present application provides a target angle estimation method using a large array antenna for high angle resolution, which receives an input signal matrix through an array antenna, the input signal matrix being the transmitted signal or reflected signal of at least one target object, and the target angle estimation method using a large multiple-input multiple-output array antenna for high angle resolution comprises the following steps: step S1: inputting the input signal matrix into a first operation model and inputting a complex and angle-dependent tracking matrix for comparison, when the complex tracking matrix satisfies the condition of the first operation model, obtaining a target object number and a coarse target angle corresponding to the position of the target object; step S2: inputting the target object number and the input signal matrix into a second operation model for singular value decomposition, and obtaining a noise matrix; step S3: obtaining an iterative angle range through the coarse target angle obtained in step S1; step S4: inputting the complex tracking matrix corresponding to the iterative angle range and the noise matrix into a third operation model and performing angle range iteration, thereby obtaining an accurate target angle.

[0009] In another embodiment of the present application, step S1 further comprises step S1a, the first operation model compares the correlation between the input signal matrix and the complex tracking matrix, if one of the complex tracking matrices has no correlation with the input signal matrix, step S1a is repeated to compare the correlation between the next complex tracking matrix and the input signal matrix; if one of the complex tracking matrices has correlation with the input signal matrix, step S1b is performed: calculating a residual value based on the tracking matrix with correlation and the input signal matrix.

[0010] In another embodiment of the present application, in step S1b, the tracking matrix with correlation with the input signal matrix is defined as a first tracking matrix, and the residual value is calculated using the first tracking matrix and the input signal matrix.

[0011] In another embodiment of the present application, step S1 further comprises step S1c, comparing the residual value with a residual threshold value, if the residual value is reduced by a large amount at one time and is lower than or equal to the residual threshold value, the first operation model obtains the target object number and the coarse target angle; if the residual value is greater than the residual threshold value, step S1d is performed: comparing the reduction amplitude of the residual value.

[0012] In another embodiment of the present application, if the reduction of the residual value is greater than a significant threshold in step SId, a step Se is performed: the first tracking matrix is stored, and the input signal matrix is converted into a second input signal matrix which is irrelevant to the first tracking matrix, and the process returns to step Sla to compare the second input signal matrix with other complex tracking matrices; if the reduction of the residual value is less than the significant threshold, the process directly returns to step Sla to compare the input signal matrix with other complex tracking matrices.

[0013] In another embodiment of the present application, the angle range of the complex tracking matrix is from -90 degrees to 90 degrees, and the corresponding complex tracking matrix is sequentially compared with the input signal matrix in a first interval angle unit in step Sla.

[0014] In another embodiment of the present application, the first interval angle is between 1 / 180 and 1 / 18 of the angle range of the complex tracking matrix.

[0015] In another embodiment of the present application, the first operation model performs the operation of the residual value by the least square method in step Sib.

[0016] In another embodiment of the present application, the input signal matrix is converted into a co-variant matrix before being singular value decomposed with the target quantity in step S2.

[0017] In another embodiment of the present application, the iteration angle range is obtained by adding and subtracting a tolerance angle from the estimated target angle in step S3.

[0018] In another embodiment of the present application, the tolerance angle is less than the first interval angle.

[0019] In another embodiment of the present application, the iteration angle range is sequentially input to the third operation model in a second interval angle unit and iterated with the noise matrix in step S4.

[0020] In another embodiment of the present application, the second interval angle is between 1 / 100 and 1 / 10 of the iteration angle range.

[0021] In another embodiment of the present application, the third operation model iterates the iteration angle range with the noise matrix by the orthogonal operation method.

[0022] In another embodiment of the present application, an angle-power curve is obtained after the iteration of the third operation model in step S4, and the angle corresponding to the highest peak value on the curve is the accurate target angle.

[0023] In another embodiment of the present application, the input signal matrix is obtained from the reflected signals of the target objects by the radar array antenna arranged on the vehicle body.

[0024] Therefore, the present application can obtain the number and angle of the target objects more accurately by combining the first, second and third operation models, and can improve the operation speed, so that the present application can be applied to more different types of devices. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 is a flowchart of the target object angle estimation method using a large array antenna for high angle resolution according to an embodiment of the present application; Figure 2 is a detailed flowchart of step S1 according to an embodiment of the present application; Figure 3 is a state diagram of the first embodiment of the present application, which is used to represent the case where there is only one target object; Figure 4 is a state diagram of the second embodiment of the present application, which is used to represent the case where there are more than two target objects; Figure 5 is a graph of the second embodiment of the present application, which is used to represent the positions of the target objects at angles of -12 degrees and 21 degrees, respectively; Figure 6 is a graph of another embodiment of the present application, which is used to represent the positions of the target objects at angles of -1 degree and 1 degree, respectively.

[0026] REFERENCE NUMERALS 100: target object angle estimation method using a large array antenna for high angle resolution S1-S4: steps S1a-S1e: steps 1: vehicle radar 2: object 3a: first object 3b: second object L: detection reference line DETAILED DESCRIPTION

[0027] To facilitate the explanation of the central idea represented in the above summary of the present application, the specific embodiments are now expressed. In the embodiments, various different objects are drawn in the proportions suitable for illustration, not in the proportions of actual components.

[0028] Radar array antennas have many uses, such as detecting objects, observing weather, tracking objects, etc. Radar array antennas used for detecting objects also use many different algorithms, such as orthogonal matching pursuit (OMP), multiple signal classification (MUSIC), iterative sparse approximate minimum variance (SAMV), etc., to determine and calculate signals.

[0029] The orthogonal matching pursuit (OMP) object detection method first compares the correlation between the input signal matrix (the reflection signal of the target object received by the radar array antenna) and the angle-dependent steering vector, and calculates the residual value by residue analysis in regression analysis according to the correlation between the input signal matrix and the steering vector, to obtain the number of objects and the angle of the objects. Orthogonal matching pursuit (OMP) has the ability to calculate quickly, but in order to achieve the ability to calculate quickly, the detection accuracy of orthogonal matching pursuit (OMP) is not high, and orthogonal matching pursuit (OMP) cannot obtain accurate angle information of the target object, but only knows that the target object is located within an angle range.

[0030] The multiple signal classification (MUSIC) object detection method first converts the covariance matrix of the reflection signal of the target object received by the radar array antenna, and then performs singular value decomposition (SVD) to obtain a noise matrix according to the number of target objects and the covariance matrix. Finally, the noise matrix and an angle-dependent steering vector are iterated to obtain the angle information of the target object. Multiple signal classification (MUSIC) can obtain high-precision angle information, but in order to obtain high-precision angle information, multiple signal classification (MUSIC) requires a very large amount of processing, resulting in very slow calculation speed of multiple signal classification (MUSIC), and multiple signal classification (MUSIC) needs to know the number of target objects before calculation, resulting in limited range of multiple signal classification (MUSIC) application.

[0031] Please refer to Figures 1-2As shown, the target angle estimation method 100 using a large array antenna for high angle resolution according to embodiments of the present application is a target angle estimation method using a multi-input multi-output large array antenna for high angle resolution. The method receives an input signal matrix through an array antenna. The input signal matrix is a transmitted signal or a reflected signal of at least one target. In embodiments of the present application, the array antenna is a multi-input multi-output (MIMO) array antenna. The MIMO array antenna can use multiple antennas at the transmitting end to independently transmit signals and use multiple antennas at the receiving end to receive and recover the original information at the same time. The input signal matrix is obtained by a radar array antenna arranged on a vehicle body receiving the reflected signal of the target. The method can be used as an advanced driver assistance system (ADAS) of a vehicle body, such as a forward collision warning (FCW) and blind spot detection (BSD), but is not limited thereto. The method can be used for any part that needs to detect the number, position, and angle of a target. For example, the method can also be used for connection judgment between a mobile phone and a base station. The mobile phone needs to detect the connection of multiple base stations to determine the best base station for connection due to the connection signal. Therefore, the method is also suitable for the method of the present application. The present application includes the following steps: Step S1: inputting the input signal matrix into a first operation model (such as Orthogonal Matching Pursuit, OMP) and inputting a complex angle-related tracking matrix for comparison. When the complex tracking matrix satisfies the condition of the first operation model, a target number and a coarse target angle corresponding to the position of the target can be obtained. The complex tracking matrix is composed of steering vectors related to angles. The steering vector can be represented by formula (1).

[0032] (1), where v(k) is the steering vector.

[0033] In embodiments of the present application, whether the complex tracking matrix satisfies the condition of the first operation model is determined by step S1a, step S1b, step S1c, step S1d, and step S1e. The contents of steps S1a, S1b, S1c, S1d, and S1e are described as follows: Step Sla: The first operation model compares the input signal matrix with the complex tracking matrices for correlation. If one of the complex tracking matrices has no correlation with the input signal matrix, step Sla is repeated to compare the next complex tracking matrix with the input signal matrix for correlation. If one of the complex tracking matrices has correlation with the input signal matrix, step Slb is performed.

[0034] Further, the angle range of the complex tracking matrices is from -90 degrees to 90 degrees, and a first interval angle is used to sequentially compare the corresponding complex tracking matrices with the input signal matrix for correlation. The comparison of correlation is the comparison of the angle correlation between the complex tracking matrices and the input signal matrix. The value of the first interval angle is between 1 / 180 and 1 / 18 (1 degree to 10 degrees) of the angle range of the complex tracking matrices. The user can set the first interval angle according to the requirement.

[0035] Step Sib: The tracking matrix having correlation with the input signal matrix is defined as a first tracking matrix, and a residual value is calculated using the first tracking matrix and the input signal matrix. In this embodiment, the first operation model calculates the residual value by the least square method.

[0036] Step Sic: The residual value is compared with a residual threshold. If the residual value is greatly reduced at one time and is lower than or equal to the residual threshold (indicating that the first tracking matrix satisfies the condition of the first operation model), the first operation model can obtain the target quantity and the estimated target angle. If the residual value is greater than the residual threshold, step Sid is performed. If the residual value is greater than the residual threshold, it means that the target quantity can be multiple, or the first tracking matrix has correlation with the input signal matrix, but the correlation is low.

[0037] Step Sid: The reduction amplitude of the residual value is compared. If the reduction amplitude of the residual value is greater than a significant threshold, it means that the first tracking matrix has high correlation with the input signal matrix, and step Sie is continued. If the reduction amplitude of the residual value is less than the significant threshold, step Sla is directly returned to continue the comparison of the input signal matrix with other complex tracking matrices for correlation until the residual value is lower than or equal to the residual threshold, so that the first operation model can obtain the target quantity and the estimated target angle. If the reduction amplitude of the residual value is less than the significant threshold, it means that the first tracking matrix has correlation with the input signal matrix, but the correlation is low.

[0038] Step S1e: storing the first tracking matrix, converting the input signal matrix into a second input signal matrix irrelevant to the first tracking matrix, and returning to step S1a to compare the second input signal matrix with other complex tracking matrices until the residual value is lower than or equal to the residual threshold, so that the first operation model can obtain the target quantity and the estimated target angle.

[0039] Step S2: converting the input signal matrix into a covariance matrix, inputting the covariance matrix and the target quantity obtained by the first operation model into a second operation model to perform singular value decomposition, and obtaining a noise matrix. The input signal matrix is converted into the covariance matrix by formula (2).

[0040] R = E{yy H}…(2), wherein R is the covariance matrix; y is the input signal matrix.

[0041] Step S3: obtaining an iteration angle range by the estimated target angle obtained in step S1, wherein the iteration angle range is obtained by adding and subtracting a permissible angle to the estimated target angle, and the permissible angle is smaller than the first interval angle.

[0042] Step S4: inputting the complex tracking matrix corresponding to the iteration angle range and the noise matrix into a third operation model (such as MUltiple SIgnal Classification, MUSIC), and performing iteration of the angle range, so as to obtain a precise target angle. The iteration angle range is sequentially input into the third operation model in units of a second interval angle, and iteration of the angle range is performed with the noise matrix.

[0043] Further, the second interval angle is between 1 / 100 and 1 / 10 of the iteration angle range; the third operation model performs iteration of the iteration angle range and the noise matrix by orthogonal operation, and after iteration by the third operation model, a curve of angle and power is obtained, and the angle corresponding to the obvious peak value on the curve is the precise target angle.

[0044] Please refer to Figure 3 and Figure 4 , there is a vehicle radar 1, which has a detection reference line L, the detection reference line L is perpendicular to the vehicle radar 1, the detection reference line L is negative in the upward direction, and the detection reference line L is positive in the downward direction. As shown in Figure 3 , it is the first embodiment of the present application, assuming that there is an object 2, and taking the detection reference line L as the reference, the angle of the object 2 is -12 degrees; as shown inFigure 4 As shown in FIG. 1, a first object 3a and a second object 3b are assumed, and the angle of the first object 3a is -12 degrees and the angle of the second object 3b is 21 degrees based on the detection reference line L. The actual detection steps of the present application in the first embodiment are described as follows: Please refer to FIG. 1 Figures 1-3 As shown in FIG. 1, in step S1, the input signal matrix is inputted into the first operation model, and the complex tracking matrix is inputted for comparison. When the complex tracking matrix satisfies the condition of the first operation model, the target object quantity and the estimated target angle can be obtained. The input signal matrix is obtained by receiving the reflection signal of the object 2 by the vehicle radar 1. The complex tracking matrix is composed of steering vectors related to the angle, and the range of the complex tracking matrix is from -90 degrees to 90 degrees. Step S1 judges whether the complex tracking matrix satisfies the condition of the first operation model through steps S1a to S1e.

[0045] In step S1a, the first operation model compares the complex tracking matrix with the input signal matrix in sequence from -90 degrees to 90 degrees in units of the first interval angle. If one of the complex tracking matrices has no correlation with the input signal matrix, step S1a is repeated.

[0046] Taking the first embodiment as an example, the first interval angle is set to 5 degrees. For example, when the first operation model compares the correlation between the tracking matrix with the angle of -40 degrees and the input signal matrix, since the angle of the object 2 (-12 degrees) is greatly different from the angle of the tracking matrix (-40 degrees), the first operation model determines that the tracking matrix with the angle of -40 degrees has no correlation with the input signal matrix, and thus the first operation model repeats step S1a to sequentially compare the correlation between the complex tracking matrix with the next angle and the input signal matrix.

[0047] On the contrary, if one of the complex tracking matrices has correlation with the input signal matrix, step S1b is performed. Taking the first embodiment as an example, when the first operation model compares the correlation between the tracking matrix with the angle of -10 degrees and the input signal matrix, since the angle of the object 2 (-12 degrees) is very close to the angle of the tracking matrix (-10 degrees), the first operation model determines that the tracking matrix with the angle of -10 degrees has correlation with the input signal matrix, and continues to perform step S1b.

[0048] In step S1b, the tracking matrix related to the input signal matrix is defined as the first tracking matrix, and the first operation model performs the residual value operation by the least square method according to the first tracking matrix and the input signal matrix.

[0049] For example, in the first embodiment, the tracking matrix with an angle of -10 degrees is defined as the first tracking matrix, and the residual value operation is performed. The initial value of the residual value is 100%.

[0050] In step S1c, the residual value and the residual threshold are compared. If the residual value is greatly reduced at one time and is lower than or equal to the residual threshold, it means that the first tracking matrix meets the condition of the first operation model, and the first operation model can obtain the target object quantity and the estimated target angle. If the residual value is greater than the residual threshold, step S1d is performed.

[0051] For example, in the first embodiment, the residual threshold is set to 10%, and the calculated residual value is 5%. Since object 2 is only one and the angle (-12 degrees) is very close to the angle (-10 degrees) of the first tracking matrix, the residual value is greatly reduced at one time and is lower than the residual threshold (5%<10%), so the first operation model can obtain the estimated target angle of -10 degrees. Moreover, since the residual value is greatly reduced at one time and is lower than the residual threshold (5%<10%), the first operation model can obtain the target object quantity of one.

[0052] Since the first embodiment has obtained the target object quantity (one) and the estimated target angle (-10 degrees) in step S1c, it means that the purpose of step S1 (obtaining the target object quantity and the estimated target angle) has been achieved, so steps S1d and S1e do not need to be performed, and step S2 can be continued.

[0053] In step S2, the input signal matrix is converted into the covariance matrix, and the covariance matrix and the target object quantity are input into the second operation model to perform singular value decomposition (SVD) to obtain the noise matrix. For example, in the first embodiment, the covariance matrix and the target object quantity (one) are input into the second operation model to perform singular value decomposition to obtain the noise matrix.

[0054] In step S3, the estimated target angle is added and subtracted by the allowable angle to obtain the iteration angle range, and the allowable angle is less than the first interval angle. In the first embodiment, the allowable angle is set to 3 degrees, and the estimated target angle (-10 degrees) is added and subtracted by the allowable angle (3 degrees) to obtain the iteration angle range of -7 degrees to -13 degrees. The allowable angle (3 degrees) is less than the first interval angle (5 degrees).

[0055] In step S4, the complex tracking matrix corresponding to the iteration angle range and the noise matrix are input into the third operation model, and the third operation model performs angle iteration of the complex tracking matrix and the noise matrix by using the orthogonal operation method to obtain the accurate target angle. The iteration angle range is in units of the second interval angle, and the corresponding complex tracking matrix is sequentially input into the third operation model.

[0056] In the first embodiment, the second interval angle is set to 0.1 degrees. The complex tracking matrix corresponding to the iteration angle range (-7 degrees to -13 degrees) is sequentially input into the third operation model (such as MUltiple SIgnal Classification, MUSIC) in units of the second interval angle (0.1 degrees) and the noise matrix, and the third operation model performs angle iteration of the complex tracking matrix corresponding to the iteration angle range (-7 degrees to -13 degrees) and the noise matrix by using the orthogonal operation method to obtain the accurate target angle (-12 degrees).

[0057] After iteration by the third operation model, the angle-power curve is obtained. Since the object 2 is located at -12 degrees, the curve only has a significant peak at -12 degrees, indicating that -12 degrees is the accurate target angle.

[0058] As described above, as shown in the first embodiment, the vehicle radar 1 can detect an object 2 located at -12 degrees by using steps S1 to S4 of the present application, and there is no other object at other angles. Figure 3

[0059] The actual detection steps in the second embodiment of the present application are described below. Please refer to Figures 1-4 ​As shown, in step S1, the input signal matrix is input to the first operation model, and the complex tracking matrix is input for comparison. When the complex tracking matrix satisfies the condition of the first operation model, the target quantity and the coarse estimated target angle can be obtained. The input signal matrix is obtained by receiving the reflection signals of the first object 3a and the second object 3b by the vehicle radar 1. The complex tracking matrix is composed of steering vectors related to angles, and the range of the complex tracking matrix is from -90 degrees to 90 degrees. Step S1 judges whether the complex tracking matrix satisfies the condition of the first operation model through steps S1a to S1e.

[0060] In step S1a, the first operation model compares the correlation between the complex tracking matrix and the input signal matrix in sequence from -90 degrees to 90 degrees in units of the first interval angle. If one of the complex tracking matrices has no correlation with the input signal matrix, step S1a is repeated.

[0061] For example, when the first operation model compares the correlation between the tracking matrix with an angle of -40 degrees and the input signal matrix, since the angle (-12 degrees) of the first object 3a and the angle (21 degrees) of the second object 3b are greatly different from the angle (-40 degrees) of the tracking matrix, the first operation model determines that the tracking matrix with an angle of -40 degrees has no correlation with the input signal matrix. Therefore, the first operation model repeats step S1a to sequentially compare the correlation between the complex tracking matrix with the next angle and the input signal matrix.

[0062] On the contrary, if one of the complex tracking matrices has correlation with the input signal matrix, step S1b is performed. For example, when the first operation model compares the correlation between the tracking matrix with an angle of -15 degrees and the input signal matrix, since the angle (-12 degrees) of the first object 3a is close to the angle (-15 degrees) of the tracking matrix, the first operation model determines that the tracking matrix with an angle of -15 degrees has correlation with the input signal matrix, and step S1b is continued.

[0063] In step S1b, the tracking matrix having correlation with the input signal matrix is defined as the first tracking matrix, and the first operation model performs the operation of the residual value by the least square method according to the first tracking matrix and the input signal matrix.

[0064] Taking the second embodiment as an example, the first operational model defines the tracking matrix with an angle of -15 degrees as the first tracking matrix, and performs the calculation of the residual value. The initial value of the residual value is 100%.

[0065] In step S1c, the residual value and the residual threshold are compared. If the residual value decreases significantly in one go and is lower than or equal to the residual threshold, it means that the first tracking matrix meets the conditions of the first operation model, and the first operation model can obtain the number of target objects and the estimated target angle. If the residual value is greater than the residual threshold, then step S1d is performed.

[0066] Taking the second embodiment as an example, the residual threshold is set to 10%, and the residual value is calculated to be 90%. The residual value is still greater than the residual threshold (90%>10%), so the next step (step S1d) must be performed.

[0067] In step S1d, the reduction rate of the residual value is compared. If the reduction rate of the residual value is less than the significant threshold, the process returns directly to step S1a to continue comparing the correlation between the input signal matrix and other complex tracking matrices until the residual value is lower than the residual threshold; if the reduction rate of the residual value is greater than the significant threshold, the process proceeds to step S1e.

[0068] Taking the second embodiment as an example, the significance threshold is set to 35%. The residual value is calculated to be 90%, but the reduction of the residual value (reduction of 10%) is less than the significance threshold (35%), which means that the first tracking matrix (angle of -15 degrees) is correlated with the input signal matrix, but not highly correlated. Therefore, the first operation model will return to step S1a to perform the next correlation comparison between the tracking matrix (angle of -10 degrees) and the input signal matrix.

[0069] Continuing from the above, the first operational model repeats steps S1a to S1d. In step S1a, since the difference between the angle (-12 degrees) of the first object 3a and the angle (-10 degrees) of the tracking matrix is ​​small, the first operational model determines that the tracking matrix with an angle of -10 degrees is correlated with the input signal matrix. In step S1b, the tracking matrix with an angle of -10 degrees is defined as the first tracking matrix, and the residual value is calculated. At this time, the initial value of the residual value is 90%. In step S1c, the calculated residual value is 50%, which is still greater than the residual threshold (50% > 10%), so step S1d must be performed. In step S1d, the reduction in the residual value (from 90% to 50%, a reduction of 40%) is greater than the significant threshold (35%), so step S1e is continued. Wherein, the reduction in the residual value (reduction of 40%) is greater than the significant threshold, indicating that the first tracking matrix (angle of -10 degrees) has a high correlation with the input signal matrix.

[0070] In step S1e, the first tracking matrix is ​​stored, and the input signal matrix is ​​converted into a second input signal matrix that is independent of the first tracking matrix. Then, the process returns to step S1a to compare the correlation between the second input signal matrix and other complex tracking matrices until the residual value is lower than or equal to the residual threshold, so that the first calculation model can obtain the number of target objects and the estimated target angle.

[0071] Taking the second embodiment as an example, the first tracking matrix (angle of -10 degrees) is stored, and the input signal matrix is ​​converted into a second input signal matrix (unrelated to -10 degrees) that is independent of the first tracking matrix (angle of -10 degrees). Then, the process returns to step S1a to compare the correlation between the next tracking matrix (angle of -5 degrees) and the second input signal matrix, until the residual value is lower than or equal to the residual threshold (10%). Here, storing the first tracking matrix (angle of -10 degrees) represents one of the estimated target angles being -10 degrees.

[0072] As described above, in the second embodiment, the first calculation model, through steps S1a to S1e, can finally obtain that the number of target objects is 2 (the first object 3a and the second object 3b), and the estimated target angles are -10 degrees and 20 degrees.

[0073] In step S2, the input signal matrix is ​​converted into the covariance matrix, and the covariance matrix and the number of target objects are input into the second computational model for Singular Value Decomposition (SVD) to obtain the noise matrix. Taking the second embodiment as an example, the covariance matrix and the number of target objects (2) are input into the second computational model for Singular Value Decomposition to obtain the noise matrix.

[0074] In step S3, the iterative angle range is obtained by adding and subtracting the allowable angle from the coarsely estimated target angles, and the allowable angle is less than the first interval angle. Taking the second embodiment as an example, the allowable angle is set to 3 degrees, and the coarsely estimated target angles (-10 degrees and 20 degrees) are added and subtracted from the allowable angle (3 degrees) to obtain the iterative angle range of -7 degrees to -13 degrees and 17 degrees to 23 degrees. The allowable angle (3 degrees) is less than the first interval angle (5 degrees).

[0075] In step S4, the complex tracking matrix corresponding to the iterative angle range and the noise matrix are input to the third operational model. The third operational model then performs angle iteration on the complex tracking matrix and the noise matrix using orthogonal operations to obtain the precise target angle. The iterative angle range is defined in units of the second interval angle, and the corresponding complex tracking matrices are sequentially input to the third operational model.

[0076] Taking the second embodiment as an example, the second interval angle is set to 0.1 degrees. The complex tracking matrix corresponding to the iterative angle range (-7 degrees to -13 degrees and 17 degrees to 23 degrees) is sequentially input to the third operation model along with the noise matrix in units of the second interval angle (0.1 degrees). The third operation model then performs orthogonal operations sequentially on the complex tracking matrix and the noise matrix corresponding to the iterative angle range (-7 degrees to -13 degrees and 17 degrees to 23 degrees) to obtain the precise target angles (-12 degrees and 21 degrees).

[0077] Please refer to Figure 5 After the third calculation model iteration, the curve of angle and power can be obtained. Since the angle of the first object 3a is -12 degrees and the angle of the second object 3b is 21 degrees, the curve will only have obvious peak values ​​at -12 degrees and 21 degrees, indicating that -12 degrees and 21 degrees are the precise target angles.

[0078] As mentioned above, such as Figure 4As shown, in the second embodiment, the vehicle radar 1 can detect a first object 3a at an angle of -12 degrees and a second object 3b at an angle of 21 degrees through steps S1 to S4 of the present invention, and there are no other objects at other angles.

[0079] For further details, please refer to Figure 6 In other embodiments of the present invention, after the calculations in steps S1 to S4 of the present invention, it is found that the curve has obvious peak values ​​at angles of -1 degree and 1 degree, representing that there is one target object at each of the angles of -1 degree and 1 degree. More specifically, as... Figure 6 As shown, using the method of the present invention, even if the angles between the target objects are very close (angles of -1 degree and 1 degree), the curve can still accurately show a clear peak at the corresponding angle positions (angles of -1 degree and 1 degree). Therefore, the user can directly and quickly know that there is one target object at the positions of -1 degree and 1 degree respectively through the curve.

[0080] Therefore, the present invention has the following advantages: 1. The present invention uses the first calculation model to perform preliminary calculations over a large range, excluding most angles that are irrelevant to the input signal matrix, and thereby obtains the number of target objects and the estimated target angles, which further improves the calculation speed of the present invention.

[0081] 2. This invention, through the combination of the second and third calculation modules, decomposes the coarsely estimated target angle into a more detailed value and obtains the precise target angle through angle iteration. This enables the invention to achieve a more accurate determination of the target object's position.

[0082] 3. The present invention uses the noise matrix calculated by singular value decomposition (SVD) and the angle iteration using orthogonal operation method so that the curve obtained by the third operation module will only have a significant peak value at the precise target angle, thereby allowing the user to more clearly know the angle information of the target object.

[0083] Although the present invention has been described with reference to a preferred embodiment, those skilled in the art can make various modifications without departing from the spirit and scope of the invention. The embodiments described above are merely illustrative of the invention and are not intended to limit its scope. All modifications or alterations made without departing from the spirit of the invention are within the scope of the patent applications of this invention.

Claims

1. A method for high-angle resolution target angle estimation using a large array antenna, comprising receiving an input signal matrix through an array antenna, wherein the array antenna is installed within a vehicle radar, and the input signal matrix is ​​a transmitted or reflected signal from at least one target, characterized in that... The method includes the following steps: Step S1: Input the input signal matrix into a first operation model, and input a complex tracking matrix composed of angle-related guidance vectors for comparison. When the complex tracking matrix satisfies the conditions of the first operation model, a rough estimate of the target angle corresponding to the location of the target object is obtained. Step S2: Input the input signal matrix into a second operational model to perform singular value decomposition and obtain a noise matrix; Step S3: Obtain an iterative angle range using the coarsely estimated target angle obtained in step S1; and Step S4: Input the complex tracking matrix and the noise matrix corresponding to the iterative angle range into a third operation model and iterate the angle range to obtain a precise target angle.

2. The target angle estimation method using a large array antenna for high-angle resolution according to claim 1, characterized in that, Step S1 further includes a step S1a, in which the first operational model compares whether the input signal matrix and the complex tracking matrix are correlated. If one of the complex tracking matrices is not correlated with the input signal matrix, step S1a is repeated to perform another correlation comparison between the complex tracking matrix and the input signal matrix. If one of the complex tracking matrices is correlated with the input signal matrix, step S1b is performed: a residual value is calculated based on the correlated tracking matrix and the input signal matrix.

3. The target angle estimation method using a large array antenna for high-angle resolution according to claim 2, characterized in that, In step S1b, the tracking matrix that is related to the input signal matrix is ​​defined as a first tracking matrix, and the residual value is calculated using the first tracking matrix and the input signal matrix.

4. The target angle estimation method using a large array antenna for high-angle resolution according to claim 3, characterized in that, In step S1, there is also a step S1c, which compares the residual value with a residual threshold. If the residual value decreases significantly in one go and is lower than or equal to the residual threshold, the first calculation model obtains the rough target angle. If the residual value is greater than the residual threshold, a step S1d is performed: the reduction rate of the residual value is compared.

5. The target angle estimation method using a large array antenna for high-angle resolution according to claim 4, characterized in that, In step S1d, if the reduction in the residual value is greater than a significant threshold, proceed to step S1e: store the first tracking matrix, convert the input signal matrix into a second input signal matrix that is independent of the first tracking matrix, return to step S1a, and compare the correlation between the second input signal matrix and other complex tracking matrices; if the reduction in the residual value is less than the significant threshold, directly return to step S1a and compare the correlation between the input signal matrix and other complex tracking matrices.

6. The target angle estimation method using a large array antenna for high-angle resolution according to claim 3, characterized in that, In step S1b, the first operational model performs the calculation of the residual value using the least squares method.

7. The target angle estimation method using a large array antenna for high-angle resolution according to claim 1, characterized in that, In step S1, when the complex tracking matrix satisfies the conditions of the first operation model, a target quantity is obtained; in step S2, the input signal matrix is ​​first converted into a common variance matrix, and then singular value decomposition is performed on the target quantity.

8. The target angle estimation method using a large array antenna for high-angle resolution according to claim 1, characterized in that, In step S3, the range of the iterative angle is determined by adding and subtracting the roughly estimated target angles by an allowable angle.

9. The method for estimating the angle of a target object using a large array antenna with high-angle resolution according to claim 1, characterized in that, In step S4, the third operation model iterates the iterative angle range and the noise matrix using orthogonal operation.

10. The method for high-angle resolution target angle estimation using a large array antenna according to claim 1, characterized in that, In step S4, after the third calculation model is iterated, a curve of angle and power can be obtained, and the angle corresponding to the highest peak on the curve is the precise target angle.

Images