Fault detection method for digital phased array

Abstract

The invention discloses a fault detection method for a digital phased array, and the method comprises the following steps: S1, enabling a plurality of fault-free digital phased arrays and a plurality of digital phased arrays with known fault types to carry out the signal transmission at a transmission position according to a preset configuration, and then carrying out the signal receiving at a receiving position through employing a detection receiver; s2, constructing signal characteristics of each digital phased array by using the signals received by the detection receiver; s3, constructing a training sample set for fault detection of the digital phased arrays according to the signal features of the digital phased arrays; s4, constructing a fault detection model by adopting a machine learning algorithm, and training the fault detection model by utilizing the training sample set; and S5, performing fault detection on the to-be-detected digital phased array by using the fault detection model. According to the method, signal features are constructed based on signal detection of digital phased arrays of different fault types and fault-free digital phased arrays, and the fault detection model is trained, so that the fault type of the to-be-detected digital phased array can be rapidly detected.

Description

Technical Field

[0001] The invention relates to the field of digital phased arrays, and in particular to a fault detection method for digital phased arrays. Background Art

[0002] Digital phased arrays accurately control the beam pointing and forming of antennas in a digital way and are commonly used equipment in the communications field. However, when it comes to digital phased array fault detection, it is often necessary to disassemble the digital phased array and check its internal structure one by one. The detection is relatively complicated and it is impossible to quickly locate the fault type of the digital phased array. It is not convenient for maintenance personnel to perform maintenance accurately and the detection will waste a lot of manpower and time costs. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a digital phased array fault detection method. Based on signal detection of digital phased arrays with different fault types and fault-free digital phased arrays, signal features are constructed, and a fault detection model is trained through machine learning, so as to realize rapid detection of the fault type of the digital phased array to be tested.

[0004] The objective of the present invention is achieved through the following technical solution: A digital phased array fault detection method comprises the following steps:

[0005] S1. Setting the transmitting position and receiving position, for multiple fault-free digital phased arrays and multiple digital phased arrays with known fault types, respectively transmitting signals at the transmitting position according to the preset configuration, and then using the detection receiver to respectively receive signals at the receiving position;

[0006] S2. constructing the signal characteristics of each digital phased array using the signal received by the detection receiver;

[0007] S3. Constructing a training sample set for digital phased array fault detection based on the signal characteristics of each digital phased array;

[0008] S4. Use a machine learning algorithm to build a fault detection model, and use a training sample set to train the fault detection model to obtain a trained fault detection model;

[0009] S5. For the digital phased array to be detected, it is set at the transmitting position to transmit the signal according to the preset configuration, and then the detection receiver is used to receive the signal at the receiving position. Then, the signal characteristics corresponding to the digital phased array to be detected are constructed and sent to the trained fault detection model to obtain the fault detection result.

[0010] The beneficial effects of the present invention are as follows: (1) based on the signal detection of digital phased arrays with different fault types and fault-free digital phased arrays, signal features are constructed, and a fault detection model is trained by machine learning, so that the fault type of the digital phased array to be tested can be quickly detected;

[0011] (2) When constructing signal features, the present invention selects the average value of the detection results of multiple fault-free digital phased arrays as a benchmark, and then calculates the difference between the detection results of the digital phased arrays without faults and those with known fault types and the benchmark as the signal feature. The signal feature can reflect the difference with the benchmark under various fault types, so that a more accurate fault detection model can be trained.

[0012] (3) The present application comprehensively considers the frequency, amplitude and phase characteristics, which can better reflect the difference between the received signal corresponding to the digital phased array of known fault types and the reference, and can further improve the accuracy of the fault detection model. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0014] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0015] like Figure 1 As shown, a digital phased array fault detection method is characterized by comprising the following steps:

[0016] S1. Setting the transmitting position and receiving position, for multiple fault-free digital phased arrays and multiple digital phased arrays with known fault types, respectively transmitting signals at the transmitting position according to the preset configuration, and then using the detection receiver to respectively receive signals at the receiving position;

[0017] The preset configuration includes: signal amplitude configuration, signal phase configuration and signal frequency configuration;

[0018] For a normal digital phased array and a digital phased array with multiple faults, the preset configuration is the same.

[0019] The step S1 comprises:

[0020] S101. Using the i-th fault-free digital phased array, it transmits signals at a transmitting position according to a preset configuration for T consecutive moments, uses a detection receiver to receive signals at a receiving position, and detects the frequency, amplitude, and phase of the signal received by the receiver within T moments:

[0021] Among them, the frequency of the received signal within T moments is recorded as:

[0022] The amplitude of the received signal within T moments is recorded as:

[0023] The phase of the received signal within T moments is recorded as:

[0024] S102. When i=1, 2, ..., N, repeat step S101, where N represents the number of fault-free digital phased arrays;

[0025] S103. Using the j-th digital phased array of known fault type, it transmits signals at a transmitting position according to a preset configuration for T consecutive moments, uses a detection receiver to receive signals at a receiving position, and detects the frequency, amplitude, and phase of the signal received by the receiver within T moments:

[0026] Among them, the frequency of the received signal within T moments is recorded as:

[0027] The amplitude of the received signal within T moments is recorded as:

[0028] The phase of the received signal within T moments is recorded as:

[0029] S104. When j=1, 2, ..., M, repeat step S103, where M represents the number of digital phased arrays of known fault types.

[0030] S2. constructing the signal characteristics of each digital phased array using the signal received by the detection receiver;

[0031] The step S2 comprises:

[0032] S201. Calculate the frequency average, amplitude average, and phase average of N fault-free digital phased arrays within T time points:

[0033] (1) At the kth moment, the average frequency of N fault-free digital phased arrays is recorded as: The average amplitude of N fault-free digital phased arrays is recorded as The average amplitude of N fault-free digital phased arrays is recorded as

[0034] (2) When k = 1, 2, ..., T, repeat (1) to obtain the average frequency, average amplitude and average phase within T time points, which are recorded as

[0035] S203. For the i-th fault-free digital phased array, construct its signal feature, which is recorded as:

[0036]

[0037] S204. When i=1, 2, ..., N, repeat step S203 to obtain all fault-free digital phased array signal characteristics;

[0038] S205. For the j-th digital phased array with known fault type, construct its signal feature, which is recorded as:

[0039]

[0040] S206. When j=1, 2, ..., M, repeat step S205 to obtain all fault-free digital phased array signal characteristics.

[0041] S3. Constructing a training sample set for digital phased array fault detection based on the signal characteristics of each digital phased array;

[0042] The step S3 comprises:

[0043] S301. For each signal feature of a fault-free digital phased array, add a label 0 to it;

[0044] S302. Assume that there are M digital phased arrays with known fault types, and there are P fault types in total. Each fault type has one or more digital phased arrays, and the P fault types are numbered 1 to P;

[0045] For each signal feature of the digital phased array with a known fault type, a label is added to it, and the label is the fault type number;

[0046] S303. Add all labeled signal features into the same set to form a training sample set for digital phased array fault detection.

[0047] S4. Use a machine learning algorithm to build a fault detection model, and use a training sample set to train the fault detection model to obtain a trained fault detection model;

[0048] In step S4, the machine learning algorithm includes but is not limited to one of naive Bayes, support vector machine or BP neural network algorithm.

[0049] When the step S4 is performed for training, the signal features in the training sample set are used as the input of the fault detection model, and the labels corresponding to the signal features are used as the expected outputs. When the training of all signal features is completed, a trained fault detection model is obtained.

[0050] S5. For the digital phased array to be detected, it is set at the transmitting position to transmit the signal according to the preset configuration, and then the detection receiver is used to receive the signal at the receiving position. Then, the signal characteristics corresponding to the digital phased array to be detected are constructed and sent to the trained fault detection model to obtain the fault detection result.

[0051] The step S5 comprises:

[0052] S501. The digital phased array to be tested is set at the transmitting position, and the signal is continuously transmitted for T time periods according to the preset configuration. The detection receiver is used to receive the signal at the receiving position, and the frequency, amplitude and phase of the signal received by the receiver within T time periods are detected:

[0053] Among them, the frequencies of the received signals within T moments are recorded as: f1,f2,...,f T ;

[0054] The amplitude of the received signal within T moments is recorded as: A1, A2, ..., A T ;

[0055] The phase of the received signal within T moments is recorded as: θ1,θ2,...,θ T ;

[0056] S502. For the digital phased array to be tested, construct its signal characteristics, which are recorded as:

[0057] [(f1-f1,f2-f2,...,f T -f T ),(A1-A1,A2-A2,...,A T -A T ),(θ1-θ1,θ2-θ2,...,θ T -θ T )]

[0058] S503. Input the signal characteristics of the digital phased array to be tested into the trained fault detection model. If the fault detection model outputs 0, it is considered that the digital phased array to be tested has no fault; if the fault detection model outputs any number from 1 to P, it is considered that the fault type corresponding to the number is the fault type of the digital phased array.

[0059] The above is a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A digital phased array fault detection method, characterized in that: The following steps are involved: S1. Setting the transmitting position and receiving position, for multiple fault-free digital phased arrays and multiple digital phased arrays with known fault types, respectively transmitting signals at the transmitting position according to the preset configuration, and then using the detection receiver to respectively receive signals at the receiving position; S2. constructing the signal characteristics of each digital phased array using the signal received by the detection receiver; S3. Constructing a training sample set for digital phased array fault detection based on the signal characteristics of each digital phased array; S4. Use a machine learning algorithm to build a fault detection model, and use a training sample set to train the fault detection model to obtain a trained fault detection model; S5. For the digital phased array to be detected, it is set at the transmitting position to transmit the signal according to the preset configuration, and then the detection receiver is used to receive the signal at the receiving position. Then, the signal characteristics corresponding to the digital phased array to be detected are constructed and sent to the trained fault detection model to obtain the fault detection result.

2. A digital phased array fault detection method according to claim 1, characterized in that: The preset configuration includes: signal amplitude configuration, signal phase configuration and signal frequency configuration; For a normal digital phased array and a digital phased array with multiple faults, the preset configuration is the same.

3. The digital phased array fault detection method according to claim 1, characterized in that: The step S1 comprises: S101. Using the i-th fault-free digital phased array, it transmits signals at a transmitting position according to a preset configuration for T consecutive moments, uses a detection receiver to receive signals at a receiving position, and detects the frequency, amplitude, and phase of the signal received by the receiver within T moments: Among them, the frequency of the received signal within T moments is recorded as: The amplitude of the received signal within T moments is recorded as: The phase of the received signal within T moments is recorded as: S102. When i=1, 2, ..., N, repeat step S101, where N represents the number of fault-free digital phased arrays; S103. Using the j-th digital phased array of known fault type, it transmits signals at a transmitting position according to a preset configuration for T consecutive moments, uses a detection receiver to receive signals at a receiving position, and detects the frequency, amplitude, and phase of the signal received by the receiver within T moments: Among them, the frequency of the received signal within T moments is recorded as: The amplitude of the received signal within T moments is recorded as: The phase of the received signal within T moments is recorded as: S104. When j=1, 2, ..., M, repeat step S103, where M represents the number of digital phased arrays of known fault types.

4. The digital phased array fault detection method according to claim 1, characterized in that: The step S2 comprises: S201. Calculate the frequency average, amplitude average, and phase average of N fault-free digital phased arrays within T time points: (1) At the kth moment, the average frequency of N fault-free digital phased arrays is recorded as: The average amplitude of N fault-free digital phased arrays is recorded as The average amplitude of N fault-free digital phased arrays is recorded as (2) When k = 1, 2, ..., T, repeat (1) to obtain the average frequency, average amplitude and average phase within T time points, which are recorded as S203. For the i-th fault-free digital phased array, construct its signal feature, which is recorded as: S204. When i=1, 2, ..., N, repeat step S203 to obtain all fault-free digital phased array signal characteristics; S205. For the j-th digital phased array with known fault type, construct its signal feature, which is recorded as: S206. When j=1, 2, ..., M, repeat step S205 to obtain all fault-free digital phased array signal characteristics.

5. The digital phased array fault detection method according to claim 1, characterized in that: The step S2 includes: the step S3 includes: S301. For each signal feature of a fault-free digital phased array, add a label 0 to it; S302. Assume that there are M digital phased arrays with known fault types, and there are P fault types in total. Each fault type has one or more digital phased arrays, and the P fault types are numbered 1 to P; For each signal feature of the digital phased array with a known fault type, a label is added to it, and the label is the fault type number; S303. Add all labeled signal features into the same set to form a training sample set for digital phased array fault detection.

6. A digital phased array fault detection method according to claim 1, characterized in that: The step S2 includes: in the step S4, the machine learning algorithm includes but is not limited to one of naive Bayes, support vector machine or BP neural network algorithm.

7. The digital phased array fault detection method according to claim 1, characterized in that: The step S2 includes: when the step S4 is trained, the signal features in the training sample set are used as the input of the fault detection model, and the labels corresponding to the signal features are trained as the expected outputs. When all signal features are trained, a trained fault detection model is obtained.

8. The digital phased array fault detection method according to claim 1, characterized in that: The step S2 includes: the step S5 includes: S501. The digital phased array to be tested is set at the transmitting position, and the signal is continuously transmitted for T time periods according to the preset configuration. The detection receiver is used to receive the signal at the receiving position, and the frequency, amplitude and phase of the signal received by the receiver within T time periods are detected: Among them, the frequencies of the received signals within T moments are recorded as: f1,f2,...,f T ; The amplitude of the received signal within T moments is recorded as: A1, A2, ..., A T ; The phase of the received signal within T moments is recorded as: θ1,θ2,...,θ T ; S502. For the digital phased array to be tested, construct its signal characteristics, which are recorded as: S503. Input the signal characteristics of the digital phased array to be tested into the trained fault detection model. If the fault detection model outputs 0, it is considered that the digital phased array to be tested has no fault; if the fault detection model outputs any number from 1 to P, it is considered that the fault type corresponding to the number is the fault type of the digital phased array.

Images