Automatic parameter optimization direction finding method and device based on data learning and medium

Abstract

The invention discloses an automatic parameter optimization direction finding method based on data learning, and the method comprises the steps: S1, carrying out the multiple times of sample measurement, and obtaining a plurality of measurement sample values; s2, calculating a first mean value of a plurality of current samples based on a plurality of current measured sample values; s3, performing sample measurement to obtain a current latest sample; s4, setting a weighting coefficient; and S5, according to the weighting coefficient and the sample value of the current newest sample, calculating a second mean value of the current samples. According to the method, a large amount of sample data is received, continuous learning and convergence are carried out to obtain a result approaching to reality, the optimization of the direction finding result is realized, and the engineering realization problem of high-precision direction finding in the field of electronic countermeasure reconnaissance is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to an automatic parameter optimization and direction finding method, device and medium based on data learning. Background Art

[0002] At present, the common direction-finding algorithms in the direction-finding system are multi-beam amplitude comparison and interferometer direction-finding. Multi-beam amplitude comparison mainly compares the output amplitudes of each adjacent beam, calculates the amplitude values ​​of the maximum beam and the second largest beam, and obtains the angle information by matching the amplitude ratio through table lookup. It has the advantages of mature technology, fast processing speed, and simple engineering implementation, but its direction-finding accuracy is inversely proportional to the beam width. If you want to improve the direction-finding accuracy, you need to reduce the beam width, but the reduction of the beam width requires increasing the number of array elements or increasing the array element spacing, which requires a lot of hardware resources and is difficult to implement in engineering. Interferometer direction-finding determines its direction by measuring the phase difference between the wide beam antennas of the same radiation source signal reaching the antenna array. Interferometer direction-finding has the advantage of high direction-finding accuracy, but it is easily affected by factors such as phase error, frequency measurement error, and baseline length error.

[0003] During the signal transmission process, tropospheric scattering, low signal-to-noise ratio, azimuth broadening or multipath may easily cause large fluctuations or false direction finding results, and existing technologies cannot solve these problems. Summary of the invention

[0004] To solve the above problems, the present invention provides a method, device and medium for automatic parameter optimization and direction finding based on data learning. By receiving a large amount of sample data, continuous learning and convergence are performed to obtain results close to the truth, thereby optimizing the direction finding results and solving the engineering implementation problem of high-precision direction finding in the field of electronic countermeasure reconnaissance.

[0005] The present invention provides an automatic parameter optimization direction finding method based on data learning, and the specific technical scheme is as follows:

[0006] S1: perform multiple sample measurements to obtain multiple measurement sample values;

[0007] S2: Calculate the first mean of the current number of samples based on the sample values ​​obtained by the current multiple measurements;

[0008] S3: perform sample measurement and obtain the latest sample;

[0009] S4: Set weighting coefficient;

[0010] S5: Calculate the second mean of the current number of samples according to the weighting coefficient and the sample value of the current latest sample.

[0011] Furthermore, in step S1, multiple measurements of electromagnetic parameters are performed on the same electromagnetic radiation source to obtain the multiple measurement samples;

[0012] The electromagnetic radiation source for obtaining the sample measurement in step S3 is the same as the electromagnetic radiation source for performing the sample measurement in step S1.

[0013] Furthermore, the first mean is calculated as follows:

[0014]

[0015] Among them, m k represents the mean of the current k samples, k represents the number of samples measured in step S2, X i Represents the i-th sample value.

[0016] Furthermore, in step S4, the weighting coefficient is set as follows:

[0017]

[0018] Among them, k represents the number of samples in the current time series, and ρ represents the weighting coefficient.

[0019] Further, in step S5, the second mean is calculated as follows:

[0020] m k+1 =(1-ρ)·m k +ρ·X k+1

[0021] Among them, m k+1 represents the mean of k+1 samples based on the updated sample time series, X k+1 Represents the sample value of the k+1th sample.

[0022] Furthermore, in step S5, the calculation process of the second mean is specifically as follows:

[0023] S6: Set sample error;

[0024] S7: updating weighting coefficients based on the sample error;

[0025] S8: Calculate the second mean of the current number of samples according to the updated weighting coefficient and the sample value of the current latest sample k+1.

[0026] Furthermore, the sample error is calculated as follows:

[0027] e k+1 =X k+1 -m k

[0028] Among them, ek+1 represents the sample error of the current k+1 sample corresponding to step S4, represents the sample value of the k+1th sample corresponding to step S4, m k Represents the first mean of the k samples corresponding to step S2.

[0029] Furthermore, in step S7, the weighted coefficient is updated as follows:

[0030]

[0031] Among them, e k+1 represents the sample error of the current k+1 sample corresponding to step S4, and δ represents the direction finding accuracy of the device performing the sample measurement.

[0032] The present invention also provides an automatic parameter optimization and direction finding device based on data learning, and the automatic parameter optimization and direction finding device based on data learning includes: a memory, a processor, and an automatic parameter optimization and direction finding program based on data learning stored in the memory and executable on the processor. When the automatic parameter optimization and direction finding program based on data learning is executed by the processor, the steps of the automatic parameter optimization and direction finding method based on data learning are implemented.

[0033] The present invention also provides a computer storage medium, on which is stored an automatic parameter optimization and direction finding program based on data learning. When the automatic parameter optimization and direction finding program based on data learning is executed by a processor, the steps of the automatic parameter optimization and direction finding method based on data learning are implemented.

[0034] The beneficial effects of the present invention are as follows:

[0035] 1. The present invention receives a large amount of sample data, calculates the first mean of the first k samples, sets a weighting coefficient, and calculates the second mean of the k+1 sample according to the weighting coefficient and the sample value of the next sample k+1. The second mean is calculated using a closed analytical expression, and a result close to the truth is obtained through continuous learning and convergence, thereby optimizing the direction finding result and solving the engineering implementation problem of high-precision direction finding in the field of electronic countermeasure reconnaissance.

[0036] 2. After setting the weighting coefficient, the present invention also introduces sample error. The weighting coefficient is updated through the sample error. Then, the second mean of the k+1 sample is calculated based on the updated weighting coefficient and the sample value of the next sample k+1. The weighting coefficient is corrected using the sample error, thereby improving the accuracy and precision of the optimized direction finding result. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the method flow of Example 1 of the present invention.

[0038] Figure 2 It is a schematic diagram of the method flow of Example 2 of the present invention.

[0039] Figure 3 It is a schematic diagram of the distribution of original direction finding values ​​of a direction finding system based on an example of the method of the present invention.

[0040] Figure 4 It is a statistical schematic diagram of a histogram of the distribution of original direction finding values ​​of a direction finding system based on an example of the method of the present invention.

[0041] Figure 5 It is a schematic diagram of the distribution of data learning results based on probability statistics based on the example original direction finding values ​​of the method of the present invention. DETAILED DESCRIPTION

[0042] The following description clearly and completely describes the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] In the description of the embodiments of the present invention, it should be noted that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the invention product is usually placed when in use, or the orientation or positional relationship commonly understood by those skilled in the art, or the orientation or positional relationship in which the invention product is usually placed when in use, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first" and "second" are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0044] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] Example 1

[0046] Embodiment 1 of the present invention discloses an automatic parameter optimization direction finding method based on data learning, such as Figure 1 As shown, the specific steps are as follows:

[0047] S1: multiple measurements of electromagnetic parameters of the same electromagnetic radiation source are performed to obtain multiple measurement sample values;

[0048] For the same electromagnetic radiation source, multiple measurement samples of electromagnetic parameters constitute a time series; generally, as long as the parameters of the electromagnetic radiation source remain unchanged, such as azimuth, time difference and position, this time series approximates a random process, and each measurement sample is relatively independent and identically distributed, following the Gaussian distribution law.

[0049] S2: Calculate the first mean of the current number of samples based on the sample values ​​obtained by the current multiple measurements;

[0050] According to the law of large numbers, for an electromagnetic radiation source with constant parameters within a certain time range, the parameter measurement sample sequence X k (k=1,2,...), the frequency of the true value converges to its mean m according to probability, that is, for any ε>0,

[0051]

[0052] In fact, for most electromagnetic radiation sources, during a shorter time duration such as milliseconds or seconds, there will be no drastic changes in a short period of time, and their orientation and position will remain approximately unchanged. The same is true for the sequence of time differences measured by multiple stations.

[0053] Therefore, in this embodiment, the first mean is calculated as follows:

[0054]

[0055] Among them, m k represents the mean of the current k samples, k represents the number of samples measured in step S2, X i Represents the i-th sample value.

[0056] S3: perform sample measurement and obtain the latest sample;

[0057] Specifically, when measuring the sample, the electromagnetic radiation source for obtaining the sample measurement is the same as the electromagnetic radiation source for performing the sample measurement in step S1.

[0058] S4: Set weighting coefficient;

[0059] According to the law of large numbers, the calculation formula of the second mean in this embodiment has a convergence process. Only when ρ is large enough can ρ maintain m k+1 Will not deviate from the true value;

[0060] As a preferred embodiment, the weighting coefficient is set as follows:

[0061]

[0062] Among them, k represents the number of samples in the current time series, and ρ represents the weighting coefficient.

[0063] The weighting coefficient is set in three intervals, corresponding to a small number of data samples, a medium number of data samples and a large number of data samples.

[0064] S5: Calculate the second mean of the current number of samples according to the weighting coefficient and the sample value of the current latest sample.

[0065] Based on the first mean value calculation formula in step S2, it can be known that the mean value of k+1 samples is:

[0066]

[0067] in,

[0068] You can get the recursive formula: m k+1 =(1-ρ k+1 )·m k +ρ k+1 ·X k+1 ;

[0069] According to the law of large numbers, as k→∞, ρ k >ρ k+1 →0; that is, the larger k is, the smaller the weighting coefficient for the probability to converge to the true value is, which is correct for the parameter sample sequence whose true value is fixed. However, for the slowly changing radiation source parameters, such as maneuvering targets, whose direction and position change continuously, such calculation will lead to wrong results. In fact, if the statistical time is agreed to be within a shorter period, as long as its parameters do not change drastically and remain relatively stable, the recursive formula can be approximately close to the true value. At this time, ρ k+1 Instead of getting smaller and smaller, the weighted value in probability statistics can be appropriately larger. That is, as k→∞, ρ k =ρ k+1 =ρ.

[0070] In other words, as a preferred embodiment, for a slowly moving maneuvering target, the radiation source measurement parameter sequence can be used to calculate the second mean according to the following formula:

[0071] m k+1 =(1-ρ)·m k +ρ·X k+1

[0072] Among them, m k+1 represents the mean of k+1 samples based on the updated sample time series, X k+1 Represents the sample value of the k+1th sample.

[0073] Example 2

[0074] Embodiment 2 of the present invention discloses an automatic parameter optimization direction finding method based on data learning, such as Figure 2 As shown, the specific process is as follows:

[0075] S1: multiple measurements of electromagnetic parameters of the same electromagnetic radiation source are performed to obtain multiple measurement sample values;

[0076] For the same electromagnetic radiation source, multiple measurement samples of electromagnetic parameters constitute a time series; generally, as long as the parameters of the electromagnetic radiation source remain unchanged, such as azimuth, time difference and position, this time series approximates a random process, and each measurement sample is relatively independent and identically distributed, following the Gaussian distribution law.

[0077] S2: Calculate the first mean of the current number of samples based on the sample values ​​obtained by the current multiple measurements;

[0078] According to the law of large numbers, for an electromagnetic radiation source with constant parameters within a certain time range, the parameter measurement sample sequence X k (k=1,2,...), the frequency of the true value converges to its mean m according to probability, that is, for any ε>0,

[0079]

[0080] In fact, for most electromagnetic radiation sources, during a shorter time duration such as milliseconds or seconds, there will be no drastic changes in a short period of time, and their orientation and position will remain approximately unchanged. The same is true for the sequence of time differences measured by multiple stations.

[0081] Therefore, in this embodiment, the first mean is calculated as follows:

[0082]

[0083] Among them, m k represents the mean of the current k samples, k represents the number of samples measured in step S2, X i Represents the i-th sample value.

[0084] S3: perform sample measurement and obtain the latest sample;

[0085] Specifically, when measuring the sample, the electromagnetic radiation source for obtaining the sample measurement is the same as the electromagnetic radiation source for performing the sample measurement in step S1.

[0086] S4: Set weighting coefficient;

[0087] According to the law of large numbers, the calculation formula of the second mean in this embodiment has a convergence process. Only when ρ is large enough can ρ maintain mk+1 Will not deviate from the true value;

[0088] As a preferred embodiment, the weighting coefficient is set as follows:

[0089]

[0090] Among them, k represents the number of samples in the current time series, and ρ represents the weighting coefficient.

[0091] The weighting coefficient is set in three intervals, corresponding to a small number of data samples, a medium number of data samples and a large number of data samples.

[0092] S5: Calculate the second mean of the current number of samples according to the weighting coefficient and the sample value of the current latest sample.

[0093] Based on the first mean value calculation formula in step S2, it can be known that the mean value of k+1 samples is:

[0094]

[0095]

[0096] in,

[0097] You can get the recursive formula: m k+1 =(1-ρ k+1 )·m k +ρ k+1 ·X k+1 ;

[0098] According to the law of large numbers, as k→∞, ρ k >ρ k+1 →0; that is, the larger k is, the smaller the weighting coefficient for the probability to converge to the true value is, which is correct for the parameter sample sequence whose true value is fixed. However, for the slowly changing radiation source parameters, such as maneuvering targets, whose direction and position change continuously, such calculation will lead to wrong results. In fact, if the statistical time is agreed to be within a shorter period, as long as its parameters do not change drastically and remain relatively stable, the recursive formula can be approximately close to the true value. At this time, ρ k+1 Instead of getting smaller and smaller, the weighted value in probability statistics can be appropriately larger. That is, as k→∞, ρ k =ρ k+1 =ρ.

[0099] In other words, as a preferred embodiment, for a slowly moving maneuvering target, the radiation source measurement parameter sequence can be used to calculate the second mean according to the following formula:

[0100] mk+1 =(1-ρ)·m k +ρ·X k+1

[0101] Among them, m k+1 represents the mean of k+1 samples based on the updated sample time series, X k+1 Represents the sample value of the k+1th sample.

[0102] As a preferred embodiment, this embodiment also includes:

[0103] S6: Set sample error;

[0104] As a preferred embodiment, the sample error is calculated as follows:

[0105] e k+1 =X k+1 -m k

[0106] Among them, e k+1 represents the sample error of the current k+1 sample corresponding to step S4, represents the sample value of the k+1th sample corresponding to step S4, m k Represents the first mean of the k samples corresponding to step S2.

[0107] S7: updating weighting coefficients based on the sample error;

[0108] Considering that the electromagnetic radiation source parameter measurement sequence is not an ideal Gaussian distribution, it is affected by factors such as multipath reflection, tropospheric scattering, and low signal-to-noise ratio. Generally, the multipath reflection signal is weaker than the direct wave, and the parameter samples are relatively few; in the case of tropospheric scattering, the probability of the occurrence of samples of the radiation source azimuth broadening caused by strong signals is also small; during the radiation source antenna scanning cycle, the signal duration near the sensitivity is short, and the parameter measurement samples are also relatively few. In addition, the parameter flying point measured by the equipment is also a low-probability event, and its samples are relatively few.

[0109] That is, the measured values ​​of these parameters deviate greatly from the true values. At this time, when performing the second mean calculation, a more reasonable approach is to assign a smaller weight coefficient to ρ. The Gaussian distribution random sequence shows that the number of samples within 3δ accounts for more than 98%. Therefore, in this embodiment, by comparing e k+1 With 3δ, the weighting coefficients are reasonably allocated, and smaller weights are given to samples with errors exceeding 3δ, so as to minimize the impact of these points with large errors on the estimation of the true value.

[0110] As a preferred embodiment, the updating weighting coefficient is specifically as follows:

[0111]

[0112] Among them, e k+1 represents the sample error of the current k+1 sample corresponding to step S4, that is, the error between the k+1th measured sample and the estimated value of the true value of the previous k samples, and δ represents the direction finding accuracy of the device performing the sample measurement.

[0113] The samples with an error of no more than 3δ are facing normal Gaussian distribution samples, and the samples with an error of more than 3δ are facing multipath, low SNR and flying spots.

[0114] S8: Calculate the second mean of the current number of samples according to the updated weighting coefficient and the sample value of the current latest sample k+1;

[0115] The second mean is calculated as follows:

[0116] m k+1 =m k +ρ·e k+1 .

[0117] By performing a second mean calculation based on the weighted coefficients updated based on the sample error, the calculation is corrected so that the sample sequence converges to the vicinity of the true value.

[0118] Based on the method of the above embodiment, a direction finding system is used to detect a distant slowly moving radiation source target, the actual incident angle of the radiation source is 0°, and the obtained direction finding data is processed using the automatic parameter optimization method based on data learning proposed in the present invention, with a sample size of 1000.

[0119] like Figure 3 As shown in the figure, it is the distribution of the original direction finding value of the direction finding system. Figure 4 As shown in FIG, the histogram statistics of the original values ​​of the direction finding samples are performed. It can be seen that the original values ​​of the direction finding samples are random due to the influence of factors such as multipath reflection, tropospheric scattering, and low signal-to-noise ratio, but they obey the Gaussian distribution as a whole.

[0120] like Figure 5 As shown in FIG. 1 , the automatic parameter optimization algorithm based on data learning proposed by the present invention is used to Figure 3 The original direction finding values ​​in the method are used for data learning based on probability statistics. It can be seen that by receiving a large amount of sample original data, the method of the present invention can continuously learn and converge to obtain a result close to the true value.

[0121] Example 3

[0122] Embodiment 3 of the present invention discloses an automatic parameter optimization direction finding device based on data learning, and the device may be a user equipment (UE), such as a mobile phone, a smart phone, a laptop computer, a digital broadcast receiver, a personal digital assistant (PDA), a tablet computer (PAD), a handheld device, a vehicle-mounted device, a wearable device, a computing device or other processing device connected to a wireless modem, a mobile station (MS), etc., for executing the automatic parameter optimization direction finding method based on data learning. The device may be called a user terminal, a portable terminal, a desktop terminal, etc.

[0123] Typically, the device includes: at least one processor, a memory, and an automatic parameter optimization and direction finding program based on data learning stored in the memory and executable on the processor, wherein the automatic parameter optimization and direction finding program based on data learning is configured to implement the steps of the automatic parameter optimization and direction finding method based on data learning as described in any one of Embodiments 1 or 2.

[0124] The processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. The processor may also include an AI (Artificial Intelligence) processor, which is used to process calculation operations related to the automatic parameter optimization and direction finding program based on data learning, so that the automatic parameter optimization and direction finding method based on data learning can be trained and learned autonomously to improve efficiency and accuracy.

[0125] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory is used to store at least one instruction, and the at least one instruction is used to be executed by the processor to implement the automatic parameter optimization direction finding method based on data learning described in any of the above method embodiments 1 or 2.

[0126] In some embodiments, the terminal may also optionally include: a communication interface and at least one peripheral device. The processor, the memory and the communication interface may be connected via a bus or a signal line. Each peripheral device may be connected to the communication interface via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit, a display screen and a power supply.

[0127] The communication interface can be used to connect at least one peripheral device related to I / O (Input / Output) to the processor and the memory. The communication interface is used to receive the movement trajectories and other data of multiple mobile terminals uploaded by users through the peripheral device. In some embodiments, the processor, memory and communication interface are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor, memory and communication interface can be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0128] The radio frequency circuit is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit communicates with the communication network and other communication devices through electromagnetic signals, so that the movement trajectory and other data of multiple mobile terminals can be obtained. The radio frequency circuit converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and the like. The radio frequency circuit can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes but is not limited to: a metropolitan area network, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit may also include circuits related to NFC (Near Field Communication), which are not specifically limited here.

[0129] The display screen is used to display the UI (User Interface). The UI may include graphics, text, icons, videos and any combination thereof. When the display screen is a touch display screen, the display screen also has the ability to collect touch signals on the surface or above the surface of the display screen. The touch signal can be input to the processor as a control signal for processing. At this time, the display screen can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen can be one, the front panel of the electronic device; in other embodiments, the display screen can be at least two, respectively arranged on different surfaces of the electronic device or in a folding design; in some further embodiments, the display screen can be a flexible display screen, arranged on a curved surface or a folding surface of the electronic device. Even, the display screen can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode) and the like.

[0130] The power supply is used to power various components in the electronic device. The power supply can be AC, DC, a disposable battery or a rechargeable battery. When the power supply includes a rechargeable battery, the rechargeable battery can support wired charging or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0131] Example 4

[0132] Embodiment 4 of the present invention discloses a computer storage medium, which is a readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the automatic parameter optimization and direction finding method based on data learning as described in any one of Embodiments 1 or 2 of the above method are implemented.

[0133] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0134] The present invention is not limited to the above-mentioned specific embodiments, but extends to any new features or any new combination disclosed in this specification, as well as any new method or process steps or any new combination disclosed.

Claims

1. An automatic parameter optimization direction finding method based on data learning, characterized in that: include: S1: perform multiple sample measurements to obtain multiple measurement sample values; S2: Calculate the first mean of the current number of samples based on the sample values ​​obtained by the current multiple measurements; S3: perform sample measurement and obtain the latest sample; S4: Set weighting coefficient; S5: Calculate the second mean of the current number of samples according to the weighting coefficient and the sample value of the current latest sample.

2. The automatic parameter optimization direction finding method based on data learning according to claim 1 is characterized in that: In step S1, multiple measurements of electromagnetic parameters are performed on the same electromagnetic radiation source to obtain the multiple measurement samples; The electromagnetic radiation source for obtaining the sample measurement in step S3 is the same as the electromagnetic radiation source for performing the sample measurement in step S1.

3. The automatic parameter optimization direction finding method based on data learning according to claim 1 is characterized in that: The first mean is calculated as follows: Among them, m k represents the mean of the current k samples, k represents the number of samples measured in step S2, X i Represents the i-th sample value.

4. The automatic parameter optimization direction finding method based on data learning according to claim 1 is characterized in that: In step S4, the weighting coefficient is set as follows: Among them, k represents the number of samples in the current time series, and ρ represents the weighting coefficient.

5. The automatic parameter optimization direction finding method based on data learning according to claim 4 is characterized in that: In step S5, the second mean is calculated as follows: m k+1 =(1-ρ)·m k +p·X k+1 Among them, m k+1 represents the mean of k+1 samples based on the updated sample time series, X k+1 Represents the sample value of the k+1th sample.

6. The automatic parameter optimization direction finding method based on data learning according to claim 4 is characterized in that: In step S5, the calculation process of the second mean is specifically as follows: S6: Set sample error; S7: updating weighting coefficients based on the sample error; S8: Calculate the second mean of the current number of samples according to the updated weighting coefficient and the sample value of the current latest sample k+1.

7. The automatic parameter optimization direction finding method based on data learning according to claim 6 is characterized in that: The sample error is calculated as follows: e k+1 =X k+1 -m k Among them, e k+1 represents the sample error of the current k+1 sample corresponding to step S4, represents the sample value of the k+1th sample corresponding to step S4, m k Represents the first mean of the k samples corresponding to step S2.

8. The automatic parameter optimization direction finding method based on data learning according to claim 7 is characterized in that: In step S7, the weighting coefficient is updated as follows: Among them, e k+1 represents the sample error of the current k+1 sample corresponding to step S4, δ represents the direction finding accuracy of the device performing sample measurement, and ρ represents the weighting coefficient.

9. An automatic parameter optimization and direction finding device based on data learning, characterized in that: The automatic parameter optimization and direction finding device based on data learning comprises: a memory, a processor, and an automatic parameter optimization and direction finding program based on data learning stored in the memory and executable on the processor. When the automatic parameter optimization and direction finding program based on data learning is executed by the processor, the steps of the automatic parameter optimization and direction finding method based on data learning according to any one of claims 1 to 8 are implemented.

10. A computer storage medium, characterized in that: The computer storage medium stores an automatic parameter optimization and direction finding program based on data learning, and when the automatic parameter optimization and direction finding program based on data learning is executed by the processor, the steps of the automatic parameter optimization and direction finding method based on data learning according to any one of claims 1-8 are implemented.

Images