Signal spectrum sensing method, program product and equipment

Abstract

The invention discloses a signal spectrum sensing method, a program product and equipment. The method comprises the following steps: receiving a signal to be detected; obtaining a time-frequency signal and time-frequency signal spectrum estimation data based on the signal to be measured; executing time domain operation based on the time-frequency signal spectrum estimation data to obtain time domain processing data; determining target judgment data based on the time domain processing data; and determining parameter data of the to-be-detected signal based on the target judgment data.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a signal spectrum sensing method, a computer program product and a signal spectrum sensing device. Background Art

[0002] With the continuous advancement and popularization of signal source equipment (such as drones), signal source equipment is increasingly widely used in civil and commercial fields. However, the security threats brought by signal source equipment are also becoming increasingly prominent. For example, some signal source equipment (such as drones) may be used for illegal filming, disrupting aviation order, carrying dangerous goods for terrorist attacks, etc. These behaviors pose a serious threat to public safety, personal privacy and national security. The first and most important step is to find the signal source equipment in the environment.

[0003] However, the current methods for detecting signal source devices can be roughly divided into three categories: image processing-based methods, spectrum (time-frequency information)-based processing methods, and deep learning-based processing methods. Among them, the image-based and deep learning-based processing methods have obvious disadvantages: for low signal-to-noise ratio environments, image processing will be affected by large noise points or even fail; deep learning requires learning the signal library of the signal source device in advance, which faces the problem of large training data volume and the completeness of the library is difficult to guarantee. The spectrum-based processing method is relatively direct, but its spectrum detection steps are vague and does not describe the detection effect under low signal-to-noise ratio. Summary of the invention

[0004] In order to solve the existing technical problems, the present application provides a signal spectrum sensing method, a computer program product and a signal spectrum sensing device, which can accurately detect signal parameters in complex environments.

[0005] In a first aspect, a signal spectrum sensing method is provided, comprising: receiving a signal to be measured; obtaining a time-frequency signal and time-frequency signal spectrum estimation data based on the signal to be measured; performing time domain operations based on the time-frequency signal spectrum estimation data to obtain time domain processing data; determining target decision data based on the time domain processing data; and determining parameter data of the signal to be measured based on the target decision data.

[0006] In a second aspect, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the signal spectrum sensing method as described in any one of the first aspect of the present application is implemented.

[0007] In a third aspect, a signal spectrum sensing device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes a signal spectrum sensing method as described in any one of the first aspects of the present application.

[0008] The present application receives a signal to be tested, obtains a time-frequency signal and a spectrum estimation data of the time-frequency signal based on the signal to be tested, the spectrum estimation of the time-frequency signal represents the energy of the signal at a specific time and frequency, performs a time domain operation on the spectrum estimation data of the time-frequency signal, and obtains time domain processed data, which can be used to improve the signal-to-noise ratio of the signal or enhance the characteristics of the signal through the time domain operation, and then obtains target decision data based on the time-frequency processed data. Since the signal characteristics are more obvious after the time-frequency operation, the signal parameter data can be obtained through the target decision data, so that the obtained signal parameters are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 This is a diagram of an application environment of a signal spectrum sensing method in an embodiment;

[0010] Figure 2 is a flow chart of a signal spectrum sensing method in an embodiment;

[0011] Figure 3 is a schematic diagram of time domain operation in one embodiment;

[0012] Figure 4 is a schematic diagram of time domain operation in another embodiment;

[0013] Figure 5 A schematic diagram of comparison between the time-frequency signal spectrum estimation data and the amplitude-frequency diagram after time-domain accumulation in one embodiment;

[0014] Figure 6 A flowchart of determining target decision data based on time domain processing data in an embodiment;

[0015] Figure 7 is a schematic diagram of performing a frequency domain smoothing operation on a column in one embodiment;

[0016] Figure 8 A schematic diagram of performing a false alarm reduction decision operation on eight columns of data in one embodiment;

[0017] Fig. 9 A schematic diagram of a flow chart of determining target decision data based on time domain processing data in an embodiment;

[0018] Fig.10 A flowchart of determining target decision data based on time domain processing data in another embodiment;

[0019] Fig.11 A schematic diagram showing a comparison between an amplitude-frequency diagram obtained after Fourier transformation of a signal to be measured and an amplitude-frequency diagram after frequency domain smoothing in one embodiment;

[0020] Fig.12 A schematic diagram of a flow chart of determining target decision data based on time domain processing data in an embodiment;

[0021] Fig.13 A schematic diagram of an amplitude-frequency diagram determined by target decision data in one embodiment;

[0022] Fig.14 is a schematic diagram of a signal spectrum sensing device in an embodiment;

[0023] Fig.15 A schematic diagram of a signal spectrum sensing device in an embodiment. DETAILED DESCRIPTION

[0024] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the technical field of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the scope of protection of the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0026] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it should be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0027] See also Figure 1 , is an application environment diagram of a signal spectrum sensing method in an embodiment. The application environment diagram includes a signal spectrum sensing device 10 and a signal source device 20. The signal source device 20 can transmit a communication signal. For example, the signal source device 20 is a drone device. The signal spectrum sensing device 10 is used to receive the signal transmitted by the signal source device 20 when working, and process the received signal to be tested, and sense the signal parameters of the signal to be tested, wherein the signal parameters include but are not limited to signal strength, center frequency, bandwidth, signal power, etc.

[0028] With the continuous advancement and popularization of drone technology, drones are increasingly used in civil and commercial fields. However, the security threats brought by drones are also becoming increasingly prominent. For example, drones may be used for illegal filming, disrupting aviation order, carrying dangerous goods for terrorist attacks, etc. These behaviors pose a serious threat to public safety, personal privacy and national security. The advancement of anti-drone technology seems imminent, and the first and most important step is to detect drones in the environment.

[0029] However, the current methods for drone detection can be roughly divided into three categories: image processing-based methods, spectrum (time-frequency information)-based processing methods, and deep learning-based processing methods. Among them, the image-based and deep learning-based processing methods have obvious disadvantages: for low signal-to-noise ratio environments, image processing will be affected by large noise points or even fail; deep learning requires learning the drone signal library in advance, which faces the problem of large amounts of training data and the completeness of the library is difficult to guarantee. The spectrum-based processing method is relatively straightforward. In 2018, Huanuo Star Company applied for a patent for "A spectrum detection and signal suppression integrated device and control method", but its spectrum detection steps are vague and the detection effect under low signal-to-noise ratio is not described.

[0030] See also Figure 2 , is a flow chart of a signal spectrum sensing method provided by an embodiment of the present application. A signal spectrum sensing method is applied to a signal spectrum sensing device, and the signal spectrum sensing method includes the following steps:

[0031] S11. Receive a signal to be tested.

[0032] In this embodiment, the signal to be tested is a signal received by a signal spectrum sensing device 10. For example, the signal source device 20 is a drone. When the drone flies into the search range of a signal spectrum sensing device 10, the signal of the drone can be received by the signal spectrum sensing device 10.

[0033] S12. Based on the signal to be measured, obtain a time-frequency signal and time-frequency signal spectrum estimation data.

[0034] In this embodiment, a short-time Fourier transform (STFT) is performed on the signal to be measured to obtain a time-frequency signal, which refers to a representation method that describes the characteristics of a signal in two dimensions: time and frequency. In signal processing, the time-frequency signal provides an intuitive view of the frequency components of a signal that change over time, and the short-time Fourier transform is a common technique for generating time-frequency signals. It multiplies the signal with a short time window function and then performs Fourier transform on the product at different time points to obtain a time-frequency representation of the signal.

[0035] Time-frequency signal spectrum estimation Y(f,τ)=|X(f,τ)| 2 ; X(f,τ) represents the time-frequency signal, where τ represents time, f represents frequency, and the square of the amplitude is |X(f,τ)| 2 ∣,|X(f,τ)| 2 It is the square of the modulus of the complex number X(f,τ). The spectrum estimation of the time-frequency signal represents the energy of the signal at a specific time τ and frequency f. The energy here is a measure of the strength or power of the signal at that time-frequency point, providing the energy distribution of the signal at different times and frequencies.

[0036] S13. Based on the time-frequency signal spectrum estimation data, perform time domain operations to obtain time domain processed data.

[0037] In this embodiment, the time-frequency signal spectrum estimation data is represented by a matrix Y, for example, the dimension of the matrix Y is M

[0038] ×N, where M is the size of the frequency domain dimension, representing different frequency points; N is the size of the time domain dimension, representing different time points. In this way, the matrix Y can be regarded as the representation of the signal in the time-frequency domain. The time domain operation means processing the signal in the time domain to obtain the time domain processed data. The time domain operation can be used to improve the signal-to-noise ratio of the signal or enhance the characteristics of the signal.

[0039] S14. Determine target decision data based on the time domain processing data.

[0040] In this embodiment, the target decision data is used to determine the data of the signal parameters of the signal to be measured. The target decision data can be represented by a matrix, for example, by Dec(f,τ), where τ represents time and f represents frequency. Then, one or more amplitude-frequency diagrams can be obtained through the target decision data, for example, through each column of data in the target decision data, an amplitude-frequency diagram corresponding to each column is obtained. The amplitude-frequency diagram describes how the amplitude of the signal in the frequency domain changes with the frequency.

[0041] S15. Determine parameter data of the signal to be measured based on the target judgment data.

[0042] In this embodiment, since the target decision data can indicate multiple amplitude-frequency diagrams, parameter data of each signal can be obtained through each amplitude-frequency diagram, where the parameter data includes but is not limited to signal strength, center frequency, bandwidth, signal power, etc.

[0043] In the above embodiment, a signal to be tested is received, and a time-frequency signal and time-frequency signal spectrum estimation data are obtained based on the signal to be tested. The time-frequency signal spectrum estimation represents the energy of the signal at a specific time and frequency. Time domain operations are performed on the time-frequency signal spectrum estimation data to obtain time domain processed data. The time domain operations can be used to improve the signal-to-noise ratio of the signal or enhance the characteristics of the signal. Based on the time-frequency processed data, target decision data is obtained. Since the signal characteristics are more obvious after the time-frequency operations, signal parameter data can be obtained through the target decision data, so that the obtained signal parameters are more accurate.

[0044] In some embodiments, performing time domain operations based on the time-frequency signal spectrum estimation data to obtain time domain processed data includes at least one of the following:

[0045] Perform a time domain operation on each of the first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated column, select the maximum energy value corresponding to each frequency point from the first preset number of energy values ​​corresponding to each frequency point in each of the first preset number of time domain units, and use the maximum energy value corresponding to each frequency point as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data;

[0046] A non-coherent accumulation operation is performed on each first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated time domain unit, and the first preset number of energy values ​​corresponding to each frequency point in each first preset number of time domain units are averaged to obtain an average energy value corresponding to each frequency point, and the average energy value corresponding to each frequency point is used as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data.

[0047] In this embodiment, the updating time domain unit is the time domain unit in the time domain processing data, such as Figure 3 As shown, Figure 3 is a schematic diagram of time domain operation in an embodiment, the horizontal axis represents the time domain, the vertical axis represents the frequency domain, for each row in the first to fourth columns, that is, for the first to fourth columns, each frequency point corresponds to a row, and a row has four data, such as Figure 3 The first row in the data has a frequency value of 1.1, 1.3, 1.6, and 1.2. The maximum value is taken from these four values, and the data value of the first row in the first column of the time-domain processed data is 1.6. The other rows in the first to fourth columns are processed similarly, and each row in the fifth to eighth columns is processed similarly to obtain the time-domain processed data.

[0048] The time-frequency signal spectrum estimation data is a Y matrix with a dimension of M×N, where M is the frequency domain dimension and N is the time domain dimension. The incoherent accumulation method is adopted, and every k columns are superimposed as one column. The accumulated signal, that is, the time domain processing data, is a matrix in is the dimension. That is, the number of columns is reduced by k times, such as Figure 4 As shown, Figure 4 FIG. 1 is a schematic diagram of time domain operation in another embodiment, that is, for the first to fourth columns, each frequency point corresponds to a row, and a row has four data, such as Figure 4 The first row in the data has energy values ​​of 1.1, 1.3, 1.6, and 1.2. The four values ​​are added and averaged to get 1.3, and the data value of the first row in the first column of the time-domain processed data is 1.3. The other rows in the first to fourth columns are processed similarly, and each row in the fifth to eighth columns is processed similarly to get the time-domain processed data. Figure 4 The operation in is a non-coherent accumulation operation. The idea of ​​non-coherent accumulation is introduced into passive spectrum detection to solve the problem of spectrum estimation difficulty under low signal-to-noise ratio conditions.

[0049] like Figure 5 As shown, Figure 5 It is a schematic diagram comparing the time-frequency signal spectrum estimation data and the amplitude-frequency diagram after time domain accumulation in one embodiment; by comparing the time-frequency signal spectrum estimation data, after performing incoherent accumulation, the signal bandwidth can be clearly distinguished, which means that the incoherent accumulation technology effectively enhances the energy of the signal, making the characteristics of the signal more obvious, especially when the target signal is weak or the signal-to-noise ratio is low, the detection probability of the signal can be improved through incoherent accumulation.

[0050] In the above embodiment, by performing time domain operations on the time-frequency signal spectrum estimation data, such as incoherent accumulation technology, to obtain time domain processed data, the energy of the signal can be effectively enhanced, making the characteristics of the signal more obvious, especially when the target signal is weak or the signal-to-noise ratio is low, the detection probability of the signal can be improved through incoherent accumulation.

[0051] In some embodiments, Figure 6 As shown, Figure 6 Flow chart of determining target decision data based on time domain processing data in one embodiment, wherein determining target decision data based on the time domain processing data includes:

[0052] S61. Based on the time-domain processed data, perform a frequency-domain smoothing operation to obtain smoothed data; perform a frequency-domain binarization operation on the smoothed data to obtain first processed data; based on the first processed data, perform a false alarm reduction decision operation to obtain first decision data.

[0053] Optionally, the frequency domain smoothing operation is performed based on the time domain processed data to obtain the smoothed data, including: for each data value in each column of the time domain processed data, obtaining a second preset number of adjacent data values ​​adjacent to the data value, calculating a data mean based on the adjacent data values, updating the data value to the data mean, and obtaining the smoothed data.

[0054] Time domain processing data is a matrix Get Y ′ After the data is collected, for each data in each column, the value is replaced by the mean of the next n (n>1) adjacent data to obtain smoothed data, using the matrix It indicates that frequency domain smoothing can effectively reduce the noise in the signal. Through smoothing, the noise component can be reduced and the distinction between signal and noise can be improved. Smoothing operation can help enhance the important frequency components in the signal, making the characteristics of the signal more obvious, which is convenient for subsequent signal analysis and processing.

[0055] like Figure 7 As shown, Figure 7This is a schematic diagram of performing a frequency domain smoothing operation on a column in an embodiment. A column has ten numbered data, and the second preset number is 4. Then for this column, after performing the frequency domain smoothing operation, the first data value numbered "1" corresponding to this column in the smoothed data is the average value of the data values ​​numbered "1" to "4" corresponding to this column in the time domain processed data, and the first data value numbered "2" corresponding to this column in the smoothed data is the average value of the data values ​​numbered "2" to "5" corresponding to this column in the time domain processed data. When there are less than four data after this column, they are padded with 0.

[0056] Optionally, performing a frequency domain binarization operation on the smoothed data to obtain first processed data includes:

[0057] For each time domain unit in the smoothed data, the threshold value corresponding to each time domain unit is calculated, the data value in each time domain unit is compared with the threshold value corresponding to each time domain unit, the data value in each time domain unit that is greater than or equal to the threshold value corresponding to each time domain unit is updated to 1, and the data value in each time domain unit that is less than the threshold value corresponding to each time domain unit is updated to 0, so as to obtain the first processed data.

[0058] Each time domain unit corresponds to a column of data. For each column of data in the smoothed data Z(f,τ), the threshold value of each column is calculated, and the data value of each column is compared with the threshold value of each column. The binarization operation is performed on Z(f,τ). The data value greater than or equal to the threshold in each column is judged as 1, and the data value less than the threshold is judged as 0, and the binarized data is obtained, that is, the first processed data For example, OS-CFAR is used to calculate the threshold value. The signal can be separated from the background or noise through the binarization operation to improve the signal discernibility.

[0059] Optionally, performing a false alarm reduction decision operation based on the first processed data to obtain first decision data includes:

[0060] Add every third preset number of time domain units in the first processed data to obtain first superimposed data; update the data values ​​in each time domain unit of the first superimposed data that are greater than or equal to the preset decision threshold to 1, and update the data values ​​in each time domain unit of the superimposed data that are less than the preset decision threshold to 0 to obtain first decision data.

[0061] In this embodiment, in the matrix formed by the first processed data, every h (h>3) columns are added together to form a column to obtain a data matrix Each column of Tar(f,τ) is judged again, and the data value in each column that is greater than or equal to the preset judgment threshold is judged as 1, and the frequency point position where 1 is located is the signal position, and the data value less than the preset judgment threshold is judged as 0, and the first judgment data is obtained. like Figure 8 As shown, Figure 8 Schematic diagram of performing a false alarm reduction decision operation on eight columns of data in one embodiment, where the preset decision threshold is 3. Figure 8 The matrix formed by the first processed data is eight columns, and the value of h is 8. After accumulating these eight columns of data for each row, Tar(f,τ) is a column of data, that is, (3,2,1,4). Therefore, the first decision data Dec(f,τ) is a column of data (1,0,0,1). If the first processed data is 16 columns and the value of h is 8, then similarly Figure 8 The operations in each of the eight columns are similar Figure 8 By performing the operation once, two columns of data can be obtained.

[0062] S62. Perform a frequency domain binarization operation on the time domain processed data to obtain second processed data, and perform a false alarm reduction decision operation based on the second processed data to obtain second decision data.

[0063] Optionally, performing a frequency domain binarization operation on the time domain processed data to obtain the second processed data includes:

[0064] For each time domain unit in the time domain processed data, the threshold value corresponding to each time domain unit is calculated, the data value in each time domain unit is compared with the threshold value corresponding to each time domain unit, the data value in each time domain unit that is greater than or equal to the threshold value corresponding to each time domain unit is updated to 1, and the data value in each time domain unit that is less than the threshold value corresponding to each time domain unit is updated to 0, so as to obtain the first processed data.

[0065] Optionally, performing a false alarm reduction decision operation based on the second processed data to obtain second decision data includes:

[0066] Add every third preset number of time domain units in the second processed data to obtain second superimposed data; update the data values ​​in each time domain unit of the second superimposed data that are greater than or equal to the preset decision threshold to 1, and update the data values ​​in each time domain unit of the second superimposed data that are less than the preset decision threshold to 0 to obtain the second decision data.

[0067] In this embodiment, for For each column of data, calculate the threshold value of each column, compare the data value of each column with the threshold value of each column, and Perform a binarization operation, in each column, the data values ​​greater than or equal to the threshold are judged as 1, and the data values ​​less than the threshold are judged as 0, to obtain the binary data, i.e. the second processed data For example, OS-CFAR is used to calculate the threshold value. The signal can be separated from the background or noise through the binarization operation to improve the signal discernibility. In the data matrix, every h (h>3) columns are added together to form a column To Tar ′ Each column of (f, τ) is judged again, and the data value greater than or equal to the preset judgment threshold in each column is judged as 1, and the frequency point position where 1 is located is the signal position, and the data value less than the preset judgment threshold is judged as 0, and the second judgment data is obtained.

[0068] S63: Merge the first judgment data and the second judgment data to obtain the target judgment data.

[0069] In this embodiment, the matrix formed by the first decision data and the matrix formed by the second decision data are added together to obtain the matrix corresponding to the target decision data. ′ The two matrices (f,τ) are added together.

[0070] like Fig. 9 As shown, as shown in 91, it is a flowchart diagram of determining target decision data based on time domain processing data in one embodiment, and the flowchart includes: receiving a signal to be measured; obtaining a time-frequency signal and a time-frequency signal spectrum estimation data based on the signal to be measured; performing a time domain operation based on the time-frequency signal spectrum estimation data to obtain time domain processing data, and then processing the time domain processing data in two ways; one way is: performing a frequency domain smoothing operation based on the time domain processing data to obtain smoothed data, performing a frequency domain binarization operation on the smoothed data to obtain first processing data, performing a false alarm reduction decision operation based on the first processing data to obtain first decision data; the other way is: performing a frequency domain binarization operation based on the time domain processing data to obtain second processing data, performing a false alarm reduction decision operation based on the second processing data to obtain second decision data. Then, the first decision data and the second decision data are combined to obtain the target decision data.

[0071] In the above embodiment, the process is performed in two ways, one of which is: performing a frequency domain smoothing operation on the time domain processed data to obtain smoothed data, which can make the signal and noise in the smoothed data more distinguishable, and performing a frequency domain binarization operation on the smoothed data. Through the binarization operation, the signal can be separated from the background or noise, and the recognizability of the signal can be improved to obtain the second processed data, and then obtain the second decision data based on the second processed data; the other way is: performing a frequency domain binarization operation on the time domain processed data, and obtaining the first processed data through the binarization operation, and then obtaining the first decision data based on the first processed data; by merging the first decision data and the second decision data, the filtering of the signal caused by excessive smoothing can be reduced, and the method of introducing multiple joint decisions can reduce the false alarm rate, which can improve the recognition accuracy of the signal.

[0072] like Fig.10 As shown, Fig.10 This is a flow chart of determining target decision data based on time domain processing data in another embodiment, wherein determining target decision data based on the time domain processing data includes the following steps:

[0073] S91. Calculate the signal-to-noise ratio of the signal to be measured based on the time-frequency signal spectrum estimation data.

[0074] In this embodiment, based on the Y(f,τ) matrix, the total average value of the accumulated data values ​​in the matrix can be calculated, and then the sum of the data values ​​higher than or equal to the total average value can be calculated in the matrix to obtain a first cumulative sum, which represents the value corresponding to the useful signal part. The sum of the data values ​​lower than the total average value can be calculated in the matrix to obtain a second cumulative sum, which represents the value corresponding to the noise part. Based on the first cumulative sum and the second cumulative sum, the signal-to-noise ratio can be calculated, for example, the ratio of the first cumulative sum to the second cumulative sum can be used as the signal-to-noise ratio.

[0075] S92. Determine target decision data based on the time domain processing data and the signal-to-noise ratio.

[0076] Optionally, the determining the target decision data based on the time-domain processed data and the signal-to-noise ratio includes at least one of the following:

[0077] When the signal-to-noise ratio is higher than a preset signal-to-noise ratio threshold, performing a frequency domain binarization operation on the time domain processed data to obtain second processed data, and performing a false alarm reduction decision operation based on the second processed data to obtain the target decision data;

[0078] When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, based on the time-domain processed data, a frequency-domain smoothing operation is performed to obtain smoothed data, a frequency-domain binarization operation is performed on the smoothed data to obtain first processed data, and based on the first processed data, a false alarm reduction decision operation is performed to obtain the target decision data;

[0079] When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, a frequency domain smoothing operation is performed based on the time domain processed data to obtain smoothed data; a frequency domain binarization operation is performed on the smoothed data to obtain first processed data, and a false alarm reduction decision operation is performed based on the first processed data to obtain first decision data; a frequency domain binarization operation is performed on the time domain processed data to obtain second processed data, and a false alarm reduction decision operation is performed based on the second processed data to obtain second decision data; and the first decision data and the second decision data are merged.

[0080] In this embodiment, when the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold, it indicates that the signal feature is relatively strong, and the frequency domain smoothing operation does not need to be performed any more. For each column of data, calculate the threshold value of each column, compare the data value of each column with the threshold value of each column, and Perform a binarization operation, in each column, the data values ​​greater than or equal to the threshold are judged as 1, and the data values ​​less than the threshold are judged as 0, to obtain the binary data, i.e. the second processed data For example, OS-CFAR is used to calculate the threshold value. The signal can be separated from the background or noise through the binarization operation to improve the signal discernibility. In the data matrix, every h (h>3) columns are added together to form a column Each column of Tar′(f,τ) is judged again, and the data value in each column that is greater than or equal to the preset judgment threshold is judged as 1, and the frequency point position where 1 is located is the signal position, and the data value less than the preset judgment threshold is judged as 0, and the second judgment data is obtained. That is, the target judgment data.

[0081] When the signal-to-noise ratio is lower than or equal to the preset signal-to-noise ratio threshold, it indicates that the signal feature part is weak and it is necessary to further perform the frequency domain smoothing operation to obtain the first processed data. For example, OS-CFAR is used to calculate the threshold value. The signal can be separated from the background or noise through the binarization operation to improve the signal discernibility. In the above equation, every h (h>3) columns are added together to form a data matrix Tar(f,τ) Each column of Tar(f,τ) is judged again. The data value in each column that is greater than or equal to the preset decision threshold is judged as 1. The frequency point where 1 is located is the signal position. The data value that is less than the preset decision threshold is judged as 0. The final decision matrix, i.e., the first decision data That is, the target judgment data.

[0082] When the signal-to-noise ratio is lower than or equal to the preset signal-to-noise ratio threshold, it indicates that the characteristic part of the signal is relatively weak and can be processed in two ways. Specifically, as shown in the flow chart 91, the first judgment data and the second judgment data are respectively obtained by processing in two ways, and the first judgment data and the second judgment data are merged to obtain the target judgment data.

[0083] like Fig.11 As shown, Fig.11 The figure is a schematic diagram comparing the amplitude-frequency diagram obtained after Fourier transform is performed on the signal to be tested and the amplitude-frequency diagram after frequency domain smoothing in one embodiment. After the frequency domain smoothing operation is performed, the distinction of the signal part is stronger, which is conducive to improving the accuracy of subsequent determination of signal parameters.

[0084] like Fig.12 The flowchart of determining the target decision data based on the time domain processing data in one embodiment is as follows: in the case of high signal-to-noise ratio, that is, the signal-to-noise ratio is higher than the preset signal-to-noise ratio threshold, the time domain processing data is subjected to a frequency domain binarization operation to obtain second processing data, and a false alarm reduction decision operation is performed based on the second processing data to obtain the target decision data. In this case, the target decision data is the second decision data. In the case of low signal-to-noise ratio, that is, the signal-to-noise ratio is lower than or equal to the preset signal-to-noise ratio threshold, a frequency domain smoothing operation is performed based on the time domain processing data to obtain smoothed data, a frequency domain binarization operation is performed on the smoothed data to obtain first processing data, and a false alarm reduction decision operation is performed based on the first processing data to obtain the target decision data, that is, the first decision data; or in the case of low signal-to-noise ratio, the operation is performed in two ways, such as Fig. 9 The two-way execution steps shown are used to obtain the target judgment data.

[0085] In the above embodiment, different processing flows are adopted according to different environmental signal-to-noise ratios. When the signal-to-noise ratio is high, the frequency domain smoothing operation is not performed. When the signal-to-noise ratio is low, the spectrum smoothing operation is added. By measuring the signal-to-noise ratio, different processing flows are adopted, especially joint processing is adopted in complex environments. This can save computing resources while maintaining good detection performance.

[0086] In some embodiments, determining the parameter data of the signal to be tested based on the target decision data includes at least one of the following:

[0087] According to the target decision data, the amplitude-frequency data of each target signal is obtained, and according to the amplitude-frequency data of each target signal, the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal are obtained, and based on the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal, the bandwidth corresponding to each target signal and / or the center frequency point corresponding to each target signal are obtained;

[0088] For each target signal within a bandwidth range, based on the time-frequency signal spectrum estimation data, the energy value corresponding to each frequency point is obtained, the energy value corresponding to each frequency point is accumulated to obtain an energy accumulation value, and the energy accumulation value is used as the signal power corresponding to each target signal.

[0089] In this embodiment, one or more target signals can be obtained according to the target decision data. The target decision data is represented by a target decision matrix. The rows of the matrix are time domain units, the columns are frequency units, and each column of the target decision matrix is ​​a target signal. Figure 8 As shown in , the target decision matrix is ​​a column of data, that is, a target signal. Fig.13 As shown in the figure, the amplitude-frequency diagram is drawn based on a target signal. Fig.13 Schematic diagram of the amplitude-frequency diagram determined by the target decision data in one embodiment, the bandwidth is (f2-f1), and the center frequency is (f2+f1) / 2. For a target signal, the energy value corresponding to each frequency point within the bandwidth range corresponding to the target signal is obtained from the time-frequency signal spectrum estimation data, and the energy value corresponding to each frequency point is accumulated to obtain an energy accumulation value, and the energy accumulation value is used as the signal power corresponding to each target signal.

[0090] In the above embodiment, the target decision data is data after multiple noise processing is performed on the signal to be tested. The amplitude-frequency diagram obtained based on the target decision data can better highlight the signal characteristics, so that the obtained signal parameters are more accurate, thereby improving the accuracy of identifying signal parameters.

[0091] In some embodiments, the method further comprises:

[0092] Based on the time domain processed data, obtaining a target energy accumulation value of a target time domain unit;

[0093] The distance range of the signal source device is determined according to the target energy accumulated value.

[0094] In this embodiment, the target time domain unit can timely process any column in the time domain data, accumulate the energy values ​​corresponding to each frequency point in the target time domain unit, and obtain a target energy accumulation value, which can indicate the distance range between the signal source device 20 and the signal spectrum perception 10, wherein the larger the target energy accumulation value, the closer the distance. Different warning methods can be divided according to different distance ranges. The closer the distance represented by the distance range, the stronger the warning method. For example, the closer the distance, the louder the warning sound.

[0095] Optionally, the method further includes:

[0096] When the distance range of the signal source device is within a preset dangerous distance range, a warning operation is performed.

[0097] In this embodiment, a danger distance range may also be set, and when the determined distance range is within the preset danger distance range, a warning operation is performed.

[0098] In the above embodiment, the strength of the signal source can be estimated based on the time domain processed data after the time domain operation, thereby estimating the distance range of the signal source device. When the distance range of the signal source device is within the preset dangerous distance range, an early warning operation is performed to reduce the interference caused by the signal source device.

[0099] The above one or more embodiments of the present application can solve the problem of spectrum detection of drone signals in complex environments. Different detection paths are adopted according to the differences in environmental signal-to-noise ratios, which effectively saves computing resources, while quickly sensing the spectrum and maintaining a low false alarm rate.

[0100] In combination with one or more embodiments of the present application described above, the present application has at least the following features: Innovation:

[0101] (1) Introducing the idea of ​​incoherent accumulation into passive spectrum detection to solve the problem of spectrum estimation difficulty under low signal-to-noise ratio conditions.

[0102] (2): Different processing flows are used for different environmental signal-to-noise ratios. Spectral smoothing is increased for low signal-to-noise ratio processes compared to high signal-to-noise ratio processes. Joint processing is used for particularly complex environments, which can save computing resources while maintaining good detection performance.

[0103] (3): Introduce multiple joint judgment methods to reduce the false alarm rate.

[0104] On the other hand, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements a signal spectrum sensing method described in any embodiment of the present application.

[0105] Among them, in the computer program product, an optional implementation form of the program module architecture of the computer program that implements each step of the signal spectrum sensing method can be a signal spectrum sensing device.

[0106] See also Fig.14 An embodiment of the present application provides a signal spectrum sensing device, including: a receiving module 1201, used to receive a signal to be tested; a computing module 1202, used to obtain a time-frequency signal and a time-frequency signal spectrum estimation data based on the signal to be tested; the computing module 1202 is also used to perform time domain operations based on the time-frequency signal spectrum estimation data to obtain time domain processing data; the computing module 1202 is also used to determine target decision data based on the time domain processing data; the computing module 1202 is also used to determine parameter data of the signal to be tested based on the target decision data.

[0107] Optionally, the calculation module 1202 is further configured to:

[0108] Perform a time domain operation on each of the first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated time domain unit, select the maximum energy value corresponding to each frequency point from the first preset number of energy values ​​corresponding to each frequency point in each of the first preset number of time domain units, and use the maximum energy value corresponding to each frequency point as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data;

[0109] A non-coherent accumulation operation is performed on each first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated time domain unit, and the first preset number of energy values ​​corresponding to each frequency point in each first preset number of time domain units are averaged to obtain an average energy value corresponding to each frequency point, and the average energy value corresponding to each frequency point is used as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data.

[0110] Optionally, the calculation module 1202 is further configured to:

[0111] Based on the time-domain processed data, a frequency-domain smoothing operation is performed to obtain smoothed data; on the smoothed data, a frequency-domain binarization operation is performed to obtain first processed data; based on the first processed data, a false alarm reduction decision operation is performed to obtain first decision data;

[0112] Based on the time-domain processed data, a frequency-domain binarization operation is performed to obtain second processed data, and based on the second processed data, a false alarm reduction decision operation is performed to obtain second decision data;

[0113] The first decision data and the second decision data are combined to obtain the target decision data.

[0114] Optionally, the calculation module 1202 is further configured to:

[0115] Calculating the signal-to-noise ratio of the signal to be measured based on the time-frequency signal spectrum estimation data;

[0116] The target decision data is determined based on the time-domain processed data and the signal-to-noise ratio.

[0117] Optionally, the calculation module 1202 is further configured to:

[0118] When the signal-to-noise ratio is higher than a preset signal-to-noise ratio threshold, performing a frequency domain binarization operation on the time domain processed data to obtain second processed data, and performing a false alarm reduction decision operation based on the second processed data to obtain the target decision data;

[0119] When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, based on the time-domain processed data, a frequency-domain smoothing operation is performed to obtain smoothed data, a frequency-domain binarization operation is performed on the smoothed data to obtain first processed data, and based on the first processed data, a false alarm reduction decision operation is performed to obtain the target decision data;

[0120] When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, a frequency domain smoothing operation is performed based on the time domain processed data to obtain smoothed data; a frequency domain binarization operation is performed on the smoothed data to obtain first processed data, and a false alarm reduction decision operation is performed based on the first processed data to obtain first decision data; a frequency domain binarization operation is performed on the time domain processed data to obtain second processed data, and a false alarm reduction decision operation is performed based on the second processed data to obtain second decision data; the first decision data and the second decision data are combined to obtain the target decision data.

[0121] Optionally, the calculation module 1202 is further configured to:

[0122] For each data value of each time domain unit in the time domain processed data, a second preset number of adjacent data values ​​adjacent to the data value are obtained, a data mean is calculated based on the adjacent data values, and the data value is updated to the data mean to obtain the smoothed data.

[0123] Optionally, the calculation module 1202 is further configured to:

[0124] For each time domain unit in the smoothed data, the threshold value corresponding to each time domain unit is calculated, the data value in each time domain unit is compared with the threshold value corresponding to each time domain unit, the data value in each time domain unit that is greater than or equal to the threshold value corresponding to each time domain unit is updated to 1, and the data value in each time domain unit that is less than the threshold value corresponding to each time domain unit is updated to 0.

[0125] Optionally, the calculation module 1202 is further configured to:

[0126] adding every third preset number of time domain units in the first processed data to obtain first superimposed data; updating data values ​​greater than or equal to a preset decision threshold in data values ​​in each time domain unit of the first superimposed data to 1, and updating data values ​​less than the preset decision threshold in data values ​​in each time domain unit of the first superimposed data to 0, to obtain first decision data;

[0127] The performing a false alarm reduction decision operation based on the second processed data to obtain second decision data comprises:

[0128] Add every third preset number of time domain units in the second processed data to obtain second superimposed data; update the data values ​​in each time domain unit of the second superimposed data that are greater than or equal to the preset decision threshold to 1, and update the data values ​​in each time domain unit of the second superimposed data that are less than the preset decision threshold to 0 to obtain the second decision data.

[0129] Optionally, the calculation module 1202 is further configured to:

[0130] According to the target decision data, the amplitude-frequency data of each target signal is obtained, and according to the amplitude-frequency data of each target signal, the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal are obtained, and based on the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal, the bandwidth corresponding to each target signal and / or the center frequency point corresponding to each target signal are obtained;

[0131] For each target signal within a bandwidth range, based on the time-frequency signal spectrum estimation data, the energy value corresponding to each frequency point is obtained, the energy value corresponding to each frequency point is accumulated to obtain an energy accumulation value, and the energy accumulation value is used as the signal power corresponding to each target signal.

[0132] Optionally, the signal spectrum sensing device further includes an early warning module 1203, which is used to:

[0133] Based on the time domain processed data, obtaining a target energy accumulation value of a target time domain unit;

[0134] The distance range of the signal source device is determined according to the target energy accumulated value.

[0135] Optionally, the early warning module 1203 is further used for:

[0136] When the distance range of the signal source device is within a preset dangerous distance range, a warning operation is performed.

[0137] See also Fig.15 On the other hand, an embodiment of the present application further provides a signal spectrum sensing device 10, including a processor 13 and a memory 14, wherein the memory 14 stores a computer program. When the computer program is executed by the processor, the processor 13 executes the steps of a signal spectrum sensing method provided in any of the above embodiments of the present application.

[0138] The processor 13 is a control center, which uses various interfaces and lines to connect various parts of the entire signal spectrum sensing device, and executes various functions and processes data of a signal spectrum sensing device by running or executing software programs and / or modules stored in the memory 14, and calling data stored in the memory 14. Optionally, the processor 13 may include one or more processing cores; preferably, the processor 13 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user pages and application programs, etc., and the modem processor mainly processes wireless communications. It can be understood that the above-mentioned modem processor may not be integrated into the processor 13.

[0139] The memory 14 can be used to store software programs and modules. The processor 13 executes various functional applications and data processing by running the software programs and modules stored in the memory 14. The memory 14 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of a signal spectrum sensing device, etc. In addition, the memory 14 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 14 may also include a memory processor to provide the processor 13 with access to the memory 14.

[0140] On the other hand, an embodiment of the present application further provides a storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of a signal spectrum sensing method provided by any of the above embodiments of the present application.

[0141] In another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, a signal spectrum sensing method as described in any embodiment of the present application is implemented.

[0142] Those of ordinary skill in the art can understand that all or part of the processes in the methods provided in the above embodiments can be completed by instructing related hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0143] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A signal spectrum sensing method, characterized in that: include: receiving a signal to be tested; Based on the signal to be measured, obtain a time-frequency signal and time-frequency signal spectrum estimation data; Based on the time-frequency signal spectrum estimation data, performing time domain operations to obtain time domain processed data; Determining target decision data based on the time-domain processed data; Based on the target decision data, parameter data of the signal to be measured is determined.

2. A signal spectrum sensing method as claimed in claim 1, characterized in that: The performing time domain operation based on the time-frequency signal spectrum estimation data to obtain time domain processed data includes at least one of the following: Perform a time domain operation on each of the first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated time domain unit, select the maximum energy value corresponding to each frequency point from the first preset number of energy values ​​corresponding to each frequency point in each of the first preset number of time domain units, and use the maximum energy value corresponding to each frequency point as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data; A non-coherent accumulation operation is performed on each first preset number of time domain units in the time-frequency signal spectrum estimation data to obtain an updated time domain unit, and the first preset number of energy values ​​corresponding to each frequency point in each first preset number of time domain units are averaged to obtain an average energy value corresponding to each frequency point, and the average energy value corresponding to each frequency point is used as the energy value corresponding to each frequency point in the updated time domain unit to obtain the time domain processed data.

3. A signal spectrum sensing method as claimed in claim 1, characterized in that: The determining target decision data based on the time domain processing data comprises: Based on the time-domain processed data, a frequency-domain smoothing operation is performed to obtain smoothed data; on the smoothed data, a frequency-domain binarization operation is performed to obtain first processed data; based on the first processed data, a false alarm reduction decision operation is performed to obtain first decision data; Based on the time-domain processed data, a frequency-domain binarization operation is performed to obtain second processed data, and based on the second processed data, a false alarm reduction decision operation is performed to obtain second decision data; The first decision data and the second decision data are combined to obtain the target decision data.

4. A signal spectrum sensing method as claimed in claim 1, characterized in that: The determining target decision data based on the time domain processing data comprises: Calculating the signal-to-noise ratio of the signal to be measured based on the time-frequency signal spectrum estimation data; The target decision data is determined based on the time-domain processed data and the signal-to-noise ratio.

5. A signal spectrum sensing method as claimed in claim 4, characterized in that: The determining the target decision data based on the time-domain processed data and the signal-to-noise ratio includes at least one of the following: When the signal-to-noise ratio is higher than a preset signal-to-noise ratio threshold, performing a frequency domain binarization operation on the time domain processed data to obtain second processed data, and performing a false alarm reduction decision operation based on the second processed data to obtain the target decision data; When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, based on the time-domain processed data, a frequency-domain smoothing operation is performed to obtain smoothed data, a frequency-domain binarization operation is performed on the smoothed data to obtain first processed data, and based on the first processed data, a false alarm reduction decision operation is performed to obtain the target decision data; When the signal-to-noise ratio is lower than or equal to a preset signal-to-noise ratio threshold, a frequency domain smoothing operation is performed based on the time domain processed data to obtain smoothed data; a frequency domain binarization operation is performed on the smoothed data to obtain first processed data, and a false alarm reduction decision operation is performed based on the first processed data to obtain first decision data; a frequency domain binarization operation is performed on the time domain processed data to obtain second processed data, and a false alarm reduction decision operation is performed based on the second processed data to obtain second decision data; The first decision data and the second decision data are combined to obtain the target decision data.

6. A signal spectrum sensing method as claimed in claim 3, characterized in that: The performing a frequency domain smoothing operation based on the time domain processed data to obtain smoothed data comprises: For each data value of each time domain unit in the time domain processed data, a second preset number of adjacent data values ​​adjacent to the data value are obtained, a data mean is calculated based on the adjacent data values, and the data value is updated to the data mean to obtain the smoothed data.

7. A signal spectrum sensing method as claimed in claim 3, characterized in that: The performing a frequency domain binarization operation on the smoothed data to obtain first processed data comprises: For each time domain unit in the smoothed data, the threshold value corresponding to each time domain unit is calculated, the data value in each time domain unit is compared with the threshold value corresponding to each time domain unit, the data value in each time domain unit that is greater than or equal to the threshold value corresponding to each time domain unit is updated to 1, and the data value in each time domain unit that is less than the threshold value corresponding to each time domain unit is updated to 0.

8. A signal spectrum sensing method as claimed in claim 3, characterized in that: The performing a false alarm reduction decision operation based on the first processed data to obtain first decision data comprises: adding every third preset number of time domain units in the first processed data to obtain first superimposed data; updating data values ​​greater than or equal to a preset decision threshold in data values ​​in each time domain unit of the first superimposed data to 1, and updating data values ​​less than the preset decision threshold in data values ​​in each time domain unit of the first superimposed data to 0, to obtain first decision data; The performing a false alarm reduction decision operation based on the second processed data to obtain second decision data comprises: Add every third preset number of time domain units in the second processed data to obtain second superimposed data; update the data values ​​in each time domain unit of the second superimposed data that are greater than or equal to the preset decision threshold to 1, and update the data values ​​in each time domain unit of the second superimposed data that are less than the preset decision threshold to 0 to obtain the second decision data.

9. A signal spectrum sensing method as claimed in claim 1, characterized in that: The determining of parameter data of the signal to be tested based on the target decision data includes at least one of the following: According to the target decision data, the amplitude-frequency data of each target signal is obtained, and according to the amplitude-frequency data of each target signal, the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal are obtained, and based on the minimum frequency value and the corresponding maximum frequency value corresponding to each target signal, the bandwidth corresponding to each target signal and / or the center frequency point corresponding to each target signal are obtained; For each target signal within a bandwidth range, based on the time-frequency signal spectrum estimation data, the energy value corresponding to each frequency point is obtained, the energy value corresponding to each frequency point is accumulated to obtain an energy accumulation value, and the energy accumulation value is used as the signal power corresponding to each target signal.

10. A signal spectrum sensing method according to claim 1, characterized in that: The method further comprises: Based on the time domain processed data, obtaining a target energy accumulation value of a target time domain unit; The distance range of the signal source device is determined according to the target energy accumulated value.

11. A signal spectrum sensing method according to claim 10, characterized in that: The method further comprises: When the distance range of the signal source device is within a preset dangerous distance range, a warning operation is performed.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, a signal spectrum sensing method as claimed in any one of claims 1 to 11 is implemented.

13. A signal spectrum sensing device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes a signal spectrum sensing method as claimed in any one of claims 1 to 11.

Images