Broadband radar single-bit data return and sparse imaging method

Abstract

The invention discloses a broadband radar single-bit data return and sparse imaging method, which belongs to the technical field of radar imaging, and comprises the following steps: S1, converting broadband radar echoes and thresholds into binary data with the same digit through data conversion; s2, processing is carried out according to the positive and negative characteristics of the threshold value; s3, arranging the extracted digits to form an extraction mode according to the sequence of the sampling points; s4, extracting one-bit data from the broadband radar echo according to the sequence of the sampling points according to the extraction mode in the step S3, and arranging the extracted one-bit data according to the sampling point data to form single-bit echo data, namely realizing variable-threshold single-bit quantization; and S5, performing imaging by using a single-bit radar two-dimensional sparse imaging algorithm. By adopting the broadband radar single-bit data return and sparse imaging method, the hardware overhead can be reduced, the transmission and storage pressure of a broadband radar system can be relieved, and the imaging quality can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of radar imaging, and particularly to a method for broadband radar single-bit data transmission and sparse imaging. Background Art

[0002] At present, synthetic aperture radar (SAR) has become an indispensable electronic equipment due to its all-weather and all-time imaging capabilities. When applied, SAR is usually required to have high resolution in order to obtain detailed information of the target scene. High-precision imaging depends on high-bit sampling quantization, high sampling rate and large-scale data processing. However, a large amount of echo data will bring heavy burdens on data acquisition, storage, transmission and processing, greatly increasing the system complexity and cost. Single-bit radar imaging technology can effectively alleviate this problem, not only relieve the pressure of data storage and transmission, but also reduce the system cost, so it has important research value in both military and civilian fields.

[0003] Currently, the single-bit radar hardware system has not yet developed to a mature stage. In order to achieve the single-bit quantization effect, usually the sign bit of the broadband radar echo is directly extracted to generate single-bit echo data. However, this method will cause the amplitude information of the scatterer to be completely lost, thus affecting the effective recognition of the target and the imaging quality. Research shows that by applying a variable threshold to the broadband radar echo and then extracting the sign bit, the amplitude information of the scatterer can be restored to a certain extent. However, implementing a variable threshold requires additional hardware resources to generate a dynamic threshold signal.

[0004] If a Gaussian threshold is used, its random characteristics not only significantly increase the hardware burden, but also make it difficult to reproduce the threshold information, affecting the operability and stability. In contrast, a single-frequency threshold only requires three key parameters: amplitude, frequency and phase, to generate a determinable single-frequency signal, thus avoiding the reproduction problem caused by randomness. However, in the dechirp receiving acquisition system, the broadband radar echo is affected by the single-frequency threshold, and a significant spectral line will be introduced in the range domain, thus reducing the imaging quality. In addition, the single-frequency threshold still relies on additional hardware resources to generate and maintain, thus increasing the hardware burden of the system and being unfavorable for the efficient deployment of small platforms.

[0005] In recent years, the research on small-platform broadband radar systems has been continuously heating up and has become one of the hot spots in the field of broadband radar technology. With the wide application of unmanned aerial vehicles, portable broadband radar devices and embedded systems, the demand for low-power, lightweight and high-performance broadband radar solutions is increasing day by day. However, limited by the computing power, storage resources and energy consumption budget of small platforms, how to achieve high-quality radar imaging under limited hardware conditions has become the key challenge in current research. Summary of the Invention

[0006] The object of the present invention is to provide a wideband radar single-bit data transmission back and sparse imaging method, which can not only reduce the hardware cost, relieve the transmission and storage pressure of the wideband radar system, but also achieve single-frequency threshold single-bit quantization of the wideband radar echo data, effectively restore the target scattering intensity information, and ensure the imaging quality.

[0007] To achieve the above object, the present invention provides a wideband radar single-bit data transmission back and sparse imaging method, including the following steps: Step S1: Convert the wideband radar echo and the threshold into binary data with the same number of bits through data conversion; Step S2: Process according to the positive or negative nature of the threshold; When the threshold is positive, extract the number of bits of the highest non-zero bit in the threshold binary data; When the threshold is negative, extract the number of bits of the highest zero bit in the threshold binary data; Step S3: According to the order of the sampling points, arrange the extracted number of bits to form a decimation method; Step S4: Decimate one bit of the radar echo in the order of the sampling points according to the decimation method in Step S3, and arrange the extracted one-bit data in the order of the sampling point data to form single-bit echo data, that is, realize variable-threshold single-bit quantization; Step S5: Use the single-bit radar two-dimensional sparse imaging algorithm for imaging.

[0008] Preferably, in Step S1, it is set that the wideband radar emits a linear frequency modulation signal, the number of pulses emitted by the wideband radar is N , the number of sampling points of each pulse is M , that is, the radar imaging scene is grid points, and the wideband radar echo is expressed as: ; Among them, is the fast time; is the slow time; and are the scattering coefficient and distance of the th sampling point in the th pulse; is the pulse width; is the frequency modulation rate; is the carrier frequency; is the speed of light; is the reference distance; is the imaginary unit; is the unit rectangular function, , ; .

[0009] Preferably, the wideband radar echo is written in matrix form and is expressed as: ; wherein, , the -th element of is , the -th element of is , the -th element of is: wherein, is the width of the imaging scene in the range direction; is the distance of the -th grid point in the width of the imaging scene in the range direction, ; is the -th moment; , the -th element of is , is the -th Doppler frequency; is expressed as the slow time at the -th moment; , , , .

[0010] Preferably, noise is added to the wideband radar echo matrix form, , and it is written in real matrix form and is expressed as: ; wherein, the specific expressions of each term are: .

[0011] Preferably, the threshold signal in step S1 is represented by a Fourier series, including a single-frequency threshold, a sawtooth threshold, and a triangular threshold, and its general expression is: ; wherein, is the coefficient of the Fourier series, is the frequency of the signal.

[0012] Preferably, when generating a single-frequency signal in negative complex form, the coefficients of its Fourier series are expressed as: ; where is the amplitude of the single-frequency signal, then the specific expression of the single-frequency threshold is: .

[0013] Preferably, when generating a sawtooth signal, the coefficients of the Fourier series are: ; When generating a sawtooth threshold of a real signal, the specific expression of the sawtooth threshold is: .

[0014] Preferably, when generating a triangular signal, the coefficients of the Fourier series are: ; When generating a triangular threshold in real form, the specific expression of the triangular threshold is: ; where is an odd number.

[0015] Furthermore, set the threshold amplitude information.

[0016] The amplitude of the threshold needs to be set according to the signal threshold ratio to ensure the effective recovery of the amplitude information of the signal. The specific expression of the signal threshold ratio is: ; where is the amplitude of the echo signal. When , the amplitude , , of the single-bit quantized signal has a linear relationship, that is: ; To better maintain the integrity of the signal amplitude information, usually needs to be done to have a good effect on recovering the signal amplitude information. Then 's expression is: .

[0017] Preferably, the single-bit echo data obtained in step S4 is expressed as: ; in, Indicates that the data is looped through steps S1-S4, and the single-bit echo data is Written in matrix form , expressed as: ; Right now, .

[0018] Preferably, in step S5, Under the conditions The probability density function of The probability density function of , the objective function is obtained according to the maximum a posteriori probability, and the objective function is simplified; specifically: Setting noise It is an independent and identically distributed Gaussian random variable with mean 0 and variance ,get under conditions The probability density function of : ; in, for No. elements, for No. OK, for No. List, , ; , z , t is the variable of the function; Set Vector Elements obeys an independent and identically distributed Laplace distribution, then The probability density function of for: ; in, is the scale factor, for No. elements; for No. elements; the objective function is obtained according to the maximum a posteriori probability formula, and the objective function is simplified to: .

[0019] Therefore, the present invention adopts the above-mentioned wideband radar single-bit data transmission and sparse imaging method, and the technical effects are as follows: (1) The present invention proposes a variable threshold extraction method, and uses this method to extract one-bit data from the wideband radar echo, and uses this one-bit data as the single-bit echo data, thereby realizing variable threshold single-bit quantization, avoiding the use of additional circuits, and greatly reducing the hardware overhead.

[0020] (2) The present invention proposes a method for extracting one-bit data from the radar echo, transforming the high-precision radar echo data into single-bit radar echo data, which can relieve the transmission and storage pressure of the wideband radar system; (3) The present invention proposes a variable threshold extraction method, which can effectively restore the scattering intensity information of the scattering points, thereby improving the imaging quality; (4) The present invention proposes a single-bit radar two-dimensional sparse reconstruction algorithm, which effectively avoids the problem that the reconstruction accuracy in a certain dimension of the one-dimensional sparse reconstruction is not high, resulting in a decrease in the reconstruction accuracy of the second dimension, and realizes fine imaging of the single-bit radar.

[0021] The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments. Brief Description of the Drawings

[0022] Figure 1 is a flowchart of an embodiment of the wideband radar single-bit data transmission and sparse imaging method of the present invention; Figure 2 is the two-dimensional sparse imaging result of the high-precision radar; Figure 3 is the two-dimensional sparse imaging result of the zero-threshold single-bit radar; Figure 4 is the two-dimensional sparse imaging result of the single-frequency threshold single-bit radar; Figure 5 is the two-dimensional sparse imaging result of the single-bit radar based on single-frequency extraction; Figure 6 is the two-dimensional sparse imaging result of the single-bit radar based on sawtooth extraction; Figure 7 is the two-dimensional sparse imaging result of the single-bit radar based on triangular extraction. Detailed Embodiments

[0023] The technical solution of the present invention will be further described below with reference to the drawings and embodiments.

[0024] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention belongs.

[0025] Embodiment 1 AsFigure 1 As shown in the figure, the present invention provides a broadband radar single-bit data backhaul and sparse imaging method, including the following steps: Step S1: Convert the broadband radar echo and the threshold into binary data with the same number of bits through data conversion to ensure the consistency of calculation. Set the broadband radar to transmit a linear frequency modulation signal. The number of pulses transmitted by the broadband radar is N , and the number of sampling points for each pulse is M , that is, the radar imaging scene is grid points. The broadband radar echo is expressed as: ; Among them, is the fast time; is the slow time; and are the scattering coefficient and distance of the th sampling point within the th pulse; is the pulse width; is the frequency modulation rate; is the carrier frequency; is the speed of light; is the reference distance; is the imaginary unit; is the unit rectangular function, , ; ; Write the broadband radar echo in matrix form , which is expressed as: ; Among them, , The th element of is , The th element of is , The th element of is: Among them, is the width of the imaging scene in the range direction; is the distance of the th grid point in the width of the imaging scene in the range direction, ; is The th moment of , The th element of is , being the th Doppler frequency; is expressed as the slow time at the th moment; , , , .

[0026] Adding noise to the wideband radar echo in matrix form, , and writing it in real matrix form is expressed as: ; wherein, the specific expressions of each term are: .

[0027] Among them, the threshold signal is represented by Fourier series, including single-frequency threshold, sawtooth threshold, and triangular threshold, and its general expression is: ; wherein, are the coefficients of the Fourier series, is the frequency of the signal; When generating a single-frequency signal in negative complex form, the coefficients of its Fourier series are expressed as: ; wherein, is the amplitude of the single-frequency signal, then the specific expression of the single-frequency threshold is: ; The coefficients of the Fourier series of the sawtooth wave signal are: ; When generating a sawtooth threshold of a real signal, the specific expression of the sawtooth threshold is: ; When generating a triangular signal, the coefficients of the Fourier series are: ; When generating a triangular threshold in real form, the specific expression of the triangular threshold is: ; wherein, is an odd number.

[0028] Set the threshold amplitude information.

[0029] The amplitude of the threshold needs to be set according to the signal threshold ratio to ensure the effective recovery of the signal amplitude information. The specific expression of the signal threshold ratio is: ; where, is the amplitude of the echo signal. When , the amplitude of the signal after single-bit quantization , , has a linear relationship, that is: ; To better maintain the integrity of the signal amplitude information, usually is required to have a good effect of recovering the signal amplitude information. Then the expression of is: .

[0030] Step S2: Process according to the positive or negative nature of the threshold; When the threshold is positive, extract the number of bits of the highest non-zero bit in the threshold binary data; for example: the threshold signal is represented by a 16-bit signed binary as , and the number of bits corresponding to the highest non-zero bit is 12, that is, the bits corresponding to in the data; When the threshold is negative, extract the number of bits of the highest zero bit in the threshold binary data; for example: the threshold signal is represented by a 16-bit signed binary as , and the number of bits corresponding to the highest non-zero bit is 11, that is, the bits corresponding to in the data; Step S3: According to the order of the sampling points, arrange the extracted number of bits to form a decimation method; Step S4: Decimate one bit of data from the wideband radar echo according to the decimation method in Step S3 in the order of the sampling points, and arrange the extracted one-bit data in the order of the sampling point data to form a single-bit echo data, that is, realize variable-threshold single-bit quantization. The single-bit echo data obtained in Step S4 is expressed as: ; where, represents the loop operation of Steps S1 - S4 on the data. Write the single-bit echo data Y in matrix form , which is expressed as: ; That is, 。

[0031] Step S5: Use the single-bit radar two-dimensional sparse imaging algorithm for imaging, specifically by deriving the probability density function under the condition of and the probability density function of , and obtaining the objective function according to the maximum a posteriori probability and simplifying the objective function; specifically: Set the noise to be an independent and identically distributed Gaussian random variable with a mean of 0 and a variance of , and obtain the probability density function of under the condition of : : ; where is the -th element of , is the -th row of , is the -th column of , , ; , z , t are the variables of the function; Set the elements of the vector to follow an independent and identically distributed Laplace distribution, then the probability density function of is: ; where is the scale factor, is the -th element of ; is the -th element of ; Obtain the objective function according to the maximum a posteriori probability formula and simplify the objective function to get: 。

[0032] The effects of the present invention are further illustrated by the results of simulation experiments. The software used in the simulation experiments is MATLAB.

[0033] The specific simulation experiment is as follows: ​The experimental conditions for simulating aircraft point targets are as follows: carrier frequency is 10 GHz, pulse repetition frequency is 400 Hz, bandwidth is 400 MHz, pulse width , sampling rate is 4 MHz, coherent processing interval is , target rotation speed is , and signal-to-noise ratio is 15 dB.

[0034] As Figure 2 shown, it is the two-dimensional sparse imaging result of a high-precision radar. Since the two-dimensional sparse reconstruction algorithm can obtain the target scattering intensity information more accurately using high-precision data, therefore, this experimental result is used as a control group in the present invention to illustrate the effectiveness of other algorithms.

[0035] As Figure 3 shown, it is the two-dimensional sparse imaging result of a zero-threshold single-bit radar. It can be seen from the figure that since single-bit quantization is non-linear quantization, it leads to the loss of target scattering intensity information, and due to the harmonic and intermodulation components brought by single-bit quantization, the imaging quality decreases.

[0036] As Figure 4 shown, it is the two-dimensional sparse imaging result of a single-frequency threshold single-bit radar. Since a variable threshold is introduced during single-bit quantization, this algorithm can effectively recover the target scattering intensity. However, since the radar is in a dechirp receiving mode, introducing a single-frequency threshold will cause a spectral line in the range direction, resulting in a decrease in imaging quality.

[0037] As Figure 5 , Figure 6 , Figure 7 shown, they are respectively the two-dimensional sparse imaging results of single-bit radars based on single-frequency decimation, sawtooth decimation, and triangular decimation. It can be seen that this method can not only recover the target scattering intensity information, but also there will be no spectral line in the range direction, thus improving the imaging quality.

[0038] Therefore, by adopting the above-mentioned wideband radar single-bit data transmission and sparse imaging method, the present invention can not only reduce the hardware cost, relieve the transmission and storage pressure of the wideband radar system, but also achieve single-frequency threshold single-bit quantization of the wideband radar echo data, effectively recover the target scattering intensity information, and ensure the imaging quality.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements do not make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A broadband radar single-bit data transmission back and sparse imaging method, characterized in that It includes the following steps: Step S1: Convert the broadband radar echo and the threshold into binary data with the same number of bits through data conversion; Step S2: Process according to the positive or negative nature of the threshold; When the threshold is positive, extract the number of bits of the highest non-zero bit in the threshold binary data; When the threshold is negative, extract the number of bits of the highest zero bit in the threshold binary data; Step S3: According to the order of the sampling points, arrange the extracted number of bits to form a decimation method; Step S4: Decimate one bit of the broadband radar echo in the order of the sampling points according to the decimation method in Step S3, and arrange the extracted one-bit data in the order of the sampling point data to form single-bit echo data, that is, realize variable-threshold single-bit quantization; Step S5: Use the single-bit radar two-dimensional sparse imaging algorithm for imaging.

2. The broadband radar single-bit data transmission and sparse imaging method according to claim 1, characterized in that In step S1, a broadband radar is set to transmit a linear frequency modulation signal, and the number of pulses transmitted by the broadband radar is N , and the number of sampling points for each pulse is M , that is, the broadband radar imaging scene is grid points, and the broadband radar echo is expressed as: ； Among them, is the fast time; is the slow time; and is the th scattering coefficient and distance of the th sampling point within the th pulse; is the pulse width; is the frequency modulation rate; is the carrier frequency; is the speed of light; is the imaginary unit; is the unit rectangular function, , ; .

3. The broadband radar single-bit data feedback and sparse imaging method according to claim 2, characterized in that Write the broadband radar echo in matrix form , which is expressed as: ； Among them, , the th element of ; , the th element of ; , the th element is: ； Among them, is the width of the imaging scene in the range direction; is the distance of the th grid point in the width of the imaging scene in the range direction, ; is at the th moment; , at the th element of is at the th Doppler frequency; is expressed as the slow time at the th moment; is the Doppler frequency, , , , .

4. The broadband radar single-bit data feedback and sparse imaging method according to claim 3, characterized in that In the broadband radar echo Add noise to the matrix form , , and write it in the form of a real matrix It is expressed as: ； Among them, the specific expressions of each item are: 。 5. The broadband radar single-bit data backhaul and sparse imaging method according to claim 4, characterized in that In Step S1, the threshold signal is represented by a Fourier series, including a single-frequency threshold, a sawtooth threshold, and a triangular threshold, and its general expression is: ； wherein, is the coefficient of the Fourier series, is the frequency of the signal.

6. The broadband radar single-bit data feedback and sparse imaging method according to claim 5, characterized in that When generating a single-frequency signal in the form of a negative complex number, the coefficients of its Fourier series are expressed as: ； Among them, is the amplitude of the single-frequency signal, and the specific expression of the single-frequency threshold is as follows: 。 7. The broadband radar single-bit data feedback and sparse imaging method according to claim 5, characterized in that When generating a sawtooth wave signal, the coefficients of the Fourier series are: ； When generating a sawtooth threshold of a real signal, the specific expression of the sawtooth threshold is: 。 8. The broadband radar single-bit data backhaul and sparse imaging method according to claim 5, characterized in that When generating a triangular signal, the coefficients of the Fourier series are: ； When generating a triangular threshold in the form of a real number, the specific expression of the triangular threshold is: ； Among them, is an odd number.

9. The broadband radar single-bit data backhaul and sparse imaging method according to claim 4, characterized in that The single-bit echo data obtained in step S4 is expressed as: ； Among them, represents performing a loop operation on the data from step S1 to step S4, and writing the single-bit echo data in matrix form , which is expressed as: ； That is, .

10. The broadband radar single-bit data feedback and sparse imaging method according to claim 9, characterized in that In step S5, deduce the probability density function under the condition of and deduce the probability density function of . Obtain the objective function according to the maximum a posteriori probability and simplify the objective function. Specifically: ​ Set noise Gaussian random variables that are independent and identically distributed, with a mean of 0 and a variance of , obtain Under the condition of The probability density function of : ； Among them, is the th element of is the th row of is the th column of , ; , z , t are variables of the function; Set vector whose elements follow an independent and identically distributed Laplace distribution, then the probability density function is as follows: ； Among them, is the scale factor, is the th element of; is the th element of; Obtain the objective function according to the maximum a posteriori probability formula, and simplify the objective function to get: 。

Images