A frequency offset estimation method based on compression sensing and frequency multiplication joint estimation

Abstract

This invention relates to the field of communication technology, specifically to a frequency offset estimation method based on compressed sensing and frequency doubling joint estimation. The method includes the following steps: S1, establishing a sparse representation model of the signal frequency offset, constructing a sparse dictionary of the signal frequency offset matching the frequency offset characteristics of short-time burst signals based on the sparse representation model, performing a coarse frequency offset estimation using a compressed sensing algorithm, and outputting a coarse frequency offset estimate; S2: based on the coarse frequency offset estimate obtained in step S1, performing digital down-conversion (DDC) processing on the received short-time burst signal to obtain a baseband signal that eliminates the principal components of the coarse frequency offset estimate and retains only the residual small frequency offset; S3: performing a frequency doubling joint estimation on the baseband signal obtained in step S2 using the four-fold relationship of the signal frequency offset to achieve a fine estimation of the frequency offset of the short-time burst signal, and outputting the final accurate frequency offset estimate. This invention solves the technical problems of high sample size requirements and lack of effective connection between coarse and fine estimation in short-time burst signal frequency offset estimation.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and specifically to a frequency offset estimation method based on compressed sensing and frequency doubling joint estimation. Background Technology

[0002] In communication fields such as wireless burst communication, low-orbit satellite communication, and IoT short-frame communication, signals are susceptible to carrier frequency offset during transmission due to factors such as the Doppler effect and channel multipath fading. Frequency offset can disrupt the synchronization between the received signal and the local carrier, directly reducing signal demodulation accuracy and deteriorating communication link quality. Therefore, carrier frequency offset estimation is a core key technology in the carrier synchronization stage of communication signal processing. Especially in short burst signal scenarios, due to the short signal duration and limited data samples, stringent requirements are placed on the small sample adaptability, estimation accuracy, and real-time performance of frequency offset estimation.

[0003] Existing methods for frequency offset estimation are mainly divided into two categories. One category is the traditional coarse-fine estimation combined method, which completes the coarse estimation of frequency offset through matched filtering, Fourier transform, etc., and then achieves fine estimation by means of frequency sweep search, second harmonic joint estimation, etc. The other category is the frequency offset estimation method based on compressed sensing, which utilizes the sparsity characteristics of frequency offset in frequency space to transform frequency offset estimation into a sparse signal reconstruction problem, and achieves coarse estimation of frequency offset under small sample conditions. Some technologies also attempt to combine compressed sensing with second harmonic estimation and apply it to frequency offset estimation scenarios of continuous signals and MIMO systems.

[0004] However, existing frequency offset estimation methods still have significant technical shortcomings. Traditional methods rely on a large number of data samples for the fine estimation stage, while frequency sweeping methods have high computational complexity and slow convergence. The accuracy improvement of second harmonic estimation is limited, and none of these methods can meet the small sample requirements of short-term burst signals. Single-step methods based on compressed sensing can only achieve coarse estimation, which is difficult to meet the accuracy requirements of engineering applications. A few techniques that integrate compressed sensing and frequency harmonic estimation have not been customized for short-term burst signals, lack an effective signal preprocessing stage to connect coarse and fine estimation, and the integration of techniques is rigid, failing to meet the engineering requirements of small sample size, high accuracy, and low computational complexity. Therefore, there is an urgent need for a frequency offset estimation method adapted to short-term burst signals to solve the above-mentioned problems of existing technologies. Summary of the Invention

[0005] This invention provides a frequency offset estimation method based on compressed sensing and frequency doubling joint estimation to solve the technical problems in the prior art of frequency offset estimation of short-time burst signals, such as high sample size requirements and lack of effective connection between coarse and fine estimation.

[0006] To address the above problems, the frequency offset estimation method based on compressed sensing and frequency doubling joint estimation provided by this invention adopts the following technical solution: A frequency offset estimation method based on compressed sensing and frequency doubling joint estimation is used for frequency offset estimation of short-time burst signals, comprising the following steps: S1. Establish a sparse representation model of signal frequency offset, construct a sparse dictionary of signal frequency offset based on the sparse representation model to match the frequency offset characteristics of short-time burst signals, use compressed sensing algorithm to perform coarse frequency offset estimation on the received short-time burst signals, and output coarse frequency offset estimation value. S2: Based on the coarse frequency offset estimate obtained in step S1, perform digital down-conversion (DDC) processing on the received short burst signal to obtain a baseband signal that eliminates the coarse frequency offset principal component and retains only the residual small frequency offset. S3: For the baseband signal obtained in step S2, perform frequency multiplication joint estimation using the four-fold relationship of the signal frequency offset to achieve accurate estimation of the frequency offset of short-time burst signals and output the final accurate frequency offset estimate.

[0007] The beneficial effects of the above technical solution are as follows: by applying it to the frequency offset estimation scenario of short-term burst signals, it can achieve efficient and high-precision estimation of signal frequency offset under small sample data conditions. Specifically, compressed sensing coarse estimation can quickly lock the approximate range of frequency offset, significantly narrowing the search interval for subsequent fine estimation; DDC processing, as a key link between coarse and fine estimation, can effectively eliminate the influence of the principal components of the frequency offset in coarse estimation, obtaining a baseband signal containing only residual small frequency offsets, simplifying the processing difficulty of fine estimation; and quadruple harmonic estimation can accurately calculate the residual small frequency offsets. The three form an organic closed loop, which not only takes into account the requirements of small sample size and low computational complexity for short-term burst signals, but also ensures the overall accuracy of frequency offset estimation, while improving the real-time performance of frequency offset estimation, making the whole method highly feasible in short-term burst signal processing.

[0008] Furthermore, in step S1, when establishing a sparse representation model of the signal frequency offset, based on the sparse distribution characteristics of the frequency offset in the frequency space, the frequency offset estimation problem of the short-time burst signal is transformed into a sparse signal reconstruction problem. A sparse representation mathematical model containing the frequency offset parameters to be estimated is constructed, with the frequency offset parameters as sparse variables and the received short-time burst signal as the observation variable.

[0009] The beneficial effects of the above technical solution are: it clarifies the construction logic and variable definition of the sparse representation model, enables the model to accurately match the frequency offset characteristics of short-term burst signals, provides a clear basis for the construction of sparse dictionaries, reduces invalid calculations in modeling, and improves the modeling efficiency and scene adaptability of compressed sensing coarse estimation.

[0010] Furthermore, in step S1, when constructing the signal frequency offset sparse dictionary, a set of frequency candidate basis vectors covering the preset frequency offset search range is constructed based on the modulation method, sampling rate, and preset frequency offset search range of the short-time burst signal. The frequency candidate basis vectors are normalized to form a signal frequency offset sparse dictionary that highly matches the frequency offset characteristics of the short-time burst signal.

[0011] The beneficial effects of the above technical solution are: by constructing and normalizing a sparse dictionary based on the inherent characteristics of short burst signals, the dictionary basis vectors can fully cover the frequency offset range to be estimated, thereby improving the accuracy of matching the received signal with the dictionary projection, quickly and accurately locking the candidate value of the frequency offset, and effectively reducing the search complexity in the coarse estimation stage.

[0012] Furthermore, in step S1, when using the compressed sensing algorithm to perform a coarse frequency offset estimation on the received short burst signal, the Orthogonal Matching Pursuit (OMP), Basis Pursuit (BP), or Sparse Bayesian Learning (SBL) algorithm is used to perform projection matching and sparse reconstruction of the received short burst signal with the frequency offset sparse dictionary of the signal, and to select frequency offset candidate values ​​from the frequency candidate set of the sparse dictionary, and output them as the coarse frequency offset estimation value.

[0013] The beneficial effects of the above technical solutions are: the selected OMP, BP, and SBL algorithms are adapted to the characteristics of small samples, can efficiently complete the matching and reconstruction of signals and dictionaries, and improve the computational efficiency and reliability of frequency offset coarse estimation; the selection of multiple algorithms also allows the method to adapt to the differentiated needs of different scenarios and enhances the flexibility of engineering applications.

[0014] Furthermore, in step S2, when performing digital down-conversion (DDC) processing, the following steps are included: S21, Generate corresponding in-phase and quadrature local carriers based on the coarse frequency offset estimate; S22, the received short burst signal is mixed with the in-phase and quadrature local carriers respectively to obtain the mixed signal; S23, perform low-pass filtering on the mixed signal to filter out high-frequency noise and spurious components; S24 performs decimation and speed-down processing on the filtered signal to obtain the baseband signal.

[0015] The beneficial effects of the above technical solution are: systematic optimization of the signal is achieved through step-by-step DDC processing, the baseband shift of the signal is accurately completed, high-frequency noise and spurious components are effectively filtered out, the amount of data to be processed in the subsequent process is reduced, the signal-to-noise ratio of the baseband signal is improved, the computational load of the fine estimation stage is significantly reduced, and a high-quality signal foundation is laid for high-precision fine estimation.

[0016] Furthermore, in step S3, when performing frequency harmonic joint estimation using the fourfold relationship of the signal frequency offset, the baseband signal obtained in step S2 is subjected to fourth harmonic component extraction, the phase change of the fourth harmonic component is calculated by the phase dewinding algorithm, the residual frequency offset value is solved based on the linear correspondence between the phase change and the frequency offset, and the residual frequency offset value is fused with the coarse frequency offset estimate value in step S1 to obtain the accurate frequency offset estimate value.

[0017] The beneficial effects of the above technical solution are: by extracting and amplifying the residual small frequency offset features through the fourth harmonic component, the weak frequency offset is more easily detected and calculated; combined with the phase dewinding algorithm, the residual frequency offset value can be accurately solved, and after being fused with the coarse estimate, the advantages of the two-stage estimation are integrated, which significantly improves the overall accuracy of frequency offset estimation.

[0018] Furthermore, when extracting the fourth harmonic component from the baseband signal, an autocorrelation operation is performed on the baseband signal to suppress additive noise in the signal. Then, a Fourier transform is performed on the autocorrelation result to extract the phase and amplitude features corresponding to the fourth harmonic position in the transform result. The effective components after filtering out noise interference are taken as the fourth harmonic components.

[0019] The beneficial effects of the above technical solution are: firstly, by suppressing additive noise through autocorrelation operation, the purity of the baseband signal is improved; then, the effective components of the fourth harmonic are extracted in a targeted manner to avoid the interference of noise on the frequency domain features, providing accurate feature basis for solving the residual frequency offset, and effectively reducing the error in the harmonic estimation process.

[0020] Furthermore, when fusing the residual frequency offset value with the coarse frequency offset estimate, a linear fusion algorithm is used to compensate for the system error of the coarse frequency offset estimate. The compensated coarse frequency offset estimate is then superimposed with the residual frequency offset value to obtain the calibrated accurate frequency offset estimate.

[0021] The beneficial effects of the above technical solution are: to compensate for the systematic error of the coarse estimate and eliminate the inherent bias of the coarse estimate; the linear fusion algorithm is simple to operate, efficient and has no additional computational burden; after fusion calibration, the final frequency offset estimate is more in line with reality, and the accuracy of the estimation result is further improved.

[0022] Furthermore, this method is implemented based on a hardware carrier, which is a signal processing chip of a field-programmable gate array (FPGA), a digital signal processor (DSP), or a microcontroller unit (ARM).

[0023] Furthermore, the short burst signal is a short burst communication signal transmitted in low-orbit satellite communication, wireless burst communication, or Internet of Things short frame communication scenarios.

[0024] The beneficial effects of the frequency offset estimation method based on compressed sensing and frequency doubling joint estimation provided by this invention are as follows: For short-time burst signals, when performing frequency offset estimation, compressed sensing and fourth-harmonic generation joint estimation techniques are integrated to construct a closed-loop process of coarse estimation-DDC processing-fine estimation, achieving efficient frequency offset estimation even under small sample conditions. Compressed sensing quickly locks the frequency offset range, significantly reducing the computational complexity of subsequent calculations; DDC processing effectively optimizes signal characteristics and improves the signal-to-noise ratio, achieving efficient connection between coarse and fine estimation; fourth-harmonic generation joint estimation amplifies residual frequency offset characteristics, significantly improving estimation accuracy. The detailed design of each step of the scheme ensures computational efficiency and result reliability, can be implemented on mainstream hardware, and is adaptable to various scenarios such as low-Earth orbit satellite communication, balancing real-time performance and engineering feasibility.

[0025] In summary, through the above-mentioned settings, this invention solves the technical problems in the prior art of estimating the frequency offset of short-time burst signals, such as the high requirement for sample size and the lack of effective connection between coarse and fine estimation. Attached Figure Description

[0026] Figure 1 This is a flowchart of the frequency offset estimation method based on compressed sensing and frequency doubling joint estimation provided by the present invention. Detailed Implementation

[0027] The principles and spirit of the present invention will be explained in detail below with reference to several representative embodiments.

[0028] An embodiment of the frequency offset estimation method based on compressed sensing and frequency doubling joint estimation provided by this invention: like Figure 1 As shown, a frequency offset estimation method based on compressed sensing and frequency doubling joint estimation is used for frequency offset estimation of short-time burst signals, including the following steps: S1. Establish a sparse representation model of signal frequency offset, construct a sparse dictionary of signal frequency offset based on the sparse representation model to match the frequency offset characteristics of short-time burst signals, use compressed sensing algorithm to perform coarse frequency offset estimation on the received short-time burst signals, and output coarse frequency offset estimation value. S2: Based on the coarse frequency offset estimate obtained in step S1, perform digital down-conversion (DDC) processing on the received short burst signal to obtain a baseband signal that eliminates the coarse frequency offset principal component and retains only the residual small frequency offset. S3: For the baseband signal obtained in step S2, perform frequency multiplication joint estimation using the four-fold relationship of the signal frequency offset to achieve accurate estimation of the frequency offset of short-time burst signals and output the final accurate frequency offset estimate.

[0029] Specifically, in step S1, when establishing the sparse representation model of the signal frequency offset, based on the sparse distribution characteristics of the frequency offset in the frequency space, the frequency offset estimation problem of short-time burst signals is transformed into a sparse signal reconstruction problem. A sparse representation mathematical model containing the frequency offset parameters to be estimated is constructed, with the frequency offset parameters as sparse variables and the received short-time burst signals as observed variables. This clarifies the construction logic and variable definitions of the sparse representation model, ensuring accurate matching between the model and the frequency offset characteristics of short-time burst signals. This provides a clear basis for the construction of the sparse dictionary, reduces invalid computations in modeling, and improves the modeling efficiency and scenario adaptability of compressed sensing coarse estimation.

[0030] Simultaneously, when constructing the sparse dictionary of signal frequency offset, a set of frequency candidate basis vectors covering the preset frequency offset search range is built based on the modulation method, sampling rate, and preset frequency offset search range of the short-time burst signal. These frequency candidate basis vectors are then normalized to form a sparse dictionary of signal frequency offset that highly matches the frequency offset characteristics of the short-time burst signal. By constructing and normalizing the sparse dictionary in conjunction with the inherent characteristics of the short-time burst signal, the dictionary basis vectors fully cover the frequency offset range to be estimated, improving the accuracy of matching the received signal with the dictionary projection. This enables rapid and accurate locking of frequency offset candidate values ​​and effectively reduces the search complexity in the coarse estimation stage.

[0031] Furthermore, when using compressed sensing algorithms to coarsely estimate the frequency offset of received short-duration burst signals, the Orthogonal Matching Pursuit (OMP), Basis Pursuit (BP), or Sparse Bayesian Learning (SBL) algorithms are employed. These algorithms perform projection matching and sparse reconstruction between the received short-duration burst signal and the sparse frequency offset dictionary. Candidate frequency offset values ​​are then selected from the frequency candidate set of the sparse dictionary and output as the coarse frequency offset estimate. The selected OMP, BP, and SBL algorithms are well-suited to small sample sizes and can efficiently complete the matching and reconstruction of the signal and dictionary, improving the computational efficiency and reliability of the coarse frequency offset estimation. The choice of multiple algorithms also allows the method to adapt to the differentiated needs of different scenarios, enhancing the flexibility of engineering applications.

[0032] In step S2, the digital down-conversion (DDC) process includes the following steps: S21, Generate corresponding in-phase and quadrature local carriers based on the coarse frequency offset estimate; S22, the received short burst signal is mixed with the in-phase and quadrature local carriers respectively to obtain the mixed signal; S23, perform low-pass filtering on the mixed signal to filter out high-frequency noise and spurious components; S24 performs decimation and speed-down processing on the filtered signal to obtain the baseband signal.

[0033] By implementing step-by-step DDC processing, the signal is systematically optimized, accurately shifting the baseband signal, effectively filtering out high-frequency noise and spurious components, reducing the amount of data to be processed in the subsequent stages, improving the signal-to-noise ratio of the baseband signal, and significantly reducing the computational load in the fine estimation stage, thus laying a high-quality signal foundation for high-precision fine estimation.

[0034] In this embodiment, in step S3, when performing joint frequency offset estimation using the fourfold relationship of the signal frequency offset, the baseband signal obtained in step S2 is subjected to fourth harmonic component extraction. The phase change of the fourth harmonic component is calculated using a phase dewinding algorithm. Based on the linear correspondence between the phase change and the frequency offset, the residual frequency offset value is solved. The residual frequency offset value is then fused with the coarse frequency offset estimate from step S1 to obtain a precise frequency offset estimate. By extracting the fourth harmonic component, the residual small frequency offset features are amplified, making weak frequency offsets easier to detect and calculate. Combined with the phase dewinding algorithm, the residual frequency offset value can be accurately solved. After fusion with the coarse estimate, the advantages of the two-stage estimation are integrated, significantly improving the overall accuracy of frequency offset estimation.

[0035] When extracting the fourth harmonic components from the baseband signal, an autocorrelation operation is performed on the baseband signal to suppress additive noise. Then, a Fourier transform is performed on the autocorrelation result to extract the phase and amplitude features corresponding to the fourth harmonic position. The effective components after filtering out noise interference are taken as the fourth harmonic components. First, additive noise is suppressed through autocorrelation to improve the purity of the baseband signal. Then, the effective fourth harmonic components are extracted selectively, avoiding noise interference with frequency domain features. This provides accurate feature basis for solving the residual frequency offset and effectively reduces the error in the harmonic estimation process.

[0036] In this embodiment, when fusing the residual frequency offset value with the coarse frequency offset estimate, a linear fusion algorithm is used to compensate for systematic errors in the coarse frequency offset estimate. The compensated coarse frequency offset estimate is then superimposed with the residual frequency offset value to obtain the calibrated accurate frequency offset estimate. Compensating for systematic errors in the coarse estimate eliminates inherent biases. The linear fusion algorithm is simple to operate, highly efficient, and requires no additional computational burden. After fusion calibration, the final frequency offset estimate is more closely aligned with reality, further improving the accuracy of the estimation results.

[0037] In this embodiment, the entire method is implemented based on a hardware platform, which is a signal processing chip such as a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), or a Microcontroller Unit (ARM). Furthermore, the short-time burst signal is a short-frame burst communication signal transmitted in low-Earth orbit satellite communication, wireless burst communication, or IoT short-frame communication scenarios.

[0038] The working principle of the frequency offset estimation method based on compressed sensing and frequency doubling joint estimation provided by this invention is as follows: First, based on the sparse distribution characteristics of frequency offset, a sparse representation model of the signal frequency offset is established, and a sparse frequency offset dictionary matching the characteristics of short-time burst signals is constructed. The received signal is then processed using a compressed sensing algorithm to complete a coarse frequency offset estimate and output the coarse estimate value. Subsequently, based on this coarse estimate value, digital down-conversion processing is performed on the received short-time burst signal. Through operations such as generating a local carrier, mixing, low-pass filtering, and decimation, a baseband signal is obtained that eliminates the principal components of the coarse frequency offset estimate and retains only the residual small frequency offset. Finally, the fourth harmonic component is extracted from the baseband signal, and the phase change is calculated using a phase dewinding algorithm to solve for the residual frequency offset value. The residual frequency offset value is then fused and calibrated with the coarse estimate value to finally output a precise estimate of the signal frequency offset, completing the entire frequency offset estimation process.

[0039] Based on the above description in this specification, those skilled in the art will also understand that the following terms, such as "upper," "lower," "front," "rear," "left," "right," "width," "horizontal," "top," "bottom," "inner," and "outer," which indicate orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings of this specification. They are only for the purpose of facilitating the explanation of the present invention and simplifying the description, and do not explicitly or implicitly suggest that the device or element involved must have the specific orientation, or be constructed and operated in a specific orientation. Therefore, the above-mentioned orientation or positional relationship terms should not be understood or interpreted as limitations on the present invention.

[0040] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A frequency offset estimation method based on compressed sensing and frequency doubling joint estimation, characterized in that, Frequency offset estimation for short-duration burst signals includes the following steps: S1. Establish a sparse representation model of signal frequency offset, construct a sparse dictionary of signal frequency offset based on the sparse representation model to match the frequency offset characteristics of short-time burst signals, use compressed sensing algorithm to perform coarse frequency offset estimation on the received short-time burst signals, and output coarse frequency offset estimation value. S2: Based on the coarse frequency offset estimate obtained in step S1, perform digital down-conversion (DDC) processing on the received short burst signal to obtain a baseband signal that eliminates the coarse frequency offset principal component and retains only the residual small frequency offset. S3: For the baseband signal obtained in step S2, perform frequency multiplication joint estimation using the four-fold relationship of the signal frequency offset to achieve accurate estimation of the frequency offset of short-time burst signals and output the final accurate frequency offset estimate.

2. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to claim 1, characterized in that: In step S1, when establishing a sparse representation model of signal frequency offset, based on the sparse distribution characteristics of frequency offset in frequency space, the frequency offset estimation problem of short-time burst signal is transformed into a sparse signal reconstruction problem. A sparse representation mathematical model containing the frequency offset parameters to be estimated is constructed, with the frequency offset parameters as sparse variables and the received short-time burst signal as the observation variable.

3. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to claim 2, characterized in that: In step S1, when constructing the signal frequency offset sparse dictionary, a set of frequency candidate basis vectors covering the preset frequency offset search range is constructed based on the modulation method, sampling rate, and preset frequency offset search range of the short-time burst signal. The frequency candidate basis vectors are normalized to form a signal frequency offset sparse dictionary that highly matches the frequency offset characteristics of the short-time burst signal.

4. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to claim 3, characterized in that: In step S1, when using the compressed sensing algorithm to perform a coarse frequency offset estimation on the received short burst signal, the Orthogonal Matching Pursuit (OMP), Basis Pursuit (BP), or Sparse Bayesian Learning (SBL) algorithm is used to perform projection matching and sparse reconstruction of the received short burst signal with the frequency offset sparse dictionary of the signal. Frequency offset candidate values ​​are selected from the frequency candidate set of the sparse dictionary and output as the coarse frequency offset estimate.

5. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to any one of claims 1 to 4, characterized in that, In step S2, the digital down-conversion (DDC) process includes the following steps: S21, Generate corresponding in-phase and quadrature local carriers based on the coarse frequency offset estimate; S22, the received short burst signal is mixed with the in-phase and quadrature local carriers respectively to obtain the mixed signal; S23, perform low-pass filtering on the mixed signal to filter out high-frequency noise and spurious components; S24 performs decimation and speed-down processing on the filtered signal to obtain the baseband signal.

6. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to any one of claims 1 to 4, characterized in that, In step S3, when performing frequency harmonic joint estimation using the fourfold relationship of the signal frequency offset, the baseband signal obtained in step S2 is subjected to fourth harmonic component extraction. The phase change of the fourth harmonic component is calculated by the phase dewinding algorithm. The residual frequency offset value is solved based on the linear correspondence between the phase change and the frequency offset. The residual frequency offset value is fused with the coarse frequency offset estimate from step S1 to obtain the accurate frequency offset estimate.

7. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to claim 6, characterized in that, When extracting the fourth harmonic component from the baseband signal, an autocorrelation operation is performed on the baseband signal to suppress additive noise in the signal. Then, a Fourier transform is performed on the autocorrelation result to extract the phase and amplitude features corresponding to the fourth harmonic position in the transform result. The effective components after filtering out noise interference are taken as the fourth harmonic components.

8. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to claim 7, characterized in that: When fusing the residual frequency offset value with the coarse frequency offset estimate, a linear fusion algorithm is used to compensate for the system error of the coarse frequency offset estimate. The compensated coarse frequency offset estimate is then superimposed with the residual frequency offset value to obtain the calibrated accurate frequency offset estimate.

9. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to any one of claims 1 to 4, characterized in that: This method is implemented based on a hardware carrier, which is a signal processing chip such as a field-programmable gate array (FPGA), a digital signal processor (DSP), or a microcontroller unit (ARM).

10. The frequency offset estimation method based on compressed sensing and frequency doubling joint estimation according to any one of claims 1 to 4, characterized in that: The short burst signal refers to the short burst communication signal transmitted in low-orbit satellite communication, wireless burst communication, or Internet of Things short frame communication scenarios.

Images