Chirp signal acquisition method and system based on dft frequency estimation correction

Abstract

The application relates to the technical field of wireless communication, and discloses a Chirp signal capturing method and system based on DFT frequency estimation correction, which comprises the following steps: searching on a received data stream to detect the existence of a Chirp signal and preliminarily positioning the starting position of the Chirp signal; based on the preliminarily positioned starting position, obtaining a first frequency estimation value and a second frequency estimation value, and correcting the first frequency estimation value and the second frequency estimation value by using an iterative frequency estimation algorithm to obtain a first-stage corrected frequency value; calculating a first time offset according to the first-stage corrected frequency value and updating the starting position by using the first time offset; based on the updated starting position, obtaining a new first frequency estimation value and a new second frequency estimation value again, and correcting the new first frequency estimation value and the new second frequency estimation value by using the iterative frequency estimation algorithm to obtain a second-stage corrected frequency value; and calculating and outputting a final frequency offset and an accurate timing position according to the second-stage corrected frequency value. The application avoids correction failure caused by excessively large initial deviation, and has high initial positioning accuracy.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and specifically to a method and system for acquiring Chirp signals based on DFT frequency estimation correction. Background Technology

[0002] In wireless communication systems, carrier frequency deviation (frequency offset) and sampling time deviation (time offset, also known as timing deviation) between the receiver and transmitter can severely degrade communication performance. Therefore, fast and accurate timing synchronization and frequency synchronization are prerequisites for reliable communication.

[0003] Chirp signals, due to their constant envelope, low peak-to-average power ratio (PAPR), and ideal autocorrelation characteristics, have been widely used in radar, sonar, and in recent years, low-power wide-area Internet of Things (IoT) applications such as LoRa. Their unique characteristic of linear frequency variation over time allows for the joint estimation of time and frequency offsets through simple processing.

[0004] Currently, a typical synchronization method based on Chirp signals has the following basic principle (corresponding to the original...). Figure 1 ):

[0005] 1. The receiving end generates a local up-scan frequency signal. and downsweep frequency signal .

[0006] 2. Compare the received signal y(t) with... and Perform correlation calculations to obtain two correlation signals. and .

[0007] 3. To and Perform Fast Fourier Transform (FFT) on each, and find the frequency value corresponding to the peak value of the spectrum. and .

[0008] 4. Finally, through the formula and Calculate the frequency offset and time bias .

[0009] However, this existing technology has the following significant drawbacks:

[0010] 1. Limited estimation accuracy: The core of this scheme is to perform FFT operations on the correlated signal and find the peak. The frequency resolution of FFT itself (…) ,in The sampling rate (where N is the number of FFT points) limits the accuracy of frequency estimation, which directly leads to frequency offset. and time bias The estimated values ​​have a large inherent error, especially under low signal-to-noise ratio conditions, where the error is more significant.

[0011] 2. Insufficient detection reliability: This method typically sets a fixed threshold value to determine whether the chirp signal has arrived. In actual changing channel environments, a fixed threshold is difficult to adjust adaptively: a threshold that is too high will lead to missed detections, while a threshold that is too low will generate false alarms, which reduces the reliability of synchronous acquisition.

[0012] 3. Insensitive to signal start position: The above process assumes that the start time of the chirp signal is known accurately. However, in actual communication, the arrival time of the signal is unknown. A simple sliding window search may result in the window not being precisely aligned with the signal, leading to dispersion of relevant peak energy and further exacerbating the error in frequency estimation.

[0013] Therefore, there is an urgent need in this field for a Chirp signal acquisition method that can overcome the above-mentioned defects, so as to achieve high-precision and high-reliability timing and frequency synchronization even in complex channel environments. Summary of the Invention

[0014] This invention provides a Chirp signal acquisition method and system based on DFT frequency estimation correction to solve the above-mentioned problems.

[0015] This invention is achieved through the following technical solution:

[0016] A chirp signal acquisition method based on DFT frequency estimation correction includes the following steps:

[0017] The presence of the Chirp signal is detected and the initial position of the Chirp signal is preliminarily located by searching on the received data stream;

[0018] Based on the initial location, a first frequency estimate and a second frequency estimate are obtained, and an iterative frequency estimation algorithm is used to correct the first frequency estimate and the second frequency estimate respectively to obtain a first-level corrected frequency value.

[0019] The first time offset is calculated based on the first-level corrected frequency value, and the starting position is updated using the first time offset; based on the updated starting position, a new first frequency estimate and a new second frequency estimate are obtained again, and the iterative frequency estimation algorithm is used to correct the new first frequency estimate and the new second frequency estimate respectively to obtain the second-level corrected frequency value.

[0020] Based on the second-level corrected frequency value, the final frequency offset and precise timing position are calculated and output.

[0021] As an optimization, the specific process of detecting the presence of the Chirp signal and initially locating its starting position by searching the received data stream is as follows:

[0022] A sliding search is performed on the received data stream with a predetermined window length and sliding step;

[0023] For each sliding window, calculate the up-and-down frequency sweep correlation results of the received signal within that sliding window, and obtain the first peak frequency, the first peak intensity, and the preliminary time offset after up-sweeping.

[0024] Calculate the signal energy of the received signal within the sliding window;

[0025] When the first peak intensity is not less than the adaptive threshold determined by the signal energy, it is determined that a Chirp signal has been detected.

[0026] By comparing the peak intensities of adjacent sliding windows that satisfy the adaptive threshold, the window head position corresponding to the sliding window with the larger peak intensity is determined as the candidate position ID, and the initial time offset corresponding to this window is recorded. ;

[0027] According to the formula The candidate positions are corrected to obtain the initial positioning starting position; where Ts is the sampling period.

[0028] As an optimization, the sliding step is half the predetermined window length, which is L*h, where L is the length of the Chirp signal and h is the sampling factor.

[0029] As an optimization, the iterative frequency estimation algorithm is a frequency estimation algorithm based on the improved Discrete Fourier Transform (DFT).

[0030] As an optimization, the frequency estimation algorithm based on the improved DFT is used to perform M iterations on the frequency estimate to be corrected, and the following steps are executed:

[0031] The frequency estimate to be corrected is used as the initial value for iteration. ;

[0032] For m = 0, 1, 2, ..., M-1, perform the m-th iteration sequentially, where M is a positive integer greater than 1. Each iteration includes:

[0033] a) Calculate the two improved DFT coefficients , :

[0034] ;

[0035] ;

[0036] n represents the sequence number of the sampling point. This represents the sampling sequence of the received data stream. This represents the frequency estimate in the m-th iteration. Where N is the sampling period and N is the number of DFT points;

[0037] b) Calculate the ratio , ;

[0038] c) Update frequency estimate: ,in, The sampling frequency;

[0039] After M iterations, the output is This is the corrected frequency value.

[0040] As an optimization, the specific process of calculating the first time offset based on the first-level correction frequency value and updating the starting position using the first time offset is as follows:

[0041] The first time offset is calculated based on the first-level correction frequency value. : ,in, , These are the first and second frequency values ​​that constitute the first-level correction frequency value, respectively.

[0042] According to the formula Update the starting position;

[0043] Where u is the chirp signal modulation frequency, Ts is the sampling period, and fix(⋅) is the floor function.

[0044] As an optimization, the specific formulas for calculating the final frequency offset and precise timing position are as follows:

[0045] ;

[0046] ;

[0047] ;

[0048] in, , These are the final frequency offset and precise timing position calculated based on the second-level corrected frequency value, respectively. , These are the first and second frequency values ​​that constitute the second-level correction frequency value, respectively, and u is the modulation frequency of the Chirp signal.

[0049] As an optimization, it also includes performing third or more levels of synchronization operations, with each level updating the starting position and performing frequency estimation and correction based on the timing results of the previous level.

[0050] This invention also discloses a Chirp signal acquisition system based on DFT frequency estimation correction, used to execute the aforementioned Chirp signal acquisition method based on DFT frequency estimation correction, comprising:

[0051] The signal detection and preliminary positioning module is used to detect the presence of the Chirp signal and preliminarily locate the starting position of the Chirp signal by searching on the received data stream.

[0052] The first-level synchronization module is used to obtain a first frequency estimate and a second frequency estimate based on the initial positioning, and to use an iterative frequency estimation algorithm to correct the first frequency estimate and the second frequency estimate respectively to obtain a first-level corrected frequency value.

[0053] The second-level synchronization module is used to calculate the first time offset based on the first-level corrected frequency value, and update the starting position using the first time offset; based on the updated starting position, a new first frequency estimate and a new second frequency estimate are obtained again, and the iterative frequency estimation algorithm is used to correct the new first frequency estimate and the new second frequency estimate respectively to obtain the second-level corrected frequency value.

[0054] The synchronization result output module is used to calculate and output the final frequency offset and precise timing position based on the second-level correction frequency value.

[0055] As an optimization, it also includes:

[0056] The Q-level synchronization module is located after the second-level synchronization module. It is used to update the starting position and perform frequency estimation and correction based on the timing results of the previous level. Q is a positive integer greater than 2.

[0057] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0058] This invention breaks through the traditional linear process of direct output from a single estimation, creatively proposing a collaborative correction architecture of first-level synchronization, timing feedback, and second-level synchronization. In traditional methods, inaccurate timing leads to inaccurate frequency estimation, which in turn affects the calculation of timing deviation, creating a vicious cycle. This architecture, through the preliminary results of first-level synchronization, prioritizes the correction of timing position, providing a cleaner signal observation window for second-level synchronization. This breaks the error coupling relationship between timing and frequency estimation, solving the mutually restrictive accuracy bottleneck. Furthermore, this architecture makes frequency deviation estimation accuracy and timing estimation accuracy no longer independent indicators, but rather mutually reinforcing and alternately increasing, ultimately achieving a high level of accuracy that a single synchronization process cannot reach, realizing a synergistic leap in accuracy indicators.

[0059] The improved DFT coefficient iterative estimation algorithm proposed in this invention breaks through the frequency resolution limit of FFT. Instead of using the conventional method of calculating DFT at a single frequency point, this invention calculates two improved DFT coefficients simultaneously on both sides of the current estimated frequency in each iteration step, and uses their amplitude ratio to accurately determine and correct the direction and magnitude of the frequency error. This method effectively overcomes the theoretical limitations of the Nyquist sampling theorem and DFT frequency resolution, enabling the estimation of the true frequency of the signal with a much higher accuracy than FFT, achieving super-resolution frequency estimation, which is the core foundation for the high accuracy of this scheme. Compared to directly performing a high-resolution search across the entire frequency band, this iterative algorithm only requires a small number of iterations (e.g., M=9 times) near the initial estimate to quickly converge to the true frequency, achieving an excellent balance between accuracy and computational complexity, and possessing the advantages of high computational efficiency and fast convergence.

[0060] The adaptive threshold and half-window peak comparison joint detection mechanism designed in this invention improves the robustness and accuracy of initial localization. It combines an adaptive energy threshold with peak comparison under half-window sliding to jointly complete signal detection and initial localization. The adaptive threshold replaces the fixed threshold. The adaptive threshold can dynamically adjust according to the energy of the received signal (related to channel noise and fading), effectively reducing the false alarm probability in low signal-to-noise ratio environments and maintaining a high detection probability in high-noise environments, significantly improving robustness in complex wireless channels and exhibiting strong environmental adaptability.

[0061] Half-window sliding and peak comparison are used to accurately locate the starting window after a signal is detected. Half-window sliding ensures that the chirp signal at any location can be covered by a complete window, while the peak comparison mechanism can intelligently determine the starting moment when the signal energy is strongest, laying a good foundation for subsequent high-precision correction and avoiding correction failure due to excessive initial deviation, thus resulting in high initial positioning accuracy.

[0062] In summary, this invention breaks through the theoretical accuracy limit through an improved DFT algorithm, solves the problem of systematic errors through a collaborative correction architecture, and ensures the reliability of initial conditions through adaptive joint detection. Attached Figure Description

[0063] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0064] Figure 1 This is the method used in existing technologies to obtain Chirp signals. and Basic block diagram;

[0065] Figure 2 This is a flowchart of a Chirp signal acquisition method based on DFT frequency estimation correction according to the present invention;

[0066] Figure 3 This is a comparison chart of the root mean square error curves for timed estimation. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0068] The expression for the Chirp signal in the prior art is as follows: ,in, The transmitted Chirp signal, where j is the imaginary unit. For random phase, For mask function, For frequency modulation, t is the time variable, and T is the duration of the chirp signal. The chirp signal parameters can be designed according to actual needs.

[0069] The receiver uses the chirp signal to obtain the frequency offset signal. and time offset signal The basic block diagram is as follows Figure 1 As shown, firstly, correlation is performed using the local up-sweep and down-sweep signals and the received signal, respectively; then, a fast Fourier transform is performed on the correlated signal to solve for the peak frequency; finally, timing synchronization is completed based on the two peak frequencies, and the carrier frequency difference is solved.

[0070] Specifically, the up-sweep frequency signal is: Assuming the received signal is y(t), multiply the received chirp signal by the up-sweep signal. Then, perform an FFT calculation to find the frequency value corresponding to the highest peak. The downsweep frequency signal is: Multiply the received chirp signal by the down-sweep frequency signal. Then, perform an FFT calculation to find the frequency value corresponding to the highest peak. The frequency offset and time offset can be obtained as follows: , Within each FFT calculation cycle, there will always be a maximum frequency, but it cannot be determined whether a chirp signal has been sent, nor the correlation between the local chirp signal and the received chirp signal. Therefore, most existing algorithms set a threshold value based on the reception conditions. When the peak frequency exceeds this threshold, it is considered that a chirp signal has been received, and then the output is... and .

[0071] The existing technology has the problems mentioned in the background. Based on this, Embodiment 1 provides a chirp signal acquisition method based on DFT frequency estimation correction, including the following steps:

[0072] S1. By searching on the received data stream, the presence of the Chirp signal is detected and the starting position of the Chirp signal is initially located;

[0073] S2. Based on the initial position of the preliminary positioning, obtain the first frequency estimate and the second frequency estimate, and use an iterative frequency estimation algorithm to correct the first frequency estimate and the second frequency estimate respectively to obtain the first-level corrected frequency value.

[0074] S3. Calculate the first time offset based on the first-level corrected frequency value, and update the starting position using the first time offset; based on the updated starting position, obtain a new first frequency estimate and a new second frequency estimate again, and use the iterative frequency estimation algorithm to correct the new first frequency estimate and the new second frequency estimate respectively to obtain the second-level corrected frequency value.

[0075] S4. Calculate and output the final frequency offset and precise timing position based on the second-level correction frequency value.

[0076] Next, we will explain the implementation process of each step in detail.

[0077] In some embodiments, the specific process of S1 is as follows:

[0078] S1.1 Perform a sliding search on the received data stream with a predetermined window length and a sliding step; the sliding step is half of the predetermined window length, and the predetermined window length is L*h, where L is the length of the Chirp signal and h is the sampling factor.

[0079] S1.2 For each sliding window, calculate the up-and-down frequency sweep correlation results of the received signal within that sliding window, and obtain the first peak frequency after up-sweep (as described above). ), first peak intensity and initial time offset.

[0080] S1.3 Calculate the signal energy of the received signal within the sliding window.

[0081] S1.4 When the first peak intensity is not less than the adaptive threshold determined by the signal energy, it is determined that a Chirp signal has been detected.

[0082] S1.5. By comparing the peak intensities of adjacent sliding windows that satisfy the adaptive threshold, the window head position corresponding to the sliding window with the larger peak intensity is determined as the candidate position ID, and the initial time offset corresponding to this window is recorded. .

[0083] S1.6, According to the formula The candidate positions are corrected to obtain the initial positioning starting position; where Ts is the sampling period.

[0084] Taking a 187-symbol Chirp sequence as an example, the symbol rate is 16 ksps, and 4x sampling is used.

[0085] The window length is set to L*h=187*4=748 sampling points, and the sliding step is half the window length, 374 sampling points.

[0086] As attached Figure 2 As shown, the signal energy e(i) and the peak intensity after up-sweep frequency correlation are calculated for each sliding window. , And the time deviation at this moment, .

[0087] When satisfied When a candidate Chirp signal is detected, This is a threshold coefficient determined through simulation, for example, 0.6.

[0088] Continue sliding the half-window, compare the peak intensities of adjacent windows, record the window head position of the window with the greater intensity as the candidate position ID, and record its initial time offset. .

[0089] According to the formula Calculate the initial position of the preliminary positioning.

[0090] In other words, when hour( (As obtained through simulation), continue sliding the half-window forward and continue performing up and down frequency sweeps, recording the peak value at this point. Calculations yielded .

[0091] if Then, record the window head position corresponding to window i+1 as the chirp starting candidate position id; record the time offset at this time. Otherwise, record the window header position corresponding to the i-window as the chirp starting candidate position ID; record the time offset at this time. .

[0092] Therefore, the actual candidate positions are , T is the symbol period, proceed to the next step.

[0093] when At that point, continue sliding forward half a window to detect the next window.

[0094] In some embodiments, the iterative frequency estimation algorithm designed in steps S2 and S3 is a frequency estimation algorithm based on the improved Discrete Fourier Transform (DFT).

[0095] The frequency estimation algorithm based on the improved DFT is used to perform M iterations on the frequency estimate to be corrected, and the following steps are followed:

[0096] A1. Use the frequency estimate to be corrected as the initial value for iteration. ;

[0097] For m = 0, 1, 2, ..., M-1, perform the m-th iteration sequentially, where M is a positive integer greater than 1. Each iteration includes:

[0098] a) Calculate the two improved DFT coefficients , :

[0099] ;

[0100] ;

[0101] n represents the sequence number of the sampling point. This represents the sampling sequence of the received data stream. This represents the frequency estimate in the m-th iteration. Where N is the sampling period and N is the number of DFT points;

[0102] b) Calculate the ratio , ;

[0103] c) Update frequency estimate: ,in, The sampling frequency;

[0104] A2. After M iterations, the output is This is the corrected frequency value.

[0105] Take the execution of the first-level synchronization operation as an example.

[0106] The first frequency estimate was obtained using the DFT-based frequency estimation algorithm. Second frequency estimate Make corrections.

[0107] 1) From Start retrieving data, retrieve a window of data. First, correlate it with the up-sweep frequency signal to calculate the result. ,remember .

[0108] 2) Calculate the improved DFT coefficients ,

[0109] ;

[0110] ;

[0111] 3) Calculation :

[0112] ;

[0113] 4) Calculate the correction frequency:

[0114] ;

[0115] 5) m = 0, 1, 2, ..., M-1, where M is the iteration number. Repeat steps 2) to 4) recursively. Let the final frequency value be... .

[0116] Similarly, by performing frequency correlation calculations, we obtain... ,remember Repeat steps 2) to 5) of the recursive operation above to obtain the final result. .

[0117] The above refers to the first-level corrected frequency value obtained by performing the first-level synchronization operation, including the first frequency value. Second frequency value .

[0118] Next, the second-level synchronization operation is performed.

[0119] The specific process of calculating the first time offset based on the first-level correction frequency value and updating the starting position using the first time offset is as follows:

[0120] S3.1. Calculate the first time offset based on the first-level corrected frequency value: ,in, For the first time deviation, , These are the first and second frequency values ​​that constitute the first-level correction frequency value, respectively.

[0121] S3.2, According to the formula Update the starting position; where u is the chirp signal modulation frequency and Ts is the sampling period. This is the floor function.

[0122] Then, based on the new starting position, perform steps 1) to 5) above to obtain the second-level corrected frequency value, including the first frequency value. Second frequency value .

[0123] The final frequency offset and precise timing position are calculated based on the final obtained first and second frequency values. In this embodiment, the specific formula for calculating the final frequency offset and precise timing position is as follows:

[0124] ;

[0125] ;

[0126] ;

[0127] in, , These are the final frequency offset and precise timing position calculated based on the second-level corrected frequency value, respectively. , These are the first and second frequency values ​​that constitute the second-level correction frequency value, respectively, and u is the modulation frequency of the Chirp signal.

[0128] Example: Take a chirp sequence with 187 symbols as an example, the symbol rate is f = 16 ksps.

[0129] ;

[0130] in, , T is the symbol duration, which is 1 / 16ksps = 0.0625us.

[0131] With 4x sampling, M=9 iterations in the DFT-base algorithm, and a frequency offset of f / 5, simulation was performed for 100,000 frames. The time root mean square error and frequency root mean square error comparison curves between the conventional and modified algorithms are shown below. Figure 3 .

[0132] The difference between Example 2 and Example 1 is that, between S3 and S4, a third or more levels of synchronization operation are also included. Each level updates the starting position and performs frequency estimation and correction based on the timing result of the previous level.

[0133] The specific formulas for calculating the final frequency offset and precise timing position are then derived from the first and second frequency values ​​obtained in the final iteration.

[0134] Example 3 discloses a Chirp signal acquisition system based on DFT frequency estimation correction, used to execute the Chirp signal acquisition method based on DFT frequency estimation correction described in Example 1, including:

[0135] The signal detection and preliminary positioning module is used to detect the presence of the Chirp signal and preliminarily locate the starting position of the Chirp signal by searching on the received data stream.

[0136] The first-level synchronization module is used to obtain a first frequency estimate and a second frequency estimate based on the initial positioning, and to use an iterative frequency estimation algorithm to correct the first frequency estimate and the second frequency estimate respectively to obtain a first-level corrected frequency value.

[0137] The second-level synchronization module is used to calculate the first time offset based on the first-level corrected frequency value, and update the starting position using the first time offset; based on the updated starting position, a new first frequency estimate and a new second frequency estimate are obtained again, and the iterative frequency estimation algorithm is used to correct the new first frequency estimate and the new second frequency estimate respectively to obtain the second-level corrected frequency value.

[0138] The synchronization result output module is used to calculate and output the final frequency offset and precise timing position based on the second-level correction frequency value.

[0139] Example 4 adds a Q-level synchronization module based on Example 3. It is set after the second-level synchronization module and is used to update the starting position and perform frequency estimation and correction based on the timing results of the previous level. Q is a positive integer greater than 2.

[0140] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A Chirp signal acquisition method based on DFT frequency estimation correction, characterized in that, The method comprises the following steps: detecting the existence of a Chirp signal and preliminarily locating a starting position of the Chirp signal by searching on a received data stream; The specific process is: performing a sliding search on the received data stream with a predetermined window length and a sliding step; for each sliding window, calculating the up-and-down sweep frequency correlation result of the received signal in the sliding window, and obtaining a first peak frequency after up-sweeping, a first peak intensity, and a preliminary time offset; calculating the signal energy of the received signal in the sliding window; when the first peak intensity is not less than an adaptive threshold determined by the signal energy, it is determined that a Chirp signal is detected; By comparing the peak intensities of the adjacent sliding windows satisfying the adaptive threshold, the window head position corresponding to the sliding window with greater peak intensity is determined as the candidate position id, and the preliminary time offset corresponding to the window is recorded ; According to the formula The candidate position is corrected to obtain a starting position of the preliminary positioning; wherein Ts is a sampling period, and fix(·) is an integer function. based on the preliminarily located starting position, obtaining a first frequency estimation value and a second frequency estimation value, and correcting the first frequency estimation value and the second frequency estimation value respectively by using an iterative frequency estimation algorithm to obtain a first-level corrected frequency value; calculating a first time offset according to the first-level corrected frequency value, and updating the starting position by using the first time offset; based on the updated starting position, obtaining a new first frequency estimation value and a new second frequency estimation value, and correcting the new first frequency estimation value and the new second frequency estimation value respectively by using the iterative frequency estimation algorithm to obtain a second-level corrected frequency value; calculating a first time offset according to the first-level corrected frequency value, and updating the starting position by using the first time offset; based on the updated starting position, obtaining a new first frequency estimation value and a new second frequency estimation value, and correcting the new first frequency estimation value and the new second frequency estimation value respectively by using the iterative frequency estimation algorithm to obtain a second-level corrected frequency value; calculating the first time offset according to the first level correction frequency value : wherein, , are respectively a first frequency value and a second frequency value constituting the first level correction frequency value. According to the formula updating the starting position; The specific process of calculating a first time offset according to the first-level corrected frequency value and updating the starting position by using the first time offset is: wherein u is the frequency modulation rate of the Chirp signal, and Ts is the sampling period; 2. The Chirp signal acquisition method based on DFT frequency estimation correction according to claim 1, characterized in that, according to the second-level corrected frequency value, calculating and outputting a final frequency offset and a precise timing position.

3. The Chirp signal acquisition method based on DFT frequency estimation correction according to claim 1, characterized in that, The sliding step is half of the predetermined window length, and the predetermined window length is L*h, wherein L is the length of the Chirp signal, and h is the sampling multiple.

4. The Chirp signal acquisition method based on DFT frequency estimation correction according to claim 3, characterized in that, The iterative frequency estimation algorithm is a frequency estimation algorithm based on an improved discrete Fourier transform (DFT). using the corrected frequency estimate as an iteration initial value ; The frequency estimation algorithm based on the improved discrete Fourier transform (DFT) performs M iterations on the frequency estimation value to be corrected, and performs the following steps: a) calculating two modified DFT coefficients 、 : ； ； n denotes the serial number of the sampling point, denotes the sequence of the sampling points of the received data stream, denotes the frequency estimate of the mth iteration, is the sampling period, and N is the number of DFT points. b) calculating a ratio , ; c) updating the frequency estimate value: wherein, is the sampling frequency; After iterating M times, the output is That is, the corrected frequency value.

5. The Chirp signal acquisition method based on DFT frequency estimation correction according to claim 1, characterized in that, for m=0, 1, 2, …, M-1, sequentially perform the mth iteration, and each iteration includes: ； ； ； wherein, , are a final frequency deviation and a final time position, respectively, calculated from the second level corrected frequency value; , are a first frequency value and a second frequency value, respectively, constituting the second level corrected frequency value, and u is the frequency modulation of the Chirp signal.

6. The Chirp signal acquisition method based on DFT frequency estimation correction according to claim 1, characterized in that, The specific formula for calculating the final frequency offset and the precise timing position is:

7. A Chirp signal acquisition system based on DFT frequency estimation refinement for performing a Chirp signal acquisition method according to any one of claims 1 to 6, characterized in that, It also includes performing a third-level or more-level synchronization operation, and each level updates the starting position based on the timing result of the previous level and performs frequency estimation and correction. It comprises: a signal detection and preliminary location module, configured to detect the existence of a Chirp signal and preliminarily locate a starting position of the Chirp signal by searching on a received data stream; The specific process is: performing a sliding search on the received data stream with a predetermined window length and a sliding step; for each sliding window, calculating the up-and-down sweep frequency correlation result of the received signal in the sliding window, and obtaining a first peak frequency after up-sweeping, a first peak intensity, and a preliminary time offset; calculating the signal energy of the received signal in the sliding window; when the first peak intensity is not less than an adaptive threshold determined by the signal energy, it is determined that a Chirp signal is detected; By comparing the peak intensities of the adjacent sliding windows satisfying the adaptive threshold, the window head position corresponding to the sliding window with greater peak intensity is determined as the candidate position id, and the preliminary time offset corresponding to the window is recorded ; According to the formula The candidate position is corrected to obtain a starting position of the preliminary positioning; wherein Ts is a sampling period; The first-stage synchronization module is configured to obtain a first frequency estimation value and a second frequency estimation value based on the initial position of the preliminary positioning, and correct the first frequency estimation value and the second frequency estimation value respectively by using an iterative frequency estimation algorithm to obtain a first-stage corrected frequency value; The second-stage synchronization module is configured to calculate a first time offset based on the first-stage corrected frequency value, update the initial position by using the first time offset, obtain a new first frequency estimation value and a new second frequency estimation value based on the updated initial position, and correct the new first frequency estimation value and the new second frequency estimation value respectively by using the iterative frequency estimation algorithm to obtain a second-stage corrected frequency value; The specific process of calculating a first time offset based on the first-stage corrected frequency value and updating the initial position by using the first time offset is as follows: calculating the first time offset according to the first level correction frequency value : wherein, , are a first frequency value and a second frequency value, respectively, constituting the first level correction frequency value. According to the formula updating the starting position; wherein u is a frequency modulation rate of the Chirp signal, Ts is a sampling period, and fix(·) is a rounding function; The synchronization result output module is configured to calculate and output a final frequency offset and an accurate timing position based on the second-stage corrected frequency value.

8. A Chirp signal acquisition system based on DFT frequency estimation correction according to claim 7, characterized in that, Further comprising: The Qth-stage synchronization module is configured to update the initial position based on a timing result of a previous stage, and perform frequency estimation and correction, wherein Q is a positive integer greater than 2.

Images