Unmanned aerial vehicle orthogonal frequency division multiplexing transmission time-frequency synchronization method and system

Abstract

The invention discloses a time-frequency synchronization method and system for orthogonal frequency division multiplexing transmission of an unmanned aerial vehicle, and belongs to the field of communication. The method comprises the steps of obtaining a baseband receiving signal transmitted by an unmanned aerial vehicle end; the baseband receiving signal is generated according to a superposed signal generated by a pseudo-random noise sequence and a preset power distribution factor; generating a synchronization function according to the baseband receiving signal and a preset local time domain training sequence; performing coarse estimation on the phase difference between the cyclic prefix and the symbol main body according to the symbol starting point at the moment when the peak value of the synchronization function exceeds the preset threshold value to obtain coarse estimation frequency offset, performing coarse compensation correction on the baseband receiving signal, and performing fine estimation on the subcarrier phase rotation amount by the baseband receiving signal after coarse compensation to obtain fine estimation frequency offset; and performing frequency offset correction on the baseband receiving signal according to the total frequency offset obtained by adding the coarse estimation frequency offset and the fine estimation frequency offset. According to the invention, the problem that broadband offset coverage and high synchronization precision cannot be satisfied simultaneously in the prior art is solved.

Description

A method and system for time-frequency synchronization of orthogonal frequency division multiplexing transmission for unmanned aerial vehicles (UAVs) Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a time-frequency synchronization method and system for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles (UAVs). Background Technology

[0002] The widespread application of drones in disaster monitoring and aerial photography places high demands on downlink communication performance, requiring high data rates, low latency, and resistance to multipath interference. Orthogonal Frequency Division Multiplexing (OFDM) technology plays a crucial role in drone wireless transmission due to its high spectral efficiency and multipath fading resistance; however, this technology is highly dependent on time-frequency synchronization, with carrier frequency offset estimation being a critical step.

[0003] The Doppler frequency shift generated by the high-speed maneuvering of UAVs, coupled with the multipath effect in complex electromagnetic environments, presents a contradiction between coverage and estimation accuracy in traditional frequency offset estimation algorithms. Taking the cyclic prefix synchronization algorithm as an example, frequency offset estimation is achieved by leveraging the correlation between the symbol tail and the cyclic prefix. However, to balance coverage, the training sequence and the cyclic prefix need to be superimposed, which disrupts the integrity of the OFDM symbol's temporal structure and produces a cumulative synchronization offset effect in multipath environments. If the training sequence is shortened to suppress the offset and improve accuracy, the estimation range will be compressed, making it difficult to cope with wide frequency offsets and failing to meet the synchronization reliability requirements of UAV scenarios.

[0004] Therefore, existing technologies, on the one hand, expand the estimation range to accommodate the wide frequency offset caused by high-speed motion, which leads to a decrease in subcarrier spacing measurement resolution and inaccuracy; on the other hand, optimizing the algorithm's fineness to improve accuracy weakens the ability to capture wide frequency offsets and makes it impossible to track rapid frequency offset jumps. This contradiction leads to synchronization interruption, exacerbating inter-symbol interference (ISI) and inter-carrier interference (ICI), and degrading synchronization performance. Summary of the Invention

[0005] This invention provides a method and system for time-frequency synchronization of UAV orthogonal frequency division multiplexing transmission, which can effectively solve the problem that existing technologies cannot simultaneously meet the requirements of wideband offset coverage and improved synchronization accuracy.

[0006] An embodiment of the present invention provides a time-frequency synchronization method for orthogonal frequency division multiplexing (OFDM) transmission of unmanned aerial vehicles (UAVs), applicable to the receiving end of an OFDM transmission time-frequency synchronization system for UAVs; the UAV OFDM transmission time-frequency synchronization system further includes a UAV terminal; the UAV OFDM transmission time-frequency synchronization method includes: acquiring a baseband received signal transmitted by the UAV terminal; wherein, the baseband received signal is generated by analog-to-digital conversion of a superimposed signal generated by the UAV terminal; the superimposed signal is generated based on a pseudo-random noise sequence and a preset power allocation factor; and a sliding cross-correlation operation is performed based on the baseband received signal and a preset local time-domain training sequence. The system calculates and generates a synchronization function. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. Based on the symbol start point, the phase difference between the cyclic prefix and the symbol body in the baseband received signal is coarsely estimated to obtain a coarsely estimated frequency offset. Based on the coarsely estimated frequency offset, the baseband received signal is coarsely compensated to obtain a coarsely compensated baseband received signal. Based on the coarsely compensated baseband received signal, the subcarrier phase rotation is finely estimated to obtain a finely estimated frequency offset. The coarsely estimated frequency offset and the finely estimated frequency offset are added to obtain a total frequency offset. Based on the total frequency offset, the baseband received signal is frequency offset corrected to obtain a corrected signal.

[0007] Further, the generation of the superimposed signal includes: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol according to a preset power allocation factor to generate the superimposed signal.

[0008] Further, a sliding cross-correlation operation is performed based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. This includes: moving the preset local time-domain training sequence on the baseband received signal based on a preset sliding window, calculating the cross-correlation value at each moving position; assembling the cross-correlation values ​​of all moving positions into a sequence and normalizing it to generate the synchronization function; extracting the amplitude of all local peaks based on the synchronization function, and taking the moment corresponding to the local peak whose amplitude exceeds the preset threshold as the symbol start point.

[0009] Further, based on the symbol start point, a coarse estimate of the phase difference between the cyclic prefix and the symbol body in the baseband received signal is obtained to obtain a coarse frequency offset, including: determining the time domain positions of the cyclic prefix and the symbol body in the baseband received signal based on the symbol start point; wherein the cyclic prefix is ​​located in the interval after the symbol start point, and the symbol body is located in the time domain length interval of the symbol body after the cyclic prefix; extracting the sampled data of the cyclic prefix and the sampled data of the symbol body as data to be compared, calculating the cross-correlation phase difference between the two sets of data to be compared; and calculating the coarse frequency offset based on the cross-correlation phase difference and the time domain length of the symbol body.

[0010] Furthermore, the subcarrier phase rotation is precisely estimated based on the coarsely compensated baseband received signal to obtain the precisely estimated frequency offset. This includes: truncating the coarsely compensated baseband received signal and retaining the main body of the symbol after the symbol start point; performing a Fast Fourier Transform on the truncated main body of the symbol to convert it into a frequency domain subcarrier signal; calculating the phase rotation of adjacent subcarriers based on the phase values ​​corresponding to the frequency domain subcarrier signals; and linearly fitting the phase rotation to calculate the precisely estimated frequency offset.

[0011] Furthermore, the quadrature amplitude modulation mapping is a 16QAM modulation mapping.

[0012] Furthermore, the UAV orthogonal frequency division multiplexing transmission time-frequency synchronization method is applicable to the UAV end of the UAV orthogonal frequency division multiplexing transmission time-frequency synchronization system; the UAV end is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; after performing analog-to-digital conversion on the superimposed signal, a baseband receiving signal is generated, and the baseband receiving signal is transmitted to the receiving end.

[0013] As an improvement to the above solution, another embodiment of the present invention provides an orthogonal frequency division multiplexing (OFDM) transmission time-frequency synchronization system for unmanned aerial vehicles (UAVs), comprising: a receiver and a UAV; the UAV is configured to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; generate a baseband received signal after analog-to-digital conversion based on the superimposed signal; and transmit the baseband received signal to the receiver; the receiver is configured to acquire the baseband received signal transmitted by the UAV; perform sliding cross-correlation calculation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function, and synchronize the time-frequency signal. When the peak value of the function exceeds a preset threshold, the corresponding time is taken as the symbol start point. Based on the symbol start point, the phase difference between the cyclic prefix and the symbol body in the baseband received signal is coarsely estimated to obtain a coarse frequency offset. Based on the coarse frequency offset, the baseband received signal is coarsely compensated to obtain a coarsely compensated baseband received signal. Based on the coarsely compensated baseband received signal, the subcarrier phase rotation is finely estimated to obtain a fine frequency offset. The coarse frequency offset and the fine frequency offset are added to obtain the total frequency offset. Based on the total frequency offset, the baseband received signal is frequency offset corrected to obtain the corrected signal.

[0014] Furthermore, the UAV terminal is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor, including: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol based on the preset power allocation factor to generate a superimposed signal.

[0015] Further, the receiving end is used to acquire the baseband received signal transmitted by the UAV; perform sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function, and take the time corresponding to when the peak value of the synchronization function exceeds a preset threshold as the symbol start point, including: moving the preset local time-domain training sequence on the baseband received signal based on a preset sliding window, calculating the cross-correlation value at each moving position; assembling the cross-correlation values ​​of all moving positions into a sequence and performing normalization processing to generate the synchronization function; extracting the amplitude of all local peaks based on the synchronization function, and taking the time corresponding to the local peak whose amplitude exceeds a preset threshold as the symbol start point.

[0016] By implementing this invention, at least the following beneficial effects are achieved: This invention provides a method and system for time-frequency synchronization of UAV orthogonal frequency division multiplexing (OFDM) transmission. The method is applicable to the receiving end of an OFDM transmission time-frequency synchronization system for UAVs. The baseband received signal originates from the superimposed signal generated by the UAV end based on a pseudo-random noise sequence and a power allocation factor. The strong autocorrelation of the pseudo-random noise sequence and the optimized configuration of signal energy by the power allocation factor significantly enhance the anti-interference capability of the baseband signal. Furthermore, by generating a synchronization function through the sliding cross-correlation operation between the baseband signal and the local time-domain training sequence, the symbol start point corresponding to the peak can be accurately located, avoiding synchronization deviation caused by misjudgment of the start point. This provides a reliable time-domain reference for subsequent frequency offset estimation and effectively reduces synchronization interruptions caused by frequency offset capture failure or start point deviation. In the coarse estimation stage, the frequency offset is roughly estimated by calculating the phase difference between the cyclic prefix and the symbol body in the baseband signal. This covers the wide frequency offset range caused by the high-speed movement of the UAV without sacrificing resolution, avoiding the problem of weakening the wide frequency offset acquisition capability and failing to track rapid frequency offset jumps due to optimizing accuracy. After coarse compensation, the subcarrier phase rotation is precisely estimated based on the signal that has eliminated most of the frequency offset interference, focusing on improving measurement resolution without compromising accuracy for the extended range, thus breaking the core contradiction of reducing accuracy for the extended range. Through total frequency offset correction, the frequency offset effect caused by the high-speed movement of the UAV can be accurately offset, avoiding subcarrier phase shift and inter-symbol energy crosstalk caused by uncompensated frequency offset, and significantly reducing inter-symbol interference (ISI) and inter-carrier interference (ICI). At the same time, the anti-interference synchronization signal design and accurate symbol start point positioning further ensure the continuity and reliability of the synchronization process. Ultimately, the UAV orthogonal frequency division multiplexing transmission system can maintain excellent time and frequency synchronization performance in high-speed movement scenarios, meet the requirements for stable transmission, and improve the accuracy of time and frequency synchronization while covering a wide frequency offset. Attached Figure Description

[0017] Figure 1 is a flowchart illustrating a time-frequency synchronization method for orthogonal frequency division multiplexing (OFDM) transmission of an unmanned aerial vehicle (UAV) according to an embodiment of the present invention; Figure 2 is a structural diagram illustrating a time-frequency synchronization system for OFDM transmission of an UAV according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Referring to Figure 1, to address the problem that existing technologies cannot simultaneously satisfy wideband offset coverage and improved synchronization accuracy, an embodiment of the present invention provides a flowchart of a UAV orthogonal frequency division multiplexing (OFDM) transmission time-frequency synchronization method. This method is applicable to the receiving end of an OFDM transmission time-frequency synchronization system. The OFDM transmission time-frequency synchronization system also includes a UAV terminal. Specifically, the UAV terminal of the OFDM transmission time-frequency synchronization system acts as the transmitting end, responsible for generating an OFDM signal containing synchronization information. The receiving end, such as a ground station or relay equipment, is responsible for synchronization processing and demodulation. The core function is to achieve high-speed data transmission and time-frequency synchronization through OFDM technology. The receiving end, as the execution entity, needs to regenerate a training sequence consistent with that of the UAV terminal to complete signal synchronization and frequency offset correction.

[0020] The UAV orthogonal frequency division multiplexing (OFDM) transmission time-frequency synchronization method includes: S1, acquiring the baseband received signal transmitted by the UAV; wherein, the baseband received signal is generated by analog-to-digital conversion of the superimposed signal generated by the UAV; the superimposed signal is generated based on a pseudo-random noise sequence and a preset power allocation factor; specifically, the baseband received signal refers to the baseband signal finally acquired by the receiver, which is obtained by analog-to-digital conversion (ADC) of the superimposed signal generated by the UAV, and in actual transmission, it also needs to be transmitted through a wireless channel. The superimposed signal is a time-domain training sequence generated by modulation and transformation of the pseudo-random noise sequence, linearly superimposed with the cyclic prefix of the OFDM symbol according to the preset power allocation factor, and is the carrier of synchronization information. The pseudo-random noise sequence refers to a deterministic sequence with approximately random statistical characteristics, such as the Gold sequence, used to generate the training sequence to ensure the sharp autocorrelation of the synchronization function. The preset power allocation factor refers to a parameter that controls the superposition ratio of the time-domain training sequence and the cyclic prefix, balancing synchronization accuracy and the multipath resistance capability of the cyclic prefix.

[0021] Preferably, the generation of the superimposed signal includes: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol according to a preset power allocation factor to generate the superimposed signal.

[0022] Specifically, the pseudo-random sequence is the original sequence used to generate the pseudo-random noise sequence, preferably a Gold sequence, which exhibits sharp autocorrelation and low cross-correlation. Orthogonal amplitude modulation mapping represents the modulation method that converts the binary sequence into complex symbols, which can be 16QAM. Information is transmitted by simultaneously adjusting the amplitude and phase of the carrier; each 16QAM symbol corresponds to 4 binary bits. The complex symbol is the output of the orthogonal amplitude modulation, composed of real and imaginary parts, corresponding to the amplitude and phase information of the OFDM subcarrier. Matrix transformation is an orthogonal transformation that maps the complex symbols to the frequency domain, which can be a Discrete Fourier Transform (DFT), ensuring that the mapped frequency domain sequence is orthogonally compatible with the OFDM subcarriers. The frequency domain sequence is the output of the matrix transformation, with a length consistent with the number of OFDM subcarriers, and each element corresponding to the frequency domain value of one subcarrier. The cyclic prefix sampling point of the OFDM symbol is the smallest unit of the cyclic prefix; the time-domain training sequence is superimposed with these sampling points to embed synchronization information.

[0023] Schematic, the quadrature amplitude modulation mapping is a 16QAM modulation mapping.

[0024] In a preferred embodiment of the present invention, a Gold sequence (pseudo-random sequence) is used as the basis for generating the pseudo-random noise (PN) sequence. The Gold sequence is generated by a pass-through operation of two preferred m-sequences, and its length is N / 2, where N is the OFDM symbol length (e.g., N=2048). The Gold sequence has good autocorrelation and cross-correlation, which can effectively reduce the influence of multipath interference. The generated PN sequence... The symbols are mapped to complex symbols using 16QAM modulation. The resulting graph after 16QAM modulation contains 16 points, with each symbol carrying 4 bits of information. The modulated complex symbols are: ,in and These are the real and imaginary parts, respectively, and their values ​​range from 1 to 2. j is the imaginary unit; The modulated complex symbol; the modulated complex symbol Frequency domain sequences are generated by mapping to the frequency domain using matrix transformation rules. : ,in, The normalization coefficient is used to ensure that the power of the time-domain training sequence matches the data symbols; k is the index of the frequency-domain subcarrier, used to identify the frequency-domain sequence. Each element corresponds to a frequency domain position (subcarrier index). This mapping method makes the time-domain training sequence sparse in the frequency domain, placing non-zero values ​​only on odd-numbered subcarriers, thereby avoiding frequency domain overlap with data symbols and reducing the frequency spectrum usage of the time-domain training sequence. The real part of the generated frequency domain sequence is even-symmetric, and the imaginary part is odd-symmetric.

[0025] The frequency domain sequence is subjected to a 64-point inverse fast Fourier transform (IFFT) to generate a time-domain training sequence. : Where N=64, n=0,1,…,63. The generated time-domain training sequence is self-antisymmetric, a property that gives it good autocorrelation in the time domain while avoiding the signal integrity degradation caused by clipping operations. Through optimized design, the peak-to-average power ratio (PAPR) of the time-domain training sequence is below 6dB, reducing out-of-band radiated power by 40% compared to traditional CAZAC sequences. The autocorrelation function of the time-domain training sequence peaks at zero delay and decays rapidly at other delays, effectively resisting multipath interference. Through sparse frequency domain mapping and power normalization design, the spectral resource occupancy of the training sequence is only 0.8%, significantly lower than the 15% of traditional methods. The time-domain self-antisymmetry and frequency-domain sparsity enable the training sequence to maintain high-precision synchronization performance under multipath channels. Using a 64-point IFFT transform, the computational cost is reduced by 37.5% compared to traditional methods, making it suitable for real-time implementation on embedded platforms.

[0026] The time-domain training sequence p(n) is embedded into the cyclic prefix (CP) segment in an optimized manner, minimizing the occupation of spectrum resources while preserving the correlation between the CP and the data portion. A preset power allocation factor β is used to control the power proportion of the training sequence p(n) in the superimposed signal, and its value ranges from β∈[0.05,0.1]. For example, simulation experiments have shown that when β=0.08, i.e., the time-domain training sequence accounts for 8% of the total power of the superimposed signal, the detection probability of the time-domain training sequence and the SNR of the data symbols are optimally balanced. The lower limit constraint of 0.05 ensures that the power of the time-domain training sequence is large enough to be accurately detected by the receiver, while the upper limit constraint of 0.1 avoids the time-domain training sequence power being too large, which would lead to a decrease in the signal-to-noise ratio (SNR) of the data symbols.

[0027] The time-domain training sequence p(n) is linearly superimposed onto the first 64 sampling points of the cyclic prefix according to the power allocation factor β to generate the superimposed signal. : Where x(n) represents the data symbols, and N is the OFDM symbol length (e.g., N=2048). The total power of the superimposed signals is... : ,in For data symbol power, Power for the time-domain training sequence. Ensured by a power allocation factor. , The cross-correlation coefficient between the CP segment and the data tail in the superimposed signal remains above 0.92, ensuring the accuracy of time-frequency synchronization. The time-domain training sequence only occupies 8% of the CP segment's power, compressing the additional spectral overhead to below 0.8%. By embedding the time-domain training sequence into the CP segment, no additional pilot symbols are needed, saving 15% of spectral resources and increasing effective data throughput by 12.6%. The power allocation between the time-domain training sequence and data symbols is optimized, resulting in a detection probability of ≥99.3% (SNR=8 dB) for the superimposed signal in multipath channels, an improvement of 28 percentage points compared to traditional methods. The superposition process involves only simple linear operations, with a computational complexity of O(N), making it suitable for real-time implementation on embedded platforms.

[0028] Schematic, the baseband received signal r(n) can be expressed as ,in In order to transmit signals, For channel impulse response, The noise is additive white Gaussian noise (AWGN). After the received signal is converted from analog to digital (ADC), it is sampled at a sampling period Ts to obtain the discrete baseband received signal r(n).

[0029] S2. Perform a sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. Specifically, the local time-domain training sequence refers to the training sequence regenerated by the receiver based on parameters negotiated with the UAV (such as pseudo-random sequence type and modulation rules), which is consistent with the time-domain training sequence generated by the UAV. The sliding cross-correlation operation uses the local time-domain training sequence as a window, sliding it point by point on the baseband received signal to calculate the correlation value at each position, which is used to capture synchronization information. The preset threshold is a criterion used to filter the peak value of the synchronization function to avoid false peaks caused by multipath interference. The symbol start point is the moment when the peak value of the synchronization function exceeds the preset threshold, marking the beginning of an OFDM symbol and serving as the time reference for subsequent frequency offset estimation and demodulation.

[0030] Preferably, the process involves performing a sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. This includes: moving the preset local time-domain training sequence over the baseband received signal based on a preset sliding window, calculating the cross-correlation value at each moving position; assembling the cross-correlation values ​​of all moving positions into a sequence and normalizing it to generate the synchronization function; extracting the amplitude of all local peaks based on the synchronization function, and taking the moment corresponding to the local peak whose amplitude exceeds a preset threshold as the symbol start point.

[0031] Specifically, the preset sliding window is a window with the same length as the local time-domain training sequence (e.g., 32 sampling points), used to slide point-by-point across the baseband received signal to capture the position of the training sequence. The cross-correlation value is an indicator of the similarity between the local time-domain training sequence and local segments of the baseband received signal; the higher the similarity, the larger the cross-correlation value. Normalization processing involves energy normalization of the cross-correlation value to eliminate the influence of signal amplitude fluctuations on the synchronization function.

[0032] In a preferred embodiment of the present invention, the cross-correlation function between the baseband received signal r(n) and the local training sequence p(n) is calculated using a sliding correlator, and the conjugate function of the local time-domain training sequence is used for the calculation. To eliminate the influence of received signal power fluctuations, the correlation function is normalized to obtain the synchronization function. : The normalized synchronization function has a value range of [0,1], and its peak position corresponds to the symbol start point. d is the delay offset of the sliding correlation. The sliding correlator calculates the correlation between the received signal and the local training sequence at different delay positions by changing the value of d, thereby traversing to find the synchronization point. L is the length of the local training sequence (time-domain training sequence). A detection threshold Tm = 0.9 is set. When the value of the synchronization function M(d) exceeds the preset threshold, the time corresponding to the local peak whose amplitude exceeds the preset threshold is taken as the symbol start point. Utilizing the self-antisymmetry of the local time-domain training sequence, the correlation peak offset caused by multipath channels is eliminated. Through normalization, the impact of noise on correlation peak detection is reduced. Through sliding correlation operation and threshold decision, the time synchronization algorithm can maintain a high detection probability even under low signal-to-noise ratio conditions. Utilizing the self-antisymmetry and normalization of the time-domain training sequence, the influence of multipath interference and noise is effectively suppressed. The computational complexity of the sliding correlation operation is O(N), making it suitable for real-time implementation on embedded platforms.

[0033] Time synchronization determines the symbol start point through the synchronization function M(d), which is a prerequisite for frequency offset compensation. Only after correctly identifying the symbol start point can the receiver avoid the influence of inter-symbol interference and inter-carrier interference, accurately sample and process the signal, and thus perform frequency offset estimation. Time synchronization effectively suppresses the influence of multipath interference and noise through autocorrelation and normalization processing of the time-domain training sequence, thereby improving the accuracy of frequency offset estimation.

[0034] S3. Based on the symbol start point, a coarse estimate of the phase difference between the cyclic prefix and the symbol body in the baseband received signal is performed to obtain a coarse frequency offset estimate. Specifically, the cyclic prefix is ​​the beginning part of the OFDM symbol, which is cyclically copied from the tail data of the symbol body and is used to absorb multipath interference and ensure subcarrier orthogonality. The symbol body is the core data part of the OFDM symbol, carrying valid service data, such as aerial images taken by UAVs. The coarse frequency offset estimate is the frequency offset value estimated based on the phase difference between the cyclic prefix and the symbol body, and is used to initially compensate for large frequency offsets.

[0035] Preferably, the coarse estimation of the phase difference between the cyclic prefix and the symbol body in the baseband received signal based on the symbol start point to obtain a coarse frequency offset includes: determining the time-domain positions of the cyclic prefix and the symbol body in the baseband received signal based on the symbol start point; wherein the cyclic prefix is ​​located in the interval after the symbol start point, and the symbol body is located in the time-domain length interval of the symbol body after the cyclic prefix; extracting the sampled data of the cyclic prefix and the sampled data of the symbol body as data to be compared, calculating the cross-correlation phase difference between the two sets of data to be compared; and calculating the coarse frequency offset based on the cross-correlation phase difference and the time-domain length of the symbol body.

[0036] Specifically, the cross-correlation phase difference is the argument (phase) of the result after cross-correlation of two sets of correlated data (cyclic prefix and symbol body). It reflects the phase difference between the two sets of data caused by frequency offset and is the core basis for rough estimation of frequency offset.

[0037] In a preferred embodiment of the present invention, frequency offset compensation aims to eliminate the Doppler frequency shift caused by the dynamic flight of the UAV and ensure the orthogonality between OFDM subcarriers. Frequency offset compensation employs a two-stage estimation mechanism, including coarse estimation and fine estimation. The goal of coarse estimation is to quickly capture a wide range of frequency offsets, covering the maximum Doppler frequency shift of the UAV (e.g., ±15.625 kHz). Due to the high-speed motion and multipath effects of the UAV, the frequency offset range may far exceed the capture capability of traditional algorithms; therefore, the design of coarse estimation is crucial. The initial frequency offset is calculated using the correlation between the first and second halves of the time-domain training sequence, i.e., the coarse frequency offset estimation. : in, The sampling period is This represents the maximum correlation value between the first and second halves of the time-domain training sequence. This represents the maximum correlation between the first and second halves of the training sequence. The first half of the time-domain training sequence typically contains the first L / 2 elements, where L is the total length of the time-domain training sequence. The second half of the time-domain training sequence contains the last L / 2 elements. The maximum correlation between the first and second halves is calculated from the first half to the second half, reflecting the similarity between them. If the time-domain training sequence has good autocorrelation, this value will reach its maximum when the frequency offset is zero. The maximum correlation between the second and first halves is calculated from the second half to the first half; however, the calculation formula differs depending on the order of calculation.

[0038] The roughly estimated frequency offset range is: when The coarse estimation range is ±15.625 kHz. This range is sufficient to cover the maximum Doppler shift generated by the UAV during high-speed flight.

[0039] S4. Perform coarse compensation correction on the baseband received signal based on the coarsely estimated frequency offset to obtain the coarsely compensated baseband received signal. Specifically, the coarse compensation correction is a process of multiplying the baseband received signal by a phase compensation factor corresponding to the coarsely estimated frequency offset to cancel out most of the frequency offset. The coarsely compensated baseband received signal is a baseband signal with a significantly reduced residual frequency offset after coarse compensation, providing high-quality input for fine frequency offset estimation.

[0040] S5. Based on the baseband received signal after coarse compensation, the subcarrier phase rotation is precisely estimated to obtain the precise frequency offset. Specifically, the subcarrier phase rotation is the phase difference between subcarriers in the frequency domain after coarse compensation due to the residual frequency offset, and it is the core basis for the precise frequency offset estimation. The precise frequency offset estimation is based on the residual frequency offset value estimated by the subcarrier phase rotation, and it has high accuracy, used to compensate for the residual error of the coarse estimation.

[0041] Preferably, the fine estimation of the subcarrier phase rotation based on the coarsely compensated baseband received signal to obtain the finely estimated frequency offset includes: truncating the coarsely compensated baseband received signal and retaining the main body of the symbol after the symbol start point; performing a fast Fourier transform on the truncated main body of the symbol to convert it into a frequency domain subcarrier signal; calculating the phase rotation of adjacent subcarriers based on the phase values ​​corresponding to the frequency domain subcarrier signals; and linearly fitting the phase rotation to calculate the finely estimated frequency offset.

[0042] Specifically, the Fast Fourier Transform (FFT) is an efficient operation that converts time-domain signals into frequency-domain signals. It is the core of OFDM demodulation, converting the main time-domain symbols into frequency-domain subcarrier signals. Linear fitting refers to performing linear regression on a series of data points (such as subcarrier phase rotation) to obtain the optimal fitting line. Its slope reflects the magnitude of the residual frequency offset. Weighted least squares fitting is preferred (assigning higher weights to subcarriers with high signal-to-noise ratios).

[0043] In a preferred embodiment of the present invention, the goal of fine estimation is to eliminate the residual frequency offset after coarse estimation, achieving an accuracy of 1% (e.g., 0.78 Hz) in the subcarrier spacing. While coarse estimation can capture a wide range of frequency offsets, its accuracy is insufficient to meet the requirements of OFDM systems; therefore, fine estimation is necessary for further refinement. The residual frequency offset is calculated based on the autocorrelation operation between the cyclic prefix (CP) and the data tail, i.e., the finely estimated frequency offset. : in The length of the cyclic prefix. This is the received signal after coarse compensation. The finely estimated frequency offset range is: when At that time, the precise estimation range is ±0.78Hz.

[0044] S6. Add the coarse frequency offset estimate and the fine frequency offset estimate to obtain the total frequency offset; and perform frequency offset correction on the baseband received signal based on the total frequency offset to obtain the corrected signal.

[0045] Specifically, the total frequency offset is the sum of the coarse and fine frequency offset estimates, reflecting the true frequency offset of the baseband received signal and serving as the basis for final correction. The corrected signal represents the signal after total frequency offset correction, where subcarrier orthogonality is restored and inter-symbol interference is eliminated.

[0046] In a preferred embodiment of the present invention, the results of the coarse estimation and the fine estimation are added together to obtain the total frequency offset. : Frequency offset correction is performed on the baseband received signal, and the corrected signal : The combination of coarse and fine estimation covers the maximum Doppler frequency shift of the UAV and achieves 1% accuracy in subcarrier spacing. Utilizing the autocorrelation of the training sequence and the autocorrelation of the cyclic prefix, it effectively suppresses the effects of multipath interference and noise. Based on the above algorithms, a time synchronization accuracy of ≥99.5% (SNR≥6dB) can be achieved in UAV channels, with a frequency offset estimation range of ±1.5 times the subcarrier spacing and an accuracy better than 0.1 subcarrier spacing. Under the same channel conditions, the bit error rate is reduced by 1-2 orders of magnitude compared to traditional algorithms.

[0047] In a preferred embodiment of the present invention, the UAV orthogonal frequency division multiplexing transmission time-frequency synchronization method is applicable to the UAV end of the UAV orthogonal frequency division multiplexing transmission time-frequency synchronization system; the UAV end is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; after performing analog-to-digital conversion on the superimposed signal, a baseband receiving signal is generated, and the baseband receiving signal is transmitted to the receiving end.

[0048] By implementing this embodiment, the baseband received signal originates from the superimposed signal generated by the UAV based on a pseudo-random noise sequence and a power allocation factor. The strong autocorrelation of the pseudo-random noise sequence and the optimized configuration of signal energy by the power allocation factor significantly enhance the anti-interference capability of the baseband signal. Furthermore, by generating a synchronization function through sliding cross-correlation between the baseband signal and the local time-domain training sequence, the symbol start point corresponding to the peak can be accurately located, avoiding synchronization deviations caused by misjudgment of the start point. This provides a reliable time-domain reference for subsequent frequency offset estimation, effectively reducing synchronization interruptions caused by frequency offset capture failure or start point deviation. In the coarse estimation stage, the frequency offset is coarsely estimated by calculating the phase difference between the cyclic prefix and the symbol body in the baseband signal. This covers the wide frequency offset range caused by the high-speed movement of the UAV without sacrificing resolution, avoiding the problem of weakening wide frequency offset capture capability and failing to track rapid frequency offset jumps due to accuracy optimization. After coarse compensation, the subcarrier phase rotation is precisely estimated based on the signal that has eliminated most frequency offset interference, focusing on improving measurement resolution without compromising accuracy for extended range, thus breaking the core contradiction of reduced accuracy with extended range. By correcting the total frequency offset, the frequency offset effect caused by the high-speed movement of the UAV can be accurately offset, avoiding subcarrier phase shift and inter-symbol energy crosstalk caused by uncompensated frequency offset, and significantly reducing inter-symbol interference (ISI) and inter-carrier interference (ICI). At the same time, the anti-interference synchronization signal design and accurate symbol start point positioning further ensure the continuity and reliability of the synchronization process. Ultimately, the UAV orthogonal frequency division multiplexing transmission system can maintain excellent time and frequency synchronization performance in high-speed movement scenarios, meet the requirements for stable transmission, and improve the accuracy of time and frequency synchronization while covering a wide frequency offset.

[0049] Referring to Figure 2, it is a schematic diagram of the structure of an orthogonal frequency division multiplexing (OFDM) transmission time-frequency synchronization system for unmanned aerial vehicles (UAVs) according to an embodiment of the present invention, including a receiver and a UAV. The UAV is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; to generate a baseband received signal after analog-to-digital conversion based on the superimposed signal; and to transmit the baseband received signal to the receiver. The receiver is used to acquire the baseband received signal transmitted by the UAV; to perform sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function; and to transmit the synchronization function... When the peak value of the number exceeds a preset threshold, the corresponding time is taken as the symbol start point; based on the symbol start point, the phase difference between the cyclic prefix and the symbol body in the baseband received signal is coarsely estimated to obtain a coarsely estimated frequency offset; based on the coarsely estimated frequency offset, the baseband received signal is coarsely compensated to obtain a coarsely compensated baseband received signal; based on the coarsely compensated baseband received signal, the subcarrier phase rotation is finely estimated to obtain a finely estimated frequency offset; the coarsely estimated frequency offset and the finely estimated frequency offset are added to obtain a total frequency offset; and based on the total frequency offset, the baseband received signal is frequency offset corrected to obtain a corrected signal.

[0050] Specifically, the UAV terminal is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor, including: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol based on the preset power allocation factor to generate a superimposed signal.

[0051] Specifically, the receiving end is used to acquire the baseband received signal transmitted by the UAV; perform sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function, and take the time corresponding to when the peak value of the synchronization function exceeds a preset threshold as the symbol start point, including: moving the preset local time-domain training sequence on the baseband received signal based on a preset sliding window, calculating the cross-correlation value at each moving position; assembling the cross-correlation values ​​of all moving positions into a sequence and performing normalization processing to generate the synchronization function; extracting the amplitude of all local peaks based on the synchronization function, and taking the time corresponding to the local peak whose amplitude exceeds a preset threshold as the symbol start point.

[0052] This invention provides a time-frequency synchronization system for UAV orthogonal frequency division multiplexing (OFDM) transmission. The system utilizes a baseband received signal source generated by the UAV from a superimposed signal based on a pseudo-random noise sequence and a power allocation factor. The strong autocorrelation of the pseudo-random noise sequence and the optimized energy configuration of the signal by the power allocation factor significantly enhance the anti-interference capability of the baseband signal. Furthermore, a synchronization function is generated through sliding cross-correlation between the baseband signal and a local time-domain training sequence. This accurately locates the symbol start point corresponding to the peak value, avoiding synchronization deviations caused by misjudgment of the start point and providing a reliable time-domain reference for subsequent frequency offset estimation. This effectively reduces synchronization interruptions caused by frequency offset capture failure or start point deviation. In the coarse estimation stage, the frequency offset is roughly estimated by calculating the phase difference between the cyclic prefix and the symbol body in the baseband signal. This covers the wide frequency offset range caused by the high-speed movement of the UAV without sacrificing resolution, avoiding the problem of weakening wide frequency offset capture capability and failing to track rapid frequency offset jumps due to accuracy optimization. After coarse compensation, the subcarrier phase rotation is precisely estimated based on the signal that has eliminated most frequency offset interference. This focuses on improving measurement resolution without compromising accuracy for extended range, thus breaking the core contradiction of reduced accuracy with extended range. By correcting the total frequency offset, the frequency offset effect caused by the high-speed movement of the UAV can be accurately offset, avoiding subcarrier phase shift and inter-symbol energy crosstalk caused by uncompensated frequency offset, and significantly reducing inter-symbol interference (ISI) and inter-carrier interference (ICI). At the same time, the anti-interference synchronization signal design and accurate symbol start point positioning further ensure the continuity and reliability of the synchronization process. Ultimately, the UAV orthogonal frequency division multiplexing transmission system can maintain excellent time and frequency synchronization performance in high-speed movement scenarios, meet the requirements for stable transmission, and improve the accuracy of time and frequency synchronization while covering a wide frequency offset.

[0053] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0054] Those skilled in the art will understand that, for convenience and brevity, the specific working process of the system described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0055] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A time-frequency synchronization method for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles, characterized in that, A receiver is applicable to an orthogonal frequency division multiplexing (OFDM) transmission time-frequency synchronization system for unmanned aerial vehicles (UAVs). The UAV OFDM transmission time-frequency synchronization system also includes a UAV terminal. The UAV OFDM transmission time-frequency synchronization method includes: acquiring a baseband received signal transmitted by the UAV terminal; wherein the baseband received signal is generated by analog-to-digital conversion of a superimposed signal generated by the UAV terminal; the superimposed signal is generated based on a pseudo-random noise sequence and a preset power allocation factor; performing a sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function; and then... When the peak value of the number exceeds a preset threshold, the corresponding time is taken as the symbol start point; based on the symbol start point, the phase difference between the cyclic prefix and the symbol body in the baseband received signal is coarsely estimated to obtain a coarsely estimated frequency offset; based on the coarsely estimated frequency offset, the baseband received signal is coarsely compensated to obtain a coarsely compensated baseband received signal; based on the coarsely compensated baseband received signal, the subcarrier phase rotation is finely estimated to obtain a finely estimated frequency offset; the coarsely estimated frequency offset and the finely estimated frequency offset are added to obtain a total frequency offset; and based on the total frequency offset, the baseband received signal is frequency offset corrected to obtain a corrected signal.

2. The time-frequency synchronization method for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles as described in claim 1, characterized in that, The generation of the superimposed signal includes: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol according to a preset power allocation factor to generate the superimposed signal.

3. The method for time-frequency synchronization of UAV orthogonal frequency division multiplexing transmission as described in claim 1, characterized in that, A synchronization function is generated by performing a sliding cross-correlation operation based on the baseband received signal and a preset local time-domain training sequence. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. This includes: moving the preset local time-domain training sequence on the baseband received signal based on a preset sliding window, calculating the cross-correlation value at each moving position; assembling the cross-correlation values ​​of all moving positions into a sequence and normalizing it to generate the synchronization function; extracting the amplitude of all local peaks based on the synchronization function, and taking the moment corresponding to the local peak whose amplitude exceeds a preset threshold as the symbol start point.

4. The method for time-frequency synchronization of UAV orthogonal frequency division multiplexing transmission as described in claim 1, characterized in that, Based on the symbol start point, a coarse estimate of the phase difference between the cyclic prefix and the symbol body in the baseband received signal is obtained, resulting in a coarse frequency offset estimate. This includes: determining the time-domain positions of the cyclic prefix and the symbol body in the baseband received signal based on the symbol start point; wherein the cyclic prefix is ​​located in the interval after the symbol start point, and the symbol body is located in the time-domain length interval of the symbol body after the cyclic prefix; extracting the sampled data of the cyclic prefix and the sampled data of the symbol body as data to be compared, and calculating the cross-correlation phase difference between the two sets of data to be compared; and calculating the coarse frequency offset estimate based on the cross-correlation phase difference and the time-domain length of the symbol body.

5. The time-frequency synchronization method for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles as described in claim 1, characterized in that, The fine-estimated frequency offset is obtained by performing a fine estimation of the subcarrier phase rotation based on the coarsely compensated baseband received signal. This includes: truncating the coarsely compensated baseband received signal and retaining the main body of the symbol after the symbol start point; performing a Fast Fourier Transform on the truncated main body of the symbol to convert it into a frequency domain subcarrier signal; calculating the phase rotation of adjacent subcarriers based on the phase values ​​corresponding to the frequency domain subcarrier signals; and performing a linear fitting of the phase rotation to calculate the fine-estimated frequency offset.

6. The time-frequency synchronization method for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles as described in claim 2, characterized in that, The quadrature amplitude modulation mapping is a 16QAM modulation mapping.

7. The method for time-frequency synchronization of UAV orthogonal frequency division multiplexing transmission as described in claim 1, characterized in that, A UAV terminal is applicable to an orthogonal frequency division multiplexing transmission time-frequency synchronization system for UAVs; the UAV terminal is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; after performing analog-to-digital conversion on the superimposed signal, a baseband receiving signal is generated, and the baseband receiving signal is transmitted to the receiving end.

8. A time-frequency synchronization system for orthogonal frequency division multiplexing transmission of unmanned aerial vehicles, characterized in that, It includes a receiver and a drone; the drone is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor; and to generate a baseband received signal after analog-to-digital conversion based on the superimposed signal. The baseband received signal is transmitted to the receiving end; the receiving end is used to acquire the baseband received signal transmitted by the UAV. Sliding cross-correlation is performed based on the baseband received signal and a preset local time-domain training sequence to generate a synchronization function. The moment when the peak value of the synchronization function exceeds a preset threshold is taken as the symbol start point. The phase difference between the cyclic prefix and the symbol body in the baseband received signal is coarsely estimated based on the symbol start point to obtain a coarsely estimated frequency offset; the baseband received signal is then coarsely compensated and corrected based on the coarsely estimated frequency offset to obtain a coarsely compensated baseband received signal. The subcarrier phase rotation is precisely estimated based on the coarsely compensated baseband received signal to obtain the finely estimated frequency offset. The coarse frequency offset is added to the fine frequency offset to obtain the total frequency offset; The baseband received signal is then frequency offset corrected based on the total frequency offset to obtain the corrected signal.

9. The orthogonal frequency division multiplexing transmission time-frequency synchronization system for unmanned aerial vehicles as described in claim 8, characterized in that, The UAV terminal is used to generate a superimposed signal based on a pseudo-random noise sequence and a preset power allocation factor, including: generating a pseudo-random noise sequence based on a preset pseudo-random sequence; performing orthogonal amplitude modulation mapping on the pseudo-random noise sequence to generate complex symbols; performing matrix transformation mapping on the complex symbols to the frequency domain to generate a frequency domain sequence; performing inverse fast Fourier transform on the frequency domain sequence to generate a time domain training sequence; and linearly superimposing the time domain training sequence onto the cyclic prefix sampling points of a preset OFDM symbol according to the preset power allocation factor to generate a superimposed signal.

10. The orthogonal frequency division multiplexing transmission time-frequency synchronization system for unmanned aerial vehicles as described in claim 8, characterized in that, The receiving end is used to acquire the baseband received signal transmitted by the UAV. Based on the baseband received signal and the preset local time domain training sequence, a sliding cross-correlation operation is performed to generate a synchronization function. When the peak value of the synchronization function exceeds a preset threshold, the corresponding time is taken as the symbol start point. This includes: moving the preset local time domain training sequence on the baseband received signal based on a preset sliding window, and calculating the cross-correlation value at each moving position. The cross-correlation values ​​of all moving positions are sequenced and normalized to generate a synchronization function. The amplitude of all local peaks is extracted based on the synchronization function, and the time corresponding to the local peak whose amplitude exceeds a preset threshold is taken as the symbol start point.

Images