Pre-feedback combined burst signal fast symbol synchronization method
By combining a front-feedback approach with an initial NCO, FIFO buffer, O&M algorithm, and Gardner loop, rapid symbol synchronization of burst signals in satellite communication is achieved. This solves the problems of slow synchronization speed and high overhead in traditional methods, and improves the stability and accuracy of synchronization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-26
- Publication Date
- 2026-03-31
AI Technical Summary
Traditional methods for fast symbol synchronization of burst signals suffer from slow convergence speed, high symbol overhead, and weak noise immunity, making it particularly difficult to achieve efficient symbol synchronization in satellite communications.
A combined front-feedback approach is adopted, which samples at 4 times the symbol rate through the initial NCO, and after split processing, uses FIFO buffer and O&M algorithm to estimate timing deviation. The interpolation control NCO is started and interpolation is performed through Farrow structure interpolation filter. Feedback control is combined with Gardner loop to achieve fast symbol synchronization.
It improves the speed and accuracy of fast symbol synchronization for burst signals, ensures the stability and timeliness of synchronization, reduces symbol overhead, and enhances noise immunity.
Smart Images
Figure CN121770958A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of digital signal processing and satellite communication technology, specifically to a method for fast symbol synchronization of burst signals using a combination of front feedback and backfeedback. Background Technology
[0002] In the field of satellite communication, TDMA technology is often used to achieve multiple access. TDMA allocates time slots to different users to achieve the sharing of wireless resources. In TDMA, the transmitted signal is generally bursty. Burst signals have high requirements for signal detection and synchronization speed, which makes implementation more difficult. In addition, during satellite communication signal transmission, since the transmitting and receiving clocks are independent of each other, there is a frequency and phase deviation between them. Therefore, symbol synchronization technology is needed to determine the optimal sampling time of the received signal. The result of symbol synchronization directly affects the quality of the received signal.
[0003] Traditional methods for fast symbol synchronization of burst signals mainly employ feedback and feedforward synchronization techniques. Feedback synchronization typically uses a time error detector to detect timing deviations and then uses the error value, processed by loop filtering, to control the digital oscillator. Feedforward synchronization generally only estimates the timing deviation once at the start of the signal, directly estimating the sampling timing deviation based on the characteristics of the received signal. Clearly, these methods for fast symbol synchronization of burst signals have at least the following shortcomings: 1. Because the signal clock changes slowly relative to the symbol rate, feedforward techniques only estimate the timing deviation once at the start of the signal. This is easily affected by noise and other factors, leading to estimation errors or clock drift. Therefore, timely correction of the timing estimate is impossible, resulting in a decrease in reception quality.
[0004] 2. Feedback-based symbol synchronization suffers from slow convergence speed and requires significant symbol overhead. Furthermore, feedback loops often use the Gardner algorithm as the TED (Transmission Traceability) algorithm. Gardner exhibits substantial self-noise when processing signals with small roll-off factors, leading to difficulty in loop convergence. A very small roll-off factor is required to ensure stable loop convergence, further increasing the symbol overhead required for symbol synchronization. Therefore, feedback-based symbol synchronization methods have significant limitations in processing burst signals, especially short burst signals. Summary of the Invention
[0005] To address the aforementioned technical shortcomings, the present invention aims to provide a method for fast symbol synchronization of burst signals that combines front feedback.
[0006] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a fast symbol synchronization method for burst signals with front feedback, including: S1, sampling: using an initial NCO to sample the received signal after matched filtering at 4 times the symbol rate.
[0007] S2. Timing Deviation Estimation: Obtain the sampled values from step S1 and divide them into two paths. One path is sent to the FIFO memory for buffering, and the other path uses the O&M algorithm to estimate the timing deviation.
[0008] S3. Start the interpolation control NCO: Obtain the timing deviation estimation result in step S2, determine the time when the FIFO structure starts reading data interpolation, the initial value of the oscillation of the interpolation control NCO and the initial value of the interpolation coefficient, and start the interpolation control NCO.
[0009] S4. Interpolation Processing: Obtain the initial values of the interpolation coefficients from step S3, and based on these initial values, send the data read from the FIFO register into the Farrow structure interpolation filter for interpolation processing.
[0010] S5. Feedback Control and Output: Obtain the interpolation result from step S4, start the Gardner loop based on the interpolation result, and send the loop error to the interpolation control NCO. Repeat steps S4-S5 to achieve feedback control and output the interpolation result from step S4.
[0011] The beneficial effects of this invention are as follows: 1. This invention provides a fast symbol synchronization method for burst signals with front feedback. First, the received signal after matched filtering is sampled at 4 times the symbol rate using an initial NCO. Then, the sampled value is divided into two paths: one path is sent to the FIFO register for buffering, and the other path uses the O&M algorithm to estimate the timing deviation. Next, based on the timing deviation estimation result, the time when the FIFO register starts reading data, the initial value of the interpolation control NCO oscillation, and the initial value of the interpolation coefficients are determined, and the interpolation control NCO is started. Then, based on the initial value of the interpolation coefficients, the data read from the FIFO register is sent to the Farrow structure interpolation filter for interpolation processing. Finally, the Gardner loop is started to perform feedback control on the interpolation control NCO and output the interpolation result, which improves the speed and accuracy of fast symbol synchronization for burst signals and ensures the stability and timeliness of fast symbol synchronization for burst signals.
[0012] 2. This invention obtains the sampled values of each sampling point in the observation area and the length of each observation area. Then, it calculates the deviation estimate based on the sampled values of each sampling point in the observation area and the length of each observation area. The deviation estimate is divided into a fractional part and an integer part. The integer part is used to control the read enable of the FIFO, and the fractional part is used to configure the oscillation initial value and the initial value of the interpolation control NCO. The interpolation control NCO is enabled by a pulse of 4 times the symbol rate and generates a symbol enable signal of 2 times the symbol rate, which improves the fast symbol synchronization speed of burst signals and ensures the timeliness of fast symbol synchronization of burst signals.
[0013] 3. This invention is based on front feedback for fast symbol synchronization of burst signals. Since the feedback symbol overhead is negligible, the loop bandwidth of the loop filter in the Gardner loop can be set very small without causing loop convergence difficulties, thus ensuring the stability of fast symbol synchronization of burst signals.
[0014] 4. This invention uses the double symbol rate enable signal as the clock for the Gardner loop. When the Farrow filter outputs the interpolation processing result and the Gardner loop can receive the double symbol rate enable signal output by the interpolation control NCO, the Gardner loop is started. After the Gardner loop is started, Gardner error detection is performed to output the timing error. Then, random noise in the timing error is suppressed and a control signal is output. The output control signal is sent to the interpolation control NCO for parameter update, which improves the accuracy of fast symbol synchronization for burst signals. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention.
[0017] Figure 2 This is a block diagram of the fast symbol synchronization system of the present invention.
[0018] Figure 3 This is the constellation diagram after the feedback is combined with symbol timing in this invention.
[0019] Figure 4 The O&M algorithm of this invention is used to estimate the performance curve at regular intervals.
[0020] Figure 5 This is a structural diagram of the interpolation filter of the present invention.
[0021] Figure 6 This is a comparison of the interpolation coefficient convergence curves of the Gardner loop and the front feedback combined scheme proposed in this invention.
[0022] Figure 7 This is the constellation diagram after the timing of the feedback loop symbol in this invention. Detailed Implementation
[0023] 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.
[0024] Please see Figure 1 , Figure 2 and Figure 3 As shown, the present invention provides a fast symbol synchronization method for burst signals with front feedback, including: S1, sampling: sampling the received signal after matched filtering at each sampling point in the observation area using an initial NCO at 4 times the symbol rate.
[0025] It should be noted that the initial NCO generates a sampling pulse at 4 times the symbol rate through accumulation. On the one hand, this reduces the amount and complexity of data processing by lowering the sampling frequency. On the other hand, it allows for flexible variation of the received signal rate depending on the configured accumulation value.
[0026] Please see Figure 4 As shown, S2, Timing Deviation Estimation: Obtain the sampled value from step S1 and divide it into two paths. One path is sent to the FIFO memory for buffering, and the other path uses the O&M algorithm to estimate the timing deviation.
[0027] It should be noted that the sampled values buffered in the FIFO memory are not output before the timing deviation estimation is completed.
[0028] In a specific embodiment, the timing bias estimation using the O&M algorithm is performed as follows: The sampled values of each sampling point in the observation area and the length of each observation area are obtained. Then, according to the calculation formula: Obtain the deviation estimate In the formula The representative index is The sampled values of the sampling points, Represents the index of each sampling point. Represents the sampling rate. Represents the length of the observation area. Index representing the observation area, Represents the imaginary unit. , Represents pi (π) All of these represent natural constants.
[0029] It should be noted that the sampling rate is 4 times the symbol rate. , .
[0030] S3. Start the interpolation control NCO: Obtain the timing deviation estimation result in step S2, determine the time when the FIFO register starts reading data, the initial value of the oscillation of the interpolation control NCO and the initial value of the interpolation coefficient, and start the interpolation control NCO.
[0031] In a specific embodiment, the process of starting the interpolation control NCO is as follows: obtain the deviation estimate value in step S2, and divide it into a fractional part and an integer part. The integer part is used to control the read enable of the FIFO register, and the fractional part is used to configure the oscillation initial value and the interpolation coefficient initial value of the interpolation control NCO.
[0032] It should be noted that, for example, when the deviation estimate is 3.3, the integer part is 3 and the decimal part is 0.3. This example is for illustrative purposes only and is not the only valid one.
[0033] The interpolation control NCO is enabled by a pulse at 4 times the symbol rate, generating a symbol enable signal at 2 times the symbol rate.
[0034] In the above, the specific process of using the integer part to control the read enable of the FIFO register is as follows: The integer part of the deviation estimate is denoted as... When the FIFO register buffers the sampled values of each sampling point, it continuously obtains the total number of sampling points buffered in the FIFO register. When the total number of sampling points buffered in the FIFO register is... And start caching the first When there are 1 sampling point, the read enable of the FIFO register is activated, and the FIFO register begins to read the sampled values of each sampling point that have been cached in the FIFO register.
[0035] In the above, the specific process for configuring the initial values of the NCO oscillation and the initial values of the interpolation coefficients is as follows: The decimal part of the deviation estimate is denoted as... And obtain the quantization bit width, according to The decimal part of the deviation estimate is quantified, where... Represents the quantization bit width. The value obtained after quantizing the decimal part of the deviation estimate.
[0036] Obtain the bit width of the phase accumulator within the interpolation control NCO and denote it as... According to Obtain the initial phase initial phase The initial value is used to control the oscillation of the NCO through interpolation.
[0037] according to The delay parameter vector required to generate the Farrow structure interpolation filter is denoted as . Obtain the fixed coefficient matrix of the Farrow structure interpolation filter, and calculate the initial values of the interpolation coefficients based on the delay parameter vector required by the Farrow structure interpolation filter and the fixed coefficient matrix of the Farrow structure interpolation filter.
[0038] It should be noted that the fixed coefficient matrix of the Farrow structure interpolation filter is... .
[0039] It should also be noted that the formula for calculating the initial value of the interpolation coefficients is: ,in This represents the initial value of the interpolation coefficients.
[0040] The specific process for generating the symbol enable signal at twice the symbol rate described above is as follows: the full scale of the phase accumulator is obtained according to the bit width of the phase accumulator in the NCO controlled by interpolation. When the phase accumulator starts working from the initial phase, whenever the phase accumulator receives a working enable pulse of 4 times the symbol rate, the phase accumulator accumulates one step and uses it as the value of the phase accumulator. At the same time, the value of the phase accumulator is obtained in real time and compared with the full scale of the phase accumulator.
[0041] It should be noted that the full scale of the phase accumulator is... .
[0042] If the value of the bit accumulator is less than the full scale of the phase accumulator, overflow will not be triggered; otherwise, overflow will be triggered. In this case, the interpolation control NCO outputs a pulse of one sampling point and subtracts the full scale of the phase accumulator from the value of the phase accumulator.
[0043] Please see Figure 5 As shown, S4, interpolation processing: Obtain the initial value of the interpolation coefficient in step S3, and according to the initial value of the interpolation coefficient, send the data read from the FIFO register into the Farrow structure interpolation filter for interpolation processing.
[0044] It should be noted that the data read from the FIFO register is the sampled value of each sampling point that was previously buffered.
[0045] In one specific embodiment, the interpolation process is as follows: The sampled values of four adjacent sampling points are read from the FIFO register and recorded as follows: , , , ,according to The index is obtained as Fractional delay compensation results for sampling points In the formula These represent the fixed coefficient parameters of the Farrow structure interpolation filter.
[0046] Please see Figure 6 As shown, S5, Feedback Control and Output: Obtain the interpolation processing result in step S4, start the Gardner loop according to the interpolation processing result, and send the loop error to the interpolation control NCO. Repeat steps S4-S5 to realize feedback control and output the interpolation processing result of step S4.
[0047] In a specific embodiment, the feedback control and output process is as follows: the double symbol rate enable signal is used as the clock for the Gardner loop to operate. When the Farrow structure interpolation filter outputs the interpolation processing result and the Gardner loop can receive the double symbol rate enable signal output by the interpolation control NCO, the Gardner loop is started.
[0048] After starting the Gardner loop, Gardner error detection is performed to output timing error, and then random noise in the timing error is suppressed to output control signal.
[0049] The output control signal is sent to the interpolation control NCO for parameter update. Steps S4-S5 are repeated to achieve feedback control and output the interpolation result of step S4.
[0050] It should be noted that the oscillation frequency and interpolation coefficients of the interpolation-controlled NCO are updated.
[0051] The above-mentioned process for performing Gardner error detection and outputting timing error is as follows: The in-phase and quadrature branch amplitudes of the interpolated signal are obtained from the interpolation processing results output by the Farrow structure interpolation filter. Based on... The index is obtained as Timing error of symbol sampling time In the formula Index of the symbol, The amplitude of the in-phase branch representing the interpolated signal. The amplitude of the orthogonal branch representing the interpolated signal. The representative index is The amplitude of the in-phase branch at the symbol sampling time. The representative index is The amplitude of the orthogonal branch at the symbol sampling time. Representing the same-phase branch, Represents orthogonal branches, The representative index is The amplitude of the in-phase branch at the symbol sampling time. The representative index is The amplitude of the orthogonal branch at the sampling time of the symbol.
[0052] The specific process for suppressing random noise in the timing error and outputting the control signal, as described above, is as follows: In the formula The representative index is The control signal at the symbol sampling time, The representative index is The control signal at the symbol sampling time, and All of these represent the Gardner loop filter scaling factor.
[0053] Example: This invention is applied to a satellite communication receiver to perform fast symbol timing recovery on burst signals received by the receiver. It includes a matched filter, an initial NCO module, an O&M algorithm timing estimation module, a FIFO buffer module, an interpolation control NCO module, an interpolation filter module, a Gardner error detection module, and a loop filter module.
[0054] Among them, the matched filter is used to match the signal output by the shaping filter in the satellite communication transmitter, so as to make the most efficient use of the limited bandwidth and achieve the best signal reception without intersymbol interference.
[0055] The initial NCO module is used to generate sampling pulse signals at 4 times the symbol rate, reducing the sampling rate while ensuring the system's flexibility for signals with different symbol rates.
[0056] The timing estimation module of the O&M algorithm calculates the arctangent value of the signal by summing the results of the DFT of the signal magnitude squared. Assuming the signal is sampled at N times the symbol rate, the observation interval length is L (i.e., NL sampling points are observed), and the received signal (complex number) is x(k), the calculation expression is: Among them, the estimated value The integer part is used for FIFO control, and the fractional part is sent to the interpolation control NCO module. Figure 2The root mean square error (RMSE) curves of timing bias estimated using the O&M algorithm for signals with different roll-off coefficients under different observation intervals L, obtained from simulations, are presented. Taking L=256 as an example, with a roll-off of 0.1, When the value is 0dB, the RMSE of the feedforward timing estimate is about 0.1 symbols, and the residual timing deviation can be quickly locked by the subsequent feedback loop.
[0057] The FIFO caching module caches data, ensuring strict synchronization between the data used to estimate timing errors and the data used for subsequent loop tracking, thus ensuring that the estimated sampling bias is applicable to subsequent data. On the other hand, caching enables the reuse of data used for feedforward estimation, reducing overall symbol overhead.
[0058] The interpolation control NCO module determines the initial state based on the fractional part of the error estimate, generates sampling pulses at twice the symbol rate, and generates corresponding interpolation coefficients.
[0059] The interpolation filter module performs fractional interpolation based on the interpolation coefficients to compensate for fractional sampling timing deviations. The interpolation process can be expressed as: .
[0060] The Gardner error detection module estimates the timing error of the interpolated output signal using the following formula: .
[0061] The loop filter module filters timing errors to reduce the impact of random errors. The output of the loop filter is: , Figure 5 A comparison of the interpolation coefficient convergence curves of the Gardner feedback loop alone and the front-feedback combined scheme proposed in this invention is given. As can be seen from the figure, the scheme of this invention eliminates the initial loop locking process, which greatly accelerates the loop convergence. Even with a small loop bandwidth and a small roll-off factor, the convergence speed of the loop does not show obvious lag. Figure 7 The constellation diagram of the 8PSK signal after timing recovery using only the Gardner feedback loop is presented. The diagram shows that even with a loop overhead of 100 symbols, some stray points still exist around the constellation points. Figure 3 The constellation diagram of the 8PSK signal after timing recovery using the scheme proposed in this invention is given. As can be seen from the diagram, when there is no symbol overhead, there is basically no spurious around the constellation point, indicating that the loop has been stably locked.
[0062] This invention first uses an initial NCO to sample the received signal after matched filtering at 4 times the symbol rate. Then, the sampled values are divided into two paths: one is sent to a FIFO register for buffering, and the other uses the O&M algorithm to estimate the timing deviation. Next, based on the timing deviation estimation result, the time when the FIFO register starts reading data, the initial value of the interpolation control NCO oscillation, and the initial value of the interpolation coefficients are determined, and the interpolation control NCO is started. Then, based on the initial value of the interpolation coefficients, the data read from the FIFO register is sent to the Farrow structure interpolation filter for interpolation processing. Finally, the Gardner loop is started to provide feedback control to the interpolation control NCO and output the interpolation result. This improves the speed and accuracy of fast symbol synchronization for burst signals and ensures the stability and timeliness of fast symbol synchronization for burst signals.
[0063] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.
[0064] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.
Claims
1. A method for fast symbol synchronization of a burst signal using feedforward and feedback combination, characterized by, The method comprises the following steps: S1, sampling: sampling the matched filtering processed received signal at 4 times symbol rate at each sampling point in the observation area using an initial NCO; S2, timing offset estimation: obtaining the sampling value in step S1 and dividing it into two paths, one of which is sent to a FIFO memory for buffering, and the other of which is used for timing offset estimation using an O&M algorithm; S3, starting the interpolation control NCO: obtaining the timing offset estimation result in step S2, determining the time when the FIFO register starts reading data, the initial value of the oscillation of the interpolation control NCO, and the initial value of the interpolation coefficient, and starting the interpolation control NCO; S4, interpolation processing: obtaining the initial value of the interpolation coefficient in step S3, and sending the data read out by the FIFO register to the Farrow structure interpolation filter for interpolation processing according to the initial value of the interpolation coefficient; S5, feedback control and output: obtaining the interpolation processing result in step S4, starting the Gardner loop according to the interpolation processing result, and sending the loop error to the interpolation control NCO, repeating steps S4-S5 to realize feedback control and output the interpolation processing result of step S4.
2. A method for fast symbol synchronization of a burst signal using feedforward combination according to claim 1, wherein, The timing offset estimation using the O&M algorithm has the following specific process: The sampling values of each sampling point in the observation area and the length of each observation area are obtained, and then the deviation estimation value is obtained according to a calculation formula: , wherein represents the sampling value of the sampling point with the index of , represents the index of each sampling point, represents the sampling rate, represents the length of the observation area, represents the index of the observation area, represents the imaginary unit, , represents the circular constant, all represent the natural constant. 3. The method of claim 1, wherein the method is a method of fast symbol synchronization of a burst signal using a feedforward combination, characterized in that, The starting of the interpolation control NCO has the following specific process: The integer part is used to control the read enable of the FIFO register, and the fractional part is used to configure the initial value of the oscillation of the interpolation control NCO and the initial value of the interpolation coefficient. The interpolation control NCO is enabled at 4 times symbol rate to generate a symbol enable signal at 2 times symbol rate.
4. A method for fast symbol synchronization of a burst signal using feedforward combination according to claim 3, wherein The integer part is used to control the read enable of the FIFO register, and the fractional part is used to configure the initial value of the oscillation of the interpolation control NCO and the initial value of the interpolation coefficient. The integer part of the bias estimate value is denoted as , the FIFO register obtains the total number of sampling points in the FIFO register in real time when buffering the sampling values of each sampling point. When the total number of sampling points in the FIFO register is and the first sampling point is buffered, the read enable of the FIFO register is activated, and the FIFO register starts reading the sampling values of each sampling point that has been buffered in the FIFO register.
5. The method for fast symbol synchronization of burst signals with front feedback as described in claim 3, characterized in that, The generation of the symbol enable signal at 2 times symbol rate has the following specific process: Let the decimal part of the bias estimate value be denoted as and the quantization bit width be obtained according to The decimal part of the bias estimate value is quantized, where represents the quantization bit width, represents the value obtained after quantization of the decimal part of the bias estimate value; The bit width of the phase accumulator in the interpolation control NCO is obtained and denoted as Then, according to the initial phase is obtained is the initial value of the oscillation of the interpolation control NCO; According to The delay parameter vector required for generating the Farrow structure interpolation filter is denoted as The fixed coefficient matrix of the Farrow structure interpolation filter is obtained, and the initial value of the interpolation coefficient is calculated according to the delay parameter vector required for the Farrow structure interpolation filter and the fixed coefficient matrix of the Farrow structure interpolation filter.
6. A method for fast symbol synchronization of a burst signal using feedforward combination according to claim 3, wherein The full scale of the phase accumulator is obtained according to the bit width of the phase accumulator in the interpolation control NCO, and when the phase accumulator starts working from the initial phase, the accumulation step of the phase accumulator is added every time the phase accumulator receives a 4 times symbol enable pulse, and the value of the phase accumulator is taken as the value of the phase accumulator at the same time. If the value of the bit accumulator is less than the full scale of the phase accumulator, no overflow is triggered, otherwise, overflow is triggered, and at this time, the interpolation control NCO outputs a sampling point pulse, and the value of the phase accumulator is reduced by the full scale of the phase accumulator. The interpolation processing has the following specific process:
7. The method of claim 2, wherein the method is a method of fast symbol synchronization of a burst signal using a feedforward combination, characterized in that, The feedback control and output have the following specific process: Four adjacent sample values are read out from the FIFO register and denoted as , , , , according to , the fractional delay compensation result of the sample point with index , where represents the fixed coefficient parameter of the Farrow structure interpolation filter. 8. The method for fast symbol synchronization of burst signals with front feedback as described in claim 1, characterized in that, The 2 times symbol rate enable signal is used as the clock of the Gardner loop, and when the Farrow structure interpolation filter outputs the interpolation processing result and the Gardner loop can receive the 2 times symbol rate enable signal output by the interpolation control NCO, the Gardner loop is started. After starting the Gardner loop, a timing error is output by detecting a Gardner error, and a control signal is output by suppressing random noise in the timing error; The output control signal is sent to the interpolation control NCO for parameter updating, and steps S4-S5 are repeated to realize feedback control and output the interpolation processing result of step S4.
9. The method of claim 8, wherein the method is a method of fast symbol synchronization of a burst signal using a feedforward combination, characterized in that, The timing error is output by detecting the Gardner error, and the specific process is as follows: The in-phase and quadrature branch amplitudes of the interpolated signal are obtained from the interpolation processing results output by the Farrow structure interpolation filter. The index is obtained as Timing error of symbol sampling time In the formula Index of the symbol, The amplitude of the in-phase branch representing the interpolated signal. The amplitude of the orthogonal branch representing the interpolated signal. The representative index is The amplitude of the in-phase branch at the symbol sampling time. The representative index is The amplitude of the orthogonal branch at the symbol sampling time. Representing the same-phase branch, Represents orthogonal branches, The representative index is The amplitude of the in-phase branch at the symbol sampling time. The representative index is The amplitude of the orthogonal branch at the sampling time of the symbol.
10. The method of claim 9, wherein the method is a method of fast symbol synchronization of a burst signal using a feedforward combination, and The control signal is output by suppressing the random noise in the timing error, and the specific process is as follows: ； wherein the control signal representing the symbol sampling instant with index , the control signal representing the symbol sampling instant with index , and both represent the Gardner loop filter scaling factor.