A frequency offset estimation and tracking method and apparatus

Abstract

The application discloses a frequency offset estimation and tracking method and device, and relates to the technical field of coherent optical communication. The method comprises the following steps: performing coarse estimation on the frequency offset of a current main channel to obtain a coarse frequency offset value Delta f1; feeding back the Delta f1 to a front-end frequency offset compensation module, so that the front-end frequency offset compensation module compensates the frequency offset of the main channel according to the Delta f1; performing synchronization in an intermittent start local correlation synchronization mode based on a pre-determined superframe preamble estimation position; after the synchronization is completed, performing fine estimation on the frequency offset by using a periodically inserted pilot to obtain a fine frequency offset value Delta f2; feeding back the Delta f2 together with the Delta f1 to the front-end frequency offset compensation module, and compensating the Delta f2 to the current main channel to realize frequency offset tracking and correction. The application can solve the problems of the prior art, such as long time consumption of frequency offset estimation, long compensation loop delay, and insufficient fast tracking and compensation capacity in a high dynamic frequency offset change scene.

Description

Technical Field

[0001] This invention relates to the field of coherent optical communication technology, specifically to a frequency offset estimation and tracking method and apparatus. Background Technology

[0002] In the field of coherent optical communication, frequency offset estimation and compensation are essential. For example... Figure 1 and Figure 2 As shown, in the existing coherent optical communication receiver link infrastructure, the frequency offset compensation module is located at the receiver front end, and the frequency offset estimation module is located after polarization demultiplexing. By feeding back the frequency offset estimation value to the frequency offset compensation module at the front end, the main path data is compensated, forming a closed-loop estimation compensation.

[0003] like Figure 2 As shown, in existing frequency offset compensation schemes, modules such as frequency offset estimation, frequency offset compensation, synchronization, and phase recovery are relatively independent and mostly performed sequentially. There is little information coordination between modules, resulting in time-consuming issues such as subsequent modules waiting for preceding modules to complete processing, and resource waste due to similar circuits being used multiple times in different modules. Furthermore, existing frequency offset compensation schemes suffer from long frequency offset estimation times and long compensation loop delays, leading to insufficient rapid tracking and compensation capabilities in scenarios with high dynamic frequency offset changes. Summary of the Invention

[0004] The purpose of this invention is to provide a frequency offset estimation and tracking method and apparatus, which can at least solve the problems of long frequency offset estimation time, long compensation loop delay, and insufficient fast tracking and compensation capability in scenarios with high frequency offset dynamic changes in the prior art.

[0005] To achieve the above objectives, in a first aspect, embodiments of the present invention provide a frequency offset estimation and tracking method, the method comprising: A coarse estimate of the frequency offset of the current main path is obtained to obtain a coarse frequency offset value Δf1; Δf1 is fed back to the frequency offset compensation module at the front end so that the frequency offset compensation module at the front end can perform frequency offset feedback compensation on the main path based on Δf1. Synchronization is performed by intermittently starting local correlation synchronization based on the pre-determined superframe preamble prediction position; After synchronization is completed, the frequency offset is precisely estimated using periodically inserted pilots to obtain the precise frequency offset value Δf2. Δf2 and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 is also compensated into the current main path to achieve frequency offset tracking correction.

[0006] In conjunction with the first aspect, in one implementation, when coarsely estimating the frequency offset of the current main path, a time-domain frequency offset estimation algorithm based on phase difference and / or a frequency-domain frequency offset estimation algorithm based on FFT are used.

[0007] In conjunction with the first aspect, in one implementation, for signals where constellation points are not rectangularly distributed, when using a frequency offset estimation algorithm based on FFT in the frequency domain for coarse estimation, a ring-rotation enhancement method is used: the second inner ring of the constellation diagram is selected for uniform angle rotation, increasing the initial phase by... / 4+k After calculating the constellation point ratio by 2, the frequency offset is estimated to the fourth power or a multiple of four angles.

[0008] In conjunction with the first aspect, in one implementation, the uniform rotation angle is 0.4636 rad or / 8.

[0009] In conjunction with the first aspect, in one implementation, when using a time-domain frequency offset estimation algorithm based on phase difference for coarse estimation, a pre-screening method is used: the pre-screening method is used to find the superframe preamble prediction position, and the frequency offset of the received one-frame data at this moment is estimated based on phase difference.

[0010] In conjunction with the first aspect, in one embodiment, the pre-screening method includes density pre-screening and energy pre-screening.

[0011] In conjunction with the first aspect, in one implementation, the method of intermittently starting local correlation synchronization includes: determining whether the superframe preamble prediction position has been successfully determined; if so, starting the preamble local symbol sliding correlation synchronization operation based on the superframe preamble prediction position; otherwise, not performing the symbol sliding correlation synchronization operation.

[0012] In conjunction with the first aspect, in one implementation, the method further includes: pausing the coarse estimation of frequency offset after successful synchronization; and restarting the coarse estimation of frequency offset once a loss of synchronization occurs.

[0013] In conjunction with the first aspect, in one implementation, while compensating Δf2 into the current main path, the coarse phase noise ΔL at the pilot is also calculated through the pilot, and ΔL is used to perform coarse phase noise compensation on the current main path.

[0014] In conjunction with the first aspect, in one implementation, the method further includes: accurately estimating and compensating for the phase noise.

[0015] In conjunction with the first aspect, in one implementation, the frequency offset is precisely estimated using periodically inserted pilot signals to obtain a precise frequency offset value Δf2. At the same time, a residual large frequency offset is estimated using a known superframe preamble sequence to obtain a residual large frequency offset value Δf3. Δf3, Δf2, and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 and Δf3 are compensated to the current main path to achieve frequency offset tracking correction.

[0016] In conjunction with the first aspect, in one implementation, the method further includes: collecting N pilot data points, performing pilot quadruple angle FFT frequency offset estimation on them to obtain a high-precision frequency offset value Δf4; compensating Δf4 into the current main path, and feeding Δf4 back to the front-end frequency offset compensation module.

[0017] In a second aspect, the present invention also provides a frequency offset estimation and tracking device for implementing the method in the first aspect embodiment, including a first-level frequency offset estimation module, a synchronization module, a second-level frequency offset estimation module, and a fast compensation module; The first-level frequency offset estimation module is used to: coarsely estimate the frequency offset of the current main path to obtain a coarse frequency offset value Δf1; and feed Δf1 back to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf1. The synchronization module is used to: perform synchronization based on the pre-determined superframe preamble estimated position by intermittently starting local correlation synchronization; The second-level frequency offset estimation module is used to: after synchronization is completed, use periodically inserted pilots to perform fine frequency offset estimation to obtain a fine frequency offset value Δf2; feed Δf2 back to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf2 and Δf1; and send Δf2 to the fast compensation module. The fast compensation module is used to compensate Δf2 into the current main path to achieve frequency offset tracking correction.

[0018] The beneficial effects of the technical solutions provided in this application include: (1) The embodiments of this application can realize a frequency offset hierarchical estimation and joint tracking scheme. It can not only coarsely estimate the frequency offset of the current main path to obtain a coarse frequency offset value, but also use periodically inserted pilots after synchronization to finely estimate the frequency offset to obtain a fine frequency offset value. By feeding the fine frequency offset value and the coarse frequency offset value back to the front-end frequency offset compensation module, frequency offset feedback compensation can be realized, which can compensate for frequency offset changes. At the same time, the fine frequency offset value is also compensated to the current main path, which can quickly realize frequency offset tracking correction and achieve the purpose of fast frequency offset compensation. Through this synergy of fast frequency offset compensation and frequency offset feedback compensation, the problems of long frequency offset estimation time, long compensation loop delay, and insufficient fast tracking compensation capability in high frequency offset dynamic change scenarios in the prior art can be effectively solved. Meanwhile, during synchronization, it is based on the pre-determined superframe preamble prediction position and uses the intermittent start of local correlation synchronization to achieve synchronization. This can not only quickly realize synchronization and save time, but also significantly reduce power consumption and complexity, improve overall processing efficiency, and meet the needs of practical applications.

[0019] (2) In this embodiment of the application, the frequency offset estimation scheme based on FFT can also be applied to non-rectangular constellations by using the ring rotation enhancement method, and the estimated value can be obtained effectively. This solves the problem that the frequency offset estimation algorithm based on FFT in the frequency domain is not applicable to non-rectangular constellations or cannot quickly and effectively estimate the frequency offset.

[0020] (3) In this application embodiment, a coarse estimation scheme based on pre-screening method is also provided. By using constellation characteristics for pre-screening, not only can a fast coarse estimation of frequency offset be achieved, but also information support can be provided for subsequent synchronization operations, forming an effective synergy between coarse frequency offset estimation and synchronization processing, thereby achieving fast synchronization with low power consumption and low complexity.

[0021] (4) In this embodiment, coarse phase noise can also be calculated by pilot signal, and high dynamic frequency offset change and coarse phase noise can be quickly compensated by using fine frequency offset and coarse phase noise. The compensation circuit can be reused, which solves the problem of time-consuming and resource-intensive existing technologies. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of a coherent optical communication receiver link infrastructure. Figure 2 A schematic diagram of the conventional estimation and compensation loop for intermediate frequency offset in existing technology; Figure 3 This is a flowchart illustrating an embodiment of the frequency offset estimation and tracking method of this application; Figure 4 The first phase in the QPSK constellation chart is / 4+k / 2 constellation point diagram; Figure 5 The first phase in the 16QAM constellation chart is / 4+k / 2 constellation point diagram; Figure 6 The first phase in the 32QAM constellation chart is / 4+k / 2 constellation point diagram; Figure 7 A schematic diagram of the 32QAM constellation diagram; Figure 8 This is a schematic diagram of the rotation of the constellation points in the second ring of the 32QAM constellation diagram; Figure 9 Non-rotating constellations Figure 4 A schematic diagram of the distribution after the angle is doubled; Figure 10 For a constellation that rotates 0.4636 rad Figure 4 A schematic diagram of the distribution after the angle is doubled; Figure 11 For rotation / 8 Zodiac Signs Figure 4 A schematic diagram of the distribution after the angle is doubled; Figure 12 This is a schematic diagram of the estimation results of the original signal without rotation enhancement; Figure 13 This is a schematic diagram of the FFT estimation results after the second rotation enhancement. Figure 14 This is a schematic diagram of a basic frame structure; Figure 15 This is a schematic diagram of the density pre-screening method; Figure 16 This is a schematic diagram of the energy pre-screening method; Figure 17 This is a schematic diagram showing the result of initiating local correlation synchronization operations after using the density pre-screening method; Figure 18 This is a flowchart illustrating yet another embodiment of the frequency offset estimation and tracking method of this application; Figure 19 This is a flowchart illustrating another embodiment of the frequency offset estimation and tracking method of this application; Figure 20 This is a schematic diagram of the pilot quadruple-angle FFT frequency offset estimation result in an example; Figure 21 This is a schematic diagram of the functional modules of an embodiment of the frequency offset estimation and tracking device of this application; Figure 22 This is a schematic diagram illustrating the specific implementation of a frequency offset estimation and tracking device in an example. Figure 23 This is a schematic diagram illustrating a specific implementation of yet another embodiment of the frequency offset estimation and tracking device of this application; Figure 24 This is a schematic diagram illustrating a further embodiment of the frequency offset estimation and tracking device of this application. Detailed Implementation

[0023] To make the technical problems, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] However, it should be noted that the examples described below are merely specific examples and are not intended to limit the embodiments of the present invention to the specific steps, values, conditions, data, order, etc. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0025] Firstly, embodiments of this application provide a frequency offset estimation and tracking method. This method can be used in the frequency offset estimation process of coherent optical communication data transmission, see [link to relevant documentation]. Figure 1As shown.

[0026] In one embodiment, reference is made to Figure 3 , Figure 3 This is a flowchart illustrating an embodiment of the frequency offset estimation and tracking method of this application. Figure 3 As shown, a frequency offset estimation and tracking method includes: Step S10: Perform a coarse estimation of the frequency offset of the current main path to obtain a coarse frequency offset value Δf1; feed Δf1 back to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf1.

[0027] It is understandable that commonly used frequency offset estimation schemes include time-domain phase difference-based frequency offset estimation algorithms and frequency-domain FFT (Fast Fourier Transform)-based frequency offset estimation algorithms. In this embodiment, as a general approach, existing commonly used frequency offset estimation schemes, including time-domain phase difference-based frequency offset estimation algorithms and / or frequency-domain FFT-based frequency offset estimation algorithms, can be selected for coarse frequency offset estimation.

[0028] However, it is worth mentioning that the following special case may occur when using a frequency offset estimation algorithm based on FFT in the frequency domain.

[0029] When employing common FFT-based frequency offset estimation algorithms, such as blind estimation algorithms based on the fourth power or quadruple angle, a large number of FFT points (e.g., 256 points) is typically used to achieve high estimation accuracy. Subsampling is often employed to reduce the sampling rate while maintaining the same number of FFT points, with subsampling factors such as 1, 4, 16, 64, and 256, progressively increasing the estimation accuracy. This method can estimate the frequency offset of QPSK and 8 / 16 / 64QAM signals. However, for signals with non-rectangular constellation point distributions, such as 32QAM and 128QAM, commonly used FFT-based frequency offset estimation schemes cannot effectively obtain estimates.

[0030] Specifically, see Figures 4 to 6 As shown, the initial phases in the QPSK, 16QAM, and 32QAM constellation diagrams are respectively... / 4+k The constellation point distribution is 2 / 2. The constellation points marked with circles in the diagram, based on fourth-power or quadruple-angle algorithms, have their initial phase changes as follows: +2k The modulation phase is eliminated Other constellation points will become interference in frequency offset estimation. QPSK modulation phase can be eliminated; some points in 16QAM can be eliminated, while the remaining points will introduce interference, but the initial phase is... / 4+k Points with a ratio of 50% ( / 2) can be estimated normally; if the constellation points are not rectangularly distributed, such as in 32QAM, then the initial phase is... / 4+k The points with a ratio of 2 / 2 account for 25%, and are located in the inner circle of the constellation with low power. They are also greatly affected by interference introduced from the outer circle. It is difficult to obtain the desired estimate by directly using the fourth power or a multiple of four angles for FFT.

[0031] Therefore, to address the issue that frequency offset estimation algorithms based on FFT in the frequency domain are unsuitable for or cannot provide fast and effective frequency offset estimation for non-rectangular constellations, some possible implementations employ a ring rotation enhancement method when using frequency offset estimation algorithms based on FFT in the frequency domain for coarse estimation of signals with non-rectangular constellation points (such as 32QAM and 128QAM). This method involves selecting the second inner ring of the constellation diagram for uniform angle rotation and increasing the initial phase. / 4+k After calculating the constellation point ratio by half, a fourth-power or quadruple-angle frequency offset estimation is performed. In practical applications, the uniform rotation angle can be chosen to be 0.4636 rad or... / 8 (clockwise or counterclockwise is acceptable).

[0032] In this implementation, by using a ring rotation enhancement method, the FFT-based frequency offset estimation scheme can also be applied to non-rectangular constellations, effectively obtaining the estimated value. The specific principle and effect are explained below: See Figure 7 As shown, the 32QAM constellation diagram is divided into 5 rings, with the spacing between the rings gradually decreasing from the inside out. The spacing is particularly small between the 4th and 5th rings. When there is significant noise or interference, the rings completely overlap, making accurate ring separation almost impossible, especially after passing through fourth or quadruple angles where the noise is amplified. The initial phase of the constellation points in the 1st and 3rd rings is... / 4+k / 2, due to its low proportion (only 25%) and location in the inner circle, frequency offset estimation methods based on the fourth power or multiple angles are greatly affected by interference. Furthermore, the second circle has a larger gap with the first and third circles, making it easier to correctly segment the data compared to the outer circle. Therefore, the second circle data was extracted and rotated to increase the proportion of usable data.

[0033] Because the frequency offset constellation points have rotated, it is difficult to determine the specific initial phase of the extracted points on the second ring, making it impossible to adjust the initial phase for individual points. Therefore, the data on the second ring is rotated uniformly as a whole. Figure 8 Taking the first quadrant as an example, the initial phase of two points on the second loop is... The difference is 0.4636 rad. After rotating by 0.4636 rad (clockwise or counterclockwise), the initial phase is... / 4+k / 2 points increase by 4, after quadrupling the angle as follows Figure 9 and Figure 10 As shown, the initial phase marked by the square box is +2k The proportion of points increases. Alternatively, it can be approximated that the initial phase of the two points mentioned above in the second circle is equal to that of the first two points. / 4 difference is / 8, meaning these two points and the initial phase The line at 4 / 4 approximately divides the first quadrant angle into four equal parts. The second loop is the closest to being equally divisible among all non-QPSK loops. Therefore, after dividing the angle... After rotation / 8, four points are initially close to each other. / 4+k / 2. The four points are initially close to k / 2, after quadrupling the angle, as shown Figures 9 to 11 As shown, the initial phases marked by the square boxes are close. +2k With the increase in points, the introduced interference points marked by the circular box are initially smaller than those after a rotation of 0.4636 rad, and closer to the zero axis. The reason for not directly performing an 8x angle is that noise would be amplified more. By using optimal segmentation and angle rotation, simple calculations and convenient unified operation are achieved, increasing the proportion of effective points in the frequency offset estimation and improving the FFT estimation effect.

[0034] To illustrate the effectiveness of this implementation, we take a 32QAM signal with a baud rate of 189 GHz, a frequency offset of 1 GHz, 1x sampling, 64 FFT points, and 64 FFT output amplitude accumulations as an example. The estimation result of the original signal without rotation enhancement is as follows: Figure 12 As shown; the FFT estimation results after the second rotation enhancement are as follows Figure 13 As shown, the peak-to-average power ratio (PAPR) at 738.3 MHz is significantly enhanced, achieving coarse frequency offset estimation with a low FFT point count. Further FFT is needed to improve the estimation resolution, but the number of sampling levels is limited by the FFT point count, due to the low percentage of points used effectively during estimation. For example, a 128-point FFT allows for 1x, 4x, and 16x decimation. After 16x decimation, approximately 10 MHz of frequency offset remains, which, while significant, prevents further resolution improvement with 64x decimation. Increasing the FFT point count further, such as to 512 points, allows for 64x decimation, but this is costly. Therefore, for 32QAM, using an FFT-based algorithm and leveraging its constellation characteristics with a ring-rotation enhancement method can significantly improve the coarse frequency offset estimation results even with a low FFT point count.

[0035] In some other possible implementations, when using a time-domain frequency offset estimation algorithm based on phase difference for coarse estimation, a pre-screening method can be used: the pre-screening method is used to find the estimated position of the superframe preamble (i.e., the approximate position of the superframe preamble), and the frequency offset of the received one-shot data at this moment is estimated based on phase difference.

[0036] Understandably, for any high-order QAM modulation format such as 16QAM, 32QAM, 64QAM, 128QAM, and 256QAM, its constellation characteristics can be utilized—the preamble and pilot sequences in the frame structure are usually placed on the same circle of the constellation diagram. The pre-screening method can be used to quickly find the approximate position of the superframe preamble. The frequency offset estimation algorithm based on phase difference can be used on the received one-shot data at this moment. There is no need to pay attention to the specific values ​​and positions of a few non-preamble scattered points. Since the length of the one-shot data is relatively large, the interference it introduces in the calculation can be ignored.

[0037] Specifically, the sending end typically transmits data in fixed superframe structures as unit periods. For example... Figure 14 As shown, a basic frame structure includes a superframe preamble, data blocks, and pilots. If a superframe consists of multiple subframes, a subframe preamble may also exist. The subframe preamble is a short sequence, shorter than the superframe preamble. Pilots are inserted at fixed symbol intervals, and all preamble and pilot sequences are QPSK signal points. For the aforementioned higher-order modulation format signals, the preamble and pilot sequences are usually placed on the same circle of the constellation diagram, which can be denoted as the m-th circle, typically containing only four constellation points. For a single clock cycle of parallel data received, for example, with a length of 64 points, the approximate location of the superframe preamble can be quickly found using a pre-screening method based on its constellation characteristics.

[0038] Furthermore, in some possible implementations, the pre-screening method includes: density pre-screening and energy pre-screening.

[0039] Density pre-screening method: By comparing the sign modulus with the constellation circle radius, it quickly determines whether most of the content is within the m-th circle, for example, if the proportion is greater than the threshold th (e.g., th=80%). This density pre-screening method can quickly locate the approximate position of the superframe preamble. Figure 15 As shown. Figure 15 In the case where m=3, the number of symbols in the received data exceeding the threshold th are all located in the m-th cycle, while a few symbols are located in other cycles (e.g., Figure 15 (Scattered points within a small square frame). Density pre-screening has high accuracy and can be verified multiple times for assistance.

[0040] Energy pre-screening method: If the m-th cycle is located at or near the innermost cycle, the average energy of the data for that cycle will be significantly lower than the average energy; if the m-th cycle is located at or near the outermost cycle, the energy of the data for that cycle will be significantly higher than the average energy. These phenomena will occur periodically, and data segments typically will not have constant energy sign values; multiple checks and verifications are necessary to confirm this. Figure 16 As shown, this illustrates the change in average power of data per clock cycle when the m-th cycle is located at the outermost cycle.

[0041] When the predicted position of the superframe preamble is successfully found using the pre-screening method, a frequency offset estimation based on the phase difference can be performed on the received data at this moment to obtain the coarse frequency offset value Δf1. This process does not require attention to the specific values ​​and positions of a few non-preamble scatter points, as the interference introduced by these points in the calculation is ignored due to the large length of the data in one frame. The specific method can be as follows: first, perform a quadruple angle operation on the data in one frame, then perform differential summing and smoothing, and finally calculate the coarse frequency offset value Δf1. For example, if there is a 1 GHz frequency offset in the signal, using... Figure 15 The density pre-screening method shown uses the third round of density pre-screening with a threshold th=80%. The coarse frequency offset Δf1 obtained by the phase difference estimation is 824MHz.

[0042] In this embodiment, pre-screening using constellation characteristics not only enables rapid coarse frequency offset estimation but also provides information support for subsequent synchronization operations, forming an effective synergy between coarse frequency offset estimation and synchronization processing. This achieves rapid synchronization with low power consumption and low complexity. For a detailed analysis, please refer to the detailed explanation of synchronization processing below.

[0043] Step S20: Based on the predetermined superframe preamble prediction position, synchronization is performed by intermittently starting local correlation synchronization.

[0044] It is understood that synchronization in this embodiment requires a pre-determined estimated superframe preamble position. Therefore, if step S10 uses coarse frequency offset estimation based on constellation characteristics for density or energy pre-screening, the approximate position of the superframe preamble (i.e., the pre-determined estimated superframe preamble position) can be found during the pre-screening process. Then, in step 20, the approximate position of the superframe preamble known during the coarse frequency offset estimation process can be directly used for synchronization, forming an effective synergy between coarse frequency offset estimation and synchronization processing, thereby achieving low-power, low-complexity, and fast synchronization. Of course, if step S10 uses other coarse frequency offset estimation methods that do not reveal the approximate position of the superframe preamble, such as a frequency offset estimation algorithm based on FFT in the frequency domain, then before synchronization, any known superframe preamble estimation method or the pre-screening method proposed in this application needs to be used to determine the approximate position of the superframe preamble.

[0045] Unlike existing technologies, in step S20 of this embodiment, synchronization is performed using an intermittent local correlation synchronization method. For example, in some possible implementations, the intermittent local correlation synchronization method includes: determining whether the superframe preamble prediction position has been successfully determined; if so, initiating a preamble local symbol sliding correlation synchronization operation based on the superframe preamble prediction position; otherwise, not performing a symbol sliding correlation synchronization operation.

[0046] It is understandable that the superframe preamble is a continuous short sequence, which can tolerate a large range of residual frequency offsets, and the phase change of the end of the sequence relative to the beginning of the sequence does not exceed [a certain value]. / 2 or even The maximum peak value can be found through normal symbol correlation to determine the superframe preamble position and achieve synchronization. For example, if there is a 1 GHz frequency offset in the signal, it can be used as follows: Figure 15 The density pre-screening method shown, when the phase difference estimation yields a coarse frequency offset Δf1 = 794MHz with a residual of 206MHz, initiates the preamble local symbol sliding correlation synchronization operation, and the result is as follows. Figure 17 As shown.

[0047] In this embodiment, symbol sliding correlation synchronization is not performed when the predicted position of the superframe preamble is not successfully determined; local symbol sliding correlation is then enabled when it is successful. Compared to the traditional method of performing correlation synchronization for each symbol individually, the intermittent activation of local correlation synchronization proposed in this embodiment can significantly reduce power consumption and complexity. Furthermore, synchronization can be achieved quickly based on the pre-determined predicted position of the superframe preamble, significantly saving synchronization time. During synchronization, the accurate starting position of the superframe preamble can be calculated from the peak position, achieving low-complexity and fast synchronization.

[0048] Furthermore, in some possible implementations, step S20 further includes: after successful synchronization, pausing the execution of coarse frequency offset estimation and proceeding to the subsequent step S30; if a synchronization failure occurs, restarting the execution of coarse frequency offset estimation and returning to step S10. In this implementation, coarse frequency offset estimation and synchronization form a collaborative interaction, which not only quickly achieves coarse frequency offset estimation but also quickly achieves the effect of low-power and low-complexity synchronization. Moreover, there is mutual control between coarse frequency offset estimation and synchronization to achieve timely and stable switching of coarse frequency offset estimation mode, synchronization strategy, etc.

[0049] Step S30: After synchronization is completed, the frequency offset is precisely estimated using the periodically inserted pilot signal to obtain the precise frequency offset value Δf2. Δf2 and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 is compensated to the current main path to quickly achieve frequency offset tracking correction.

[0050] Understandably, this embodiment uses periodically inserted pilots for precise frequency offset estimation after synchronization. The sufficient number of pilots, preambles, and pilots with periodic insertions provides the capability to quickly track high-dynamic frequency offset changes. The precise frequency offset value Δf2 and the coarse frequency offset value Δf1 are fed back to the front-end frequency offset compensation module to achieve frequency offset feedback compensation, thus compensating for frequency offset changes. Simultaneously, the precise frequency offset value Δf2 is compensated into the current main path, enabling rapid frequency offset tracking and correction, achieving the goal of rapid frequency offset compensation. This embodiment, through the synergy of rapid frequency offset compensation and frequency offset feedback compensation, effectively solves the problems of long frequency offset estimation time, long compensation loop delay, and insufficient rapid tracking compensation capability in existing technologies under high-dynamic frequency offset scenarios.

[0051] Furthermore, in some possible implementations, step S30 further includes: calculating the coarse phase noise ΔL at the pilot location using the pilot signal, and using ΔL to quickly compensate for the coarse phase noise of the current main path. The specific calculation method for the coarse phase noise can employ existing commonly used algorithms, such as: multiplying the received pilot signal and the local pilot signal by their conjugates to calculate the phase angle; the smoothed phase angle is then the estimated coarse phase noise value. This embodiment does not impose specific limitations. It is understood that in this embodiment, Δf2 and ΔL can be simultaneously used after loop processing to simultaneously compensate for frequency offset and phase noise in the main path. ( The complex rotation factor that needs to be compensated for the signal is Δf, where Δf is the frequency offset and θ is the phase noise. The compensation circuit is reused to achieve simultaneous and rapid compensation for high dynamic frequency offset changes and coarse phase noise. Because the same compensation circuit can be reused and both frequency offset and phase noise are compensated simultaneously, the time-consuming and resource-intensive problems of existing technologies are solved.

[0052] Building upon this, in some possible implementations, the method further includes: precisely estimating and compensating for the phase noise to achieve precise phase recovery. For example, in practical applications, the corresponding constellation points can be obtained using sign-decision method, and the current phase noise can be precisely estimated and compensated. It is understood that in this implementation, phase compensation can be performed in two steps: first, using pilot signals to calculate and compensate for coarse phase noise in the previous implementation to achieve coarse phase recovery; second, the corresponding constellation points can be obtained using sign-decision method, and the current phase noise can be precisely estimated and compensated to achieve fine phase recovery. This two-stage compensation method, from coarse to fine, can quickly lock large-range phase errors and accurately track residual jitter, resulting in better practical application performance.

[0053] In another embodiment, reference is made to Figure 18 As shown, Figure 18 This is a flowchart illustrating another embodiment of the frequency offset estimation and tracking method of this application. Figure 18As shown, the difference in this embodiment lies in step S30': after synchronization is completed, the frequency offset is precisely estimated using the periodically inserted pilot to obtain the precise frequency offset value Δf2; the residual large frequency offset is estimated using the known superframe preamble sequence to obtain the residual large frequency offset value Δf3; Δf3, Δf2 and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 and Δf3 are compensated to the current main path to quickly achieve frequency offset tracking correction.

[0054] In this embodiment, while using periodically inserted pilots for precise frequency offset estimation, a known superframe preamble sequence is also used to estimate the residual large frequency offset. During frequency offset feedback compensation, the residual large frequency offset value Δf3, the precise frequency offset value Δf2, and the coarse frequency offset value Δf1 are fed back to the front end, forming a multi-level frequency offset feedback compensation. Simultaneously, during rapid frequency offset compensation, the residual large frequency offset value Δf3 and the precise frequency offset value Δf2 are also compensated to the current main path to quickly achieve frequency offset tracking correction. In this way, the preamble can be used to quickly pull back large frequency offsets (preventing flyaway), while the pilots can be used to continuously track small errors (maintaining accuracy). The two complement each other, ensuring both a large dynamic range and steady-state accuracy, resulting in good compensation and tracking effects.

[0055] In another embodiment, referring to Figure 19 As shown, Figure 19 This is a flowchart illustrating another embodiment of the frequency offset estimation and tracking method of this application. Figure 19 As shown, the difference in this embodiment is that the method further includes: Step S40: Collect N pilot data points, perform pilot quadruple angle FFT frequency offset estimation on them to obtain a high-precision frequency offset value Δf4; compensate Δf4 into the current main path and feed Δf4 back to the front-end frequency offset compensation module. Typically, the number of FFT points is a power of 2, such as 32 points, 64 points, 128 points, 256 points, etc.; therefore, N here is also a power of 2, such as 32, 64, 128, 256, etc.

[0056] Understandably, after performing fast frequency offset compensation (or, in some implementations, coarse phase recovery), N pilot data points can be collected, for example, N can be 64 points or other values. Performing a pilot four-fold angle FFT frequency offset estimation on these data points yields a high-precision frequency offset value Δf4. At this point, since the data involved in the calculation are all pilot data with an initial phase of... / 4+k / 2, without interference from other constellation points, the performance is expected to be further improved, and the obtained high-precision frequency offset Δf4 is more ideal. Furthermore, the N-point pilot data is obtained after the synchronization module. Since neither the pilot-based four-fold phase difference method nor the pilot-based four-fold FFT method requires knowledge of the pilot's specific position in the pilot sequence, as long as its initial phase is... / 4+k The pilot signal is only half the length of the standard pilot signal, meaning that it's not always necessary to use a method of first synchronizing the frame and then acquiring the pilot signal. Pilot detection methods, such as pilot energy period detection, can be used to determine the pilot signal without synchronization, thus shortening the estimation time. For example, when the pilot signal is placed on the outermost ring, high energy values ​​of single symbols will appear periodically. The pre-stage coarse frequency offset estimation, pilot-based fine frequency offset estimation, or even pre-leader-based large frequency offset estimation, have already limited the residual frequency offset to a small range. This reduces the need for loop filtering or smoothing operations to improve accuracy based on phase difference-based frequency offset estimation results, reducing processing latency and resources. A smaller FFT (e.g., N=64 points) can be used to obtain a more accurate estimate. The pilot signal is equivalent to sampling the signal. If the pilot signal is inserted at fixed intervals, such as 64 intervals, the sampling factor is 64, and the symbol rate of the N-point pilot data is reduced to 1 / 64 of the original signal. After estimation, the frequency offset (Δf4) at the peak is immediately transmitted to the main path for compensation, and it is also fed back to the front-end module for future data compensation.

[0057] For example, assuming a 1 GHz frequency offset in the signal, a coarse frequency offset estimate based on pre-screening yields 738.3 MHz, and a fine estimate based on pilot signals yields 249.6 MHz. Even after reducing loop filtering or smoothing to improve accuracy, the residual offset is relatively large at 12.1 MHz. Further frequency offset estimation using 64-point pilot data at a four-fold angle (FFT) yields a peak frequency offset of 11.5 MHz. Figure 20 As shown, the total residual frequency offset for the third level is estimated to be 0.6MHz.

[0058] Secondly, embodiments of this application also provide a frequency offset estimation and tracking device for implementing the method in the first aspect embodiment.

[0059] In one embodiment, reference is made to Figure 21 , Figure 21 This is a functional module diagram of an embodiment of the frequency offset estimation and tracking device of this application. Figure 21 As shown, a frequency offset estimation and tracking device includes a first-level frequency offset estimation module, a synchronization module, a second-level frequency offset estimation module, and a fast compensation module.

[0060] The first-level frequency offset estimation module is used to: coarsely estimate the frequency offset of the current main path to obtain a coarse frequency offset value Δf1; and feed Δf1 back to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf1.

[0061] Specifically, when performing a coarse estimation of frequency offset, the first-level frequency offset estimation module can select from existing commonly used frequency offset estimation schemes, including time-domain frequency offset estimation algorithms based on phase difference and / or frequency-domain frequency offset estimation algorithms based on FFT.

[0062] When using a frequency offset estimation algorithm based on FFT in the frequency domain, for signals with non-rectangular constellation point distributions (such as 32QAM and 128QAM), a ring-rotation enhancement method can be used: rotate the second inner ring of the constellation diagram at a uniform angle to increase the initial phase. / 4+k After calculating the constellation point ratio by 2, the frequency offset is estimated to the fourth power or a multiple of four angles.

[0063] When using a time-domain frequency offset estimation algorithm based on phase difference for coarse estimation, a pre-screening method can be used: The pre-screening method is used to find the estimated position of the superframe preamble (i.e., the approximate position of the superframe preamble), and then the frequency offset of the received data at that moment is estimated based on the phase difference. The pre-screening method includes density pre-screening and energy pre-screening.

[0064] The synchronization module is used to perform synchronization based on a pre-determined superframe preamble prediction position using an intermittent local correlation synchronization method. Specifically, the synchronization module uses an intermittent local correlation synchronization method, which may include the following operations: determining whether the superframe preamble prediction position has been successfully determined; if so, initiating a preamble local symbol sliding correlation synchronization operation based on the superframe preamble prediction position; otherwise, not performing a symbol sliding correlation synchronization operation.

[0065] The second-level frequency offset estimation module is used to: after synchronization is completed, use periodically inserted pilots to perform a fine estimation of frequency offset to obtain a fine frequency offset value Δf2; feed Δf2 back to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf2 and Δf1; and send Δf2 to the fast compensation module.

[0066] The fast compensation module is used to compensate Δf2 into the current main path to quickly achieve frequency offset tracking correction.

[0067] As described above, the frequency offset estimation and tracking device of this embodiment can realize a hierarchical frequency offset estimation and joint tracking scheme. It can not only coarsely estimate the frequency offset of the current main path using the first-level frequency offset estimation module, but also, after synchronization, finely estimate the frequency offset using the periodically inserted pilot signal using the second-level frequency offset estimation module. Frequency offset feedback compensation is achieved by feeding both the fine and coarse frequency offset values ​​back to the front-end frequency offset compensation module, thus compensating for frequency offset variations. Simultaneously, the fine frequency offset value can be compensated for in the current main path using the fast compensation module, enabling rapid frequency offset tracking correction and achieving the goal of rapid frequency offset compensation. Through this synergy between rapid frequency offset compensation and frequency offset feedback compensation, the problems of long frequency offset estimation time, long compensation loop delay, and insufficient rapid tracking compensation capability in scenarios with high frequency offset dynamic changes, as present in existing technologies, can be effectively solved.

[0068] Furthermore, in some possible implementations, the fast compensation module is also used to: calculate the coarse phase noise ΔL at the pilot frequency using the pilot signal, and use ΔL to perform coarse phase noise compensation on the current main path. In this embodiment, the fast compensation module can simultaneously use Δf2 and ΔL through loop processing to simultaneously compensate for frequency offset and phase noise on the main path, reusing the compensation circuit, thereby achieving simultaneous and rapid compensation for high dynamic frequency offset changes and coarse phase noise. Since the same compensation circuit can be reused and frequency offset and phase noise can be compensated simultaneously, the time-consuming and resource-intensive problems existing in the prior art are solved.

[0069] Furthermore, in some possible implementations, the device also includes a phase fine recovery module. This phase fine recovery module is used to: precisely estimate and compensate for phase noise to achieve precise phase recovery. In this embodiment, phase compensation is performed in two steps: first, a fast compensation module is used to calculate and compensate for coarse phase noise to achieve coarse phase recovery; second, a phase fine recovery module is used to finely calculate and compensate for the current phase noise, thereby achieving fine phase recovery. This two-stage compensation method, from coarse to fine, can quickly lock in a wide range of phase errors and accurately track residual jitter, resulting in better practical application performance.

[0070] For example, to better understand the frequency offset estimation and tracking device and its technical effects in this embodiment, the following will use coarse frequency offset estimation based on density pre-screening as an example, and illustrate the specific implementation process of the frequency offset estimation and tracking device in this embodiment with reference to the accompanying drawings. See Figure 22 As shown, Figure 22 This is a schematic diagram illustrating the specific implementation of a frequency offset estimation and tracking device in an example. For example... Figure 22 As shown, the specific implementation process of this frequency offset estimation and tracking device includes: 1. After the first-level frequency offset estimation module successfully finds the predicted position of the superframe preamble using the density pre-screening method, it performs frequency offset estimation based on the phase difference on the received one-frame data at this moment to obtain a coarse frequency offset value Δf1. Δf1 is fed back to the front-end frequency offset compensation module, which performs frequency offset feedback compensation on the main path based on Δf1.

[0071] 2. The first-level frequency offset estimation module will use the superframe preamble prediction position information (such as...) Figure 22 The position shown in the image is passed to the synchronization module; the synchronization module performs synchronization based on the predicted position information from the superframe preamble, using an intermittent local correlation synchronization method; after synchronization is complete, the synchronization information (such as...) is transmitted to the synchronization module. Figure 22 The synchronization information (as shown in the figure) is fed back to the first-level frequency offset estimation module. This synchronization information is divided into synchronization success information and out-of-step information. When the first-level frequency offset estimation module receives the synchronization information as synchronization success information, it pauses the execution of coarse frequency offset estimation. When it receives the synchronization information as out-of-step information, it restarts the execution of coarse frequency offset estimation.

[0072] 3. After synchronization, the second-level frequency offset estimation module uses the periodically inserted pilot signals to perform a fine estimation of the frequency offset, obtaining the fine frequency offset value Δf2. The second-level frequency offset estimation module feeds Δf2 back to the front-end frequency offset compensation module and simultaneously sends Δf2 to the fast compensation module. It can be understood that in this example, when the second-level frequency offset estimation module feeds Δf2 back to the front-end frequency offset compensation module, it does so along with the first-level frequency offset estimation module. In practical applications, the second-level frequency offset estimation module can also directly feed Δf2 back to the front-end frequency offset compensation module without going through the first-level frequency offset estimation module. This example is for illustrative purposes only and does not constitute a specific limitation.

[0073] 4. The front-end frequency offset compensation module performs frequency offset feedback compensation on the main path based on Δf2 and Δf1 (e.g., Figure 22 As shown in the figure, f_new = f_old + Δf1 + Δf2, where f_new refers to the frequency offset compensation value updated at the current moment and f_old refers to the frequency offset compensation value at the previous moment; at the same time, the fast compensation module compensates Δf2 into the current main path to quickly realize frequency offset tracking correction and complete the fast frequency offset compensation.

[0074] In this example, Δf2 and Δf1 are fed back to the front-end frequency offset compensation circuit to form a compensation and estimation loop; however, the new data after compensation needs to pass through the intermediate modules of the link before reaching the back-end module. Therefore, immediately compensating Δf2 into the current main path can achieve fast frequency offset compensation and tolerate high dynamic frequency offset changes such as at least 10MHz / µs and above. Figure 2 In the illustrated link, if the frequency offset is estimated using 5 levels of sampling FFT, it requires not only data storage but also a long clock cycle to complete data extraction, as well as delays in intermediate modules, making it difficult to achieve fast frequency offset tracking. In this example, however, frequency offset feedback compensation and fast frequency offset compensation are performed simultaneously, effectively solving the problem of tracking high dynamic frequency offset changes.

[0075] In another embodiment, reference is made to Figure 23 , Figure 23 This is a schematic diagram illustrating a specific implementation of another embodiment of the frequency offset estimation and tracking device of this application. Figure 23 As shown, the difference in this embodiment is that in this frequency offset estimation and tracking device, the synchronization module is further used to: estimate the residual large frequency offset using a known superframe preamble sequence to obtain the residual large frequency offset value Δf3; feed Δf3 back to the front-end frequency offset compensation module, and simultaneously send Δf3 to the fast compensation module. At this time, the front-end frequency offset compensation module will perform frequency offset feedback compensation on the main path based on Δf3, Δf2, and Δf1 (e.g., ...). Figure 23As shown in the figure, f_new = f_old + Δf1 + Δf2 + Δf3); at the same time, the fast compensation module will compensate Δf2 and Δf3 together into the current main path to quickly achieve frequency offset tracking correction and complete the fast frequency offset compensation.

[0076] In another embodiment, referring to Figure 24 , Figure 24 This is a schematic diagram illustrating another embodiment of the frequency offset estimation and tracking device of this application. Figure 24 As shown, this embodiment differs in that the frequency offset estimation and tracking device also includes a third-level frequency offset estimation module. This third-level frequency offset estimation module is used for: collecting N pilot data points after processing by the fast compensation module, performing pilot four-fold FFT frequency offset estimation on them to obtain a high-precision frequency offset value Δf4; feeding back Δf4 to the front-end frequency offset compensation module, and simultaneously sending Δf4 to the fast compensation module. At this time, the front-end frequency offset compensation module will perform frequency offset feedback compensation on the main path based on Δf4, Δf3, Δf2, and Δf1 (e.g., ...). Figure 24 As shown in the figure, f_new = f_old + Δf1 + Δf2 + Δf3 + Δf4); at the same time, the fast compensation module will compensate Δf2, Δf3 and Δf4 together into the current main path to quickly realize frequency offset tracking correction and complete the fast frequency offset compensation.

[0077] It should be noted that the various variations and specific examples in the foregoing method embodiments are also applicable to the device in this embodiment. Through the detailed description of the foregoing method, those skilled in the art can clearly understand the implementation method of the device in this embodiment. Therefore, for the sake of brevity, they will not be described in detail here.

[0078] Note: The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0079] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0080] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0081] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0083] The above description has been given for illustrative and descriptive purposes. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein. Moreover, anything not described in detail in this specification is prior art well known to those skilled in the art.

Claims

1. A frequency offset estimation and tracking method, characterized in that, The method includes: A coarse estimate of the frequency offset of the current main path is obtained to obtain a coarse frequency offset value Δf1; Δf1 is fed back to the frequency offset compensation module at the front end so that the frequency offset compensation module at the front end can perform frequency offset feedback compensation on the main path based on Δf1. Synchronization is performed by intermittently starting local correlation synchronization based on the pre-determined superframe preamble prediction position; After synchronization is completed, the frequency offset is precisely estimated using periodically inserted pilots to obtain the precise frequency offset value Δf2. Δf2 and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 is also compensated into the current main path to achieve frequency offset tracking correction.

2. The frequency offset estimation and tracking method as described in claim 1, characterized in that: When making a coarse estimate of the frequency offset of the current main path, a time-domain frequency offset estimation algorithm based on phase difference and / or a frequency-domain frequency offset estimation algorithm based on FFT are used.

3. The frequency offset estimation and tracking method as described in claim 2, characterized in that, For signals with non-rectangular constellation point distributions, when using a frequency offset estimation algorithm based on FFT in the frequency domain for coarse estimation, a ring rotation enhancement method is employed: Choose to rotate the second inner circle of the constellation chart at a uniform angle to increase the initial phase. / 4+k After calculating the constellation point ratio by 2, the frequency offset is estimated to the fourth power or a multiple of four angles.

4. The frequency offset estimation and tracking method as described in claim 3, characterized in that: The uniform rotation angle is 0.4636 rad or / 8.

5. The frequency offset estimation and tracking method as described in claim 2, characterized in that, When using a time-domain frequency offset estimation algorithm based on phase difference for coarse estimation, a pre-screening method is employed: The pre-screening method is used to find the predicted position of the superframe preamble, and the frequency offset is estimated based on the phase difference of the received data at this moment.

6. The frequency offset estimation and tracking method as described in claim 5, characterized in that: The pre-screening methods include density pre-screening and energy pre-screening.

7. The frequency offset estimation and tracking method as described in claim 1, characterized in that, The method of intermittently starting local correlation synchronization for synchronization includes: Determine whether the superframe preamble prediction position has been successfully determined. If so, initiate the preamble local symbol sliding correlation synchronization operation based on the superframe preamble prediction position; otherwise, do not perform the symbol sliding correlation synchronization operation.

8. The frequency offset estimation and tracking method as described in claim 1, characterized in that, The method also includes: pausing the coarse estimation of frequency offset after successful synchronization; and restarting the coarse estimation of frequency offset once a loss of synchronization occurs.

9. The frequency offset estimation and tracking method as described in claim 1, characterized in that: While compensating Δf2 into the current main path, the coarse phase noise ΔL at the pilot is also calculated through the pilot, and ΔL is used to compensate the coarse phase noise of the current main path.

10. The frequency offset estimation and tracking method as described in claim 9, characterized in that, The method also includes: accurately estimating and compensating for phase noise.

11. The frequency offset estimation and tracking method as described in claim 1, characterized in that: The frequency offset is precisely estimated using periodically inserted pilot signals to obtain the precise frequency offset value Δf2. At the same time, the residual large frequency offset is estimated using the known superframe preamble sequence to obtain the residual large frequency offset value Δf3. Δf3, Δf2 and Δf1 are fed back to the frequency offset compensation module at the front end, and Δf2 and Δf3 are compensated to the current main path to achieve frequency offset tracking correction.

12. The frequency offset estimation and tracking method as described in claim 1, characterized in that, The method also includes: collecting N pilot data points, performing pilot quadruple angle FFT frequency offset estimation on them to obtain a high-precision frequency offset value Δf4; compensating Δf4 into the current main path and feeding Δf4 back to the front-end frequency offset compensation module.

13. A frequency offset estimation and tracking apparatus for implementing the method of any one of claims 1 to 12, characterized in that: The frequency offset estimation and tracking device includes a first-level frequency offset estimation module, a synchronization module, a second-level frequency offset estimation module, and a fast compensation module; The first-level frequency offset estimation module is used to: coarsely estimate the frequency offset of the current main path to obtain a coarse frequency offset value Δf1; and feed back Δf1 to the front-end frequency offset compensation module so that the front-end frequency offset compensation module can perform frequency offset feedback compensation on the main path based on Δf1. The synchronization module is used to: perform synchronization by intermittently starting local correlation synchronization based on a pre-determined superframe preamble estimated position; The second-level frequency offset estimation module is used to: after synchronization is completed, use periodically inserted pilots to perform fine frequency offset estimation to obtain the fine frequency offset value Δf2; Δf2 is fed back to the frequency offset compensation module at the front end, so that the frequency offset compensation module at the front end can perform frequency offset feedback compensation on the main path based on Δf2 and Δf1; and Δf2 is sent to the fast compensation module. The fast compensation module is used to compensate Δf2 into the current main path to achieve frequency offset tracking correction.

Images