Coarse frequency offset estimation compensation method and device

Abstract

The invention discloses a coarse frequency offset estimation compensation method and device, and relates to the technical field of communication, and the method comprises the steps: in a clock period, selecting a continuous sequence of one polarization from two polarizations; performing frequency offset compensation on the sequence according to the initial frequency offset estimation value in the positive and negative frequency offset compensation directions or the frequency offset estimation value transmitted in the previous clock period to obtain two output sequences in the positive and negative frequency offset compensation directions; determining two frequency point positions with the closest frequency point sequence number interval and the maximum amplitude difference value, and determining a first module value difference and a second module value difference of the two frequency points in the positive and negative frequency offset compensation direction according to the two frequency point positions and the two output sequences; and determining the direction and size of frequency offset compensation in the next clock period according to the first module value difference and the second module value difference, repeatedly performing frequency offset compensation until an iteration condition is met, and outputting a frequency offset estimation value referred by the first module value difference or the second module value difference of the corresponding clock period. According to the invention, coarse frequency offset estimation and compensation functions in an optical transmission channel can be realized, and the realization complexity is low.

Description

Technical Field

[0001] This application relates to the field of communication technology, specifically to a coarse frequency offset estimation and compensation method and apparatus. Background Technology

[0002] In the field of fiber optic communication technology, when using coherent optical communication systems, it is required that the lasers at the receiving and transmitting ends operate at the same frequency. However, due to limitations in laser manufacturing processes, the center frequencies cannot be made exactly the same, resulting in a certain frequency deviation. Low-cost transceiver lasers may produce even larger frequency deviations, potentially exceeding the compensation capabilities of current traditional fourth-power algorithms. If frequency deviations are not pre-estimated and compensated, adaptive equalization modules may malfunction, thereby negatively impacting the receiver's equalization capabilities.

[0003] Coherent optical communication systems mainly use phase increment algorithms and frequency offset estimation algorithms based on spectral power for frequency offset estimation.

[0004] The phase increment algorithm uses a fourth-power operation to remove the modulated data phase and calculates the frequency offset based on the average of the phase increments of adjacent symbols. However, the frequency offset estimation range of this algorithm is only ±Rs / 8 (where Rs is the baud rate), which is relatively small.

[0005] Frequency offset estimation algorithms based on spectral power utilize FFT (Fast Fourier Transform) to transform the oversampled data at the receiver to the frequency domain, and then obtain the power of the symmetrical spectrum through filtering, thereby estimating the frequency offset. However, this algorithm has high implementation complexity, making it difficult to balance complexity and system performance. Summary of the Invention

[0006] This application provides a coarse frequency offset estimation and compensation method and apparatus, which can realize coarse frequency offset estimation and compensation functions in optical transmission channels, and has low implementation complexity.

[0007] In a first aspect, embodiments of this application provide a coarse frequency offset estimation and compensation method, which includes the following steps: Within this clock cycle, a continuous sequence of one polarization is selected from the two polarizations. Based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, frequency offset compensation is performed on the sequence of the current clock cycle to obtain two output sequences in the positive and negative frequency offset compensation directions. Determine the positions of the two frequency points with the closest frequency index and the largest amplitude difference. Based on the positions of the two frequency points and the two output sequences, determine the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction. The direction and magnitude of frequency offset compensation for the next clock cycle are determined based on the first and second modulus differences. Frequency offset compensation is repeated until the iteration condition is met. The frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle is output as the final coarse frequency offset estimate.

[0008] In conjunction with the first aspect, in one implementation, determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction, based on the two frequency point positions and the two output sequences, includes: Call the dispersion compensation module to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; Based on the FFT output, select two frequency points of the signal and calculate the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction, respectively. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction.

[0009] In conjunction with the first aspect, in one implementation, determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction, based on the two frequency point positions and the two output sequences, includes: Multiply the two output sequences by the butterfly transform tables at the two frequency points respectively, sum them, and take the magnitude to obtain the magnitudes of the two frequency points in the positive frequency offset compensation direction and the two frequency points in the negative frequency offset compensation direction. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction. The butterfly transform table for two frequency points is as follows:

[0010] in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

[0011] In conjunction with the first aspect, in one implementation method, According to the formula: Determine the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; in, The roll-off factor of the pulse shaping filter. The oversampling rate of the analog-to-digital converter (ADC) The length of the selected polarization signal.

[0012] In conjunction with the first aspect, in one implementation, the step of determining the direction and magnitude of frequency offset compensation for the next clock cycle based on the first modulus difference and the second modulus difference, repeating frequency offset compensation until the iteration condition is met, and outputting the frequency offset estimate indicated by the first modulus difference or the second modulus difference for the corresponding clock cycle as the final coarse frequency offset estimate includes: Determine the maximum value between the first modulus difference and the second modulus difference. If the maximum value is greater than the first threshold, set the frequency offset compensation step size to a small step size. If the maximum value is less than or equal to the first threshold, set the frequency offset compensation step size to a large step size. Based on the initial frequency offset estimate or the frequency offset estimate passed from the previous clock cycle, and the determined frequency offset compensation step size, determine the frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle. If the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, the iteration condition is satisfied, and the frequency offset estimate indicated by the first or second modulus value of the corresponding clock cycle is taken as the final coarse frequency offset estimate. If the following conditions are not met: the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, it is determined that the iteration condition is not met. The frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle are then passed to the next clock cycle to continue the coarse frequency offset estimation and compensation process.

[0013] In conjunction with the first aspect, in one implementation, the small step size is 300MHz and the large step size is 3GHz.

[0014] In conjunction with the first aspect, in one embodiment, before determining the first modulus difference between two frequency points in the positive frequency offset compensation direction and the second modulus difference between two frequency points in the negative frequency offset compensation direction, the method further includes: The steps for filtering the magnitude values ​​of two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of two frequency points in the negative frequency offset compensation direction are as follows: If the number of filtering iterations has not been reached, the frequency offset estimate for this clock cycle will remain consistent with the frequency offset estimate for the previous clock cycle. If the number of filtering iterations is reached, the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction are calculated.

[0015] Secondly, embodiments of this application provide a coarse frequency offset estimation and compensation device, the coarse frequency offset estimation and compensation device comprising: The frequency offset compensation module is used to select a continuous sequence of one polarization from two polarizations, and to perform frequency offset compensation on the sequence of the current clock cycle based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, so as to obtain two output sequences in the positive and negative frequency offset compensation directions. The Fourier transform module is used to determine the positions of the two frequency points with the closest frequency index and the largest amplitude difference. Based on the two frequency point positions and the two output sequences, it determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction. The frequency offset estimation module determines the direction and magnitude of frequency offset compensation for the next clock cycle based on the first and second modulus differences. It repeats the frequency offset compensation until the iteration condition is met, and outputs the frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

[0016] In conjunction with the second aspect, in one implementation, the Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Call the dispersion compensation module to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; Based on the FFT output, select two frequency points of the signal and calculate the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction, respectively. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction.

[0017] In conjunction with the second aspect, in one implementation, the Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Multiply the two output sequences by the butterfly transform tables at the two frequency points respectively, sum them, and take the magnitude to obtain the magnitudes of the two frequency points in the positive frequency offset compensation direction and the two frequency points in the negative frequency offset compensation direction. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction. The butterfly transform table for two frequency points is as follows:

[0018] in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

[0019] The beneficial effects of the technical solutions provided in this application include at least the following: The coarse frequency offset estimation and compensation method in this application involves selecting a continuous sequence of two polarizations within the current clock cycle; performing frequency offset compensation on the sequence of the current clock cycle based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, resulting in two output sequences; determining the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences; determining the direction and magnitude of frequency offset compensation for the next clock cycle based on the first modulus difference and the second modulus difference; repeating the frequency offset compensation until the iteration condition is met; and outputting the frequency offset estimate indicated by the first modulus difference or the second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

[0020] This application determines the direction of the next frequency offset compensation by judging the power in the positive and negative directions at two frequency points. It eliminates the need for real-time frequency offset calculation, providing only a coarse frequency offset estimate and direction to achieve coarse frequency offset estimation and compensation in optical transmission channels, thus reducing complexity. Attached Figure Description

[0021] Figure 1 A schematic diagram of a module applicable to the field of optical fiber communication technology; Figure 2 This is a flowchart illustrating an embodiment of the coarse frequency offset estimation and compensation method of this application; Figure 3 This is a schematic diagram of the entire process of the coarse frequency offset estimation and compensation method in this application; Figure 4 This is a structural block diagram of an embodiment of the coarse frequency offset estimation and compensation device of this application; Figure 5 This is a graph showing the estimated value versus the number of iterations for the coarse frequency offset estimation algorithm in the coarse optical communication system presented in this application; Figure 6 This is a spectrum diagram of frequency offset estimation based on dual-frequency power on a single-sided spectrum in a coherent optical communication system as shown in this application; Figure 7 This is a convergence effect diagram of the magnitude estimation based on IIR filters in a coherent optical communication system as shown in this application. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] It is worth noting that this application can be applied to digital coherent receivers, and is suitable for coherent receivers in the field of fiber optic communication technology, such as... Figure 1 As shown, the typical DSP (Digital Signal Processing) workflow of a digital coherent receiver includes coarse frequency offset estimation and compensation, polarization demultiplexing, clock recovery, phase recovery, and other processes and corresponding modules. In the field of optical fiber communication technology, the dispersion compensation FFT module can be multiplexed, and the coarse frequency offset estimation and compensation module can be coupled with the dispersion compensation module, thereby improving module utilization and reducing complexity. Figure 1 The location for estimating and compensating for the coarse frequency offset is the location where this application can be placed.

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0029] In a first aspect, embodiments of this application provide a coarse frequency offset estimation and compensation method.

[0030] In one embodiment, reference is made to Figure 2 , Figure 2 This is a schematic flowchart of an embodiment of the coarse frequency offset estimation and compensation method of this application. Figure 2 As shown, the coarse frequency offset estimation compensation method includes: S1. Within this clock cycle, select a continuous sequence of one polarization from the two polarizations.

[0031] Specifically, the selected continuous polarization signal can be either H-polarization or V-polarization, with a length that is a power of 2, i.e., the length can be 64, 128, 256, 512, 1024, etc. The longer the length, the higher the implementation complexity and the longer the estimation time, but the more accurate the coarse frequency offset estimate.

[0032] S2. Based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, perform frequency offset compensation on the sequence of the current clock cycle to obtain two output sequences in the positive and negative frequency offset compensation directions.

[0033] In step S2, frequency offset compensation is performed based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions, or the frequency offset estimate passed in the previous clock cycle in the positive and negative frequency offset compensation directions. The phase deviation caused by the frequency offset compensation values ​​in the two directions can be expressed in exponential form. and .

[0034] S3. Determine the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference. Based on the positions of the two frequency points and the two output sequences, determine the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction.

[0035] Specifically, it can be done according to the formula: Determine the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; in, The roll-off factor of the pulse shaping filter. The oversampling rate of the analog-to-digital converter (ADC) The length of the selected polarization signal.

[0036] Since it is not necessary to perform Fourier transform on all sequences, only two frequency points are used, and the complexity is greatly reduced.

[0037] When applied to the field of optical fiber communication technology, the dispersion compensation module can be directly invoked to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; based on the FFT outputs, signals at two frequency points are selected, and the magnitudes of the two frequency points in the positive frequency offset compensation direction and the magnitudes of the two frequency points in the negative frequency offset compensation direction are calculated respectively; the first magnitude difference is calculated based on the magnitudes of the two frequency points in the positive frequency offset compensation direction, and the second magnitude difference is calculated based on the magnitudes of the two frequency points in the negative frequency offset compensation direction.

[0038] Alternatively, the two output sequences can be multiplied by the butterfly transform table of the two frequency points respectively, and the summation can be taken to obtain the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction. The first magnitude difference is calculated based on the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction, and the second magnitude difference is calculated based on the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction.

[0039] The butterfly transformation table for two frequency points can be generated in advance in the hardware implementation and obtained by looking up the table, without the need for real-time calculation in the hardware, thus saving hardware resources.

[0040] This table can be generated using the following formula:

[0041] in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

[0042] In practice, the power of the signal is not calculated directly, but the magnitude is used instead. The magnitude can be calculated using the Cordic (coordinate rotation algorithm). The same-direction and quadrature components of each signal are input into the Cordic circuit, which calculates the magnitude of the signal.

[0043] It is worth noting that, in order to obtain a coarse estimate of the frequency offset even at a very low signal-to-noise ratio, before determining the first magnitude difference between the two frequency points in the positive frequency offset compensation direction and the second magnitude difference between the two frequency points in the negative frequency offset compensation direction, the following steps are included: filtering the magnitudes of the two frequency points in the positive and negative frequency offset compensation directions. If the number of filtering iterations has not been reached, the frequency offset estimate for the current clock cycle remains consistent with the frequency offset estimate for the previous clock cycle; if the number of filtering iterations has been reached, the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction are calculated.

[0044] S4. Determine the direction and magnitude of frequency offset compensation for the next clock cycle based on the first and second modulus differences. Repeat the frequency offset compensation until the iteration condition is met. Output the frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

[0045] Specifically, step S4 includes: S41. Determine the maximum value between the first modulus difference and the second modulus difference. If the maximum value is greater than the first threshold, set the frequency offset compensation step size to a small step size. If the maximum value is less than or equal to the first threshold, set the frequency offset compensation step size to a large step size. Preferably, the small step size is 300MHz and the large step size is 3GHz. It is understood that, of course, these can be reasonably set according to actual circumstances, and this embodiment does not impose any limitations.

[0046] S42. Based on the initial frequency offset estimate or the frequency offset estimate passed from the previous clock cycle, and the determined frequency offset compensation step size, determine the frequency offset value estimated in the positive and negative frequency offset compensation directions for this clock cycle. S43. If the maximum value is greater than the second threshold, and the absolute value of the difference between the first modulus difference and the second modulus difference is less than the third threshold, it is determined that the iteration condition is met, and the frequency offset estimate indicated by the first modulus difference or the second modulus difference of the corresponding clock cycle is taken as the final coarse frequency offset estimate. S44. If the following conditions are not met: the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, it is determined that the iteration condition is not met. The frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle are passed to the next clock cycle, and the coarse frequency offset estimation and compensation process continues.

[0047] In other words, if either of the following two conditions is not met, the iteration condition is determined not to be satisfied: the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold.

[0048] See Figure 3As shown, the following is a concrete and complete example to further illustrate the above steps: Step 500: Within this clock cycle, select one of the two polarizations, and select a continuous sequence of that polarization according to the preset Fourier transform length.

[0049] The selected continuous polarization signal can be either H-polarization or V-polarization, with a length that is a power of 2, i.e., the length can be 64, 128, 256, 512, 1024, etc. The longer the length, the higher the implementation complexity, but the more accurate the coarse frequency offset estimate. For example, the selected Fourier transform length is 128.

[0050] Step 501: Multiply the phase deviation values ​​caused by the frequency offset compensation values ​​in the two frequency offset compensation directions by the continuous sequence respectively.

[0051] The phase deviation caused by the frequency offset compensation values ​​in both directions can be expressed in exponential form. and The results of multiplying a continuous sequence can be denoted as sig1 and sig2.

[0052] Step 502: Multiply the two output sequences by the butterfly transform table, sum them, and calculate the power to obtain four values.

[0053] The two frequency points for Fourier transform can be selected using the following formula. Since it is not necessary to perform Fourier transform on all sequences, only two frequency points are used, which greatly reduces the complexity. This is the roll-off factor for the pulse shaping filter. A more realistic filter roll-off can also be obtained based on the actual channel. The oversampling rate of the ADC. The length of the polarization signal selected in step 500. The calculated positions of the two frequency points can be rounded to integer values.

[0054] The two frequency points can be set as point(1) and point(2), for example, positions 28 and 37.

[0055] This table can be generated using the following formula, where point(1) and point(2) are the two frequency points mentioned above.

[0056]

[0057] in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

[0058] Specifically, the table with a length of 2 and a depth of 128 is shown below:

[0059] When applied to the field of optical fiber communication technology, the dispersion compensation module can be directly called, and its FFT output can be used to generate a Fourier transform to compensate for positive and negative frequency offsets, and then its magnitude can be taken. That is, the FFT module can be used to obtain the FFT outputs of sig1 and sig2, and the signals at point(1) and point(2) can be selected and their magnitudes calculated.

[0060] Alternatively, you can multiply sig1 and sig2 by the butterfly transform table of the two frequencies respectively, then sum them, take the magnitude, and finally get four values: 1. Frequency 1 in the positive direction, ; 2. Frequency 1 in the negative direction, ; 3. Two frequency points in the positive direction. ; 4. Two frequency points in the negative direction. ; In practice, the power of the signal is generally not calculated directly, but the magnitude is used instead. The power and magnitude mentioned below refer to the magnitude of the signal. The magnitude can be calculated using the Cordic (coordinate rotation algorithm). The same-direction and quadrature components of each signal are input into the Cordic circuit, which calculates the magnitude of the signal.

[0061] Step 503: IIR filtering yields 4 outputs.

[0062] The signal magnitude can be filtered using a simple first-order IIR filter. The time-domain expression of the filter for signal filtering is as follows:

[0063] in, Indicates the first iteration This is the decay factor; the smaller the value, the slower the convergence and the smaller the steady-state error. A typical value can be 1e-3. There are 4 outputs. The four modulus values ​​obtained in step 502 are: .

[0064] Step 504: Determine whether the number of IIR filter iterations has been reached.

[0065] If the number of iterations set by the system is not reached, execute step 510: end the frequency offset estimation process for this clock cycle, and keep the frequency offset estimation value for this clock cycle consistent with the frequency offset estimation value for the previous clock cycle.

[0066] When the system-set number of iterations is reached, proceed to step 505.

[0067] Step 505: Calculate the power difference between the two frequency offset compensation directions.

[0068] The power difference between the two directions refers to the difference between frequency 1 and frequency 2 in the positive direction, and the difference between frequency 1 and frequency 2 in the negative direction. and They are M1 and M2, respectively.

[0069] Step 506: Select the frequency offset compensation step size based on the power difference.

[0070] When the maximum value of M1 and M2 is greater than the first threshold, the frequency offset compensation step size for the next clock cycle is set to a small step size; otherwise, the frequency offset compensation step size for the next clock cycle is set to a large step size. The selectable small step size is 300MHz, and the selectable large step size is 3GHz.

[0071] Step 507: Determine the final coarse frequency offset estimate based on preset conditions.

[0072] During the first frequency offset estimation, the frequency offset values ​​estimated in the two positive and negative directions within this clock cycle are the positive and negative step values. That is, the frequency offset values ​​estimated in the two positive and negative directions within this clock cycle = ±step. When step is a large step size, the estimated frequency offset values ​​in the two positive and negative directions at this time = ±3GHz.

[0073] The second frequency offset estimation begins. The frequency offset values ​​estimated in the two positive and negative directions in this clock cycle are the frequency offset values ​​indicated by the maximum value of M1 or M2, plus or minus the step size. Assuming M1 is the maximum value, it refers to the positive frequency offset value obtained in the previous clock cycle. That is, the frequency offset values ​​estimated in the two positive and negative directions in this clock cycle = the positive frequency offset value obtained in the previous clock cycle ± step.

[0074] Assuming the current clock cycle is still a large step of 3GHz, the frequency offset values ​​in the positive and negative directions obtained in the previous clock cycle are ±3GHz. M1 is the maximum value in the current clock cycle. Therefore, the frequency offset values ​​in the two positive and negative directions estimated in the current clock cycle are 3GHz ± 3GHz = [6GHz 0GHz].

[0075] When the maximum value of M1 and M2 is greater than the second threshold and the absolute value of the difference between M1 and M2 is less than the third threshold, the preset condition is met, and step 508 is executed.

[0076] Step 508: End the frequency offset estimation for this clock cycle, and also end the coarse frequency offset estimation. Output the frequency offset estimate value indicated by M1 or M2 as the final coarse frequency offset estimate value.

[0077] Otherwise, it is considered that the preset conditions have not been met, and step 509 is executed.

[0078] Step 509: End the frequency offset estimation for this clock cycle, and pass the frequency offset values ​​estimated in the two positive and negative directions for this clock cycle to the next clock cycle to continue the coarse frequency offset estimation process.

[0079] In summary, the coarse frequency offset estimation and compensation method in this application involves selecting a continuous sequence of one polarization from two polarizations within the current clock cycle; performing frequency offset compensation on the sequence of the current clock cycle based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, resulting in two output sequences in the positive and negative frequency offset compensation directions; determining the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences; determining the direction and magnitude of frequency offset compensation for the next clock cycle based on the first modulus difference and the second modulus difference; repeating the frequency offset compensation until the iteration condition is met; and outputting the frequency offset estimate indicated by the first modulus difference or the second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

[0080] This application determines the direction of the next frequency offset compensation by judging the power in the positive and negative directions at two frequency points. It eliminates the need for real-time frequency offset calculation, providing only a coarse frequency offset estimate and direction, thus reducing complexity. Furthermore, it suppresses white noise using IIR filtering, enabling the acquisition of a coarse frequency offset estimate even at very low signal-to-noise ratios. In addition, this application employs blind estimation, eliminating the need for prior synchronization and reliance on known training sequences or pilots, resulting in a wide frequency offset estimation range, exceeding the maximum ±Rs / 8 range.

[0081] Secondly, embodiments of this application also provide a coarse frequency offset estimation compensation device.

[0082] In one embodiment, reference is made to Figure 4 , Figure 4 This is a functional module diagram of an embodiment of the coarse frequency offset estimation and compensation device of this application. Figure 4 As shown, the coarse frequency offset estimation and compensation device includes: The frequency offset compensation module is used to select a continuous sequence of one polarization from two polarizations, and to perform frequency offset compensation on the sequence of the current clock cycle based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, so as to obtain two output sequences in the positive and negative frequency offset compensation directions. The Fourier transform module is used to determine the positions of the two frequency points with the closest frequency index and the largest amplitude difference. Based on the two frequency point positions and the two output sequences, it determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction. The frequency offset estimation module determines the direction and magnitude of frequency offset compensation for the next clock cycle based on the first and second modulus differences. It repeats the frequency offset compensation until the iteration condition is met, and outputs the frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

[0083] Further, in one embodiment, the Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Call the dispersion compensation module to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; Based on the FFT output, select two frequency points of the signal and calculate the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction, respectively. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction.

[0084] Furthermore, in another embodiment, the Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Multiply the two output sequences by the butterfly transform tables at the two frequency points respectively, sum them, and take the magnitude to obtain the magnitudes of the two frequency points in the positive frequency offset compensation direction and the two frequency points in the negative frequency offset compensation direction. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction. The butterfly transform table for two frequency points is as follows:

[0085] in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

[0086] Furthermore, in one embodiment, the Fourier transform module is used for: According to the formula: Determine the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; in, The roll-off factor of the pulse shaping filter. The oversampling rate of the analog-to-digital converter (ADC) The length of the selected polarization signal.

[0087] Further, in one embodiment, the frequency offset estimation module determines the direction and magnitude of the frequency offset compensation for the next clock cycle based on the first modulus difference and the second modulus difference, repeats the frequency offset compensation until the iteration condition is met, and outputs the frequency offset estimate value indicated by the first modulus difference or the second modulus difference for the corresponding clock cycle as the final coarse frequency offset estimate value, including: Determine the maximum value between the first modulus difference and the second modulus difference. If the maximum value is greater than the first threshold, set the frequency offset compensation step size to a small step size. If the maximum value is less than or equal to the first threshold, set the frequency offset compensation step size to a large step size. Based on the initial frequency offset estimate or the frequency offset estimate passed from the previous clock cycle, and the determined frequency offset compensation step size, determine the frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle. If the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, the iteration condition is satisfied, and the frequency offset estimate indicated by the first or second modulus value of the corresponding clock cycle is taken as the final coarse frequency offset estimate. If the following conditions are not met: the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, it is determined that the iteration condition is not met. The frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle are then passed to the next clock cycle to continue the coarse frequency offset estimation and compensation process.

[0088] Furthermore, in one embodiment, the small step size is 300MHz and the large step size is 3GHz.

[0089] Furthermore, in one embodiment, a filtering module is also included. Before determining the first modulus difference between two frequency points in the positive frequency offset compensation direction and the second modulus difference between two frequency points in the negative frequency offset compensation direction, the filtering module is used to: The steps for filtering the magnitude values ​​of two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of two frequency points in the negative frequency offset compensation direction are as follows: If the number of filtering iterations has not been reached, the frequency offset estimate for this clock cycle will remain consistent with the frequency offset estimate for the previous clock cycle. If the number of filtering iterations is reached, the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction are calculated.

[0090] The functions of each module in the coarse frequency offset estimation and compensation device correspond to the steps in the above-mentioned coarse frequency offset estimation and compensation method embodiment, and their functions and implementation processes will not be described in detail here.

[0091] This application also underwent simulation verification through mathematical modeling. The simulation conditions included a QPSK modulation format, a baud rate of 32 Gbaud, a channel SNR of 8 dB, a data block Fourier transform length of 1024, and a transmit / receive laser frequency offset greater than 7 GHz.

[0092] Figure 5 This is a graph showing the coarse frequency offset estimation algorithm for a coarse optical communication system as a function of the number of iterations. The horizontal axis represents the number of iterations for frequency offset estimation, and the vertical axis represents the estimated frequency offset value. `Freoffset est` is the estimated frequency offset value, and `Freoffset set` is the set value of the frequency offset in the simulation, i.e., the target value for frequency offset estimation. It can be seen that in scenarios with very low signal-to-noise ratios and frequency offsets exceeding ±Rs / 8, a relatively accurate coarse frequency offset estimate can be obtained after 20 iterations, with an error not exceeding 100MHz.

[0093] Figure 6 This is a spectrum diagram of frequency offset estimation based on dual-frequency power on a single-sided spectrum in a coherent optical communication system as shown in this application. The horizontal axis represents the number of FFT points, which is 128 points in the diagram; the vertical axis represents the power value at each frequency point. Figure 6 The left image shows the spectrum in the positive direction, and the right image shows the spectrum in the negative direction. Because the frequency offset compensation is more accurate in the positive direction, the power difference between the two frequency points (the two vertical lines) is larger. In contrast, because the frequency offset compensation is less accurate in the negative direction, the power difference between the two frequency points (the two vertical lines) is smaller. After subtracting the two power differences, the next frequency offset estimation direction will still be consistent with the positive direction spectrum.

[0094] Figure 7 This is a convergence effect diagram of the magnitude estimation based on IIR filters in a coherent optical communication system as shown in this application. The attenuation factor is set to 3e-4. It can be seen that after multiple iterations, the influence of white noise can be effectively suppressed, and the magnitudes in the positive and negative directions at two frequency points can be accurately estimated.

[0095] Where the horizontal axis represents the number of iterations of the IIR filter; the vertical axis represents the power value estimated in each iteration; Frequency point 1, direction +: Positive frequency offset compensation direction for frequency point 1; Frequency point 1, direction -: Negative frequency offset compensation direction for frequency point 1; Frequency point 2, direction +: Positive frequency offset compensation direction for frequency point 2; Frequency point 2, direction -: the direction of negative frequency offset compensation at frequency point 2.

[0096] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for coarse frequency offset estimation and compensation, characterized in that, The coarse frequency offset estimation and compensation method includes the following steps: Within this clock cycle, a continuous sequence of one polarization is selected from the two polarizations. Based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, frequency offset compensation is performed on the sequence of the current clock cycle to obtain two output sequences in the positive and negative frequency offset compensation directions. Determine the positions of the two frequency points with the closest frequency index and the largest amplitude difference. Based on the positions of the two frequency points and the two output sequences, determine the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction. The direction and magnitude of frequency offset compensation for the next clock cycle are determined based on the first and second modulus differences. Frequency offset compensation is repeated until the iteration condition is met. The frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle is output as the final coarse frequency offset estimate.

2. The coarse frequency offset estimation and compensation method as described in claim 1, characterized in that, The step of determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences includes: Call the dispersion compensation module to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; Based on the FFT output, select two frequency points of the signal and calculate the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction, respectively. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction.

3. The coarse frequency offset estimation and compensation method as described in claim 1, characterized in that, The step of determining the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences includes: Multiply the two output sequences by the butterfly transform tables at the two frequency points respectively, sum them, and take the magnitude to obtain the magnitudes of the two frequency points in the positive frequency offset compensation direction and the two frequency points in the negative frequency offset compensation direction. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction. The butterfly transform table for two frequency points is as follows: in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

4. The coarse frequency offset estimation and compensation method as described in claim 1, characterized in that: According to the formula: Determine the positions of the two frequency points with the closest frequency point numbers and the largest amplitude difference; in, The roll-off factor of the pulse shaping filter. The oversampling rate of the analog-to-digital converter (ADC) The length of the selected polarization signal.

5. The coarse frequency offset estimation and compensation method as described in claim 1, characterized in that, The process of determining the direction and magnitude of frequency offset compensation for the next clock cycle based on the first and second modulus differences, repeating frequency offset compensation until the iteration condition is met, and outputting the frequency offset estimate indicated by the first or second modulus difference for the corresponding clock cycle as the final coarse frequency offset estimate includes: Determine the maximum value between the first modulus difference and the second modulus difference. If the maximum value is greater than the first threshold, set the frequency offset compensation step size to a small step size. If the maximum value is less than or equal to the first threshold, set the frequency offset compensation step size to a large step size. Based on the initial frequency offset estimate or the frequency offset estimate passed from the previous clock cycle, and the determined frequency offset compensation step size, determine the frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle. If the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, the iteration condition is satisfied, and the frequency offset estimate indicated by the first or second modulus value of the corresponding clock cycle is taken as the final coarse frequency offset estimate. If the following conditions are not met: the maximum value is greater than the second threshold, and the absolute value of the difference between the first and second modulus values ​​is less than the third threshold, it is determined that the iteration condition is not met. The frequency offset values ​​of the positive and negative frequency offset compensation directions estimated in this clock cycle are then passed to the next clock cycle to continue the coarse frequency offset estimation and compensation process.

6. The coarse frequency offset estimation and compensation method as described in claim 5, characterized in that: The small step size is 300MHz, and the large step size is 3GHz.

7. The coarse frequency offset estimation and compensation method as described in claim 1, characterized in that: Before determining the first modulus difference between two frequency points in the positive frequency offset compensation direction and the second modulus difference between two frequency points in the negative frequency offset compensation direction, the following is also included: The steps for filtering the magnitude values ​​of two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of two frequency points in the negative frequency offset compensation direction are as follows: If the number of filtering iterations has not been reached, the frequency offset estimate for this clock cycle will remain consistent with the frequency offset estimate for the previous clock cycle. If the number of filtering iterations is reached, the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction are calculated.

8. A coarse frequency offset estimation and compensation device, characterized in that, The coarse frequency offset estimation and compensation device includes: The frequency offset compensation module is used to select a continuous sequence of one polarization from two polarizations, and to perform frequency offset compensation on the sequence of the current clock cycle based on the initial frequency offset estimate in the positive and negative frequency offset compensation directions or the frequency offset estimate passed from the previous clock cycle, so as to obtain two output sequences in the positive and negative frequency offset compensation directions. The Fourier transform module is used to determine the positions of the two frequency points with the closest frequency index and the largest amplitude difference. Based on the two frequency point positions and the two output sequences, it determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction. The frequency offset estimation module determines the direction and magnitude of frequency offset compensation for the next clock cycle based on the first and second modulus differences. It repeats the frequency offset compensation until the iteration condition is met, and outputs the frequency offset estimate indicated by the first or second modulus difference of the corresponding clock cycle as the final coarse frequency offset estimate.

9. The coarse frequency offset estimation and compensation device as described in claim 8, characterized in that, The Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Call the dispersion compensation module to obtain the Fast Fourier Transform (FFT) outputs of the two output sequences; Based on the FFT output, select two frequency points of the signal and calculate the magnitude values ​​of the two frequency points in the positive frequency offset compensation direction and the magnitude values ​​of the two frequency points in the negative frequency offset compensation direction, respectively. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction.

10. The coarse frequency offset estimation and compensation device as described in claim 8, characterized in that, The Fourier transform module determines the first modulus difference between the two frequency points in the positive frequency offset compensation direction and the second modulus difference between the two frequency points in the negative frequency offset compensation direction based on the two frequency point positions and the two output sequences, including: Multiply the two output sequences by the butterfly transform tables at the two frequency points respectively, sum them, and take the magnitude to obtain the magnitudes of the two frequency points in the positive frequency offset compensation direction and the two frequency points in the negative frequency offset compensation direction. The first modulus difference is calculated based on the modulus values ​​of the two frequency points in the positive frequency offset compensation direction, and the second modulus difference is calculated based on the modulus values ​​of the two frequency points in the negative frequency offset compensation direction. The butterfly transform table for two frequency points is as follows: in, The length of the selected polarization signal, The location of frequency point 1. The location of the 2 frequency point. Indicates from 0 to .

Images