Method and apparatus for deciding transmission of data

Abstract

The invention discloses a method and a device for judging transmission data. The method comprises the following steps of: inputting an equalizer output signal which is subjected to signal processing on transmission data by a feedback equalizer into a decision device so as to generate a decision signal of the equalizer output signal; determining a current decision error according to the equalizer output signal and the decision signal; updating a current error weight for the feedback equalizer according to the current decision error; and performing signal processing on a next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until a decision error for the next equalizer output signal is within a set error interval. According to the invention, equalizer errors can be reduced, a digital equalization system is converged, and signal processing precision is ensured.

Description

Technical Field

[0001] The present invention relates to the field of signal transmission technology, and in particular to a method and device for determining transmission data. Background Art

[0002] The digital equalization system, i.e. the digital equalizer, uses digital signal processing technology to equalize the signal. It can automatically adjust parameters and algorithms according to the characteristics of the signal and user needs to achieve a more accurate and efficient equalization effect. The digital equalization system generally includes a forward feedback equalizer, a decision maker and a decision feedback equalizer. The decision feedback equalizer uses the decision symbol generated by the decision maker as input to eliminate the inter-symbol interference generated by the channel. Since the decision feedback equalizer belongs to a feedback circuit, when there is an error in the symbol generated by the decision maker, the error will be fed back to the input of the decision feedback equalizer, thereby affecting the feedback output of the next decision feedback equalizer, and ultimately affecting the subsequent symbol decision. When the decision error is too large, it is easy for the error to circulate repeatedly in the decision feedback equalizer, resulting in continuous accumulation of errors, which continuously affects subsequent decisions. In severe cases, it may even cause the decision feedback equalizer to fail to converge, affecting the accuracy of signal processing. Summary of the invention

[0003] The present invention provides a method and a device for determining transmission data, which can reduce the error of an equalizer, make a digital equalization system converge, and ensure the accuracy of signal processing.

[0004] In one aspect of the present invention, a method for determining transmission data is provided. The method comprises: inputting an equalizer output signal after a feedback equalizer performs signal processing on transmission data into a determiner to generate a decision signal of the equalizer output signal; determining a current decision error according to the equalizer output signal and the decision signal; updating a current error weight for the feedback equalizer according to the current decision error; and performing signal processing on a next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until the decision error for the next equalizer output signal is within a set error interval.

[0005] In another aspect of the present invention, a device for determining transmission data is provided. The device includes: a feedback equalizer configured to perform signal processing on the transmission data to generate an equalizer output signal; a determiner configured to perform digital decision based on the equalizer output signal to generate a decision signal; and an error processor configured to: determine a current decision error according to the equalizer output signal and the decision signal; update a current error weight for the feedback equalizer according to the current decision error; and perform signal processing on a next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until the decision error for the next equalizer output signal is within a set error interval.

[0006] According to the technical solution of the present invention, the feedback equalizer performs signal processing on the transmission data to obtain an equalizer output signal, which is input to the decision device, and the decision device performs decision processing on the equalizer output signal to obtain a decision signal. The decision error between the equalizer output signal and the decision signal indicates the gap between the expected value and the actual value of the equalization system, and the error weight is dynamically configured according to the decision error, and the next equalizer output signal is signal processed based on the decision error and the dynamically configured error weight to narrow the gap between the expected value and the actual value, so that the decision error of the next equalizer output signal is within the set error range, thereby achieving convergence of the digital equalization system and ensuring signal processing accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A flowchart of a method for determining transmission data according to an embodiment of the present invention; Figure 2 A schematic diagram of a method for generating a decision signal by a decision device according to an embodiment of the present invention; Figure 3 A block diagram of signal processing according to an embodiment of the present invention; Figure 4 is a schematic diagram of a shift register according to an embodiment of the present invention; Figure 5 is a block diagram of an apparatus for determining transmission data according to an embodiment of the present invention; Figure 6 Schematic diagram of the structure of a multi-stage forward feedback equalizer according to an embodiment of the present invention; Figure 7 It is a schematic diagram of the structure of a multi-stage decision feedback equalizer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0008] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0009] In the prior art, a digital equalization system generally includes a forward feedback equalizer, a decision maker, and a decision feedback equalizer. The decision feedback equalizer uses the decision symbol generated by the decision maker as input to eliminate the inter-symbol interference generated by the channel. Since the decision feedback equalizer belongs to a feedback circuit, when there is an error in the symbol generated by the decision maker, the error will be fed back to the input of the decision feedback equalizer, thereby affecting the feedback output of the next decision feedback equalizer, and ultimately affecting the subsequent symbol decisions. When the decision error is too large, it is easy for the error to circulate repeatedly in the decision feedback equalizer, resulting in continuous accumulation of errors, which continuously affects subsequent decisions. In severe cases, it may even cause the decision feedback equalizer to fail to converge, affecting the accuracy of signal processing.

[0010] In order to solve at least the above technical problems, the present disclosure provides a method for determining transmission data. According to the present disclosure, the feedback equalizer performs signal processing on the transmission data to obtain an equalizer output signal, which is input to the determiner, and the determiner performs decision processing on the equalizer output signal to obtain a decision signal. The decision error between the equalizer output signal and the decision signal indicates the gap between the expected value and the actual value of the equalization system. The error weight is dynamically configured according to the decision error, and the next equalizer output signal is signal processed based on the decision error and the dynamically configured error weight to narrow the gap between the expected value and the actual value, so that the decision error of the next equalizer output signal is within the set error interval, thereby achieving digital equalization system convergence and ensuring signal processing accuracy. In this way, according to an embodiment of the present disclosure, the error weight is dynamically configured based on the decision error, so that the error weight can be updated with the decision error during the adaptive process, thereby reducing the error of the equalization signal through the dynamically configured error weight smoothing, improving the filter performance, and effectively accelerating the system convergence speed.

[0011] According to an embodiment of the present disclosure, a technical solution for prediction weights and an error predictor is provided. By predicting future errors, the error predictor allows the filter to adapt to changes in the signal more quickly, thereby accelerating convergence to the optimal tap coefficient. Incorporating the prediction error into the current weight update can reduce future errors, thereby improving the performance of the filter. In the presence of noise or other interference, prediction correction can improve the algorithm's resistance to interference, thereby maintaining good performance. For non-stationary signals, that is, the statistical characteristics of the signal change over time, prediction correction can help the algorithm better track changes in the signal. In this system, the use of prediction weights makes the LMS algorithm more flexible and adaptable to a variety of different signals and environmental conditions. The use of prediction weights can improve the quality of the received signal and reduce the impact of interference and noise. Because in the presence of noise or non-stationary signals, directly adjusting the filter coefficient based on the current error prediction value may result in over-adjustment, modifying the weights can reduce the impact of over-adjustment by smoothing such adjustments.

[0012] Hereinafter, the technical solution according to the present disclosure will be described with reference to specific embodiments and in conjunction with the accompanying drawings.

[0013] Figure 1 1 is a flow chart showing a method 100 for determining transmission data according to an embodiment of the present disclosure. Figure 1 The method 100 includes the following steps 102 to 108.

[0014] In step 102, the equalizer output signal after the feedback equalizer performs signal processing on the transmission data is input into the decision device to generate a decision signal of the equalizer output signal.

[0015] In some embodiments, the feedback equalizer includes a forward feedback equalizer and a decision feedback equalizer. The equalizer output signal is a digital signal obtained by eliminating inter-symbol interference based on tap coefficients of the forward feedback equalizer and the decision feedback equalizer.

[0016] Specifically, Figure 2 1 is a schematic diagram showing a method for a decision device to generate a decision signal according to an embodiment of the present disclosure. A decision reference signal is provided in the decision device, and the decision device compares the equalized signal with the decision reference signal and generates a decision signal according to the comparison result. Figure 2 , the horizontal coordinate in the figure represents the decision signal output by the decision device in each cycle, and the vertical coordinate in the figure represents the value of the equalization signal and the decision reference signal. When the equalization signal is higher than the decision reference signal A, the decision signal output is 1. When the equalization signal is lower than the decision reference signal B, the decision signal output is -1. When the equalization signal is lower than the decision reference signal A and higher than the decision reference signal B, the decision signal output is 0. Among them, the value range of the decision reference signal A is [0.3,0.7], which is usually set to 0.5; the value range of the decision reference signal B is [-0.7,-0.3], which is usually set to -0.5.

[0017] In step 104, a current decision error is determined based on the equalizer output signal and the decision signal.

[0018] In some embodiments, determining the current decision error according to the equalizer output signal and the decision signal includes: calculating the difference between the decision signal and the equalizer output signal, and determining the difference as the current decision error. In this way, the decision signal output by the decider characterizes the relationship between the equalized signal and the decision reference value, that is, the relationship between the equalized signal and the expected signal. Therefore, when there is a difference between the equalized signal and the decision signal, it means that the equalized signal is inconsistent with the expected signal, so the error between the two is calculated, and the error is fed back to the feedback equalizer, so that the feedback equalizer can adaptively adjust the parameters, thereby reducing the error between the future equalized signal and the expected signal, and achieving system convergence.

[0019] In some embodiments, the decision error of this time can be obtained by subtracting the equalized signal input this time from the decision signal output by the decision device this time. Figure 2 , the equalized signal F is lower than the decision reference line B, then the decision signal of the equalized signal F is -1, so the decision error err=-1-F.

[0020] In step 106, a current error weight for the feedback equalizer is updated according to the current decision error.

[0021] In some embodiments, updating the current error weight for the feedback equalizer according to the current decision error includes: determining the error level to which the current decision error belongs according to a preset error interval; and configuring the current error weight of the feedback equalizer as a standard weight corresponding to the error level. In this way, different standard weights can be adaptively selected according to the size of the decision error, thereby achieving synchronous update of the standard weight and the decision error, and then smoothing the equalization result of the feedback equalizer, thereby avoiding the impact caused by excessive processing of the equalization result of the feedback equalizer.

[0022] In some embodiments, configuring the current error weight of the feedback equalizer as a standard weight corresponding to the error level includes: configuring the shift value in the shift register corresponding to the feedback equalizer as the standard weight corresponding to the error level. In this way, dynamic configuration of the error weight can be achieved based on the shift register of the feedback equalizer, which can effectively improve the processing efficiency of the feedback equalizer.

[0023] In this embodiment, when the decision error is greater than the first preset value, the standard weight within the first range can be selected; when the decision error is less than or equal to the first preset value, the standard weight within the second range can be selected; the first range is greater than the second range. Specifically, the error interval of the decision error includes a first error interval, a second error interval, and a third error interval, wherein the first error interval, the second error interval, and the third error interval are all located in the interval (-1, 1). The error levels of the first error interval, the second error interval, and the third error interval are high, medium, and low, respectively, that is, the first error interval, the second error interval, and the third error interval correspond to the first weight, the second weight, and the third weight, respectively. In a specific application scenario, if the decision error is in the first error interval (-1, -0.5] or [0.5, 1), it means that the equalization system is still in the initial adaptive stage at this time, and the signal processing error is large, that is, the error level of the decision error is high, so the error weight should be configured as the first weight 0.5. If the decision error is in the second error interval (-0.5, -0.2] or [0.2, 0.5), it means that the signal processing error is moderate, that is, the error level of the decision error is medium, so the error weight should be configured as the second weight 0.25. If the decision error is in the third error interval (-0.2, 0.2), it means that the signal processing error is small, that is, the error level of the decision error is low, so the error weight should be configured as the third weight 0.125.

[0024] In some specific scenarios, if the user is already aware of the application scenarios and environmental influencing factors of the digital equalization system, and the usage scenario of the digital equalization system is single, and the environmental influencing factors have little impact on signal processing, then the user can directly configure the current error weight of the feedback equalizer through the processor, that is, the error weight at this time is a fixed value and does not need to be automatically updated according to the current judgment error.

[0025] Fixed error weights are used in relatively stable environments to reduce the difficulty of system operation and system power consumption, providing users with an additional option. Usually the system should use the adaptive error weights in the present invention, which can adjust the error weights in real time according to environmental influences. In scenarios where there is noise or non-stationary signals, if the tap coefficients of the equalizer are adjusted directly based on the current decision error, over-adjustment may result. Adjusting the tap coefficients through the current decision error and adaptively updated error weights can achieve smooth adjustment to reduce the impact of over-adjustment, while being able to quickly adapt to changes in the signal. When the signal changes or fluctuates, adjusting the error weights can track signal changes in a timely manner.

[0026] In step 108, signal processing is performed on the next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until the decision error for the next equalizer output signal is within a set error interval.

[0027] In some embodiments, signal processing of the next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error includes: determining a standard compensation error according to the standard weight and the current decision error; and error compensation of the next equalizer output signal of the feedback equalizer according to the standard compensation error. In this way, the size of the standard weight varies according to the size of the decision error. When the decision error is large, a larger standard weight can be selected to obtain a larger standard compensation error, so that the feedback equalizer performs error compensation on the next equalizer output signal based on the larger standard compensation error, thereby eliminating the current decision error to the greatest extent. When the decision error is small, a smaller standard weight can be selected to obtain a smaller standard compensation error, so that the feedback equalizer performs error compensation on the next equalizer output signal based on the smaller standard compensation error, thereby avoiding excessive adjustment of the feedback equalizer and affecting the adaptive result of the feedback equalizer.

[0028] In some embodiments, performing error compensation on the next equalizer output signal of the feedback equalizer according to the standard compensation error includes: superimposing the standard compensation error on the next equalizer output signal of the feedback equalizer. In this way, the current standard compensation error is superimposed on the next signal recovery result of the feedback equalizer, thereby eliminating the influence of the current error result on the next signal, thereby improving the accuracy of signal processing.

[0029] Figure 3 is a block diagram showing a signal processing according to an embodiment of the present disclosure. Figure 3, X[n] is the nth digital signal input to the digital equalization system, FF[n] is the digital signal after the nth equalization processing by the forward feedback equalizer, DF[n] is the digital signal after the nth equalization processing by the decision feedback equalizer, EP[n] is the standard compensation error obtained by the nth equalization processing, EQ[n] is the nth equalizer output signal obtained according to FF[n], DF[n] and EP[n-1], DEC[n] is the nth decision signal generated by the decider, and α is the standard weight obtained according to the nth decision error configuration. Based on this, the equalizer output signal EQ[n]=FF[n]+DF[n]+EP[n-1], the decision error err[n]=decision signal DEC[n]-equalizer output signal EQ[n], and the standard compensation error EP[n]=standard weight α[n]×current decision error err[n]. For example, when the decision error err[n] is 0.5, the standard weight α[n] is 0.5, and the standard compensation error EP[n]=0.25. Therefore, the standard compensation error EP[n]=standard weight α[n]×(decision signal DEC[n]-(FF[n]+DF[n]+EP[n-1])). Further, the next equalizer output signal after error compensation based on the standard compensation error EP[n] is EQ[n+1]=FF[n+1]+DF[n+1]+EP[n].

[0030] The standard weight can adjust the size of the decision error calculated by the equalizer last time. As the decision error becomes smaller and smaller, the standard weight tends to be reduced to achieve stable adjustment of the standard compensation error. Since in the presence of noise or non-stationary signals, directly adjusting the filter coefficients in the equalizer based on the current decision error may lead to over-adjustment, the standard weight can be adaptively updated to achieve smooth adjustment of the tap coefficients and reduce the impact of over-adjustment. At the same time, it can adapt to changes in the signal. When the signal changes or fluctuates, increasing the standard weight can track the signal changes in time. Increase the standard weight: When the standard weight is increased, the digital signal processing algorithm in the equalizer will pay more attention to the current error signal, which can speed up the adjustment of the tap coefficients and thus speed up the convergence. Increasing the standard weight can make the filter adapt to changes in the input signal more quickly, which is very useful in environments where signal characteristics change rapidly. However, if the standard weight is too large, it may reduce the stability of the algorithm, cause over-adjustment and oscillation, and thus affect the performance of the filter. Reduce the standard weight: When the standard weight is reduced, the digital signal processing algorithm in the equalizer becomes less sensitive to the current error signal, which helps to improve the stability of the algorithm and reduce the risk of over-adjustment and oscillation. Smaller standard weights can make the filter adjustment smoother and more gradual, which helps to achieve more stable performance in environments with large noise or infrequent signal changes. However, if the standard weight is too small, it may lead to slower convergence and make the filter unable to adapt to signal changes in time, thus affecting performance.

[0031] In some embodiments, determining the standard compensation error according to the standard weight and the current decision error includes: loading the current decision error into the shift register so that the current decision error is shifted according to the standard weight, and outputting the value after the shift processing as the standard compensation error. In this way, the feedback equalizer can obtain a new standard compensation error in each cycle, and the standard compensation error obtained by the shift processing eliminates the error accumulation of the decision maker, so that the decision maker can output the decision signal more accurately, thereby ensuring the accuracy of signal processing.

[0032] Figure 4 is a schematic diagram showing a shift register according to an embodiment of the present disclosure. Figure 4, load the current judgment error into the 10-bit shift register, select the error weight by shifting, the shifted value is the standard weight, and the result after shifting is the standard compensation error obtained by multiplying the current judgment error by the standard weight. The standard compensation error is superimposed on the next equalization signal for compensation. Therefore, a new standard compensation error can be obtained in each cycle. Through the standard compensation error, the error accumulation is eliminated, so that the decision maker can make a more accurate decision. Among them, the 16 bits in the shift register store the total weight value, and each shift reduces the weight value of the previous bit by half. Specifically, the total weight value is set to 1. When the current judgment error is not moved, the standard weight is 100% of the total weight value; when the current judgment error moves 1 bit, the standard weight is 50% of the total weight value; when the current judgment error moves 2 bits, the standard weight is 25% of the total weight value, and so on. This achieves the shift selection of different error weights.

[0033] In some embodiments, after determining the current decision error according to the equalizer output signal and the decision signal, the method may include: adjusting the tap coefficients of the feedback equalizer used for signal processing according to the current decision error, so that the feedback equalizer performs signal processing on the next equalizer output signal based on the tap coefficients. In this way, the feedback equalizer automatically adjusts the tap coefficients according to the characteristics of the signal and the needs of the user during the signal processing process, adjusts the tap coefficients through the current decision error, reduces the gap between the equalized signal processed by the tap coefficients and the decision signal, and enables the equalizer to adapt to changes in the signal more quickly, thereby accelerating the convergence efficiency of the system.

[0034] In some embodiments, adjusting the tap coefficients of the feedback equalizer for signal processing according to the current decision error includes adjusting the tap coefficients by the following formula: the tap coefficients after adjustment = the tap coefficients before adjustment + the preset coefficient adaptive weight × the current decision error × the current input signal; the current input signal is a digital signal of the equalizer output signal that has not been processed by the feedback equalizer. In this way, by minimizing the mean square error to adjust the tap coefficients for signal processing, the decision error between the equalized signal and the decision signal is finally within the set error range, thereby achieving convergence of the digital equalization system.

[0035] The decision error needs to be input into the forward feedback equalizer and the decision feedback equalizer at the same time, so as to perform the minimum mean square error according to the current decision error, thereby adaptively adjusting the tap coefficients in the feedback equalizer. Specifically, if the tap coefficient of the feedback equalizer for the nth equalization process is W(n), the decision error obtained by its equalized signal is e(n), the digital signal input to the feedback equalizer is X(n), and its coefficient adaptive weight is u, then the tap coefficient of the feedback equalizer for the n+1th equalization process is W(n+1)=W(n)+u×e(n)×X(n). It can be seen that the decision error err is not only used to adjust the standard weight to achieve the adjustment of the standard compensation error, but also used to adjust the tap coefficients of each level of filters in the feedback equalizer.

[0036] In some embodiments, until the decision error for the next equalizer output signal is within the set error interval, the process includes: inputting the current error weight and the signal processing result of the current decision error and the next equalizer output signal into the decider as the processed next equalizer output signal to generate the next decision signal; determining the next decision error according to the processed next equalizer output signal and the next decision signal; and if the next decision error is within the set error interval based on 0, then stopping updating the error weight for the feedback equalizer. In this way, the final decision error should be 0 in an ideal state, but in actual application, the decision error can only approach 0 and cannot be directly 0, so the error interval is set based on 0. As long as the decision error falls within a certain interval around 0, the system can be considered stable.

[0037] In some embodiments, the method may also include: if the next decision error is outside the set error interval, updating the next error weight for the feedback equalizer according to the next decision error; and performing signal processing on another equalizer output signal of the feedback equalizer according to the next error weight and the next decision error.

[0038] According to another aspect of the present invention, Figure 5 2 is a block diagram showing a device 200 for determining transmission data according to an embodiment of the present invention. Figure 5 The apparatus 200 includes a feedback equalizer 202, a decision device 204, and an error processor 206. The feedback equalizer 202 is configured to perform signal processing on the transmission data to generate an equalizer output signal. The decision device 204 is configured to perform digital decision based on the equalizer output signal to generate a decision signal. The error processor 206 is configured to perform various steps of a method for deciding transmission data as described above.

[0039] Feedback equalizer 202 includes forward feedback equalizer 2022 and decision feedback equalizer 2024. The outputs of forward feedback equalizer 2022, decision feedback equalizer 2024 and error processor 206 are all connected to the input of decider 204, and the output of decider 204 is connected to the input of decision feedback equalizer 2024 and error processor 206 respectively.

[0040] Figure 6 is a schematic diagram showing the structure of a multi-stage forward feedback equalizer according to an embodiment of the present invention. Figure 7 It is a schematic diagram showing the structure of a multi-stage decision feedback equalizer according to an embodiment of the present invention.

[0041] Reference Figure 6 and Figure 7 , FFE is a forward feedback equalizer, DFE is a decision feedback equalizer, F1~N represents the tap coefficients of each level of forward feedback equalizer, D1~N represents the tap coefficients of each level of decision feedback equalizer, and err represents the current decision error output by the error processor. According to the current decision error err, the tap coefficients F1~N and D1~N of the feedback equalizer FFE and DFE used for signal processing are adjusted.

[0042] In summary, the present invention provides a method and device for determining transmission data, wherein the feedback equalizer performs signal processing on the transmission data based on the current tap coefficient to obtain an equalizer output signal, and the determiner performs decision processing on the equalizer output signal to obtain a decision signal. The decision error between the equalizer output signal and the decision signal indicates the gap between the expected value and the actual value of the equalization system, that is, the signal processing effect of the current tap coefficient is determined. The error weight is dynamically configured according to the decision error, and the compensation error is determined based on the decision error and the dynamically configured error weight, and the compensation error is superimposed on the next equalizer output signal to narrow the gap between the expected value and the actual value, so that the decision error of the next equalizer output signal is within the set error interval, thereby achieving convergence of the digital equalization system and ensuring signal processing accuracy. The present invention dynamically configures the error weight based on the decision error, so that the error weight can be updated with the decision error during the adaptive process, thereby reducing the error of the equalization signal through the dynamically configured error weight smoothing, improving the filter performance, and effectively accelerating the system convergence speed.

[0043] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for determining transmission data, characterized in that: include: Inputting the equalizer output signal after the feedback equalizer performs signal processing on the transmission data into the decision device to generate a decision signal of the equalizer output signal; Determine a current decision error according to the equalizer output signal and the decision signal; updating a current error weight for the feedback equalizer according to the current decision error; as well as Signal processing is performed on a next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until the decision error for the next equalizer output signal is within a set error interval.

2. The method according to claim 1, characterized in that Until the decision error for the next equalizer output signal is within the set error interval comprises: Inputting the signal processing result of the current error weight, the current decision error and the next equalizer output signal into the decider as the processed next equalizer output signal to generate a next decision signal; Determining a next decision error according to the processed next equalizer output signal and the next decision signal; and If the next decision error is within the set error interval based on 0, then the error weight for the feedback equalizer is stopped from being updated.

3. The method according to claim 2, characterized in that Also includes: If the next decision error is outside the set error interval, updating a next error weight for the feedback equalizer according to the next decision error; as well as Signal processing is performed on another equalizer output signal of the feedback equalizer according to the next error weight and the next decision error.

4. The method according to claim 1, characterized in that: Updating a current error weight for the feedback equalizer according to the current decision error comprises: Determining the error level to which the current judgment error belongs according to a preset error interval; and The current error weight of the feedback equalizer is configured as a standard weight corresponding to the error level.

5. The method according to claim 4, characterized in that Configuring the current error weight of the feedback equalizer to a standard weight corresponding to the error level comprises: The shift value in the shift register corresponding to the feedback equalizer is configured as a standard weight corresponding to the error level.

6. The method according to claim 5, characterized in that The signal processing of the next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error comprises: Determining a standard compensation error according to the standard weight and the current decision error; and The next equalizer output signal of the feedback equalizer is error compensated according to the standard compensation error.

7. The method according to claim 6, characterized in that Determining the standard compensation error according to the standard weight and the current decision error includes: The current decision error is loaded into the shift register, so that the current decision error is shifted according to the standard weight, and the shifted value is output as the standard compensation error.

8. The method according to claim 6, characterized in that Performing error compensation on a next equalizer output signal of the feedback equalizer according to the standard compensation error comprises: The standard compensation error is superimposed on a next equalizer output signal of the feedback equalizer.

9. The method according to claim 1, characterized in that: After determining the current decision error according to the equalizer output signal and the decision signal, the method further comprises: The tap coefficients of the feedback equalizer used for signal processing are adjusted according to the current decision error, so that the feedback equalizer performs signal processing on the next equalizer output signal based on the tap coefficients.

10. The method according to claim 9, characterized in that Adjusting the tap coefficients of the feedback equalizer for signal processing according to the current decision error includes adjusting the tap coefficients by the following formula: The tap coefficient after adjustment = the tap coefficient before adjustment + the preset coefficient adaptive weight × the current decision error × the current input signal; The current input signal is a digital signal that has not been processed by the feedback equalizer.

11. The method according to claim 1, characterized in that: Determining a current decision error according to the equalizer output signal and the decision signal comprises: The difference between the decision signal and the equalizer output signal is calculated, and the difference is determined as the current decision error.

12. A device for determining transmission data, characterized in that: include: a feedback equalizer configured to perform signal processing on the transmission data to generate an equalizer output signal; a decision device configured to make a digital decision based on the equalizer output signal to generate a decision signal; as well as The error processor is configured as: Determine a current decision error according to the equalizer output signal and the decision signal; updating a current error weight for the feedback equalizer according to the current decision error; as well as Signal processing is performed on a next equalizer output signal of the feedback equalizer according to the current error weight and the current decision error until the decision error for the next equalizer output signal is within a set error interval.

Images