Link balancing method and device, computer equipment and storage medium

Abstract

The invention relates to a link balancing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring each group of link equalization parameters of a link to be equalized, and determining a signal sampling center corresponding to each group of link equalization parameters; continuously adjusting the deviation parameter of each dimension on the basis of the signal sampling center, sampling the signal transmitted on the to-be-equalized link based on each adjusted deviation parameter to obtain each signal sampling result, and determining a signal sampling range corresponding to the link equalization parameter based on each signal sampling result; determining a target link equalization parameter from each group of link equalization parameters according to each signal sampling range and a stability evaluation algorithm; the target link equalization parameter is used for performing link equalization on the to-be-equalized link. By adopting the method, the link balancing efficiency can be improved.

Description

Technical Field

[0001] This application relates to the field of data communication technology, and in particular to a link equalization method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] During the establishment of a SerDes (Serializer / Deserializer Link) link, a link equalization process typically occurs. Link equalization aims to establish a stable connection between devices / chips by adjusting the settings of the Tx (transmitter) and Rx (receiver) ends to improve signal quality, enabling the link to transmit at the most stable and faster rate. The optimal link equalization parameters are identified through the link equalization process, and link equalization is performed to ensure that the link signal quality meets the link data transmission standards.

[0003] Currently, when selecting link equalization parameters, the bit error rate (BER) of the signal is typically monitored after applying the parameters at both the transmitting and receiving ends. The BER is then used to assess whether the link equalization parameters meet the preset transmission conditions. If the link equalization parameters do not meet the transmission conditions, they are adjusted until they do.

[0004] However, traditional techniques require extensive monitoring and debugging time to continuously adjust link equalization parameters by monitoring the signal's bit error rate over a long period. Therefore, current link equalization methods suffer from insufficient real-time performance in adjusting link equalization parameters. Summary of the Invention

[0005] Therefore, it is necessary to provide a link balancing method, apparatus, computer equipment, storage medium, and computer program product to address the aforementioned technical problems.

[0006] Firstly, this application provides a link balancing method, including:

[0007] Obtain each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters;

[0008] Based on the signal sampling center, the deviation parameters of each dimension are continuously adjusted, and the signals transmitted on the link to be balanced are sampled based on the adjusted deviation parameters to obtain the signal sampling results. Based on the signal sampling results, the signal sampling range corresponding to the link equalization parameter is determined.

[0009] Based on the signal sampling range and stability evaluation algorithm, target link equalization parameters are determined from each group of link equalization parameters; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0010] In one embodiment, determining the signal sampling center corresponding to each set of link equalization parameters includes:

[0011] The transmitting end corresponding to the link to be balanced is configured based on each set of link balancing parameters; the signal transmitted on the link to be balanced is sent by the transmitting end on each set of the link to be balanced based on the set of link balancing parameters.

[0012] Based on the signal, channel adaptation is performed to obtain the signal sampling center of the link equalization parameters.

[0013] In one embodiment, each dimension includes a time dimension and a voltage dimension, wherein the deviation parameter of the time dimension is a time code, and the voltage parameter of the voltage dimension is a voltage code. The step of continuously adjusting the deviation parameters of each dimension based on the signal sampling center, sampling the signal transmitted on the link to be balanced based on the adjusted deviation parameters to obtain each signal sampling result, and determining the signal sampling range corresponding to the link equalization parameter based on each signal sampling result includes:

[0014] The time code is continuously adjusted horizontally based on the signal sampling center, and the signal transmitted on the link to be balanced is sampled based on the adjusted time code to obtain the sampling results of each signal, and the time boundary value is determined based on the sampling results of each signal and the preset error threshold.

[0015] Based on the signal sampling center, the voltage code is continuously adjusted from a vertical angle, and the signal transmitted on the link to be equalized is sampled based on the adjusted voltage code to obtain the sampling results of each signal, and the voltage boundary value is determined based on each sampling result of the signal and the error threshold.

[0016] Based on the time boundary value and the voltage boundary value, the signal sampling range corresponding to the link equalization parameter is constructed.

[0017] In one embodiment, the step of continuously adjusting the time code horizontally based on the signal sampling center, sampling the signal transmitted on the link to be equalized based on the adjusted time code to obtain each signal sampling result, and determining the time boundary value based on each signal sampling result and a preset signal threshold includes:

[0018] Centered on the signal sampling center, the time code is moved horizontally according to a preset time offset unit, and the signal transmitted on the link to be balanced is sampled based on the moved time code to obtain the signal sampling result.

[0019] If the sampling error of the signal sampling result does not reach the preset error threshold, the step of moving the time code horizontally with the signal sampling center as the center and according to the preset time offset unit is executed until the sampling error reaches the error threshold, and the sampling time is determined as the time boundary value in the time dimension.

[0020] In one embodiment, the step of continuously adjusting the voltage code from a vertical angle based on the signal sampling center, sampling the signal transmitted on the link to be equalized based on the adjusted voltage code to obtain each signal sampling result, and determining the voltage boundary value based on each signal sampling result and the error threshold includes:

[0021] Centered on the signal sampling center, the voltage code is moved vertically according to a preset voltage offset unit, and the signal transmitted on the link to be equalized is sampled based on the moved voltage code to obtain the signal sampling result.

[0022] If the sampling error of the signal sampling result does not reach the error threshold, the step of moving the voltage code in the vertical direction with the signal sampling center as the center and according to the preset voltage offset unit is executed until the sampling error reaches the error threshold, and the sampled voltage is determined as the voltage boundary value in the voltage dimension.

[0023] In one embodiment, determining the target link equalization parameter from each group of link equalization parameters based on each of the signal sampling ranges and stability evaluation algorithms includes:

[0024] Based on the stability evaluation algorithm and each of the signal sampling ranges, the eccentricity of the signal under the stability index and the effective length under the anti-interference index are determined, and the eye diagram evaluation value of the signal sampling range is determined based on the eccentricity and the effective length.

[0025] Based on the eye diagram evaluation values, the target link balancing parameters corresponding to the link to be balanced are determined from the link balancing parameters in each group.

[0026] In one embodiment, determining the eccentricity of the signal under a stability index and the effective length under an anti-interference index based on a stability evaluation algorithm and each of the signal sampling ranges, and determining the eye diagram evaluation value of the signal sampling range based on the eccentricity and the effective length, includes:

[0027] For each dimension in the signal sampling range, the eccentricity of the signal under the stability index and the effective length under the anti-interference index are determined based on the boundary values ​​of the dimension.

[0028] Based on the stability evaluation algorithm, the effective lengths, and the eccentricities, the eye diagram evaluation values ​​of the link equalization parameters are determined.

[0029] In one embodiment, determining the eccentricity of the signal under the stability index and the effective length under the anti-interference index for each dimension of the signal sampling range, based on the boundary values ​​of the dimension, includes:

[0030] Among the boundary values, a minimum boundary value is determined, and based on the minimum boundary value, the effective length corresponding to the dimension is determined; the effective length characterizes the anti-interference capability of the signal; the larger the effective length, the greater the anti-interference capability of the signal.

[0031] Determine the maximum boundary value among all the boundary values, and determine the difference between the maximum boundary value and the minimum boundary value;

[0032] The ratio between the difference and the maximum boundary value is determined as the eccentricity corresponding to the dimension; the eccentricity characterizes the stability of the signal; the smaller the eccentricity, the better the stability of the signal.

[0033] In one embodiment, determining the eye diagram evaluation value of the link equalization parameters based on the stability evaluation algorithm, each of the effective lengths, and each of the eccentricities includes:

[0034] Based on the effective length and eccentricity of each dimension, a signal sampling evaluation value for that dimension is determined; the signal sampling evaluation value is directly proportional to the effective length and inversely proportional to the eccentricity.

[0035] Based on the stability evaluation algorithm, the signal sampling evaluation values ​​of each dimension are processed to obtain the eye diagram evaluation values ​​of the link equalization parameters.

[0036] In one embodiment, determining the target link balancing parameter corresponding to the link to be balanced from each group of link balancing parameters based on each eye diagram evaluation value includes:

[0037] Among all the eye diagram evaluation values, the maximum eye diagram evaluation value is determined as the target eye diagram evaluation value;

[0038] The link balancing parameters corresponding to the target eye diagram evaluation value are determined as the target link balancing parameters.

[0039] Secondly, this application also provides a link equalization device, comprising:

[0040] The acquisition module is used to acquire each set of link equalization parameters of the link to be equalized, and to determine the signal sampling center corresponding to each set of link equalization parameters.

[0041] The sampling module is used to continuously adjust the deviation parameters of each dimension based on the signal sampling center, and to sample the signals transmitted on the link to be balanced based on the adjusted deviation parameters to obtain the sampling results of each signal, and to determine the signal sampling range corresponding to the link equalization parameter based on the sampling results of each signal.

[0042] The determining module is used to determine the target link equalization parameters from each group of link equalization parameters based on the sampling range and stability evaluation algorithm of each signal; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0043] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0044] Obtain each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters;

[0045] Based on the signal sampling center, the deviation parameters of each dimension are continuously adjusted, and the signals transmitted on the link to be balanced are sampled based on the adjusted deviation parameters to obtain the signal sampling results. Based on the signal sampling results, the signal sampling range corresponding to the link equalization parameter is determined.

[0046] Based on the signal sampling range and stability evaluation algorithm, target link equalization parameters are determined from each group of link equalization parameters; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0047] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0048] Obtain each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters;

[0049] Based on the signal sampling center, the deviation parameters of each dimension are continuously adjusted, and the signals transmitted on the link to be balanced are sampled based on the adjusted deviation parameters to obtain the signal sampling results. Based on the signal sampling results, the signal sampling range corresponding to the link equalization parameter is determined.

[0050] Based on the signal sampling range and stability evaluation algorithm, target link equalization parameters are determined from each group of link equalization parameters; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0051] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0052] Obtain each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters;

[0053] Based on the signal sampling center, the deviation parameters of each dimension are continuously adjusted, and the signals transmitted on the link to be balanced are sampled based on the adjusted deviation parameters to obtain the signal sampling results. Based on the signal sampling results, the signal sampling range corresponding to the link equalization parameter is determined.

[0054] Based on the signal sampling range and stability evaluation algorithm, target link equalization parameters are determined from each group of link equalization parameters; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0055] The aforementioned link equalization method, apparatus, computer equipment, storage medium, and computer program products acquire each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters; continuously adjust the deviation parameters of each dimension based on the signal sampling center, and sample the signal transmitted on the link to be equalized based on the adjusted deviation parameters to obtain each signal sampling result, and determine the signal sampling range corresponding to the link equalization parameter based on each signal sampling result; determine the target link equalization parameter from each set of link equalization parameters according to each signal sampling range and a stability evaluation algorithm; the target link equalization parameter is used to perform link equalization on the link to be equalized. Using this method, by actively adjusting the deviation parameters of each dimension based on the signal sampling center and continuously sampling the signal under each set of link equalization parameters, a signal sampling range representing the stability of signal sampling can be obtained, and the optimal set of link equalization parameters can be evaluated using the signal sampling range and a stability evaluation algorithm. This achieves real-time determination of the target link equalization parameter matched to the link to be equalized before applying the link equalization parameter, improving the evaluation efficiency of the target link equalization parameter, and thus improving the adjustment efficiency of the link equalization parameter. Attached Figure Description

[0056] Figure 1 This is a diagram illustrating the application environment of a link balancing method in one embodiment;

[0057] Figure 2 This is a flowchart illustrating a link balancing method in one embodiment;

[0058] Figure 3 This is a flowchart illustrating the process of determining the signal sampling center in one embodiment;

[0059] Figure 4 This is a flowchart illustrating the process of determining the signal sampling range in one embodiment;

[0060] Figure 5 This is a schematic diagram of an eye diagram in an exemplary embodiment;

[0061] Figure 6 This is a flowchart illustrating the process of determining time boundary values ​​in one embodiment;

[0062] Figure 7 This is a flowchart illustrating the process of determining voltage boundary values ​​in one embodiment;

[0063] Figure 8 This is a flowchart illustrating the process of determining target balanced link parameters in one embodiment;

[0064] Figure 9 This is a schematic diagram of the process for determining eye diagram evaluation values ​​in one embodiment;

[0065] Figure 10 This is a flowchart illustrating the process of determining the effective length and eccentricity in one embodiment;

[0066] Figure 11 This is a schematic diagram of the process for calculating eye diagram evaluation values ​​in one embodiment;

[0067] Figure 12 This is a flowchart illustrating the process of selecting target link balancing parameters in one embodiment;

[0068] Figure 13 This is a flowchart of a link balancing method in an exemplary embodiment;

[0069] Figure 14 This is a block diagram of a link equalization device in one embodiment;

[0070] Figure 15 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0072] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0073] The link balancing method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown includes a sender 110 and a receiver 120, connected via a link to be balanced 130. Before transmitting data between the sender 110 and receiver 120 via the link to be balanced 130, the receiver 120 needs to determine the target link balancing parameters of the link to be balanced using the link balancing method, so that the sender 110 can transmit data based on the target link balancing parameters. The sender 110 or receiver 120 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, or servers. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0074] In one exemplary embodiment, such as Figure 2 As shown, a link balancing method is provided, which can be applied to... Figure 1 Taking the receiver 120 (hereinafter referred to as the receiver) as an example, the explanation includes the following steps 202 to 206. Wherein:

[0075] Step 202: Obtain the link equalization parameters of each group of links to be equalized, and determine the signal sampling center corresponding to each group of link equalization parameters.

[0076] Among them, the signal sampling center corresponding to each set of link equalization parameters is the optimal signal sampling center when performing signal sampling under that set of link equalization parameters.

[0077] In implementation, the receiving end is pre-configured with various sets of link balancing parameters. Each set of link balancing parameters is a combination of parameter values ​​used for link balancing. At least one parameter in each set of link balancing parameters has a different value. The specific parameters comprising each set of link balancing parameters can be determined by those skilled in the art based on actual needs.

[0078] Each set of link equalization parameters can be configured at the transmitting end before link equalization begins. When link equalization starts, the transmitting and receiving ends, respectively, determine the corresponding signal sampling center for each set of link equalization parameters at the receiving end.

[0079] Specifically, the receiver configures the transmitter for each set of link equalization parameters. Under this set of link equalization parameters, the transmitter transmits data or signals on the equalized link. The receiver performs channel adaptation based on the signal to obtain the signal sampling center corresponding to that set of link equalization parameters.

[0080] Step 204: Based on the signal sampling center, continuously adjust the deviation parameters of each dimension, and sample the signals transmitted on the link to be balanced based on the adjusted deviation parameters to obtain the sampling results of each signal, and determine the signal sampling range corresponding to the link equalization parameters based on the sampling results of each signal.

[0081] The parameters used when sampling a signal can include parameters of different dimensions, such as sampling timing (time) and sampling voltage threshold (voltage).

[0082] In implementation, the receiver continuously adjusts the deviation parameters of each dimension with the signal sampling center as the center. Then, the receiver samples the signal transmitted on the link to be equalized based on the adjusted deviation parameters, obtains the signal sampling result, and determines whether the sampling error of the signal sampling result reaches a preset error threshold. If the sampling error of the signal sampling result reaches the error threshold, the receiver determines the signal sampling range corresponding to the link equalization parameters based on the current deviation parameters.

[0083] Specifically, for each dimension, the receiving end adjusts the deviation parameter of that dimension from the direction corresponding to the signal sampling center, and samples the signal transmitted on the link to be equalized according to the adjusted deviation parameter to obtain the signal sampling result. Then, the receiving end determines whether the sampling error of the signal sampling result reaches a preset error threshold. If the sampling error of the signal sampling result does not reach the error threshold, the receiving end performs the step of adjusting the deviation parameter of that dimension from the direction corresponding to that dimension until the sampling error of the signal sampling result reaches the error threshold. Based on the signal sampling center and the deviation parameter of that dimension at this moment, the receiving end determines the boundary value of that dimension. Then, the computer equipment constructs the signal sampling range corresponding to the link equalization parameters based on the boundary values ​​of each dimension.

[0084] Step 206: Determine the target link equalization parameters from each group of link equalization parameters based on the sampling range and stability evaluation algorithm of each signal.

[0085] Among them, the target link balancing parameter is used to perform link balancing on the links to be balanced.

[0086] In implementation, the receiver determines the eccentricity and effective length for each signal sampling range using a stability assessment algorithm. Eccentricity characterizes signal stability, while effective length characterizes the signal's anti-interference capability. Then, based on the eccentricity and effective length, the receiver determines the eye diagram evaluation value corresponding to the signal sampling range and selects target link equalization parameters based on each eye diagram evaluation value.

[0087] Specifically, the receiver determines the signal's eccentricity under stability metrics and effective length under anti-interference metrics based on the stability assessment algorithm and the sampling range of each signal. It then determines the eye diagram evaluation value for the signal sampling range based on the eccentricity and effective length. Next, based on each eye diagram evaluation value, the receiver determines the target link equalization parameters for the link to be equalized from each set of link equalization parameters. A larger eccentricity value indicates poorer anti-interference capability, while a larger effective length value indicates better signal stability.

[0088] In an optional embodiment, after the transmitting end has sent signals to the receiving end based on all groups of link equalization parameters, and the receiving end has obtained the signal sampling range of all groups of link equalization parameters, the receiving end can determine which group of link equalization parameters should be used as the target link equalization parameter for link equalization. The criteria for the receiving end to select the target link equalization parameter can be determined according to actual needs. For example, in scenarios with severe signal jitter, the preset link equalization parameter with the widest signal sampling range corresponding to the sampling timing can be selected as the target link equalization parameter. Alternatively, the receiving end can also calculate the eye diagram area based on each corresponding signal sampling range for each preset link equalization parameter, and select the group of link equalization parameters with the largest eye diagram area (i.e., the best signal sampling margin) as the target link equalization parameter. This embodiment does not specifically limit this approach.

[0089] The aforementioned link equalization method actively adjusts the deviation parameters in each dimension based on the signal sampling center and continuously samples the signal under each set of link equalization parameters. This allows it to obtain the signal sampling range, representing the stability of the signal sampling, and then evaluates which set of link equalization parameters is optimal using a signal sampling range and stability assessment algorithm. This enables real-time determination of the target link equalization parameters for the link to be equalized before applying the parameters, improving the evaluation efficiency of the target link equalization parameters and thus enhancing the adjustment efficiency of the link equalization parameters.

[0090] In one exemplary embodiment, such as Figure 3 As shown, the specific processing steps for determining the signal sampling center corresponding to each set of link equalization parameters in step 202 include steps 302 to 304. Wherein:

[0091] Step 302: Configure the sending end corresponding to the link to be balanced based on each set of link balancing parameters.

[0092] The signals transmitted on the link to be balanced are sent by the transmitting end on the link to be balanced based on each set of link balancing parameters.

[0093] In practice, the receiving end configures the corresponding transmitting end for each set of link balancing parameters. Then, the transmitting end sends a signal to the receiving end under that set of link balancing parameters.

[0094] Step 304: Based on the signal, perform channel adaptation to obtain the signal sampling center of the link equalization parameters.

[0095] In practice, when receiving signals, the receiving end performs channel adaptation based on the signals transmitted on the link, and can obtain the optimal signal sampling center when sampling signals under the current group link equalization parameters.

[0096] In this embodiment, the transmitting end corresponding to the link to be balanced is configured by configuring the link balancing parameters of each group, and the signal sampling center is determined based on the signal sent by the transmitting end. This allows us to obtain the actual signal sampling center on the current link to be balanced, which is convenient for determining the signal sampling range in the future.

[0097] In an exemplary embodiment, the dimensions include a time dimension and a voltage dimension. The deviation parameter of the time dimension is a time code, and the voltage parameter of the voltage dimension is a voltage code, such as... Figure 4 As shown, the specific processing procedure of step 204 includes steps 402 to 406. Wherein:

[0098] Step 402: Based on the signal sampling center, continuously adjust the time code from a horizontal angle, and sample the signals transmitted on the link to be equalized based on the adjusted time code to obtain the sampling results of each signal, and determine the time boundary value based on the sampling results of each signal and the preset error threshold.

[0099] In implementation, the receiver continuously adjusts the timecode horizontally, centering on the signal sampling center, and samples the signals transmitted on the link to be equalized based on the adjusted timecode, obtaining the sampling results for each signal. Then, the receiver determines whether the sampling error of the signal sampling results reaches a preset error threshold, and if the sampling error reaches the error threshold, determines the time boundary value based on the current timecode and the signal sampling center.

[0100] Specifically, the receiving end moves the time code left and right respectively, centered on the signal sampling center, and samples the signal transmitted on the link to be equalized based on the moved time code to obtain the signal sampling result. The receiving end determines whether the sampling error of the signal sampling result reaches a preset error threshold. If the sampling error of the signal sampling result does not reach the error threshold, the computer device executes the steps of moving the time code left and right respectively, and sampling the signal transmitted on the link to be equalized based on the moved time code, until the sampling error reaches the error threshold. The computer device then determines the time boundary value between the current time code and the signal sampling center.

[0101] Step 404: Based on the signal sampling center, continuously adjust the voltage code from the vertical angle, and sample the signal transmitted on the link to be equalized based on the adjusted voltage code to obtain the sampling results of each signal, and determine the voltage boundary value based on the sampling results of each signal and the error threshold.

[0102] In implementation, the receiver continuously adjusts the voltage code horizontally, centering on the signal sampling center, and samples the signals transmitted on the link to be equalized based on the adjusted voltage code, obtaining the sampling results for each signal. Then, the receiver determines whether the sampling error of the signal sampling results reaches a preset error threshold, and if the sampling error reaches the error threshold, it determines the time boundary value based on the current time code and the signal sampling center.

[0103] Specifically, the receiving end moves the voltage code up or down relative to the signal sampling center, and samples the signal transmitted on the link to be equalized based on the moved voltage code, obtaining the signal sampling result. The receiving end determines whether the sampling error of the signal sampling result reaches a preset error threshold. If the sampling error of the signal sampling result does not reach the error threshold, the computer device executes the steps of moving the voltage code up or down and sampling the signal transmitted on the link to be equalized based on the moved voltage code, until the sampling error reaches the error threshold. The computer device then determines the voltage boundary value based on the current voltage code and the signal sampling center.

[0104] Step 406: Based on the time boundary value and voltage boundary value, construct the signal sampling range corresponding to the link equalization parameters.

[0105] The signal sampling range indicates the range of parameter values ​​that the signal sampling results fall within, meeting certain requirements. Time boundary values ​​include a first time boundary value and a second time boundary value; voltage boundary values ​​include a second voltage boundary value and a first voltage boundary value.

[0106] In implementation, the computer equipment uses the signal sampling center as the center and constructs the signal sampling range corresponding to the link equalization parameters based on the first time boundary value, the second time boundary value, the first voltage boundary value, and the second voltage boundary value. The signal sampling range data includes an eye diagram of the signal under the current group of link equalization parameters.

[0107] In one exemplary embodiment, Figure 5 This is a schematic diagram of an eye diagram in an exemplary embodiment. For example... Figure 5 As shown, the horizontal axis represents time, and the vertical axis represents voltage. The margin formed by quadrilateral ABED is symmetrical in both the timing and voltage directions, while the margin formed by quadrilateral AB1CD is symmetrical only in the voltage direction. There are also margins that are not symmetrical in either the voltage or timing directions (no additional diagram is provided here).

[0108] In this embodiment, sampling is performed from both time and voltage perspectives using a signal sampling center to obtain the actual signal sampling range of the link to be balanced, thus clarifying the signal transmission stability under each set of link equalization parameters. Furthermore, by performing a destructive scan on the link to be balanced, compared to the prior art's bypass scanning of the link and correction of scanning errors, the accuracy of the signal sampling range is significantly improved. Moreover, omitting the step of correcting scanning errors also improves the efficiency of the link equalization method.

[0109] In one exemplary embodiment, such as Figure 6 As shown, the specific processing procedure of step 402 includes steps 602 to 604. Wherein:

[0110] Step 602: Using the signal sampling center as the center, move the time code horizontally according to the preset time offset unit, and sample the signal transmitted on the link to be equalized based on the moved time code to obtain the signal sampling result.

[0111] In implementation, a time offset unit is pre-set in the receiver, which is the step size of each time the signal sampling center adjusts the time. The receiver moves the time code to the left (right) of the signal sampling center according to the time offset unit, and samples the signal transmitted on the link to be equalized based on the moved time code, obtaining the signal sampling result. Then, the receiver determines whether the sampling error of the signal sampling result reaches a preset error threshold. The requirements for the sampling error can be set by those skilled in the art, such as the bit error rate being less than a certain bit error rate threshold, the difference between the bit error rate and the bit error rate of the sampling result using the initial value being greater than a certain threshold, the bit error rate being lower than the bit error rate when using the currently tested optimal preset link equalization parameters, the signal-to-noise ratio being greater than a certain threshold, etc.

[0112] It should be noted here that the time boundary values ​​include a first time boundary value and a second time boundary value. Therefore, the receiving end needs to shift the time code to the left to determine the first time boundary value, and also needs to shift the time code to the right to determine the second time boundary value.

[0113] Optionally, the time offset unit can be set by those skilled in the art according to actual needs. It should be noted that, for the first set of link equalization parameters received by the receiver, the receiver can start from the signal sampling center and obtain the signal sampling range by adjusting the time code in small steps each time; for link equalization parameters not received by the receiver, the receiver can start from the boundary of the signal sampling range of the currently tested optimal set of link equalization parameters and obtain the signal sampling range by adjusting the time code in small steps each time, thereby accelerating the speed of obtaining the signal sampling range for link equalization parameters not received by the receiver.

[0114] Step 604: If the sampling error of the signal sampling result does not reach the preset error threshold, perform the step of moving the time code horizontally with the signal sampling center as the center and according to the preset time offset unit until the sampling error reaches the error threshold, and determine the sampling time as the time boundary value in the time dimension.

[0115] The time boundary values ​​include the first time boundary value and the second time boundary value.

[0116] In implementation, if the sampling error of the signal sampling result does not reach the preset error threshold, the receiver continues to execute step 602 above until the sampling error reaches the error threshold. At this point, a boundary value in the time dimension is reached. The receiver determines the time boundary value based on the current timecode and the signal sampling center. The receiver determines whether all time boundary values ​​in the time dimension have been acquired (that is, whether the first and second time boundary values ​​have been determined). If not all time boundary values ​​have been acquired, the above process is repeated until all time boundary values ​​in the time dimension are obtained. In other words, the receiver also shifts the timecode to the right to determine the second time boundary value.

[0117] The error threshold can be a preset value (for example, the error threshold for the bit error rate can be 1e). -6 (where e is a natural constant). When the error threshold is set to a preset value, the time boundary value obtained by adjusting the timecode represents the actual boundary of the parameter, that is, a boundary within which the parameter can be stably sampled. By setting the error threshold in this way, the optimal set of link equalization parameters can be determined based on the actual boundary of the parameter.

[0118] The error threshold can also be determined based on the target sampling error, which is the error of the value currently adjusted to the signal sampling center corresponding to the previous set of link equalization parameters. The error threshold can be the target sampling error itself, or it can be the sum of the target sampling error and a preset value, etc. When setting the error threshold in this way, the signal sampling range obtained by adjusting the signal sampling center represents the range within which the current link equalization parameters can have the same performance as the candidate target link equalization parameters.

[0119] In this embodiment, before applying the link to be balanced, the deviation parameter of the time dimension is actively adjusted based on the signal sampling center, and the signal under each set of link balancing parameters is continuously sampled. This allows for the acquisition of time boundary values, clarifies the time sampling boundary of the signal, and facilitates the subsequent determination of the signal sampling range based on the time boundary values.

[0120] In one exemplary embodiment, such as Figure 7 As shown, the specific processing procedure of step 404 includes steps 702 to 704. Wherein:

[0121] Step 702: Using the signal sampling center as the center, move the voltage code in the vertical direction according to the preset voltage offset unit, and sample the signal transmitted on the link to be equalized based on the moved voltage code to obtain the signal sampling result.

[0122] In implementation, a voltage offset unit is pre-set in the receiver, which is the step size of each voltage adjustment at the signal sampling center. The receiver moves the voltage code up (or down) according to the voltage offset unit, centered on the signal sampling center, and samples the signal transmitted on the link to be equalized based on the moved voltage code, obtaining the signal sampling result. Then, the receiver determines whether the sampling error of the signal sampling result reaches a preset error threshold. This is the same as the sampling error of time in step 602 above. The requirements for the sampling error here can be set by those skilled in the art, such as the bit error rate being less than a certain bit error rate threshold, the difference between the bit error rate and the bit error rate of the sampling result using the initial value being greater than a certain threshold, the bit error rate being lower than the bit error rate when using the currently tested optimal preset link equalization parameters, the signal-to-noise ratio being greater than a certain threshold, etc.

[0123] It should be noted here that the voltage boundary values ​​include a first voltage boundary value and a second voltage boundary value. Therefore, the receiving end needs to shift the voltage code upwards to determine the first voltage boundary value, and also needs to shift the voltage code downwards to determine the second voltage boundary value.

[0124] Optionally, the step size of the voltage adjustment at the signal sampling center can be set by those skilled in the art according to actual needs. It should be noted that, for the first set of link equalization parameters received by the receiver, the receiver can start from the signal sampling center and obtain the signal sampling range by adjusting the voltage code in small steps each time; for link equalization parameters not received by the receiver, the receiver can start from the boundary of the signal sampling range of the currently tested optimal set of link equalization parameters and obtain the signal sampling range by adjusting the voltage code in small steps each time, thereby accelerating the speed of obtaining the signal sampling range for link equalization parameters not received by the receiver.

[0125] Step 704: If the sampling error of the signal sampling result does not reach the error threshold, perform the step of moving the voltage code in the vertical direction with the signal sampling center as the center and according to the preset voltage offset unit until the sampling error reaches the error threshold, and determine the sampled voltage as the voltage boundary value in the voltage dimension.

[0126] In implementation, if the sampling error of the signal sampling result does not reach the preset error threshold, the receiver continues to execute step 702 above until the sampling error reaches the error threshold. At this point, a boundary value in the voltage dimension is reached. The receiver determines the voltage boundary value based on the current voltage code and the signal sampling center. The receiver determines whether all voltage boundary values ​​in the voltage dimension have been obtained. If not, the above process is repeated until all voltage boundary values ​​in the voltage dimension are obtained. That is, the receiver also moves the voltage code to the right to determine the second voltage boundary value.

[0127] The error threshold can be a preset value (for example, the error threshold for the bit error rate can be 1e). -6 When the error threshold is set to a preset value, the voltage boundary value obtained by adjusting the voltage code represents the actual boundary of the parameter, that is, a boundary within which the parameter can be stably sampled. By setting the error threshold in this way, the optimal set of link equalization parameters can be determined based on the actual boundary of the parameter.

[0128] In this embodiment, before applying the link to be balanced, the voltage dimension deviation parameter is actively adjusted based on the signal sampling center, and the signal under each set of link balancing parameters is continuously sampled. This allows the voltage boundary value to be obtained, the voltage sampling boundary of the signal to be clearly defined, and the signal sampling range to be determined based on the voltage boundary value in the future.

[0129] In one exemplary embodiment, such as Figure 8 As shown, the specific processing procedure of step 206 includes steps 802 to 804. Wherein:

[0130] Step 802: Based on the stability evaluation algorithm and the sampling range of each signal, determine the eccentricity of the signal under the stability index and the effective length under the anti-interference index, and determine the eye diagram evaluation value of the signal sampling range based on the eccentricity and the effective length.

[0131] The eye diagram evaluation value is obtained by considering the signal sampling range corresponding to the parameters of each dimension used during signal sampling, and can reflect the signal sampling margin under the current group link equalization parameters. The eye diagram evaluation value can be the eye diagram area, or it can be any value such as eye diagram height, eye diagram width, eye diagram eccentricity, or a combination of these values. This application embodiment does not specifically limit this.

[0132] In implementation, after obtaining the signal sampling range, the receiving end calculates the eye diagram evaluation value based on the boundary values ​​within the signal sampling range and the signal sampling center. The signal sampling center is the optimal value obtained after channel adaptation by the receiving end. The boundary values ​​include the boundaries of the signal sampling range. Generally, the signal sampling range is continuous, therefore it usually contains two boundaries. However, in the special case where the signal sampling range consists of multiple discontinuous ranges, the boundaries here include the boundaries of each discontinuous range.

[0133] To illustrate with a practical example, if the dimension is the sampling voltage threshold, and the receiver performs channel adaptation and determines that the optimal value of the sampling voltage threshold is 1.0V (volts, a unit of voltage), and tests show that the signal sampling range of the sampling voltage threshold is 0.7V to 1.3V, then the signal sampling center corresponding to the sampling voltage threshold is 1.0V, and the boundary values ​​are 0.7V and 1.3V.

[0134] The receiver can calculate a numerical value representing the shape of the eye diagram as an eye diagram evaluation value based on the initial and boundary values ​​of each signal sampling range. Examples include: eye diagram area (representing the overall margin level of signal sampling), eye diagram diagonal length or side length (representing whether the margins in different dimensions are balanced), and the degree of deviation between the eye diagram center and the initial value (representing whether the margins in different dimensions are symmetrical). Alternatively, the receiver can first calculate an evaluation value for each dimension based on the initial and boundary values ​​(e.g., effective length (e.g., the difference between two boundary values, or calculating the difference between two boundary values ​​and the initial value separately, then multiplying the smaller difference by 2), eccentricity (e.g., the difference between the average of the boundary values ​​and the initial value), etc.), and then synthesize the evaluation values ​​of each dimension to obtain the eye diagram evaluation value. For example, the evaluation values ​​for each dimension can be summed or multiplied.

[0135] In an exemplary embodiment, to ensure interference resistance and stability during signal transmission, the receiving end processes each signal sampling range based on a stability evaluation algorithm to obtain the eccentricity under the stability index and the effective length under the interference resistance index, and determines the eye diagram evaluation value of the signal sampling range based on the effective length and eccentricity. A larger eye diagram evaluation value indicates stronger stability and interference resistance during signal transmission.

[0136] Step 804: Based on the eye diagram evaluation values, determine the target link balancing parameters corresponding to the link to be balanced in each group of link balancing parameters.

[0137] In practice, the eye diagram evaluation values ​​at the receiving end are used to determine the set of link equalization parameters with the strongest stability and anti-interference capability among the various sets of link equalization parameters, which are then used as the target link equalization parameters.

[0138] Specifically, the receiving end determines the maximum eye diagram evaluation value among all eye diagram evaluation values ​​as the target eye diagram evaluation value, and determines a set of link equalization parameters corresponding to the target eye diagram evaluation value as the target link equalization parameters.

[0139] In this embodiment, the optimal set of link equalization parameters is evaluated using signal sampling range and stability assessment algorithms. This enables real-time determination of the target link equalization parameters for the link to be equalized before applying the parameters, improving the evaluation efficiency of the target link equalization parameters and thus enhancing the adjustment efficiency of the link equalization parameters.

[0140] In one exemplary embodiment, such as Figure 9 As shown, the specific processing procedure of step 802 includes steps 902 to 904. Wherein:

[0141] Step 902: For each dimension in the signal sampling range, determine the eccentricity of the signal under the stability index and the effective length under the anti-interference index based on the boundary values ​​of the dimension.

[0142] The signal sampling range includes boundary values ​​for each dimension. The effective length characterizes the margin in that dimension and can be calculated by the difference between the maximum and minimum boundary values ​​for each dimension. A larger effective length results in a larger margin at the signal sampling center. Eccentricity characterizes the degree to which the parameter deviates from its initial value in different directions and can be calculated by the difference between the average of the boundary values ​​and the initial value. A smaller eccentricity indicates a more uniform deviation from the initial value in different directions, resulting in higher anti-interference capability and stability during sampling.

[0143] In practice, the receiver determines the eccentricity of the signal under the stability index and the effective length under the anti-interference index for each dimension of the signal sampling range, based on the first and second boundary values ​​of the dimension.

[0144] Specifically, for each boundary value in each dimension, the receiver determines the minimum boundary value and, based on the minimum boundary value, determines the effective length of the dimension. Then, the receiver determines the maximum boundary value and the difference between the maximum and minimum boundary values. The ratio between this difference and the maximum boundary value is then used as the eccentricity of the dimension.

[0145] Step 904: Determine the eye diagram evaluation values ​​of the link equalization parameters based on the stability evaluation algorithm, each effective length, and each eccentricity.

[0146] The signal sampling evaluation value is a numerical value calculated based on the effective length and eccentricity. It is directly proportional to the effective length and inversely proportional to the eccentricity, and is used to quantify the performance of a single dimension under the current preset link equalization parameters. For example, the signal sampling evaluation value can be obtained by calculating the ratio of the effective length to the eccentricity or the difference between the effective length and the eccentricity.

[0147] In practice, the receiver processes data on each effectiveness level and each eccentricity according to the stability assessment algorithm to obtain the eye diagram evaluation value of the set of link equalization parameters.

[0148] Specifically, the receiver determines the signal sampling evaluation value for each dimension based on the effective length and eccentricity of that dimension. Then, the receiver processes the signal sampling evaluation values ​​for each dimension using a stability evaluation algorithm to obtain the eye diagram evaluation value for that set of link equalization parameters. The eye diagram evaluation value is obtained by aggregating all signal sampling evaluation values; for example, it can be obtained by summing or multiplying all signal sampling evaluation values.

[0149] In this embodiment, the anti-interference capability and stability of the signal under the current group of link equalization parameters are quantified by calculating the effective length and eccentricity of each signal boundary value in each dimension. Then, the eye diagram evaluation value under the current group of link equalization parameters is determined by each effective length and eccentricity, which quantifies the transmission quality of the signal under the current group of link equalization parameters. This facilitates the subsequent selection of target link equalization parameters based on the eye diagram evaluation value, thereby improving the signal transmission quality.

[0150] In one exemplary embodiment, such as Figure 10 As shown, the specific processing procedure of step 902 includes steps 1002 to 1006. Wherein:

[0151] Step 1002: Determine the minimum boundary value among all boundary values, and determine the effective length corresponding to the dimension based on the minimum boundary value.

[0152] The effective length characterizes the signal's anti-interference capability; the larger the effective length, the greater the signal's anti-interference capability.

[0153] In practice, the receiving end determines the minimum boundary value for each dimension based on the difference between each boundary value and the signal sampling distance. Then, the computer device determines the effective length corresponding to that dimension based on the minimum boundary value.

[0154] Specifically, the receiver calculates the difference between each boundary value and the signal sampling center for each dimension. Then, the computer determines the boundary value with the smallest difference as the minimum boundary value for that dimension. The computer determines the effective length as twice the minimum boundary value, ensuring that the effective length takes into account the symmetry of the signal sampling range with respect to the signal sampling center, and accurately reflects the receiver's anti-interference capability against a set of link equalization parameters.

[0155] In an exemplary embodiment, the dimensions are time and voltage, respectively. The boundary value corresponding to time is the time boundary value, and the boundary value corresponding to voltage is the voltage boundary value. The time boundary value includes a first time boundary value and a second time boundary value. The voltage boundary value includes a first voltage boundary value and a second voltage boundary value. The receiving end calculates the difference between each time boundary value and the signal sampling center. Then, the computer device determines the time boundary value with the smallest difference as the minimum time boundary value for that dimension. The computer device determines twice the minimum time boundary value as the effective length. This effective length is the effective eye width of the eye diagram. The algorithm for the effective eye width is shown in the following formula (1):

[0156] (1)

[0157] In the above formula (1), For effective eye width, The first time boundary value, This is the second time boundary value. This indicates the minimum.

[0158] The receiver calculates the difference between each voltage boundary value and the signal sampling center. Then, the computer determines the voltage boundary value with the smallest difference as the minimum voltage boundary value for that dimension. The computer determines twice the minimum voltage boundary value as the effective length. This effective length is the effective eye height of the eye diagram. The algorithm for the effective eye width is shown in the following formula (2):

[0159] (2)

[0160] In the above formula (2), To effectively improve eye level, This is the first voltage boundary value. This is the second voltage boundary value. This indicates the minimum.

[0161] Step 1004: Determine the maximum boundary value among all boundary values, and determine the difference between the maximum boundary value and the minimum boundary value.

[0162] In practice, the receiver determines the maximum boundary value for each dimension based on the difference between each boundary value and the signal sampling distance. Then, the computer device determines the difference between the maximum and minimum boundary values ​​in that dimension.

[0163] Specifically, the receiving end calculates the difference between each boundary value and the signal sampling center for each dimension. Then, the computer device determines the boundary value with the largest difference as the maximum boundary value for that dimension. The computer device then calculates the difference between the maximum and minimum boundary values ​​to obtain the difference between the maximum and minimum boundary values.

[0164] In one exemplary embodiment, the dimensions are time and voltage, respectively. The boundary value corresponding to time is the time boundary value, and the boundary value corresponding to voltage is the voltage boundary value. The time boundary value includes a first time boundary value and a second time boundary value. The voltage boundary value includes a first voltage boundary value and a second voltage boundary value. The receiving end calculates the difference between each time boundary value and the signal sampling center. Then, the computer device determines the time boundary value with the largest difference as the maximum time boundary value for that dimension, and calculates the difference between the maximum and minimum time boundary values ​​to obtain the time boundary value difference. The receiving end calculates the difference between each voltage boundary value and the signal sampling center. Then, the computer device determines the voltage boundary value with the largest difference as the maximum voltage boundary value for that dimension, and calculates the difference between the maximum and minimum voltage boundary values ​​to obtain the voltage boundary value difference.

[0165] Step 1006: The ratio between the difference and the maximum boundary value is determined as the eccentricity corresponding to the dimension.

[0166] Eccentricity characterizes the stability of a signal; the smaller the eccentricity, the better the stability of the signal.

[0167] In practice, the receiving end determines the eccentricity of that dimension as the ratio between the difference and the maximum boundary value.

[0168] Specifically, the receiver determines the time eccentricity corresponding to the time dimension by the ratio between the time boundary value difference and the maximum time boundary value. Similarly, the receiver determines the voltage eccentricity corresponding to the voltage dimension by the ratio between the voltage boundary value difference and the maximum voltage boundary value. Time eccentricity is also known as eye width eccentricity, and voltage eccentricity is also known as eye height eccentricity.

[0169] The time eccentricity is shown in formula (3) and the voltage eccentricity is shown in formula (4).

[0170] T_skew=(max(T1, T2) –min(T1, T2)) / max(T1, T2) (3)

[0171] V_skew=(max(V1, V2) –min(V1, V2)) / max(V1, V2)(4)

[0172] In the above formulas (3) and (4), T_skew is the eye width eccentricity, and V_skew is the eye height eccentricity. V1 is the first voltage boundary value, and V2 is the second voltage boundary value. The minimum value is represented by T1, the second time boundary value is represented by T2, and the maximum value is represented by max.

[0173] In an optional embodiment, after calculating the ratio of the minimum difference to the maximum difference, the eccentricity can be obtained by subtracting the ratio from 1, so that the smaller the eccentricity, the more evenly the parameters deviate from their initial values ​​in different directions.

[0174] Optionally, this application only mentions one method for calculating eccentricity, but other methods may also be used. The method can be selected according to the requirements, and the embodiments of this application are not limited here.

[0175] In this embodiment, the effective length and eccentricity are calculated by the signal boundary values ​​of each dimension, which quantifies the anti-interference capability and stability of the signal under the current group of link equalization parameters. This facilitates the subsequent selection of target link equalization parameters based on the effective length and evaluation value, thereby improving the signal transmission quality.

[0176] In one exemplary embodiment, such as Figure 11As shown, the specific processing procedure of step 904 includes steps 1102 to 1104. Wherein:

[0177] Step 1102: Determine the signal sampling evaluation value for each dimension based on the effective length and eccentricity of each dimension.

[0178] Among them, the signal sampling evaluation value is directly proportional to the effective length and inversely proportional to the eccentricity.

[0179] In practice, the receiver adjusts the effective length based on the eccentricity for each dimension to obtain the signal sampling evaluation value for that dimension.

[0180] Specifically, each dimension is voltage and time. The effective length includes the effective eye width corresponding to time and the effective eye height corresponding to voltage. The eccentricity is divided into eye width eccentricity and eye height eccentricity corresponding to time. The receiver corrects the effective eye width according to the eye width eccentricity to obtain the signal sampling evaluation value of the time dimension, and corrects the effective eye height according to the eye height eccentricity to obtain the signal sampling evaluation value of the voltage dimension. The correction process of the effective eye width is shown in the following formula (5), and the correction process of the effective eye height is shown in the following formula (6):

[0181] The time dimension signal sampling evaluation value = width - T_skew*width (5)

[0182] The voltage dimension signal sampling evaluation value = height - V_skew*height (6)

[0183] In formulas (5) and (6) above, width is the effective eye width, height is the effective eye height, T_skew is the eye width eccentricity, and V_skew is the eye height eccentricity.

[0184] Step 1104: Based on the stability assessment algorithm, the signal sampling evaluation values ​​of each dimension are processed to obtain the eye diagram evaluation values ​​of the link equalization parameters.

[0185] In implementation, the receiver processes the signal sampling evaluation values ​​of each dimension according to the stability assessment algorithm to obtain the eye diagram evaluation values ​​of the link equalization parameters. For example, the receiver determines the length of the signal sampling evaluation values ​​of each dimension as a set of eye diagram evaluation values ​​for the link equalization parameters.

[0186] In one example, taking a dimension that includes sampling timing and sampling voltage threshold, the eye diagram evaluation value can be calculated as shown in formula (7):

[0187] EVAL = (width - T_skew*width)*(height - V_skew*height) (7)

[0188] In formula (7) above, width is the effective eye width, height is the effective eye height, T_skew is the eye width eccentricity, V_skew is the eye height eccentricity, and EVAL is the eye diagram evaluation value.

[0189] In an optional embodiment, in addition to considering the effect of eccentricity, the ratio of eye height to eye width is also considered. A larger eye width indicates a greater reduction in the impact of jitter, while a larger eye height indicates a greater reduction in the impact of noise. Therefore, in practical signal quality issues, the ratio of eye height to eye width needs to be adjusted according to the specific problem. Thus, this application provides another method for calculating the eye diagram evaluation value based on the ratio, as shown in the following formula (8):

[0190] newEVAL = A*(width - T_skew*width) + B*(height - V_skew*height) (8)

[0191] In the above formula (8), A and B are proportional coefficients, width is the effective eye width, height is the effective eye height, T_skew is the eye width eccentricity, V_skew is the eye height eccentricity, and newEVAL is the eye diagram evaluation value.

[0192] In this embodiment, the eye diagram evaluation value under the current group link equalization parameters is determined by each effective length and each eccentricity, which quantifies the signal transmission quality under the current group link equalization parameters. This facilitates the subsequent selection of target link equalization parameters based on the eye diagram evaluation value, thereby improving the signal transmission quality.

[0193] In one exemplary embodiment, such as Figure 12 As shown, the specific processing procedure of step 804 includes steps 1202 to 1204. Wherein:

[0194] Step 1202: Determine the maximum eye diagram evaluation value as the target eye diagram evaluation value from among all eye diagram evaluation values.

[0195] In practice, the receiving end determines the maximum eye diagram evaluation value as the target eye diagram evaluation value among all eye diagram evaluation values.

[0196] Step 1204: Determine the link balancing parameters corresponding to the target eye diagram evaluation value as the target link balancing parameters.

[0197] In implementation, the receiving end determines a set of link equalization parameters corresponding to the target eye diagram evaluation value as the target link equalization parameters. Since a larger eye diagram evaluation value indicates higher signal stability and anti-interference capability on the link to be equalized, determining a set of link equalization parameters corresponding to the maximum eye diagram evaluation value as the target link equalization parameters and performing link equalization based on these target link equalization parameters can improve signal stability and anti-interference capability.

[0198] In this embodiment, the larger the eye diagram evaluation value, the higher the stability and anti-interference capability of the signal transmission on the link to be balanced. By determining a set of link equalization parameters corresponding to the maximum eye diagram evaluation value as the target link equalization parameters, and performing link equalization on the link to be balanced based on the target link equalization parameters, the signal transmission quality of the link to be balanced can be improved.

[0199] In one exemplary embodiment, Figure 13 This is a flowchart of a link balancing method in an exemplary embodiment. Figure 13 As shown, the specific processing of this link balancing method includes:

[0200] Step 1301: Determine the number of link balancing parameter combinations. Each set of link balancing parameters (preset) can be represented as P0, P1, P2...PM. M is the number of link balancing parameter combinations.

[0201] Step 1302: Configure the nth set of link balancing parameters to the transmitting end connected to the link to be balanced. Here, n is a natural number greater than 0 and less than or equal to M.

[0202] Step 1303: The receiving end performs channel adaptation to obtain the signal sampling center corresponding to the current group link equalization parameters. The current group link equalization parameters are the same as the nth group link equalization parameters in step 1302.

[0203] Step 1304: Based on the signal sampling center, adjust the deviation parameters of each dimension and perform up, down, left, and right scans to obtain four values: T1, T2, V1, and V2. Here, T1 is the first time boundary value, T2 is the second time boundary value, V1 is the first voltage boundary value, and V2 is the second voltage boundary value. The four values ​​T1, T2, V1, and V2 constitute the signal sampling range.

[0204] Step 1305: Calculate the eye height, eye width, and eye height eccentricity based on the four values ​​T1, T2, V1, and V2. Specifically, eye width = min(T1, T2) * 2; eye height = min(V1, V2) * 2; eye width eccentricity T_skew = (max(T1, T2) – min(T1, T2)) / max(T1, T2); and eye height eccentricity V_skew = (max(V1, V2) – min(V1, V2)) / max(V1, V2).

[0205] Step 1306: Determine the eye diagram evaluation value of the link equalization parameters for the current group based on the stability evaluation algorithm, eye height and eye width, and the eccentricity of eye height and eye width. The eye diagram evaluation value newEVAL_N = (width - T_skew*width) * (height - V_skew*height).

[0206] Step 1307: Determine whether the eye diagram evaluation value of the current group link balancer parameters is greater than the eye diagram evaluation value of the candidate target link balancer parameters. The eye diagram evaluation value of the candidate target link balancer parameters is initially the eye diagram evaluation value of the previous group of link balancer parameters. If the eye diagram evaluation value of the current group link balancer parameters is greater than the eye diagram evaluation value of the candidate target link balancer parameters, proceed to step 1308. If the eye diagram evaluation value of the current group link balancer parameters is less than or equal to the eye diagram evaluation value of the candidate target link balancer parameters, proceed to step 1309.

[0207] Step 1308: Update the current group link balancing parameters to the candidate target link balancing parameters.

[0208] Step 1309: Determine whether n exceeds M. If n exceeds M, proceed to step 1311 below; if n does not exceed M, proceed to step 1310 below.

[0209] Step 1310, increment n by 1.

[0210] Step 1311: Determine the candidate target link balancing parameters as the target link balancing parameters, and apply the target link balancing parameters to the sending end for link balancing.

[0211] The above steps only list the single-lane (balanced link) process. In fact, the multi-lane process is the same as the single-lane core process. Taking the TX preset selection as an example in the single-lane process, it also applies to other TX and RX balancing parameters.

[0212] In step 1304 above, when adjusting the deviation parameter, starting from the signal sampling center, the deviation parameter is adjusted multiple times in any adjustment direction of any dimension. After the adjustment, the receiving end samples the signal transmitted on the link based on the adjusted parameter value and calculates the sampling error of the sampling result.

[0213] If the sampling error exceeds the preset error threshold, it is considered that a boundary value for that dimension has been reached. If not all boundary values ​​for that dimension have been obtained at this point, the above process is repeated until all boundary values ​​for that dimension are obtained. Then, the dimension to be adjusted is changed until the signal sampling center adjustment has been performed for all dimensions.

[0214] The error threshold can be a preset value (for example, the error threshold for the bit error rate can be 1e). -6 When the error threshold is set to a preset value, the signal sampling range obtained by adjusting the signal sampling center represents the actual boundary of the parameter, that is, a boundary within which the parameter can be stably sampled. By setting the error threshold in this way, the optimal preset link equalization parameter can be determined based on the actual boundary of the parameter.

[0215] The error threshold can also be determined based on the target sampling error, which is the error of the value currently adjusted to the signal sampling center corresponding to the candidate target link equalization parameter. The error threshold can be the target sampling error itself, or it can be the sum of the target sampling error and a preset value, etc. When setting the error threshold in this way, the signal sampling range obtained by adjusting the signal sampling center represents the range within which the current link equalization parameter can have the same performance as the candidate target link equalization parameter.

[0216] The above process is illustrated with a practical example. Let's take the sampled voltage threshold as an example of the dimension being adjusted. Assume the target boundary values ​​for the sampled voltage thresholds corresponding to the candidate target link equalization parameters are 0.8V and 1.2V, and the initial value of the current group of link equalization parameters is 1.0V. Starting from halfway between the target boundary value and the initial value (i.e., 0.9V and 1.1V), the signal sampling center is adjusted in the direction of the target boundary value. The stopping condition is set when the bit error rate of the sampling result is higher than the bit error rate of the candidate target link equalization parameter at this position, or when the current value of the parameter in this dimension exceeds the target boundary value by 0.1V.

[0217] First, signal sampling was performed at 0.9V. The bit error rate (BER) of the sampled signal was the same as that of the candidate target link equalization parameters at 0.9V, so adjustments were continued in this direction. Second, signal sampling was performed at 0.85V. The BER of the sampled signal was greater than that of the candidate target link equalization parameters at 0.85V. At this point, it was verified that the current group of link equalization parameters performed worse than the candidate target link equalization parameters in this direction. Therefore, adjustments in this direction were stopped, and adjustments were started at the next target boundary value (1.2V).

[0218] Assuming that when adjusting towards 1.2V, the bit error rate (BER) remains lower than that of the candidate target link equalization parameter at 1.25V until it reaches 1.25V, it has been verified that the current group of link equalization parameters performs sufficiently better than the candidate target link equalization parameter in this direction. Therefore, adjustment in this direction is stopped. The signal sampling range used in subsequent eye diagram evaluation calculations will be from 0.85V to 1.25V.

[0219] It's important to further clarify that numerical thresholds can also be determined in real-time. For example, regarding a specific dimension, if adjustments are made to a target boundary value corresponding to that dimension, and the adjustment termination condition is triggered before reaching the target boundary value, then the numerical thresholds for other target boundary values ​​corresponding to that dimension can be set according to the value at which adjustment stopped and the algorithm used to calculate the signal sampling evaluation value. This ensures that the sum of the value at which adjustment stopped based on this target boundary value, the other target boundary values ​​and their numerical thresholds, and the algorithm used to calculate the signal sampling evaluation value, is at least no worse than the signal sampling evaluation value corresponding to the candidate target link equalization parameter for this dimension. Alternatively, the numerical thresholds for other target boundary values ​​can be set according to the difference between the value at which adjustment stopped and the target boundary value. For example, calculate the difference between the target boundary value and the value at which adjustment stopped, and then use a multiple (e.g., 2 times, 3 times) of this difference as the numerical threshold.

[0220] Using the previous example, since adjusting to 0.85V triggers the termination condition when adjusting towards 0.8V, when adjusting towards 1.2V, we can calculate the minimum value the current link equalization parameter needs to be adjusted to in this direction to ensure its performance is not inferior to the candidate target link equalization parameter. For example, if the signal sampling evaluation value is the actual width of the signal sampling range, the current link equalization parameter needs to be adjusted to 1.25V in this direction without triggering the termination condition to ensure its performance is not inferior to the candidate target link equalization parameter. Therefore, the threshold value is 0.05V.

[0221] In steps 1307 to 1311, the candidate target link equalization parameters are the currently optimal parameters temporarily stored by the receiver. When the receiver receives the first set of current link equalization parameters, after calculating the eye diagram evaluation value of this set of current link equalization parameters, the receiver can directly use this set of current link equalization parameters as the candidate target link equalization parameters.

[0222] After this, each time the receiver receives the link equalization parameters for the current group, it calculates the eye diagram evaluation value of the current group's link equalization parameters and then compares the eye diagram evaluation value with the eye diagram evaluation values ​​of the saved candidate target link equalization parameters. If the eye diagram evaluation value corresponding to the current group's link equalization parameters is greater than the eye diagram evaluation value of the candidate target link equalization parameters, the receiver updates the current group's link equalization parameters to the new candidate target link equalization parameters; otherwise, the candidate target link equalization parameters remain unchanged.

[0223] The receiving end then checks whether all groups of link balancing parameters have been received. The sending end can inform the receiving end of the total number of preset link balancing parameters before link balancing begins. The receiving end can then determine whether all groups of link balancing parameters have been received by accumulating the number of groups of received link balancing parameters. Alternatively, the sending end can inform the receiving end of the sequence number of the current group of link balancing parameters and the total number of groups of link balancing parameters each time it transmits the current link balancing parameters. This application embodiment does not limit how to determine whether all preset link balancing parameters have been received.

[0224] If the receiving end does not receive all the link balancing parameters, it waits for the sending end to send the next set of link balancing parameters and repeats the steps in step 1302. If it has received all the preset link balancing parameters, the receiving end uses the currently temporarily saved optimal candidate target link balancing parameters as the final target link balancing parameters, sends these parameters back to the sending end, and communicates with the sending end to complete the final configuration of link balancing.

[0225] In one exemplary embodiment, examples of the stability evaluation algorithm and standard algorithm in step 1306 (or step 206 above) are provided.

[0226] Standard algorithm for eye opening area: area = (WIDTH * 1 / 35 * 31.25) * (HEIGHT * 3.7 / 1000), unit pV / cdotps. Where WIDTH is the total eye width and HEIGHT is the total eye height. Table 1 is a comparison table between the standard algorithm and the stability evaluation algorithm of this application.

[0227] Table 1

[0228]

[0229] In Table 1 above, preset represents a set of link equalization parameters. Under preset5 and preset7, two sets of T (time boundary value) and V (voltage boundary value) data are obtained respectively. T_skew represents time eccentricity, and V_skew represents voltage eccentricity.

[0230] As shown in Table 1, the eye diagram is eccentric: there is asymmetry in the vertical direction of preset7 (V2>V1), which may be caused by noise or inter-symbol interference; while the timing jitter in the horizontal direction is more obvious (T2>T1), and the signal stability is poor.

[0231] Eye diagram symmetry: The symmetry of preset5 indicates that the signal is more stable in amplitude and timing. Even if the absolute value is slightly lower, the anti-interference ability brought by the balance is stronger, which can reduce the bit error rate.

[0232] Therefore, while maximizing the eye opening area is desirable, if the eye opening areas are similar, then eye diagram symmetry should be emphasized. The stability evaluation is not intended to overturn the standard algorithm, but rather to further optimize and supplement the adaptive equalization process.

[0233] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0234] Based on the same inventive concept, this application also provides a link balancing device for implementing the link balancing method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more link balancing device embodiments provided below can be found in the limitations of the link balancing method described above, and will not be repeated here.

[0235] In one embodiment, such as Figure 14 As shown, a link equalization device 1400 is provided, including: an acquisition module 1401, a sampling module 1402, and a determination module 1403, wherein:

[0236] The acquisition module 1401 is used to acquire each set of link equalization parameters of the link to be equalized, and to determine the signal sampling center corresponding to each set of link equalization parameters.

[0237] The sampling module 1402 is used to continuously adjust the deviation parameters of each dimension based on the signal sampling center, and to sample the signals transmitted on the link to be balanced based on the adjusted deviation parameters to obtain the sampling results of each signal, and to determine the signal sampling range corresponding to the link equalization parameters based on the sampling results of each signal.

[0238] The determination module 1403 is used to determine the target link equalization parameters from each group of link equalization parameters based on the sampling range of each signal and the stability evaluation algorithm; the target link equalization parameters are used to perform link equalization on the link to be equalized.

[0239] In one embodiment, the acquisition module 1401 includes a first acquisition submodule and a first determination submodule. Specifically, the first determination submodule is used to: configure the transmitting end corresponding to the link to be balanced based on each set of link balancing parameters; the signal transmitted on the link to be balanced is sent by the transmitting end on the link to be balanced based on each set of link balancing parameters; and perform channel adaptation based on the signal to obtain the signal sampling center of the link balancing parameters.

[0240] In one embodiment, each dimension includes a time dimension and a voltage dimension, the deviation parameter of the time dimension is a time code, and the voltage parameter of the voltage dimension is a voltage code. The sampling module 1402 includes:

[0241] The first sampling submodule is used to continuously adjust the time code from a horizontal angle based on the signal sampling center, and to sample the signals transmitted on the link to be equalized based on the adjusted time code to obtain the sampling results of each signal, and to determine the time boundary value based on the sampling results of each signal and the preset error threshold.

[0242] The second sampling submodule is used to continuously adjust the voltage code from a vertical angle based on the signal sampling center, and to sample the signals transmitted on the link to be equalized based on the adjusted voltage code to obtain the sampling results of each signal, and to determine the voltage boundary value based on the sampling results of each signal and the error threshold.

[0243] The first construction submodule is used to construct the signal sampling range corresponding to the link equalization parameters based on the time boundary value and the voltage boundary value.

[0244] In one embodiment, the first sampling submodule is specifically used to: move the time code horizontally in the direction of the signal sampling center according to a preset time offset unit, and sample the signal transmitted on the link to be equalized based on the moved time code to obtain the signal sampling result; if the sampling error of the signal sampling result does not reach the preset error threshold, the step of moving the time code horizontally in the direction of the signal sampling center according to the preset time offset unit is executed until the sampling error reaches the error threshold, and the sampling time is determined as the time boundary value in the time dimension.

[0245] In one embodiment, the second sampling submodule is specifically used to: move the voltage code vertically in a preset voltage offset unit with the signal sampling center as the center, and sample the signal transmitted on the link to be equalized based on the moved voltage code to obtain the signal sampling result; if the sampling error of the signal sampling result does not reach the error threshold, perform the step of moving the voltage code vertically in a preset voltage offset unit with the signal sampling center as the center until the sampling error reaches the error threshold, and determine the sampled voltage as the voltage boundary value in the voltage dimension.

[0246] In one embodiment, the determining module 1403 includes:

[0247] The second determining submodule is used to determine the eccentricity of the signal under the stability index and the effective length under the anti-interference index according to the stability evaluation algorithm and the sampling range of each signal, and to determine the eye diagram evaluation value of the signal sampling range based on the eccentricity and the effective length.

[0248] The third determination submodule is used to determine the target link balancing parameters corresponding to the link to be balanced from each group of link balancing parameters based on the eye diagram evaluation values.

[0249] In one embodiment, the second determining submodule includes:

[0250] The fourth determination submodule is used to determine the eccentricity of the signal under the stability index and the effective length under the anti-interference index for each dimension in the signal sampling range, based on the boundary values ​​of the dimension.

[0251] The fifth determination submodule is used to determine the eye diagram evaluation values ​​of the link equalization parameters based on the stability evaluation algorithm, each effective length, and each eccentricity.

[0252] In one embodiment, the fourth determining submodule is specifically used to: determine the minimum boundary value among all boundary values, and determine the effective length corresponding to the dimension based on the minimum boundary value; the effective length characterizes the anti-interference capability of the signal; the larger the effective length, the greater the anti-interference capability of the signal; determine the maximum boundary value among all boundary values, and determine the difference between the maximum boundary value and the minimum boundary value; determine the ratio between the difference and the maximum boundary value as the eccentricity corresponding to the dimension; the eccentricity characterizes the stability of the signal; the smaller the eccentricity, the better the stability of the signal.

[0253] In one embodiment, the fifth determining submodule is specifically used to: determine the signal sampling evaluation value of each dimension based on the effective length and eccentricity of each dimension; the signal sampling evaluation value is proportional to the effective length and inversely proportional to the eccentricity; and perform data processing on the signal sampling evaluation values ​​of each dimension based on the stability evaluation algorithm to obtain the eye diagram evaluation value of the link equalization parameters.

[0254] In one embodiment, the third determining submodule is specifically used to: determine the maximum eye diagram evaluation value as the target eye diagram evaluation value among all eye diagram evaluation values; and determine the link equalization parameter corresponding to the target eye diagram evaluation value as the target link equalization parameter.

[0255] Each module in the aforementioned link equalization device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0256] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 15 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a link balancing method.

[0257] Those skilled in the art will understand that Figure 15 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0258] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0259] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0260] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0261] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0262] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0263] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0264] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A link load balancing method, characterized in that, The method includes: Obtain each set of link equalization parameters for the link to be equalized, and determine the signal sampling center corresponding to each set of link equalization parameters; Based on the signal sampling center, the deviation parameters of each dimension are continuously adjusted, and the signals transmitted on the link to be balanced are sampled based on the adjusted deviation parameters to obtain the signal sampling results. Based on the signal sampling results, the signal sampling range corresponding to the link equalization parameter is determined. Based on the signal sampling range and stability evaluation algorithm, target link equalization parameters are determined from each group of link equalization parameters; the target link equalization parameters are used to perform link equalization on the link to be equalized.

2. The method according to claim 1, characterized in that, Determining the signal sampling center corresponding to each set of link equalization parameters includes: The transmitting end corresponding to the link to be balanced is configured based on each set of the link balancing parameters; the signal transmitted on the link to be balanced is sent by the transmitting end on the link to be balanced based on each set of the link balancing parameters. Based on the signal, channel adaptation is performed to obtain the signal sampling center of the link equalization parameters.

3. The method according to claim 1, characterized in that, Each dimension includes a time dimension and a voltage dimension. The deviation parameter of the time dimension is a time code, and the voltage parameter of the voltage dimension is a voltage code. The process involves continuously adjusting the deviation parameters of each dimension based on the signal sampling center, sampling the signals transmitted on the link to be balanced based on the adjusted deviation parameters to obtain each signal sampling result, and determining the signal sampling range corresponding to the link equalization parameter based on each signal sampling result. This includes: The time code is continuously adjusted horizontally based on the signal sampling center, and the signal transmitted on the link to be balanced is sampled based on the adjusted time code to obtain the sampling results of each signal, and the time boundary value is determined based on the sampling results of each signal and the preset error threshold. Based on the signal sampling center, the voltage code is continuously adjusted from a vertical angle, and the signal transmitted on the link to be equalized is sampled based on the adjusted voltage code to obtain the sampling results of each signal, and the voltage boundary value is determined based on each sampling result of the signal and the error threshold. Based on the time boundary value and the voltage boundary value, the signal sampling range corresponding to the link equalization parameter is constructed.

4. The method according to claim 3, characterized in that, The step of continuously adjusting the time code horizontally based on the signal sampling center, sampling the signals transmitted on the link to be equalized based on the adjusted time code to obtain each signal sampling result, and determining the time boundary value based on each signal sampling result and a preset signal threshold includes: Centered on the signal sampling center, the time code is moved horizontally according to a preset time offset unit, and the signal transmitted on the link to be balanced is sampled based on the moved time code to obtain the signal sampling result. If the sampling error of the signal sampling result does not reach the preset error threshold, the step of moving the time code horizontally with the signal sampling center as the center and according to the preset time offset unit is executed until the sampling error reaches the error threshold, and the sampling time is determined as the time boundary value in the time dimension.

5. The method according to claim 3, characterized in that, The step of continuously adjusting the voltage code from a vertical angle based on the signal sampling center, sampling the signal transmitted on the link to be equalized based on the adjusted voltage code to obtain each signal sampling result, and determining the voltage boundary value based on each signal sampling result and the error threshold includes: Centered on the signal sampling center, the voltage code is moved vertically according to a preset voltage offset unit, and the signal transmitted on the link to be equalized is sampled based on the moved voltage code to obtain the signal sampling result. If the sampling error of the signal sampling result does not reach the error threshold, the step of moving the voltage code in the vertical direction with the signal sampling center as the center and according to the preset voltage offset unit is executed until the sampling error reaches the error threshold, and the sampled voltage is determined as the voltage boundary value in the voltage dimension.

6. The method according to claim 1, characterized in that, The step of determining the target link equalization parameters from each group of link equalization parameters based on the signal sampling range and stability evaluation algorithm includes: Based on the stability evaluation algorithm and each of the signal sampling ranges, the eccentricity of the signal under the stability index and the effective length under the anti-interference index are determined, and the eye diagram evaluation value of the signal sampling range is determined based on the eccentricity and the effective length. Based on the eye diagram evaluation values, the target link balancing parameters corresponding to the link to be balanced are determined from the link balancing parameters in each group.

7. The method according to claim 6, characterized in that, The step of determining the eccentricity of the signal under the stability index and the effective length under the anti-interference index according to the stability evaluation algorithm and each of the signal sampling ranges, and determining the eye diagram evaluation value of the signal sampling range based on the eccentricity and the effective length, includes: For each dimension in the signal sampling range, the eccentricity of the signal under the stability index and the effective length under the anti-interference index are determined based on the boundary values ​​of the dimension. Based on the stability evaluation algorithm, the effective lengths, and the eccentricities, the eye diagram evaluation values ​​of the link equalization parameters are determined.

8. The method according to claim 7, characterized in that, The step of determining the eccentricity of the signal under the stability index and the effective length under the anti-interference index for each dimension of the signal sampling range, based on the boundary values ​​of the dimension, includes: Among the boundary values, a minimum boundary value is determined, and based on the minimum boundary value, the effective length corresponding to the dimension is determined; the effective length characterizes the anti-interference capability of the signal; the larger the effective length, the greater the anti-interference capability of the signal. Determine the maximum boundary value among all the boundary values, and determine the difference between the maximum boundary value and the minimum boundary value; The ratio between the difference and the maximum boundary value is determined as the eccentricity corresponding to the dimension; the eccentricity characterizes the stability of the signal; the smaller the eccentricity, the better the stability of the signal.

9. The method according to claim 7, characterized in that, The step of determining the eye diagram evaluation value of the link equalization parameters based on the stability evaluation algorithm, each of the effective lengths, and each of the eccentricities includes: Based on the effective length and eccentricity of each dimension, a signal sampling evaluation value for that dimension is determined; the signal sampling evaluation value is directly proportional to the effective length and inversely proportional to the eccentricity. Based on the stability evaluation algorithm, the signal sampling evaluation values ​​of each dimension are processed to obtain the eye diagram evaluation values ​​of the link equalization parameters.

10. The method according to claim 6, characterized in that, The step of determining the target link balancing parameters corresponding to the link to be balanced from each group of link balancing parameters based on each eye diagram evaluation value includes: Among all the eye diagram evaluation values, the maximum eye diagram evaluation value is determined as the target eye diagram evaluation value; The link balancing parameters corresponding to the target eye diagram evaluation value are determined as the target link balancing parameters.

11. A link equalization device, characterized in that, The device includes: The acquisition module is used to acquire each set of link equalization parameters of the link to be equalized, and to determine the signal sampling center corresponding to each set of link equalization parameters. The sampling module is used to continuously adjust the deviation parameters of each dimension based on the signal sampling center, and to sample the signals transmitted on the link to be balanced based on the adjusted deviation parameters to obtain the sampling results of each signal, and to determine the signal sampling range corresponding to the link equalization parameter based on the sampling results of each signal. The determining module is used to determine the target link equalization parameters from each group of link equalization parameters based on the sampling range and stability evaluation algorithm of each signal; the target link equalization parameters are used to perform link equalization on the link to be equalized.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

Images