Method for determining chip parameters, chip parameter determination device, and storage medium

Abstract

The present application provides a method for determining chip parameters, a device for determining chip parameters, and a storage medium. The method is applied to the field of chips. The method includes: obtaining initial parameters of each channel in the chip under candidate configuration information, where the candidate configuration information is used to configure the transmission performance of the channel, and the initial parameters include the sampling window length of the channel; adjusting the sorting of the initial parameters based on the adjustment directions of the candidate configuration information to obtain reference parameters; performing smoothing processing on the reference parameters to obtain the target parameters of each channel; and selecting the target configuration information of each channel from the candidate configuration information based on the target parameters. This method can determine the configuration information of each channel in the chip to improve the tolerance of communication transmission.

Description

Technical Field

[0001] This application relates to the field of chips, and more particularly, to a method for determining chip parameters, a device for determining chip parameters, and a storage medium. Background Art

[0002] With the continuous development of heterogeneous integration technology, a modular system architecture based on chiplets has been widely used in high-performance computing chips.

[0003] The die-to-die (D2D) interconnection between chiplets has become one of the key links in the system-level performance bottleneck. To enhance the signal integrity and system robustness of the D2D link, modern high-speed transceivers usually integrate multiple programmable parameters; for example, pre-emphasis at the transmitter, equalizer configuration at the receiver, duty cycle correction, and cross point control of the eye diagram, etc. These parameters work together on the frequency-domain characteristics and sampling time position of the signal to adapt to the channel loss and interference under different trace conditions, thus directly affecting the effective sampling window length and bit error rate performance of the link.

[0004] In the prior art, a parameter traversal (or frequency sweeping) method is often used. By testing the sampling window length under each parameter combination, the configuration with the largest effective sampling window is selected. However, this method has significant deficiencies: due to non-ideal factors (such as noise, jitter, etc.) in the test and measurement accuracy limitations, multiple local optima may appear for different parameter combinations, resulting in the selected "optimal parameters" not being stable or representative, and having insufficient adaptability to temperature changes, power supply interference, or process fluctuations in actual applications, which easily causes the link performance boundary to tend to fail and has poor robustness. Therefore, how to determine the configuration information of each channel in the chip to improve the tolerance of communication transmission has become an urgent problem to be solved. Summary of the Invention

[0005] This application provides a method for determining chip parameters, a device for determining chip parameters, and a storage medium. This method can determine the configuration information of each channel in the chip to improve the tolerance of communication transmission.

[0006] In a first aspect, a method for determining chip parameters is provided. The method includes:

[0007] Obtain the initial parameters of each channel in the chip under the candidate configuration information; wherein, the candidate configuration information is used to configure the transmission performance of the channel; the initial parameters include the sampling window length of the channel;

[0008] Adjust the sorting of the initial parameters based on the adjustment directions of the respective candidate configuration information to obtain reference parameters;

[0009] Perform smoothing processing on the reference parameters to obtain the target parameters of the respective channels;

[0010] Based on the target parameters, select the target configuration information of the respective channels from the candidate configuration information.

[0011] In the above technical solution, by obtaining the initial parameters of each channel in the chip under the candidate configuration information, adjusting the sorting of the initial parameters based on the adjustment directions of the respective candidate configuration information to obtain reference parameters, performing smoothing processing on the reference parameters to obtain the target parameters of the respective channels, and based on the target parameters, selecting the target configuration information of the respective channels from the candidate configuration information; compared with the prior art in which the available window lengths corresponding to each configuration parameter in the chip channels are traversed and the configuration parameter corresponding to the maximum available window length is determined as the chip parameter to be adjusted, in this application, the sorting corresponding to the initial parameters is adjusted through chip characteristics to obtain reference parameters in which the physical quantities of adjacent parameters change continuously, and the reference parameters are smoothed to eliminate random measurement errors in the target parameters. On this basis, the target parameters are traversed to obtain the target configuration information corresponding to the maximum parameter in the target parameters, which can ensure the accuracy of the target configuration information of each channel and thus improve the tolerance of communication transmission.

[0012] In combination with the first aspect, in some possible implementation manners, before performing the smoothing processing on the reference parameters to obtain the target parameters of the respective channels, the method further includes:

[0013] Determine whether the target configuration information of the respective channels needs to be the same;

[0014] The performing the smoothing processing on the reference parameters to obtain the target parameters of the respective channels includes:

[0015] If the target configuration information of the respective channels needs to be the same, determine the superimposed parameter of the reference parameters of the respective channels; and perform the smoothing processing on the superimposed parameter to obtain the target parameters of the respective channels;

[0016] If the target configuration information of the respective channels does not need to be the same, perform smoothing processing on the reference parameters to obtain the target parameters of the respective channels.

[0017] In the above technical solution, by determining whether the target configuration information of each channel needs to be the same, when the target configuration information of each channel needs to be the same, the superimposed parameter of the reference parameters of each channel is determined; and the superimposed parameter is smoothed to obtain the target parameter of each channel; when the target configuration information of each channel does not need to be the same, the reference parameter is smoothed to obtain the target parameter of each channel; on the one hand, when the target configuration information of each channel needs to be the same, by superimposing and smoothing the original reference parameters, while ensuring the accuracy of the original data, the accuracy of the target parameters of each channel is ensured; on the other hand, when the target configuration information of each channel does not need to be the same, by smoothing the reference parameters of each channel, the random measurement error in the target parameters of each channel is reduced, thereby ensuring the accuracy of the target parameters of each channel.

[0018] Combined with the first aspect and the above implementation manners, in some possible implementation manners, the reference parameter is an initial matrix;

[0019] The smoothing the reference parameter to obtain the target parameter of each channel includes:

[0020] Taking the target element in the initial matrix as the matrix center point, a plurality of first matrices are collected from the initial matrix; wherein, the target element is any element in the initial matrix;

[0021] Performing an inner product operation on the target matrix and a preset weight to obtain an inner product value; wherein, the target matrix is any one of the plurality of first matrices;

[0022] Updating the target element corresponding to the target matrix based on the inner product value.

[0023] In the above technical solution, taking the target element (any element in the initial matrix) in the initial matrix as the matrix center point, a plurality of first matrices are collected from the initial matrix, and an inner product operation is performed on the target matrix (any one of the plurality of first matrices) and a preset weight to obtain an inner product value, and then based on the inner product value, the target element corresponding to the target matrix is updated; by forming a first matrix with each matrix element in the initial matrix and adjacent elements, and performing a weighting process on the first matrix, and then updating the corresponding matrix element in the initial matrix, the various matrix elements in the initial matrix can be associated with adjacent elements, ensuring the correlation between adjacent parameters in the updated target parameters. On this basis, determining the target configuration information of each channel can ensure the accuracy of the target configuration information while improving the tolerance of communication transmission.

[0024] Combined with the first aspect and the above implementation manners, in some possible implementation manners, the taking the target element in the initial matrix as the matrix center point, and collecting a plurality of first matrices from the initial matrix includes:

[0025] Taking the target element in the initial matrix as the center point of the matrix, multiple first sub-matrices are obtained by sliding a sliding window in the initial matrix;

[0026] If there is an empty matrix element in the target sub-matrix, determine the target matrix element corresponding to the empty matrix element, and fill the empty matrix element based on the target matrix element to obtain the first matrix corresponding to the target sub-matrix; wherein, the target sub-matrix is used to indicate any one of the multiple first sub-matrices; the target matrix element is used to indicate the non-empty matrix element closest to the empty matrix element;

[0027] If all the elements in the target sub-matrix are non-empty matrix elements, take the target sub-matrix as the first matrix.

[0028] In the above technical solution, taking the target element in the initial matrix as the center point of the matrix, multiple first sub-matrices are obtained by sliding a sliding window in the initial matrix. If there is an empty matrix element in the target sub-matrix (any one of the multiple first sub-matrices), determine the target matrix element corresponding to the empty matrix element, and fill the empty matrix element based on the target matrix element to obtain the first matrix corresponding to the target sub-matrix; if all the elements in the target sub-matrix are non-empty matrix elements, take the target sub-matrix as the first matrix; when there is an empty matrix element in the target sub-matrix, by filling the empty matrix element, a first matrix without empty matrix elements is obtained, and the matrix elements in the initial matrix are updated through the first matrix, thereby ensuring the accuracy of the target parameters of each channel.

[0029] Combined with the first aspect and the above implementation manner, in some possible implementation manners, the determining the superimposed parameter of the reference parameters of each channel includes:

[0030] Based on the reference parameters of each channel, determine the reference sub-parameters under the same candidate configuration information;

[0031] Perform a superimposing process on the reference sub-parameters to obtain the superimposed sub-parameters under the same candidate configuration information;

[0032] Based on the superimposed sub-parameters, generate the superimposed parameter of the reference parameters of each channel.

[0033] Based on the reference parameters of each channel, the reference sub-parameters under the same candidate configuration information are determined, the reference sub-parameters are superimposed to obtain the superimposed sub-parameters under the same candidate configuration information, and based on the superimposed sub-parameters, the superimposed parameters of the reference parameters of each channel are generated; by superimposing the reference parameters of each channel, the superimposed parameters of all channels in the chip are obtained, so that through the superimposed parameters, the accuracy of the target parameters of each channel can be ensured, thereby improving the accuracy of the target configuration information of each channel.

[0034] Combined with the first aspect and the above implementation, in some possible implementations, the smoothing the superimposed parameters to obtain the target parameters of each channel includes:

[0035] Smoothing the superimposed parameters to obtain the parameters after smoothing;

[0036] Taking the parameters after smoothing as the target parameters of each channel.

[0037] The above technical solution smooths the superimposed parameters to obtain the parameters after smoothing, and takes the parameters after smoothing as the target parameters of each channel; by smoothing the superimposed parameters, the correlation between adjacent parameters in the superimposed parameters can be established, ensuring the correlation between adjacent parameters in the target parameters of each channel, and then determining the target configuration information of each channel through the target parameters, while ensuring the accuracy of the target configuration information, improving the tolerance of communication transmission.

[0038] Combined with the first aspect and the above implementation, in some possible implementations, the selecting the target configuration information of each channel from the candidate configuration information based on the target parameters includes:

[0039] Traversing the target parameters to obtain the maximum parameter in the target parameters;

[0040] Taking the candidate configuration information corresponding to the maximum parameter as the target configuration information of each channel.

[0041] The above technical solution traverses the target parameters to obtain the maximum parameter in the target parameters, and takes the candidate configuration information corresponding to the maximum parameter as the target configuration information of each channel; by adjusting the sorting corresponding to the initial parameters through the chip characteristics, the reference parameters with continuously changing physical quantities of adjacent parameters are obtained, and the reference parameters are smoothed to eliminate the random measurement errors in the target parameters. On this basis, traversing the target parameters to obtain the target configuration information corresponding to the maximum parameter in the target parameters can ensure the accuracy of the target configuration information of each channel.

[0042] In combination with the first aspect and the above implementation manner, in some possible implementation manners, the method further includes:

[0043] Based on the target parameters of each channel, an initial image is generated; wherein the brightness value in the initial image is determined based on the value of the target parameter;

[0044] Marking the position of the maximum parameter in the initial image to generate the target image of each channel;

[0045] Output the target image of each channel.

[0046] The above technical scheme generates an initial image through the target parameters of each channel. The brightness value in the initial image is determined based on the numerical value of the target parameter, and the position of the maximum parameter in the initial image is marked to generate the target image of each channel and output the target image of each channel. By outputting the target images of multiple channels, technicians can identify and determine the abnormal channels in the chip, and then further troubleshoot the problems to avoid the impact of chip hardware problems on data transmission as much as possible. In addition, when there is no obvious abnormality in the data transmission of the chip, the target images of each channel can be used to troubleshoot the chip hardware problems as early as possible and optimize the chip design problems as early as possible.

[0047] In a second aspect, a chip parameter determination device is provided, the chip parameter determination device comprising:

[0048] An acquisition module, used to acquire initial parameters of each channel in the chip under candidate configuration information; wherein the candidate configuration information is used to configure the transmission performance of the channel; the initial parameters include the sampling window length of the channel;

[0049] An adjustment module, configured to adjust the order of the initial parameters based on the adjustment direction of each candidate configuration information to obtain reference parameters;

[0050] A processing module, used for smoothing the reference parameters to obtain target parameters of each channel;

[0051] A selection module is used to select the target configuration information of each channel from the candidate configuration information based on the target parameter.

[0052] In a third aspect, a chip parameter determination device is provided, comprising a memory and a processor, the memory being used to store executable program code; the processor being used to call and run the executable program code from the memory, so that the chip parameter determination device executes the chip parameter determination method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0053] Fourthly, a computer-readable storage medium is provided, which stores computer program code. When the computer program code runs on a computer, the computer is caused to execute the chip parameter determination method in the above first aspect or any possible implementation manner of the first aspect.

[0054] Fifthly, a computer program product is provided, which includes: computer program code. When the computer program code runs on a computer, the computer is caused to execute the chip parameter determination method in the above first aspect or any possible implementation manner of the first aspect. Description of the Drawings

[0055] Figure 1 is a schematic flowchart of a chip parameter determination method provided by an embodiment of the present application;

[0056] Figure 2 is a schematic flowchart of another chip parameter determination method provided by an embodiment of the present application;

[0057] Figure 3 is a schematic diagram of a smoothing process provided by an embodiment of the present application;

[0058] Figure 4 is a schematic diagram of another smoothing process provided by an embodiment of the present application;

[0059] Figure 5 is a visualization schematic diagram of a chip parameter determination method provided by an embodiment of the present application;

[0060] Figure 6 is a schematic structural diagram of a chip parameter determination device provided by an embodiment of the present application;

[0061] Figure 7 is a schematic structural diagram of a chip parameter determination device provided by an embodiment of the present application. Detailed Embodiments

[0062] Next, the technical solutions in the present application will be clearly and thoroughly described with reference to the drawings. Among them, in the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B: "and / or" in the text is only a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality of" means two or more than two.

[0063] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0064] In the prior art, parameter traversal (or frequency sweep) is often used to select the configuration with the largest effective sampling window by testing the sampling window length under each parameter combination. However, this method has significant shortcomings: due to non-ideal factors (such as noise, jitter, etc.) and measurement accuracy limitations in the test, different parameter combinations may have multiple local optima, resulting in the selected "optimal parameters" not being stable or representative. In actual applications, the ability to adapt to temperature changes, power supply interference or process fluctuations is insufficient, which easily causes the link performance boundary to fail and has poor robustness.

[0065] For example, in order to ensure signal integrity and transmission stability, it is usually necessary to calibrate the duty cycle of the clock signal of the transmitter and receiver. However, since the actual length and path characteristics of the link vary from channel to channel, it is difficult to meet performance requirements by configuring the clock duty cycle of all channels using a single parameter. In addition, the adjustment direction and amplitude of the duty cycle control (DCC) parameter have certain physical meanings, and simple single-point maximum value search is often interfered by measurement errors or local anomalies.

[0066] In view of this, the present application provides a chip parameter determination method, a chip parameter determination device and a storage medium. By combining the transmission performance corresponding to the adjacent information in each channel and weighted processing the neighborhood information, it is possible to highlight the current point performance while smoothing the surrounding noise, thereby finding the optimal configuration information of each channel, ensuring that the optimal configuration information of each channel is stable, and thereby improving the tolerance of communication transmission.

[0067] It can be understood that by weighting the neighborhood information, the surrounding noise can be smoothed while highlighting the performance of the current point, avoiding non-ideal factors and measurement accuracy limitations in the test, and different parameter combinations may result in multiple local optimal situations, ensuring that the optimal configuration information found is stable or representative, and has strong adaptability to temperature changes, power supply interference or process fluctuations in actual applications, thereby improving the tolerance of communication transmission.

[0068] Figure 1 It is a schematic flow chart of a chip parameter determination method provided in an embodiment of the present application.

[0069] For example, Figure 1 As shown, the chip parameter determination method 100 includes S110 - S140 .

[0070] S110, obtain the initial parameters of each channel in the chip under the candidate configuration information.

[0071] Exemplarily, the candidate configuration information is used to configure the transmission performance of the channel. It can be understood that each channel in the chip includes specific circuits; for example, during the DCC parameter adjustment process, the candidate configuration information of each channel includes 0 - 15. As the selection of the candidate configuration information corresponding to DCC for each channel is different (that is, by modifying the value of the register, the adjustment of the DCC parameters of each channel is realized), the transmission performance of each channel is different.

[0072] It can be understood that the candidate configuration information is used to configure the corresponding functional modules in the transceiver, thereby affecting the transmission performance of the channel. Continuing with the example of DCC parameter adjustment, due to the different candidate configuration information of the DCC parameters, the settings of the DCC adjustment modules of the receiver and the transmitter are changed, thereby affecting the transmission performance of the channel. Here, configuring the transmission performance of the channel means that with different selections of the candidate configuration information, the transmission performance of each channel may be different. In other words, it is not that the higher the DCC parameters are configured, the better the transmission performance of the channel is. Instead, as the DCC parameters change, the transmission performance of the channel may change, that is, the candidate configuration information will affect the transmission performance of the channel.

[0073] Exemplarily, the initial parameters include the sampling window length of the channel. The sampling window length refers to the length of the time window for accurately sampling the transmitted data; for example, the width of the eye diagram (the length of the time opening for correctly receiving data). In the entire link, the transmission and reception of the chip will be affected by many factors. Therefore, the length of the time window for each channel to accurately sample the data is different (sampling window length). The sampling window length also represents the signal transmission quality and overall performance of the link under specific parameters. Generally, the sampling window length is positively correlated with the transmission performance of the channel. The longer the sampling window length, the better the transmission performance of the channel.

[0074] It should be noted that the receiving structure of the chip in the embodiments of the present application (a combination of functional modules used to receive and process transmission signals in integrated circuit design) includes a clock delay line (delay line) and a structure that traverses the delay line to perform a specific pattern detection. This structure traverses the delay line and checks whether the data pattern can be correctly received at each delay line tap; and the length of the gear range that can continuously and accurately receive the pattern is the sampling window length, that is, the sampling window length indicates the maximum continuous effective sampling window length (the maximum continuous effective sampling span). It can be understood that the sampling window length is similar to the eye diagram width. The longer the sampling window length (the wider the eye diagram width), the higher the tolerance of the communication link to jitter and the better the bit error rate performance of the link. Therefore, the quality of the communication link and its anti-jitter ability can be evaluated through the sampling window length.

[0075] Optionally, the sampling window length of the channel can be the time period length of the window; or, the sampling window length of the channel can also be the delay amount corresponding to the gear (the delay line includes multiple delay gears, and different delay gears correspond to different delay amounts, that is, delay values). Of course, the initial parameters can also include other parameters that can represent the transmission performance of the channel, which are not specifically limited here.

[0076] Optionally, the candidate configuration information of each channel can be the configuration information corresponding to the DCC, or the pre-emphasis parameter (Pre-cursor, the influence of one or more symbols before the current symbol on the current symbol; or, Post-cursor, the influence of one or more symbols after the current symbol on the current symbol) in the Finite Impulse Response (FIR). The candidate configuration information of each channel can be determined according to the actual situation, which is not specifically limited here.

[0077] Optionally, the initial parameters of each channel under the candidate configuration information can be represented by a table; or, the initial parameters of each channel under the candidate configuration information can be represented by a matrix. The representation method of the initial parameters can be determined according to the actual situation, which is not specifically limited here.

[0078] It should be noted that the initial parameters are obtained by traversing the parameter data of the combined performance of each channel in the chip between the transmitter and the receiver. For example, during the DCC parameter adjustment process, the sampling window lengths of 0 - 15 (candidate configuration information) for each channel under all combinations of transmitter DCC (Transmit Duty Cycle Control, TX DCC) and receiver DCC (Receive Duty Cycle Control, RX DCC) are traversed, and the data matrix is constructed with the sampling window lengths according to the TX DCC and RX DCC coordinates.

[0079] S120, based on the adjustment directions of the respective candidate configuration information, adjusts the sorting of the initial parameters to obtain reference parameters.

[0080] Exemplarily, the arrangement order of the respective candidate configuration information affects the arrangement order of the initial parameters, that is, by adjusting the arrangement order of the candidate configuration information, the initial parameters can be re - sorted so that the change of the physical quantity represented by the re - sorted reference parameters is continuous and there are no mutations.

[0081] Taking the adjustment of the DCC parameters in the chip as an example, during the process of adjusting the DCC parameters of each channel in the chip, first obtain the initial parameters of each channel under the candidate configuration information (0 - 15) of DCC, and construct the initial parameters into a 16 * 16 matrix. However, due to the actual structure of the hardware DCC register (for example, the forward adjustment and reverse adjustment in the register), it is necessary to smooth the initial parameters so that the change of the physical quantity represented by the re - sorted (adjusted) reference parameters is continuous.

[0082] For example, in the actual structure of the DCC register, since the adjustment is forward between 0 - 7 and reverse between 8 - 15, the sorting of the directly obtained initial parameters is 0 - 15, and there is a data mutation between 7 and 8. Therefore, it is necessary to adjust the sorting of the initial parameters according to the adjustment directions of the respective candidate configuration information to obtain the reference parameters.

[0083] It can be understood that during the process of adjusting the DCC parameters of each channel of the chip, the initial parameter is a 16*16 matrix, and the parameters corresponding to the candidate configuration information with the rows and columns of the initial parameter sorted in ascending order from 0 to 15. According to the adjustment direction of the candidate configuration information of the DCC, the initial parameter (16*16 matrix) is adjusted to obtain a reference parameter which is also a 16*16 matrix. However, the rows and columns of the reference parameter are sorted in descending order from 7 to 0 and ascending order from 8 to 15 for the parameters corresponding to the candidate configuration information, and the parameters corresponding to 0 and 8 are adjacent parameters. That is, according to the chip characteristics, the order of the rows and columns of the matrix corresponding to the initial parameter is adjusted to obtain the matrix corresponding to the reference parameter. Among the parameters corresponding to the rows and columns of the matrix corresponding to the reference parameter, there is no sudden change in the physical quantities represented by the parameters corresponding to adjacent rows or columns.

[0084] S130, perform smoothing processing on the reference parameter to obtain the target parameter of each channel.

[0085] Exemplarily, in the case of obtaining the reference parameter, performing smoothing processing on the reference parameter can remove the random measurement error and noise interference in the reference parameter, obtain the target parameter of each channel, and thus can determine the target configuration information of each channel according to the target parameter of each channel.

[0086] In one example, the reference parameter is smoothed by a mean filtering algorithm to obtain the target parameter of each channel; or, the reference parameter is smoothed by an adaptive filtering algorithm to obtain the target parameter of each channel.

[0087] In order to make the configuration information of each channel in the chip match the chip characteristics or user requirements, the target configuration information of each channel can also be determined according to the actual requirements of the configuration information of each channel to ensure the communication transmission performance of each channel.

[0088] In another example, obtain the initial parameter of each channel in the chip under the candidate configuration information, and based on the adjustment direction of each candidate configuration information, adjust the sorting of the initial parameter to obtain the reference parameter, determine whether the target configuration information of each channel needs to be the same. If the target configuration information of each channel needs to be the same, determine the superimposed parameter of the reference parameters of each channel; and perform smoothing processing on the superimposed parameter to obtain the target parameter of each channel; if the target configuration information of each channel does not need to be the same, perform smoothing processing on the reference parameter to obtain the target parameter of each channel, and thus determine the target configuration information of each channel according to the target parameter.

[0089] Exemplarily, it is possible to determine whether the target configuration information of each channel needs to be the same according to the chip performance, or to determine whether the target configuration information of each channel needs to be the same according to the user requirements; of course, it is also possible to determine whether the target configuration information of each channel needs to be the same according to the usage scenario of the chip. The method for determining whether the target configuration information of each channel needs to be the same can be determined according to the actual situation and will not be specifically limited here.

[0090] In the above technical solution, by determining whether the target configuration information of each channel needs to be the same, when the target configuration information of each channel needs to be the same, the superimposed parameter of the reference parameters of each channel is determined; and the superimposed parameter is smoothed to obtain the target parameter of each channel; when the target configuration information of each channel does not need to be the same, the reference parameter is smoothed to obtain the target parameter of each channel; on the one hand, when the target configuration information of each channel needs to be the same, by superimposing and smoothing the original reference parameters, while ensuring the accuracy of the original data, the accuracy of the target parameters of each channel is ensured; on the other hand, when the target configuration information of each channel does not need to be the same, by smoothing the reference parameters of each channel, the random measurement error in the target parameters of each channel is reduced, thereby ensuring the accuracy of the target parameters of each channel.

[0091] The following embodiments will specifically describe the process of performing weighted neighborhood averaging on the reference parameter with the reference parameter as the initial matrix when the target configuration information of each channel does not need to be the same.

[0092] Exemplarily, with the target element in the initial matrix as the center point of the matrix, a plurality of first matrices are collected from the initial matrix, and the target element is any element in the initial matrix; and the inner product of the target matrix and the preset weight is processed to obtain an inner product value, where the target matrix is any one of the plurality of first matrices; then, based on the inner product value, the target element corresponding to the target matrix is updated.

[0093] For example, traversing with a 3×3 sliding window starting from the parameter in the first row and first column of the initial matrix as the initial element, a plurality of 3×3 first matrices are obtained, and the center point of each of the plurality of 3×3 first matrices corresponds to an element in the initial matrix, that is, any element in the initial matrix corresponds to a 3×3 first matrix.

[0094] It can be understood that when the sliding window is a 3×3 matrix, the preset weight is also a 3×3 weight matrix, and the weight values in the weight matrix can be adjusted according to the positions of the elements of the matrix. For example, the weight value of the central element of the 3×3 weight matrix is 5, and the other matrix elements are all 1. The weight values in the weight matrix can be determined according to the actual situation and will not be specifically limited here.

[0095] Further, for each matrix element in the initial matrix, an inner product operation is performed between the first matrix corresponding to each matrix element and a preset weight to obtain an inner product value, and the target element corresponding to the inner product value in the initial matrix is updated based on the inner product value. For example, the target element corresponding to the inner product value in the initial matrix can be replaced by the inner product value; or, the target element corresponding to the inner product value in the initial matrix can be replaced by the average value of the inner product values.

[0096] It should be noted that the initial matrix is updated based on the inner product value to obtain an updated matrix, and the updated matrix is used to indicate the target parameters of each channel.

[0097] In the above technical solution, taking the target element (any element in the initial matrix) in the initial matrix as the matrix center point, a plurality of first matrices are collected from the initial matrix, and an inner product operation is performed between the target matrix (any matrix among the plurality of first matrices) and a preset weight to obtain an inner product value. Then, based on the inner product value, the target element corresponding to the target matrix is updated; by forming a first matrix with each matrix element in the initial matrix and its adjacent elements and performing a weighted process on the first matrix, the corresponding matrix element in the initial matrix is further updated, which can associate each matrix element in the initial matrix with its adjacent elements, ensuring the correlation between adjacent parameters in the obtained target parameters. On this basis, determining the target configuration information for each channel can ensure the accuracy of the target configuration information while improving the tolerance of communication transmission.

[0098] It can be understood that the tolerance of communication transmission indicates the ability to resist external interference during the communication transmission process. The higher the tolerance, the smaller the impact of external interference on the communication transmission.

[0099] When the sliding window is greater than 1, there may be empty matrix elements in the multiple matrices collected from the initial matrix through the sliding window, and the empty matrix elements may affect the result of smoothing the reference parameters. By supplementing the empty matrix elements, the accuracy of the obtained target parameters can be ensured during the smoothing process of the reference parameters.

[0100] Exemplarily, taking the target element in the initial matrix as the matrix center point, a plurality of first sub-matrices are obtained by sliding the sliding window in the initial matrix, and it is determined whether there are empty matrix elements in the target sub-matrix, where the target sub-matrix is used to indicate any sub-matrix among the plurality of first sub-matrices; if there are empty matrix elements in the target sub-matrix, the target matrix element corresponding to the empty matrix element is determined, and the empty matrix element is filled based on the target matrix element to obtain the first matrix corresponding to the target sub-matrix, where the target matrix element is used to indicate the non-empty matrix element closest to the empty matrix element; if all elements in the target sub-matrix are non-empty matrix elements, the target sub-matrix is used as the first matrix.

[0101] Exemplarily, when the target element is at the edge of the initial matrix, there are empty matrix elements in the first sub-matrix. For example, for a sliding window of a 3*3 matrix, if the target element (m, n) is the center point of the matrix and m-1<0, there are empty matrix elements in the first sub-matrix. Then, determine the distances from the empty matrix elements to each non-empty matrix element, and determine the non-empty matrix element with the closest distance as the target matrix element. Fill the empty matrix elements with the corresponding non-empty matrix elements to obtain the first matrix corresponding to the first sub-matrix; when there are no non-empty matrix elements in the first sub-matrix, use the first sub-matrix as the first matrix.

[0102] In the above technical solution, with the target element in the initial matrix as the center point of the matrix, multiple first sub-matrices are obtained by sliding a sliding window in the initial matrix. If there are empty matrix elements in the target sub-matrix (any one of the multiple first sub-matrices), determine the target matrix element corresponding to the empty matrix element, and fill the empty matrix elements based on the target matrix element to obtain the first matrix corresponding to the target sub-matrix; if all elements in the target sub-matrix are non-empty matrix elements, use the target sub-matrix as the first matrix; when there are empty matrix elements in the target sub-matrix, fill the empty matrix elements to obtain a first matrix without empty matrix elements, and update the matrix elements in the initial matrix through the first matrix, thereby ensuring the accuracy of the target parameters of each channel.

[0103] The following embodiments will elaborate in detail on the process of determining the target parameters of each channel by referring to parameters when the target configuration information of each channel needs to be the same.

[0104] Exemplarily, when the target configuration information of each channel needs to be the same, based on the reference parameters of each channel, determine the reference sub-parameters under the same candidate configuration information, perform a superposition process on the reference sub-parameters to obtain the superposed sub-parameters under the same candidate configuration information, and generate the superposed parameters of the reference parameters of each channel based on the superposed sub-parameters.

[0105] It can be understood that when performing the superposition process on the reference parameters of each channel in the chip, align the reference parameters of each channel with the same candidate configuration information to obtain the reference sub-parameters under the same candidate configuration information, perform a superposition process on the reference sub-parameters under the same candidate configuration information to obtain the superposed sub-parameters under the same candidate configuration information, and generate the superposed parameters of the channels in the chip through the superposed sub-parameters.

[0106] It can be understood that when the communication channels in the chip are 12, perform a superposition process on the reference parameters of each channel to obtain the superposed parameters of the 12 channels in the chip. That is, superimpose the reference parameters of the 12 channels in the chip into 1 superposed parameter.

[0107] For example, the reference parameters of 12 channels in the chip can be superimposed to directly obtain the superimposed parameters; alternatively, the reference parameters of 12 channels in the chip can be superimposed and averaged to obtain the superimposed parameters.

[0108] Based on the reference parameters of each channel, the reference sub-parameters under the same candidate configuration information are determined, the reference sub-parameters are superimposed to obtain the superimposed sub-parameters under the same candidate configuration information, and based on the superimposed sub-parameters, the superimposed parameters of the reference parameters of each channel are generated; by superimposing the reference parameters of each channel, the superimposed parameters of all channels in the chip are obtained, so that through the superimposed parameters, the accuracy of the target parameters of each channel can be ensured, thereby improving the accuracy of the target configuration information of each channel.

[0109] Further, in the case of generating the superimposed parameters of the reference parameters of each channel, the superimposed parameters are smoothed to obtain the target parameters of each channel.

[0110] Exemplarily, the superimposed parameters are smoothed to obtain the smoothed parameters, and the smoothed parameters are used as the target parameters of each channel.

[0111] Optionally, the method of smoothing the superimposed parameters may refer to the method of smoothing the reference parameters described above, which will not be elaborated here.

[0112] It can be understood that the superimposed parameters are the parameters after superimposing each channel in the chip. After smoothing the superimposed parameters, the smoothed parameters are obtained, and the smoothed parameters are used as the target parameters of each channel.

[0113] The above technical solution smooths the superimposed parameters to obtain the smoothed parameters, and uses the smoothed parameters as the target parameters of each channel; by smoothing the superimposed parameters, the adjacent parameters in the superimposed parameters can be correlated, ensuring the correlation between adjacent parameters in the target parameters of each channel, and then determining the target configuration information of each channel through the target parameters, while ensuring the accuracy of the target configuration information, improving the tolerance of communication transmission.

[0114] S140, based on the target parameters, select the target configuration information of each channel from the candidate configuration information.

[0115] Exemplarily, according to the adjustment direction of each candidate configuration information, the sorting of the initial parameters of each channel in the chip under the candidate configuration information is adjusted to obtain the reference parameters, the reference parameters are smoothed to obtain the target parameters of each channel, and the target parameters are traversed to obtain the maximum parameter in the target parameters, and the candidate configuration information corresponding to the maximum parameter is used as the target configuration information of each channel.

[0116] The above technical scheme, by traversing the target parameters, obtains the maximum parameter among the target parameters, and uses the candidate configuration information corresponding to the maximum parameter as the target configuration information of each channel; adjusts the sorting corresponding to the initial parameters through the chip characteristics, obtains the reference parameters whose physical quantities of adjacent parameters are continuously changing, and smoothes the reference parameters to eliminate random measurement errors in the target parameters. On this basis, the target parameters are traversed to obtain the target configuration information corresponding to the maximum parameter among the target parameters, which can ensure the accuracy of the target configuration information of each channel.

[0117] Furthermore, target configuration information of each channel is obtained by selecting from candidate configuration information based on the target parameters, and a target image corresponding to the target parameters of each channel is output.

[0118] Exemplarily, based on the target parameters of each channel, an initial image is generated, the brightness value in the initial image is determined based on the numerical value of the target parameter, and the position of the maximum parameter in the initial image is marked to generate the target image of each channel, and the target image of each channel is output.

[0119] Optionally, the initial image may be a two-dimensional thermal map; or the initial image may be a three-dimensional surface map. The initial image may be determined according to actual conditions and is not specifically limited here.

[0120] It can be understood that the brightness value in the initial image is determined based on the numerical value of the target parameter. As the numerical value of the target parameter increases, the corresponding pixel points in the initial image become brighter, and as the numerical value of the target parameter decreases, the corresponding pixel points in the initial image become darker.

[0121] The above technical scheme generates an initial image through the target parameters of each channel. The brightness value in the initial image is determined based on the numerical value of the target parameter, and the position of the maximum parameter in the initial image is marked to generate the target image of each channel and output the target image of each channel. By outputting the target images of multiple channels, technicians can identify and determine the abnormal channels in the chip, and then further troubleshoot the problems to avoid the impact of chip hardware problems on data transmission as much as possible. In addition, when there is no obvious abnormality in the data transmission of the chip, the target images of each channel can be used to troubleshoot the chip hardware problems as early as possible and optimize the chip design problems as early as possible.

[0122] For example, taking the adjustment of DCC parameters in a chip as an example, when the DCC target parameters of each channel are obtained, a two-dimensional heat map of each channel is generated through the target parameters of each channel. The X-axis of the two-dimensional heat map represents TX DCC, the Y-axis represents RX DCC, and the color depth (or color gradient) represents the size of the sampling window length. Alternatively, a three-dimensional surface map of each channel is generated through the target parameters of each channel. The X-axis of the three-dimensional surface map represents TX DCC, the Y-axis represents RX DCC, and the Z-axis is the parameter value in the target parameters (the samplable window length after smoothing processing), so that users can rotate and zoom to observe the change trend.

[0123] It can be understood that images can be separately drawn for each channel, and the results of multiple channels can also be superimposed in the same visualization interface (such as multiple pictures side by side or combined into a large picture).

[0124] Furthermore, in the case of generating an initial image corresponding to the target parameters, the maximum parameter in the target parameters is automatically searched and marked in the initial image (such as highlighting, circling, or arrow indication) to generate a target image.

[0125] In the above technical solution, by obtaining the initial parameters of each channel in the chip under the candidate configuration information, based on the adjustment direction of each candidate configuration information, the sorting of the initial parameters is adjusted to obtain reference parameters, the reference parameters are smoothed to obtain the target parameters of each channel, and based on the target parameters, the target configuration information of each channel is selected from the candidate configuration information; compared with the prior art in which the available window length corresponding to each configuration parameter in the chip channel is traversed and the configuration parameter corresponding to the maximum available window length is determined as the chip parameter to be adjusted, in this application, the sorting corresponding to the initial parameters is adjusted through chip features to obtain reference parameters with continuously changing physical quantities of adjacent parameters, and the reference parameters are smoothed to eliminate random measurement errors in the target parameters. On this basis, the target parameters are traversed to obtain the target configuration information corresponding to the maximum parameter in the target parameters, which can ensure the accuracy of the target configuration information of each channel and thus improve the tolerance of communication transmission.

[0126] To more clearly describe the chip parameter determination method, the following embodiments are described by taking the adjustment of DCC parameters in the chip and the initial parameters being a matrix as an example.

[0127] Figure 2 It is a schematic flowchart of another chip parameter determination method provided by an embodiment of the present application.

[0128] Exemplarily, as Figure 2 shown, the data augmentation method 200 includes the following processes S201 - S211.

[0129] S201. Obtain the initial parameters of each channel in the chip under the candidate configuration information.

[0130] Exemplarily, the candidate configuration information is used to configure the transmission performance of the channel. Taking the DCC parameter adjustment as an example, the candidate configuration information (DCC parameter) of each channel includes 0 - 15. As the selection of the corresponding candidate configuration information of DCC for each channel is different (that is, by modifying the value of the register, the adjustment of the DCC parameter of each channel is realized), the transmission performance of each channel is different.

[0131] It can be understood that the candidate configuration information is used to configure the corresponding functional module in the transceiver, thus affecting the transmission performance of the channel. Continuing with the DCC parameter adjustment as an example, due to the different candidate configuration information of the DCC parameter, the settings of the DCC adjustment modules of the receiver and the transmitter are changed, thus affecting the transmission performance of the channel. Here, configuring the transmission performance of the channel means that with different selections of the candidate configuration information, the transmission performance of each channel may be different. In other words, it is not that the higher the DCC parameter is configured, the better the transmission performance of the channel is, but rather that as the DCC parameter changes, the transmission performance of the channel may change, that is, the candidate configuration information will affect the transmission performance of the channel.

[0132] Exemplarily, the initial parameters include the sampling window length of the channel. The sampling window length refers to the length of the time window for accurately sampling the transmitted data; for example, the width of the eye diagram (the length of the time opening for correctly receiving data). In the entire link, the transmission and reception of the chip will be affected by many factors, so the length of the time window for each channel to accurately sample the data (sampling window length) is different. The sampling window length also represents the signal transmission quality and overall performance of the link under specific parameters. Generally, the sampling window length is positively correlated with the transmission performance of the channel. The longer the sampling window length, the better the transmission performance of the channel.

[0133] Exemplarily, the initial parameters are obtained by traversing the parameter data of the combined performance of the transmitter and receiver of each channel in the chip. Specifically, during the DCC parameter adjustment process, traverse the sampling window length of each channel from 0 to 15 (candidate configuration information) under all combinations of TX DCC and RX DCC, and construct a data matrix (initial parameters) of the sampling window length according to the TX DCC and RX DCC coordinates.

[0134] S202. Based on the adjustment direction of each candidate configuration information, adjust the sorting of the initial parameters to obtain the reference parameters.

[0135] Exemplarily, during the process of adjusting the DCC parameters of each channel of the chip, first obtain the initial parameters of each channel under the candidate configuration information (0 - 15) of DCC, and construct the initial parameters into a 16 * 16 matrix. However, due to the actual structure of the hardware DCC register (for example, the forward adjustment and reverse adjustment in the register), it is necessary to smooth the initial parameters so that the change of the physical quantity represented in the re - sorted (adjusted) reference parameters is continuous.

[0136] For example, in the actual structure of the DCC register, since the forward adjustment is between 0 - 7 and the reverse adjustment is between 8 - 15, the sorting of the directly obtained initial parameters is 0 - 15. Then, there is a data mutation between 7 and 8. Therefore, it is necessary to adjust the sorting of the initial parameters according to the adjustment direction of each candidate configuration information to obtain the reference parameters. That is, perform a matrix transformation on the first matrix corresponding to the initial parameters to obtain the second matrix corresponding to the reference parameters.

[0137] It can be understood that during the process of adjusting the DCC parameters of each channel of the chip, the initial parameters are a 16 * 16 matrix. The sorting of the rows and columns of the initial parameters are the parameters corresponding to the candidate configuration information that increases sequentially from 0 - 15. And according to the adjustment direction of the candidate configuration information of DCC, the initial parameters (16 * 16 matrix) are adjusted to obtain the reference parameters which are also a 16 * 16 matrix. However, the sorting of the rows and columns of the reference parameters are the parameters corresponding to the candidate configuration information that decreases sequentially from 7 - 0 and increases sequentially from 8 - 15, and the parameters corresponding to 0 and 8 are adjacent parameters. That is, according to the chip characteristics, adjust the order of the rows and columns of the matrix corresponding to the initial parameters to obtain the matrix corresponding to the reference parameters. Among the parameters corresponding to the rows and columns of the matrix corresponding to the reference parameters, there is no mutation in the change of the physical quantity represented by the adjacent rows or columns of parameters.

[0138] S203, determine whether the target configuration information of each channel needs to be the same; if not, execute S204; if so, execute S207.

[0139] Exemplarily, it can be determined whether the target configuration information of each channel needs to be the same according to the chip performance, or it can be determined whether the target configuration information of each channel needs to be the same according to the user requirements; of course, it can also be determined whether the target configuration information of each channel needs to be the same according to the usage scenario of the chip.

[0140] Exemplarily, determine whether the target configuration information of each channel needs to be the same, and determine the determination method of the target parameters according to whether the target configuration information of each channel needs to be the same, so as to ensure the accuracy of the target parameters of each channel.

[0141] S204, with the target element in the initial matrix as the matrix center point, collect multiple first matrices from the initial matrix.

[0142] Exemplarily, when the target configuration information of each channel does not need to be the same, the reference parameters corresponding to each channel are smoothed to obtain the target parameters of each channel.

[0143] Exemplarily, with the target element in the initial matrix as the matrix center point, multiple first sub-matrices are obtained by sliding a sliding window in the initial matrix. When the target element is at the edge position, a complete first sub-matrix cannot be collected.

[0144] Specifically, with the target element in the initial matrix as the matrix center point, multiple first sub-matrices are obtained by sliding a sliding window in the initial matrix. It is determined whether there are empty matrix elements in the target sub-matrix, and the target sub-matrix is used to indicate any one of the multiple first sub-matrices; if there are empty matrix elements in the target sub-matrix, the target matrix element corresponding to the empty matrix element is determined, and the empty matrix element is filled based on the target matrix element to obtain the first matrix corresponding to the target sub-matrix, and the target matrix element is used to indicate the non-empty matrix element closest to the empty matrix element; if all elements in the target sub-matrix are non-empty matrix elements, the target sub-matrix is used as the first matrix.

[0145] Exemplarily, when the target element is at the edge of the initial matrix, there are empty matrix elements in the first sub-matrix. For example, the sliding window is a 3*3 matrix, and the target element (m, n) is the matrix center point. When m - 1 < 0, there are empty matrix elements in the first sub-matrix, then the distance between the empty matrix element and the non-empty matrix elements in the first sub-matrix is determined, and the non-empty matrix element with the closest distance is selected to fill the empty matrix element, thereby obtaining the first matrix without empty matrix elements; when there are no non-empty matrix elements in the first sub-matrix, the first sub-matrix is used as the first matrix.

[0146] S205, perform an inner product operation on the target matrix and the preset weight to obtain an inner product value.

[0147] It should be noted that the target matrix is any one of the multiple first matrices.

[0148] Exemplarily, multiple first matrices are collected from the initial matrix through a sliding window, and an inner product operation is performed on the target matrix and the preset weight to obtain an inner product value.

[0149] It can be understood that when the sliding window is a 3*3 matrix, the preset weight is also a 3*3 weight matrix, and the weight values in the weight matrix can be adjusted according to the positions of the elements of the matrix. For example, the weight value of the central element of the 3*3 weight matrix is 5, and the other matrix elements are all 1. The weight values in the weight matrix can be determined according to the actual situation and are not specifically limited here.

[0150] Further, for each matrix element in the initial matrix, an inner product process is performed between the first matrix corresponding to each matrix element and a preset weight to obtain an inner product value, that is, the inner product process between the 3×3 target matrix and the 3×3 preset weight, and the result is the inner product value.

[0151] S206. Update the target element of the target matrix based on the inner product value, and then obtain the target parameter.

[0152] Exemplarily, a plurality of first matrices are collected from the initial matrix, and an inner product process is performed between the target matrix and the preset weight to obtain an inner product value, where the target matrix is any one of the plurality of first matrices; then, based on the inner product value, the target element corresponding to the target matrix is updated.

[0153] For example, starting from the parameter at the first row and first column of the initial matrix as the initial element, a 3×3 sliding window is used to traverse, obtaining a plurality of 3×3 first matrices. The matrix center points of the plurality of 3×3 first matrices correspond to one element in the initial matrix, that is, any element in the initial matrix corresponds to a 3×3 first matrix. When the sliding window is a 3×3 matrix, the preset weight is also a 3×3 weight matrix. Through the inner product process between the 3×3 target matrix and the 3×3 preset weight, the result is the inner product value, and the target element corresponding to the initial matrix is updated based on the inner product value.

[0154] It can be understood that by updating the target element corresponding to the first matrix through the inner product value between each first matrix and the preset weight, each matrix element in the initial matrix can be updated, and then the target parameters of each channel can be obtained.

[0155] S207. Determine the superimposed parameter of the reference parameters of each channel.

[0156] Exemplarily, when the target configuration information of each channel needs to be the same, based on the reference parameters of each channel, the reference sub-parameters under the same candidate configuration information are determined, and the reference sub-parameters are superimposed to obtain the superimposed sub-parameters under the same candidate configuration information. Based on the superimposed sub-parameters, the superimposed parameter of the reference parameters of each channel is generated.

[0157] It can be understood that when superimposing the reference parameters of each channel in the chip, the reference parameters of each channel are aligned with the same candidate configuration information to obtain the reference sub-parameters under the same candidate configuration information, and the reference sub-parameters under the same candidate configuration information are superimposed to obtain the superimposed sub-parameters under the same candidate configuration information. The superimposed parameter of the channel in the chip is generated through the superimposed sub-parameters.

[0158] Exemplarily, the number of channels of the chip is 12, and the configuration information of each channel includes 4 states. Taking the abscissa as the receiver candidate configuration information (including 0 - 3) and the ordinate as the transmitter candidate configuration information (including 0 - 3), the combined reference parameters of the receiver and the transmitter are obtained, and the reference parameters are represented by a matrix (such as Figure 3 shown in 310).

[0159] Taking channel 0 corresponding to matrix 310 as an example, when the receiver candidate configuration information is 0 and the transmitter candidate configuration information is 0, the reference sub - parameter (sampling window length) is a11. And so on, determine the reference sub - parameters when the receiver candidate configuration information corresponding to each channel is 0 and the transmitter candidate configuration information is 0, superimpose the reference sub - parameters corresponding to multiple channels, obtain the superimposed sub - parameter when the receiver candidate configuration information is 0 and the transmitter candidate configuration information is 0, and so on for the superimposed sub - parameters corresponding to other candidate configuration information. Then, according to multiple superimposed sub - parameters, each superimposed parameter is generated.

[0160] It can be understood that when the number of communication channels in the chip is 12, the reference parameters of each channel are superimposed to obtain the superimposed parameters of the 12 channels in the chip. That is, the reference parameters of the 12 channels in the chip are superimposed into 1 superimposed parameter. For example, if the superimposed parameters of each channel are displayed as images, then the superimposed parameter is to superimpose 12 images into 1 image.

[0161] S208, perform smoothing processing on the superimposed parameter to obtain the parameter after smoothing processing.

[0162] Exemplarily, perform smoothing processing on the superimposed parameter to obtain the parameter after smoothing processing, and use the parameter after smoothing processing as the target parameter of each channel.

[0163] Optionally, the method of performing smoothing processing on the superimposed parameter can refer to the method of performing smoothing processing on the reference parameter as described above, and will not be elaborated here.

[0164] S209, use the parameter after smoothing processing as the target parameter of each channel.

[0165] It can be understood that since the superimposed parameter is the parameter after superimposing each channel in the chip, after performing smoothing processing on the superimposed parameter, the parameter after smoothing processing is obtained, and using the parameter after smoothing processing as the target parameter of each channel, the target parameters of each channel are the same.

[0166] S210, traverse the target parameter to obtain the maximum parameter in the target parameter.

[0167] Exemplarily, in the case of obtaining the target parameters of each channel, traverse the target parameter to obtain the maximum parameter in the target parameter.

[0168] S211. Use the candidate configuration information corresponding to the maximum parameter as the target configuration information for each channel.

[0169] Exemplarily, according to the adjustment directions of the candidate configuration information, adjust the sorting of the initial parameters of each channel in the chip under the candidate configuration information to obtain reference parameters, smooth the reference parameters to obtain the target parameters of each channel, traverse the target parameters to obtain the maximum parameter in the target parameters, and use the candidate configuration information corresponding to the maximum parameter as the target configuration information for each channel.

[0170] In the above technical solution, the sorting corresponding to the initial parameters is adjusted according to the chip characteristics to obtain reference parameters with continuously changing physical quantities of adjacent parameters, and the reference parameters are smoothed to eliminate random measurement errors in the target parameters. On this basis, the target parameters are traversed to obtain the target configuration information corresponding to the maximum parameter in the target parameters, which can ensure the accuracy of the target configuration information of each channel and thus improve the tolerance of communication transmission.

[0171] It should be noted that the method in the following embodiments takes the reference parameter as the original matrix as an example to refine step S120 in the Figure 1 illustrated embodiment, or to smooth the reference parameter / superposition parameter.

[0172] In this embodiment, the configuration information of each channel includes 4 states. Taking the abscissa as the candidate configuration information of the receiver and the ordinate as the candidate configuration information of the transmitter as an example, the combined reference parameters of the receiver and the transmitter are obtained, and the reference parameters are represented by a matrix.

[0173] Figure 3 It is a schematic diagram of a smoothing process provided by an embodiment of the present application.

[0174] Exemplarily, as Figure 3 shown, the initial matrix is 310, and the sliding window is a 3*3 matrix. Taking the target element (any element in the initial matrix) in the initial matrix 310 as the center point of the matrix, multiple 3*3 first sub-matrices are collected from the initial matrix 310 (for example, when the target element is a11 330, the corresponding first sub-matrix is 320).

[0175] It can be understood that when the target element (m, n) is the center point of the matrix, when m - 1 < 0 or m + 1 > M - 1 (M is the number of rows of the initial matrix, and M is 4 in this embodiment), there are empty matrix elements in the matrix collected by the 3*3 sliding window. Or, when n = 0 or n = N - 1 (N is the number of columns of the initial matrix, and N is 4 in this embodiment), there are empty matrix elements in the matrix collected by the 3*3 sliding window. In summary, when there are empty matrix elements in the matrix collected by the sliding window, the empty matrix elements need to be filled.

[0176] Taking the target element a11 as an example, if there are empty matrix elements b11, b12, b13, b21, b31 in the first sub-matrix 320 collected by the 3*3 sliding window, then calculate the target matrix elements of the empty matrix elements b11, b12, b13, b21, b31 respectively.

[0177] Specifically, determine that the non-empty matrix element closest to the empty matrix element b11 is a11, then determine a11 as the target matrix element of the empty matrix element b11, and fill the empty matrix element b11 with the target element a11. Determine that the non-empty matrix element closest to the empty matrix element b12 is a11, then determine a11 as the target matrix element of the empty matrix element b12, and fill the empty matrix element b12 with the target element a11. Determine that the non-empty matrix element closest to the empty matrix element b13 is a12, then determine a12 as the target matrix element of the empty matrix element b13, and fill the empty matrix element b13 with the target element a12. Determine that the non-empty matrix element closest to the empty matrix element b21 is a11, then determine a11 as the target matrix element of the empty matrix element b21, and fill the empty matrix element b21 with the target element a11. Determine that the non-empty matrix element closest to the empty matrix element b31 is a21, then determine a21 as the target matrix element of the empty matrix element b31, and fill the empty matrix element b31 with the target element a21.

[0178] Further, by filling the empty matrix element b11 with the target element a11, filling the empty matrix element b12 with the target element a11, filling the empty matrix element b13 with the target element a12, filling the empty matrix element b21 with the target element a11, and filling the empty matrix element b31 with the target element a21, the first matrix 321 is obtained.

[0179] Exemplarily, the distance between the empty matrix element and the non-empty matrix element can be determined by their distribution in the matrix. Taking the empty matrix element b13 as an example, from the empty matrix element b13 to the non-empty matrix element a12, only 1 step is required (from the first row to the second row), then determine the distance between the empty matrix element b13 and the non-empty matrix element a12 is 1; from the empty matrix element b13 to the non-empty matrix element a11, 2 steps are required (from the first row to the second row, from the third column to the second column), then determine the distance between the empty matrix element b13 and the non-empty matrix element a11 is 2, and then determine a12 as the target matrix element of the empty matrix element b13. The determination methods of the target matrix elements of other empty matrix elements can be deduced by analogy and will not be elaborated here.

[0180] Assume that the preset weight is a 3*3 weight matrix 340. Perform an inner product operation on the first matrix 321 and the weight matrix 340 to obtain an inner product value c11 350, and update a11 with c11 (that is, replace a11 with c11). For example, c11 = a11 + a11 + a12 + a11 + 5 * a11 + a12 + a21 + a21 + a22; or, c11 = (a11 + a11 + a12 + a11 + 5 * a11 + a12 + a21 + a21 + a22) / 9.

[0181] Figure 4 It is a schematic diagram of another smoothing process provided by an embodiment of the present application;

[0182] Exemplarily, as Figure 4 shown, the target element (m, n) is the center point of the matrix, and there may also be no empty matrix elements in the first sub-matrix collected by the 3*3 sliding window. For example, the initial matrix is 310, and the sliding window is a 3*3 matrix. Taking the target element (any element in the initial matrix) in the initial matrix 310 as the center point of the matrix, multiple 3*3 first sub-matrices are collected from the initial matrix 310 (when the target element is a22 370, the corresponding first sub-matrix is 360).

[0183] Taking the target element a22 as an example, all elements in the first sub-matrix 360 collected by the 3*3 sliding window are non-empty matrix elements, then the first sub-matrix 360 is used as the first matrix corresponding to a22. Assume that the preset weight is a 3*3 weight matrix 340. Perform an inner product operation on the first matrix (the first sub-matrix 360) and the weight matrix 340 to obtain an inner product value c22 380, and update a22 with c22 (that is, replace a22 with c22). For example, c22 = a11 + a12 + a13 + a21 + 5 * a22 + a23 + a31 + a32 + a33; or, c22 = (a11 + a12 + a13 + a21 + 5 * a22 + a23 + a31 + a32 + a33) / 9.

[0184] It can be understood that the weight value of the central element of the preset weight is greater than that of other elements, which can ensure that each element retains its own characteristics during the smoothing process, moderately integrates adjacent parameter information, and reduces the influence of random jitter. Further, through the above method, all elements in the initial matrix can be updated to obtain the target parameters of each channel (that is, the matrix after smoothing and updating the initial matrix).

[0185] It should be noted that the above description of the accompanying drawings (explained by taking 4 status information as the candidate configuration information as an example) is only exemplary. In fact, the data volume of the original matrix corresponding to the above initial parameters may be much more than 4. However, no matter how much the data volume of the original matrix is, the smoothing process can be carried out in the above manner.

[0186] It should be noted that the method of the following embodiments adjusts the initial parameters to reference parameters and smooths the reference parameters through an image method to obtain target parameters, and further determines the refinement of the target configuration information of each channel.

[0187] Figure 5 It is a visualization schematic diagram of a chip parameter determination method provided by an embodiment of the present application.

[0188] This embodiment is described by taking the DCC parameter adjustment process as an example. The candidate configuration information with 4 channels (channel 0, channel 1, channel 2, and channel 3) includes 0-15. The abscissa is the receiver DCC (RX DCC), the ordinate is the transmitter DCC (TX DCC), each pixel point is the sampling window length corresponding to each candidate configuration information, and the brightness of each pixel point is positively correlated with the sampling window length, that is, as the sampling window length increases, the brightness of the pixel point is greater.

[0189] Exemplarily, Figure 5 Figure (a) in it is a visualization schematic diagram of the initial parameters, Figure 5 Figure (b) in it is a visualization schematic diagram of the reference parameters, Figure 5 Figure (c) in it is a visualization schematic diagram of the target parameters.

[0190] Exemplarily, as Figure 5 shown in Figure (a) in it, obtain the initial parameters (sampling window length) of each channel in the chip under the candidate configuration information (0-15). The sampling window length refers to the window length of the transmitter and receiver under different configuration information, and perform visualization processing on the sampling window length (matrix data of 16*16) corresponding to each channel to obtain the image of channel 0, the image of channel 1, the image of channel 2, and the image of channel 3.

[0191] It should be noted that due to the actual structure of the hardware DCC register (for example, the forward adjustment and reverse adjustment in the register), it is necessary to adjust the initial parameters so that the changes in the physical quantities represented by the re-ordered (adjusted) reference parameters are continuous. Specifically, in the actual structure of the DCC register, since the forward adjustment is between 0 and 7, and the reverse adjustment is between 8 and 15, the sorting of the directly obtained initial parameters is 0 - 15. Therefore, there is a data mutation between 7 and 8. Thus, it is necessary to adjust the sorting of the initial parameters according to the adjustment directions of the respective candidate configuration information to obtain the reference parameters (16 * 16 matrix data).

[0192] Exemplarily, as Figure 5 shown in (b) therein, the reference parameters (16 * 16 matrix data) of each channel are visualized to obtain the image of channel 0, the image of channel 1, the image of channel 2, and the image of channel 3. It should be noted that the changes in the physical quantities represented by any adjacent parameters in the image of channel 0, the image of channel 1, the image of channel 2, and the image of channel 3 are continuous.

[0193] Exemplarily, as Figure 5 shown in (c) therein, the reference parameters are smoothed to obtain the target parameters (16 * 16 matrix data) of each channel. The target parameters (16 * 16 matrix data) of each channel are visualized, and the positions of the maximum parameters of each channel are marked to generate the image of channel 0, the image of channel 1, the image of channel 2, and the image of channel 3.

[0194] In the above technical solution, the present application adjusts the sorting corresponding to the initial parameters through chip features to obtain reference parameters with continuously changing physical quantities for adjacent parameters, and smooths the reference parameters to eliminate random measurement errors in the target parameters. On this basis, the target parameters are traversed to obtain the target configuration information corresponding to the maximum parameters in the target parameters, which can ensure the accuracy of the target configuration information of each channel, thereby improving the tolerance of communication transmission.

[0195] It should be noted that the above description of the drawings (explained by taking the number of chip channels as 4 as an example) is only exemplary, and the actual number of chip channels may be more. Other scenarios can be analogized accordingly and will not be elaborated here.

[0196] It should be understood that the above examples are for helping those skilled in the art to understand the embodiments of the present application, rather than limiting the embodiments of the present application to the specific numerical values or specific scenarios illustrated. Those skilled in the art can obviously make various equivalent modifications or changes according to the above examples, and such modifications or changes also fall within the scope of the embodiments of the present application.

[0197] As described above in conjunction withFigures 1 to 5 This describes in detail the method for determining chip parameters provided by the embodiments of the present application; hereinafter, it will be combined with Figure 6 and Figure 7 This describes in detail the embodiments of the chip parameter determination device of the present application. It should be understood that the chip parameter determination device in the embodiments of the present application can execute various methods of the foregoing embodiments of the present application, that is, the specific working processes of the following various products can refer to the corresponding processes in the foregoing method embodiments.

[0198] Figure 6 It is a schematic structural diagram of a chip parameter determination device provided by the embodiments of the present application.

[0199] Exemplarily, as Figure 6 shown, the chip parameter determination device 600 includes:

[0200] An acquisition module 610: configured to acquire the initial parameters of each channel in the chip under the candidate configuration information; wherein, the candidate configuration information is used to configure the transmission performance of the channel; the initial parameters include the sampling window length of the channel;

[0201] An adjustment module 620: configured to adjust the sorting of the initial parameters based on the adjustment directions of the respective candidate configuration information to obtain reference parameters;

[0202] A processing module 630: configured to perform smoothing processing on the reference parameters to obtain the target parameters of each channel;

[0203] A selection module 640: configured to select the target configuration information of each channel from the candidate configuration information based on the target parameters.

[0204] Optionally, as an embodiment, the processing module 630 is specifically configured to:

[0205] Determine whether the target configuration information of each channel needs to be the same;

[0206] If the target configuration information of each channel needs to be the same, determine the superimposed parameters of the reference parameters of each channel; and perform smoothing processing on the superimposed parameters to obtain the target parameters of each channel;

[0207] If the target configuration information of each channel does not need to be the same, perform smoothing processing on the reference parameters to obtain the target parameters of each channel.

[0208] Optionally, as an embodiment, the reference parameter is an initial matrix; the processing module 630 is specifically configured to:

[0209] Taking the target element in the initial matrix as the matrix center point, collect a plurality of first matrices from the initial matrix; wherein, the target element is any element in the initial matrix;

[0210] Perform an inner product operation on the target matrix and the preset weight to obtain an inner product value; wherein, the target matrix is any one of multiple first matrices;

[0211] Update the target element corresponding to the target matrix based on the inner product value.

[0212] Optionally, as an embodiment, the processing module 630 is specifically configured to:

[0213] Using the target element in the initial matrix as the matrix center point, slide a sliding window in the initial matrix to obtain multiple first sub-matrices;

[0214] If there are empty matrix elements in the target sub-matrix, determine the target matrix elements corresponding to the empty matrix elements, and fill the empty matrix elements based on the target matrix elements to obtain the first matrix corresponding to the target sub-matrix; wherein, the target sub-matrix is used to indicate any one of the multiple first sub-matrices;

[0215] If all the elements in the target sub-matrix are non-empty matrix elements, use the target sub-matrix as the first matrix.

[0216] Optionally, as an embodiment, the processing module 630 is specifically configured to:

[0217] Determine the reference sub-parameters under the same candidate configuration information based on the reference parameters of each channel;

[0218] Perform a superposition operation on the reference sub-parameters to obtain the superposed sub-parameters under the same candidate configuration information;

[0219] Generate the superposed parameters of the reference parameters of each channel based on the superposed sub-parameters.

[0220] Optionally, as an embodiment, the processing module 630 is specifically configured to:

[0221] Perform a smoothing operation on the superposed parameters to obtain the parameters after smoothing;

[0222] Use the parameters after smoothing as the target parameters of each channel.

[0223] Optionally, as an embodiment, the selection module 640 is specifically configured to:

[0224] Traverse the target parameters to obtain the maximum parameter in the target parameters;

[0225] Use the candidate configuration information corresponding to the maximum parameter as the target configuration information of each channel.

[0226] Optionally, as an embodiment, the chip parameter determination device 600 further includes an output module, and the output module is specifically configured to:

[0227] Generate an initial image based on the target parameters of each channel; wherein, the brightness value in the initial image is determined based on the value of the target parameter;

[0228] Perform a marking process on the position of the maximum parameter in the initial image to generate the target image of each channel;

[0229] Output the target image of each channel.

[0230] It should be noted that the above chip parameter determination device 600 is embodied in the form of functional units. The term "module" here can be implemented in software and / or hardware forms, and no specific limitation is made thereto.

[0231] For example, the "module" can be a software program, a hardware circuit, or a combination of the two that implements the above functions. The hardware circuit may include an application specific integrated circuit (ASIC), an electronic circuit, a processor (such as a shared processor, a proprietary processor, or a group of processors, etc.) for executing one or more software or firmware programs, and a memory, a combined logic circuit, and / or other suitable components that support the described functions.

[0232] Therefore, the units of each example described in the embodiments of the present application can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0233] Figure 7 It is a schematic structural diagram of a chip parameter determination device provided by an embodiment of the present application.

[0234] Exemplarily, as Figure 7 shown, the chip parameter determination device 700 includes: a memory 710 and a processor 720, wherein, an executable program code 730 is stored in the memory 710, and the processor 720 is configured to call and execute the executable program code 730 to execute a chip parameter determination method.

[0235] Exemplarily, the memory 710 can be used to store the relevant programs of the chip parameter determination method provided in the embodiments of the present application; the processor 720 can call the relevant programs of the chip parameter determination method stored in the memory 710 to execute the chip parameter determination method of the embodiments of the present application; for example, obtain the initial parameters of each channel in the chip under the candidate configuration information; wherein, the candidate configuration information is used to configure the transmission performance of the channel; the initial parameters include the sampling window length of the channel; based on the adjustment directions of the candidate configuration information, adjust the sorting of the initial parameters to obtain reference parameters; perform smoothing processing on the reference parameters to obtain the target parameters of each channel; based on the target parameters, select the target configuration information of each channel from the candidate configuration information.

[0236] In this embodiment, the device can be divided into functional modules according to the above method examples. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0237] In the case of dividing each functional module corresponding to each function, the device may further include an acquisition module, an adjustment module, a processing module, a selection module, etc. It should be noted that all relevant contents of each step involved in the above method embodiments can be cited in the function descriptions of the corresponding functional modules, and will not be repeated here.

[0238] It should be understood that the device provided in this embodiment is used to execute the above-mentioned chip parameter determination method, so the same effect as the above implementation method can be achieved.

[0239] In the case of adopting an integrated unit, the device may include a processing module and a storage module. Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logical blocks, modules, and circuits shown in combination with the disclosure of the present application. The processor can also be a combination that realizes computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0240] In addition, the device provided in the embodiments of the present application can specifically be a chip, a component, or a module. The chip can include a connected processor and a memory; wherein, the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the chip parameter determination method provided in the above embodiments.

[0241] The present application also provides a computer-readable storage medium, in which computer program code is stored. When the computer program code runs on a computer, the computer is caused to execute the above-mentioned related method steps to implement a method for determining chip parameters provided in the above embodiments. Among them, the computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, digital versatile discs (DVDs), compact disc read-only memories (CD-ROMs), microdrives, and magneto-optical discs, read-only memories (ROMs), random access memories (RAMs), erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), dynamic random access memories (DRAMs), video random access memories (VRAMs), flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0242] The present application also provides a computer program product. When the computer program product runs on a computer, the computer is caused to execute the above-mentioned related steps to implement a method for determining chip parameters provided in the above embodiments.

[0243] Among them, the computer-readable storage medium, computer program product or chip provided in the present application are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be elaborated here.

[0244] Through the description of the above embodiments, those skilled in the art can understand that, for the convenience and brevity of description, only the above-mentioned division of each functional module is used as an example for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.

[0245] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0246] The above content is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A chip parameter determination method, characterized in that: The method comprises: Acquire initial parameters of each channel in the chip under candidate configuration information; wherein the candidate configuration information is used to configure the transmission performance of the channel; the initial parameters include the sampling window length of the channel; Based on the adjustment direction of each candidate configuration information, the order of the initial parameters is adjusted to obtain reference parameters, so that the change of the physical quantity represented by the reference parameters is continuous and not abrupt; Smoothing the reference parameters to obtain target parameters of each channel; Based on the target parameter, target configuration information of each channel is obtained by selecting from the candidate configuration information.

2. The method according to claim 1, characterized in that: Before the reference parameters are smoothed to obtain the target parameters of each channel, the method further includes: Determine whether the target configuration information of each channel needs to be the same; The smoothing of the reference parameters to obtain the target parameters of each channel includes: If the target configuration information of each channel needs to be the same, determine the superposition parameters of the reference parameters of each channel; and perform the smoothing process on the superposition parameters to obtain the target parameters of each channel; If the target configuration information of each channel does not need to be the same, the reference parameter is smoothed to obtain the target parameter of each channel.

3. The method according to claim 2, characterized in that The reference parameter is an initial matrix; The step of smoothing the reference parameters to obtain the target parameters of each channel includes: Taking the target element in the initial matrix as the matrix center point, a plurality of first matrices are acquired from the initial matrix; wherein the target element is any element in the initial matrix; Performing inner product processing on the target matrix and the preset weights to obtain an inner product value; wherein the target matrix is ​​any matrix among the multiple first matrices; Based on the inner product value, the target element corresponding to the target matrix is ​​updated.

4. The method according to claim 3, characterized in that: The step of taking the target element in the initial matrix as the matrix center point and acquiring a plurality of first matrices from the initial matrix includes: Taking the target element in the initial matrix as the center point of the matrix, sliding in the initial matrix based on a sliding window to obtain a plurality of first sub-matrices; If there is an empty matrix element in the target submatrix, determine the target matrix element corresponding to the empty matrix element, fill the empty matrix element based on the target matrix element, and obtain the first matrix corresponding to the target submatrix; wherein the target submatrix is ​​used to indicate any submatrix among the multiple first submatrices; and the target matrix element is used to indicate a non-empty matrix element that is closest to the empty matrix element; If all matrix elements in the target sub-matrix are non-empty, the target sub-matrix is ​​used as the first matrix.

5. The method according to claim 2, characterized in that: The determining of the superposition parameters of the reference parameters of each channel includes: Determine reference sub-parameters under the same candidate configuration information based on the reference parameters of each channel; Performing superposition processing on the reference sub-parameters to obtain superposition sub-parameters under the same candidate configuration information; Based on the superposition sub-parameters, the superposition parameters of the reference parameters of each channel are generated.

6. The method according to claim 2, characterized in that The performing the smoothing process on the superposition parameters to obtain the target parameters of each channel includes: Performing the smoothing process on the superposition parameters to obtain smoothed parameters; The smoothed parameters are used as the target parameters of each channel.

7. The method according to any one of claims 1 to 6, characterized in that The step of selecting, based on the target parameter, the target configuration information of each channel from the candidate configuration information comprises: Traversing the target parameters to obtain the maximum parameter among the target parameters; The candidate configuration information corresponding to the maximum parameter is used as the target configuration information of each channel.

8. The method according to claim 7, characterized in that The method further comprises: Based on the target parameters of each channel, an initial image is generated; wherein the brightness value in the initial image is determined based on the value of the target parameter; Marking the position of the maximum parameter in the initial image to generate the target image of each channel; Output the target image of each channel.

9. A chip parameter determination device, characterized in that: include: A memory for storing executable program codes; A processor is used to call and run the executable program code from the memory, so that the chip parameter determination device executes the chip parameter determination method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the chip parameter determination method according to any one of claims 1 to 8 is implemented.

Images