Training Frame Parsing for Bit-Error-Tolerant SerDes Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for optimizing SerDes performance by transferring training frames between interfaces are not effective due to high bit error rates, which hinder correct parsing and adjustment of SerDes parameters.
Innovation Solution
A data processing method and apparatus that allow a first interface to correctly parse a training frame encoding sequence even when bit errors occur, by using a preset value for target sign bits prone to errors and adjusting the parsing based on these values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If training frames are transferred between interfaces to optimize SerDes performance, then SerDes performance optimization is enabled, but bit errors occur during transfer which hinder correct parsing
Solution Approach 1:
The patent changes the parameter of sign bit value interpretation. When a sign bit value differs from the preset value, the system parses the encoding sequence based on the assumption that the sign bit equals the preset value, rather than treating it as an error condition. This parameter change in interpretation logic allows the system to tolerate bit errors and correctly parse training frames despite voltage jumps during transfer.
2Ease of operation
If traditional parsing methods are used for training frames, then parsing simplicity is maintained, but bit errors cause parsing failures
Solution Approach 1:
The patent modifies the parsing parameter by introducing a conditional interpretation rule for sign bits. Instead of requiring exact matching between received and expected sign bit values, the system parses based on the preset value even when differences occur. This maintains the simplicity of the parsing process while significantly improving reliability in the presence of bit errors.
Data Source
Figure 1~2
Figure 3~4
AI summary
A data processing method is provided. The method includes: obtaining a first encoding sequence (S101), and parsing the first encoding sequence. The first encoding sequence can also be correctly parsed when a tolerable bit error occurs in the first encoding sequence. For a target sign bit in the first encoding sequence, if a value of the target sign bit is different from a preset value of the target sign bit, the first encoding sequence is parsed based on the fact that the value of the target sign bit is equal to the preset value of the target sign bit (102). The target sign bit is one or more sign bits in which a bit error is likely to occur. It can be learned that, according to this method, for the target sign bit in which a bit error is likely to occur, if a bit error occurs in the target sign bit, in other words, the value of the target sign bit is different from the preset value of the target sign bit, the first encoding sequence may also be correctly parsed. This improves efficiency of obtaining a training frame through parsing, and correspondingly improves an effect of optimizing SerDes performance by transferring a training frame between interfaces.