Bidirectional Link Equalization With Selective Re-Training After Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data processing systems face inefficiencies in performing equalization operations and recovery processes, particularly due to the need to execute all phases during initial equalization, which can be time-consuming and unnecessary in subsequent operations after error detection.
Innovation Solution
The system and method allow for skipping unnecessary phases during a second equalization operation by selectively performing only the first and second phases, or both, after an error is detected, thereby optimizing the recovery process and reducing execution time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all phases of the initial equalization operation are executed, then the equalization is performed completely and reliably, but the execution time increases and system performance decreases
Solution Approach 1:
The equalization operation is divided into multiple phases. The method segments the recovery process into initial equalization (all phases) and re-equalization (selected phases only), allowing selective execution of phase groups based on error detection results.
Solution Approach 2:
During re-equalization after error detection, the method performs only necessary phases (first and second phases) rather than all phases. This partial action is sufficient to resolve the detected error while avoiding unnecessary execution of other phases, thus reducing time loss.
2Reliability
If all phases of the initial equalization operation are executed, then the equalization is performed completely, but the system operating performance deteriorates
Solution Approach 1:
The method applies partial action by executing only the necessary first and second phases during re-equalization operations. This maintains sufficient equalization quality to ensure reliability while improving system operating performance by reducing the time spent on unnecessary phases.
Solution Approach 2:
The method skips unnecessary phases during re-equalization by directly jumping to the first and second phases when errors are detected. This rushing through of essential phases while omitting non-essential ones improves productivity without compromising the critical equalization function.
3Loss of time
If the equalization operation is performed quickly by skipping phases, then the execution time is reduced, but the risk of incomplete equalization increases
Solution Approach 1:
The method uses error detection as feedback to determine which phases to execute. When an error is detected, the system feedback-induces a re-equalization operation that selectively performs the first and second phases, ensuring completeness where needed while maintaining speed.
Solution Approach 2:
The first and second phases are performed as preliminary actions during re-equalization to address the most common error conditions. This preliminary focus on critical phases ensures reliability for the majority of error cases while maintaining fast execution time.
Data Source
AI summary
A system includes a first device including a first transmitter and a first receiver, and a second device including a second transmitter and a second receiver and configured to communicate with the first device. The two devices perform a first equalization operation by performing a first phase in which the first receiver performs a signal tuning operation on the second transmitter, a second phase in which the second receiver performs a signal tuning operation on the first transmitter, and one or more other phases. The first, second, and other phases may constitute all the phases of the first equalization operation, and some of the other phases may precede the first and second phases. In response to detection of an error after the first equalization operation, the two devices may perform a second equalization operation by performing the first phase, the second phase, or both, but not performing the other phases.