Link training method and apparatus
By using pseudo-random sequences for channel equalization and locking during link training, the compatibility issues caused by different bus encoding methods are resolved, the generalization and robustness of the link training process are improved, and the reliability and efficiency of communication between devices are enhanced.
WO2025112045A9PCT designated stage Publication Date: 2026-05-28HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- HUAWEI TECH CO LTD
- Filing Date
- 2023-12-01
- Publication Date
- 2026-05-28
AI Technical Summary
Technical Problem
Different types of buses use different encoding methods during link training, resulting in incompatibility and a lack of generalization in the link training process.
Method used
Pseudo-random sequences are used as training sequences in the link training process for channel equalization and locking, and the encoding method is not limited. Channel equalization and locking are performed through different training sequences, thereby achieving the universality of the link training process.
Benefits of technology
It improves the generalization and robustness of the link training process, ensuring communication reliability even in harsher environments and improving the efficiency of communication between devices.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN2023135908_28052026_PF_FP_ABST
Abstract
A link training method and apparatus, which relate to the technical field of communications. The method comprises: starting a link training process for a link, wherein the link comprises at least one channel; receiving a first control frame, wherein the first control frame comprises a first training sequence; on the basis of the first control frame, locking a clock of the at least one channel, and sending a first logic layer management packet, wherein the first logic layer management packet indicates that the clock of the at least one channel has been successfully locked; receiving a second control frame, wherein the second control frame comprises a second training sequence, which comprises a pseudo-random sequence; on the basis of the second control frame, executing equalization training of the at least one channel, and sending a second logic layer management packet, wherein the second logic layer management packet indicates that the equalization training of the at least one channel has succeeded; receiving a third control frame, wherein the third control frame comprises a third training sequence, which comprises a pseudo-random sequence, and the second training sequence is different from the third training sequence; and on the basis of the third control frame, locking a data boundary of the at least one channel, and sending a third logic layer management packet, wherein the third logic layer management packet indicates a channel that has been successfully locked and / or a channel that has failed to be locked among the at least one channel.
Need to check novelty before this filing date? Find Prior Art