PHY Link Status Detection Using Speed Up Mode Circuit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Gigabit Ethernet physical-layer devices require hundreds of milliseconds to indicate a link down condition, which is too long for fault-tolerant networks that need to detect a faulty link and shift data transmission within 50 ms or less.
Innovation Solution
A physical layer controller with a speed up mode circuit that indicates a link status as FAIL when the local receiver status is not OK, regardless of a timer's status, allowing for rapid transition to a LINK DOWN state without relying on the maxwait_timer, and using a loss lock timer to ensure quick detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If maxwait_timer is used to determine link down status, then link status indication is reliable and avoids premature failure detection, but link down detection time increases to hundreds of milliseconds
Solution Approach 1:
The patent introduces a dynamic mechanism where the link down detection timing is made adjustable based on operational conditions. A speed up mode circuit can override the fixed maxwait_timer delay when link quality indicators show sustained degradation, allowing the system to transition from a static timing approach to a dynamic one that adapts to actual link conditions.
Solution Approach 2:
The patent changes the timing parameter from a fixed value (maxwait_timer) to a variable approach. By introducing conditions under which the timer can be bypassed or adjusted, the system transforms the rigid timing parameter into a flexible one that can be modified based on link quality metrics and operational requirements.
2Loss of time
If maxwait_timer is reduced to achieve faster link down detection, then link failure detection time decreases to meet fault-tolerant requirements, but premature entry into LINK DOWN state may occur during initial training
Solution Approach 1:
The patent applies different detection criteria and timing requirements to different operational phases. During initial training, the system uses more stringent criteria to avoid premature failure detection, while in normal operation, faster detection mechanisms can be employed. This local differentiation of detection quality based on operational context resolves the contradiction.
Solution Approach 2:
The patent performs preliminary assessment of link quality indicators before triggering link down detection. By evaluating multiple parameters and establishing trends before declaring link failure, the system prepares in advance to distinguish between temporary training fluctuations and genuine link failures, enabling faster detection without premature actions.
3Loss of time
If speed up mode circuit is enabled for rapid link down detection, then link failure detection time decreases to less than 50 ms, but risk of premature link down indication increases
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor link quality indicators and provide information to the speed up mode circuit. This feedback loop allows the system to adjust detection aggressiveness based on actual link conditions, enabling rapid detection when truly needed while suppressing premature indications during transient issues.
Solution Approach 2:
The patent introduces protective mechanisms that cushion against premature link down indications. By establishing thresholds, hysteresis, or confirmation requirements before triggering link failure detection, the system prepares in advance to filter out false positives while maintaining the capability for rapid detection of genuine failures.
Data Source
AI summary
A system including a monitoring module and a controller. The monitoring module is configured to monitor a link. The controller is configured to, in response to the monitoring module determining that the link is up, (i) start a first timer; (ii) indicate that the link is down in response to the monitoring module determining, prior to the first timer expiring, that the link is down; and (iii) indicate that the link is up in response to the monitoring module determining, subsequent to the first time expiring, that the link is up.