PLD Configuration Memory Recovery From Soft Error Upsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Programmable logic devices (PLDs) face failures due to soft errors in volatile configuration memories, which can be caused by natural radioactivity, power fluctuations, and cosmic rays, leading to potential system failures in critical applications.
Innovation Solution
A method and circuitry within PLDs to continuously monitor and automatically recover from soft errors by using an error-detection algorithm to generate checksums, detecting discrepancies, and reconfiguring the configuration memory while maintaining normal operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If volatile configuration memory is used in programmable logic devices, then configurability and reprogrammability are improved, but reliability deteriorates due to soft errors from natural radioactivity, power fluctuations, and cosmic rays
Solution Approach 1:
The patent applies preliminary action by continuously calculating checksums of configuration data before soft errors can cause system failure. The error detection circuit proactively monitors configuration memory integrity by repeatedly processing configuration data through error-detection algorithms, enabling the system to detect and correct errors before they disrupt normal operation.
Solution Approach 2:
The patent implements feedback through a closed-loop error detection and correction system. The error detection circuit continuously monitors configuration data, compares calculated checksums with stored checksums, and triggers reconfiguration when discrepancies are detected. This feedback mechanism ensures configuration integrity is maintained by automatically responding to detected errors.
2Reliability
If continuous monitoring of configuration memory is implemented, then reliability is improved, but device complexity increases due to additional error detection and reconfiguration circuitry
Solution Approach 1:
The patent applies universality by designing the configuration engine to perform multiple functions: it configures the PLD during initialization, continuously monitors configuration data integrity through checksum calculation, and executes reconfiguration when errors are detected. This multi-functional approach eliminates the need for separate dedicated error detection circuits, reducing overall device complexity while maintaining reliability.
Solution Approach 2:
The patent implements self-service through automatic error detection and correction mechanisms. The configuration memory system monitors its own integrity using built-in error detection circuits that continuously verify configuration data. When soft errors are detected, the system automatically triggers reconfiguration without external intervention, enabling the device to self-correct and maintain reliable operation.
3Reliability
If reconfiguration is initiated upon soft error detection, then reliability is improved, but loss of time occurs due to reconfiguration process interruption
Solution Approach 1:
The patent applies preliminary action by continuously calculating and storing checksums of configuration data before soft errors occur. This proactive monitoring allows the system to detect errors immediately upon occurrence, enabling rapid reconfiguration response. The preliminary computation of error detection codes ensures that when an error is detected, the reconfiguration process can begin without delay for error analysis or diagnosis.
Solution Approach 2:
The patent implements continuity of useful action by designing the error detection and reconfiguration process to minimize disruption to normal PLD operation. The configuration engine continuously monitors configuration data integrity in the background without interrupting the device's primary functions. When reconfiguration is triggered, the system maintains operational continuity by quickly restoring correct configuration data, ensuring minimal loss of useful action.
Data Source
AI summary
A method of recovering from a soft error within configuration data stored in a configured programmable logic device. The method includes repeatedly processing the configuration data stored within configuration memory of the device using an error-detection algorithm to generate a checksum. The generated checksum is compared with a previously generated checksum to detect if a soft error exists in the configuration data. If a soft error is detected, the programmable logic device initiates a reconfiguration of the configuration memory. The configuration memory is then reconfigured with the configuration data while preventing the programmable logic device from responding to the reconfiguration as though the reconfiguration was an initial configuration of the device. An embodiment of a programmable logic device designed for practicing the method is also disclosed.