A Sopc Chip Fault Tolerance Method Based on Readback Self-Reconfiguration

Abstract

A SoPC chip fault-tolerant method based on readback self-reconfiguration, aiming at the problem that the traditional radiation-hardened FPGA will have multiple particle flips in space every day, it provides a method based on the feedback without increasing the SoPC chip area. The self-reconfiguration mode of reading realizes the method of autonomous fault detection and fault repair of the chip. The method of the present invention first reads the configuration data in the FPGA configuration memory and the original configuration data stored in the Flash, and then compares the two bit by bit, and can verify whether the read-back configuration data has a fault by comparing the difference in the file format. Locate the fault, and finally correct the fault according to the original configuration file. The method of the invention completes fault detection, fault interpretation and fault repair without increasing peripheral detection equipment and detection circuits, improves the reliability of SoPC chips in outer space environment applications, and promotes the development of SoPC chips.

Description

technical field [0001] The invention relates to an intelligent autonomous fault tolerance and fault repair of a domestic SoPC (Programing System on Chip) chip BM3109, in particular to a SoPC chip fault tolerance method based on readback self-reconfiguration, so that the SoPC chip can be used in an outer space environment. Ability to autonomously detect and repair faults. Background technique [0002] Traditional radiation-hardened FPGAs will have multiple particle flips (SEU) in space every day. For example, the XQVR300 FPGA in LEO orbit flips an average of 2.05 times a day, while the XQR4036XL in a 98-degree inclined orbit flips up to 148.5 times a day when solar flares are abnormal. Second-rate. In order to adapt to the harsh space environment, traditional space-borne electronic equipment generally adopts redundant methods to improve system reliability, such as dual-machine backup, three-machine backup and other redundancy schemes, but most of the redundancy schemes are a...

AI Technical Summary

Problems solved by technology

In order to adapt to the harsh space environment, traditional space-borne electronic equipment generally adopts redundant methods to improve system reliability, such as dual-machine backup, three-machine backup and other redundancy schemes, but most of the redundancy schemes are aimed at known Designed for failure modes, there is no effective way to deal with complex and unknown failures

In addition, even if the mechanism of triple-mode redundancy TMR is used, it can only accommodate one fault at most. When there are two or more faults, errors will occur because the data arbiter cannot give the correct arbitration result.

The triple-mode redundancy method can only filter faults, but cannot repair the faults that occur

At the same time, the satellite on a chip itself is extremely small, and the redundant design scheme will inevitably lead to an increase in volume, which is contrary to the design concept of the satellite on a chip

Images