Hamming code based method for carrying out fault tolerance on static RAM multiple bit upset

Abstract

The invention discloses a Hamming code based method for carrying out fault tolerance on static RAM (random-access memory) multiple bit upset. The method comprises the following steps: carrying out coding and decoding on the static RAM by employing Hamming code; employing two identical static RAMs with Hamming code coding and decoding to realize parallel input and output; and connecting a dual mode output selection circuit on output terminals of the two static RAMs. The invention has certain tolerance on SRAM multiple bit upset caused by high energy particles; and as no refresh or update is required, a system speed is increased substantially.

Description

technical field [0001] The invention relates to the field of static random access memory (SRAM) fault-tolerant technology, in particular to a method for fault-tolerant SRAM multi-bit flipping based on Hamming codes. Background technique [0002] Static random access memory (SRAM) is an important part of spacecraft electronics system. The study of SRAM affected by space high-energy particles is the basis for fully understanding the mechanism of particle effects and conducting research on other types of devices. By studying the particle effect of SRAM, it provides an important reference for the operating life evaluation and reinforcement design of the spacecraft electronics system. [0003] The core part of SRAM is a memory cell array, which often stores data or instructions. Due to the impact of high-energy particles in space, the contents of the storage unit will be reversed, resulting in errors in the data or instructions output by the SRAM, causing the system to collect w...

AI Technical Summary

Problems solved by technology

However, the problem with the Hamming code is that it can only realize the detection and correction of single-bit errors in one byte and the detection of double-bit errors. For double-bit and multiple-bit errors, the error cannot be corrected. The system is refreshed or built-in. Update measures to correct multiple bit errors in the

If the method of system refresh or self-update is adopted every time, it will bring a great delay to the system, occupy the timing resources of the system, and reduce the system speed

Images