Bit-Spreading Memory Layout for Radiation-Hardened ECC Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Memory systems in space-based applications suffer from high radiation-induced single-bit, multi-bit, and single event functional interrupt errors, rendering them non-functional, and existing solutions like shielding and redundant memories are impractical due to weight or cost constraints.

Innovation Solution

A bit spreading technique distributes data bits and error correction codes across multiple separated RAMs, allowing for error correction and recovery even if a single RAM is affected by radiation, reducing the likelihood of multiple bits being impacted simultaneously.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If shielding is used to reduce radiation exposure, then reliability is improved, but weight increases significantly

Engineering Contradiction:
Improveradiation resistanceVSAvoidweight
Core Design Contradiction:
ReliabilityVSWeight of moving object

Solution Approach 1:

The patent divides a single data word into multiple smaller units called sub-words, where each sub-word is stored in a separate RAM. This segmentation ensures that even if radiation affects one RAM, the other sub-words remain intact, providing error correction without requiring heavy shielding.

Inventive Principle:
Principle #1Segmentation

2Reliability

If redundant memories with majority voting are used, then reliability is improved, but cost and complexity increase

Engineering Contradiction:
Improveerror correction capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of using redundant copies of entire memory words with majority voting logic, the patent segments each word into sub-words stored in separate RAMs. This reduces the complexity of error correction by eliminating the need for majority voting circuits while still providing protection against radiation-induced errors.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses error correction codes that are distributed across multiple RAMs rather than creating full redundant copies of data. This approach provides error correction functionality without the complexity and cost of complete redundancy systems.

Inventive Principle:
Principle #26Copying

3Reliability

If ECC schemes are used, then error correction is improved, but they cannot effectively handle multi-bit and SEFI errors

Engineering Contradiction:
Improvesingle-bit error correctionVSAvoidmulti-bit and SEFI error susceptibility
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent segments data into sub-words stored in separate RAMs, which fundamentally changes how errors are handled. This segmentation approach can detect and correct multi-bit errors and single event functional interrupts because even if one RAM is completely corrupted, the other sub-words provide the information needed for recovery.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20260010430A1Bit spreading technique for radiation hardened error resistant memory system
Publication Date: 2026.01.08 BAE SYSTEMS INFORMATION ANDELECTRONIC SYSTEMS INTEGRATION INC
  • US20260010430A1 patent drawing
  • US20260010430A1 patent drawing
  • US20260010430A1 patent drawing

AI summary

Techniques are provided for an error resistant radiation hardened memory system based on spreading of data bits among multiple random access memories (RAMs). A memory system implementing the techniques according to an embodiment includes a first plurality of RAMs configured to store data bits written to the memory system, the data bits distributed over the first plurality of RAMs. The system also includes an error correction coding (ECC) circuit configured to generate ECC codes, each of the codes associated with a unique group of the data bits. The system further includes a second plurality of RAMs configured to store bits of the ECC codes such that the bits of each ECC code are distributed over the second plurality of RAMs. The system further includes a reporting circuit configured to report a single bit error correction or a double bit error detection, resulting from a read operation on the memory system.