OTP Error Detection and Correction Circuitry

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems with one-time programmable (OTP) elements, particularly magnetic tunnel junction (MTJ) devices, face challenges in error detection and correction due to wear and tear leading to erroneous logic states, which are not effectively addressed by traditional dual OTP configurations that perform logical OR operations.

Innovation Solution

Implementing error detection and correction circuitry with Single Error Correction (SEC) and Double Error Detection (DED) decoders or cyclic redundancy check (CRC) decoders in dual MTJ OTP configurations to generate flags for error detection and correction, ensuring accurate read outputs by selecting data from the least error-prone OTP element.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a dual OTP configuration with logical OR operation is used, then programming failures can be overcome, but MTJ oxide breakdown errors cannot be detected or corrected

Engineering Contradiction:
Improveerror detection and correction capabilityVSAvoiddecoder circuit complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system divides the error detection and correction function into separate decoder circuits for each OTP element bank. Each decoder independently processes data from its associated OTP bank, generating flags for error detection. This segmentation allows the system to handle different error types (single errors correctable, double errors detectable) through dedicated circuit modules rather than a monolithic complex decoder.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

ECC decoder circuits serve as intermediary components between the OTP element banks and the final output logic. These decoders act as mediators that detect and correct errors before data is combined through logical OR operations. The intermediary decoders prevent erroneous data from propagating through the system by correcting single-bit errors and detecting double-bit errors in each bank independently.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If dual OTP configuration is implemented, then redundancy is provided, but error propagation occurs with MTJ oxide breakdown

Engineering Contradiction:
Improvedata integrityVSAvoiderroneous logic one propagation
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The system converts the potentially harmful effect of MTJ oxide breakdown into a detectable and correctable condition. By implementing ECC decoders that generate error flags, the system transforms erroneous logic states into identifiable error conditions. Single-bit errors are corrected automatically, and double-bit errors are detected and flagged, preventing erroneous data propagation while utilizing the redundant OTP configuration.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

ECC decoder circuits provide feedback through error flags that indicate the status of data integrity in each OTP bank. When errors are detected (single or double bit errors), the decoders generate flags that feed back to the output selection logic. This feedback mechanism allows the system to identify which bank contains errors and select the correct data from the other bank, preventing error propagation.

Inventive Principle:
Principle #23Feedback

3Reliability

If logical OR operation is performed on dual OTP outputs, then programming failures are overcome, but post-programming errors are included

Engineering Contradiction:
Improveprogramming failure toleranceVSAvoidread output accuracy
Core Design Contradiction:
ReliabilityVSManufacturing precision

Solution Approach 1:

Error detection and correction is performed as a preliminary action before the logical OR operation combines data from dual OTP banks. ECC decoders process each bank's data independently, correcting single-bit errors and detecting double-bit errors before the results are combined. This preliminary error handling ensures that only corrected or validated data undergoes the logical OR operation, preventing post-programming errors from contaminating the final output.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2951835B1Error detection and correction of one-time programmable elements
Publication Date: 2019.10.09 QUALCOMM INC
  • EP2951835B1 patent drawingFigure 1
  • EP2951835B1 patent drawingFigure 2
  • EP2951835B1 patent drawingFigure 3

AI summary

A circuit includes a first one-time programmable (OTP) element and a second OTP element. The circuit also includes error detection circuitry coupled to receive a first representation of data from the first OTP element. The circuit further includes output circuitry responsive to an output of the error detection circuitry to output an OTP read result based on the first representation of the data or based on a second representation of the data from the second OTP element.