Feedback-Driven Fuzzing for Electronic Device Testing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fuzzing technologies often operate in an open loop manner, limiting their ability to detect interesting test cases and identify feedback beyond crash events, which restricts their effectiveness in generating further test signals, especially for electronic devices with limited measurable parameters or those operating in constrained environments.

Innovation Solution

The implementation of a system and method that provides a first test signal to an electronic device, monitors its response, determines an expected response using a response model, and adjusts subsequent test signals based on detected and expected responses, enabling the detection of feedback beyond crash events and improving the identification of interesting test cases.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If open loop fuzzing is used, then the system is simple to operate, but the ability to detect interesting test cases and identify feedback is limited

Engineering Contradiction:
Improvefeedback detection capabilityVSAvoidfuzzing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements a feedback-driven fuzzing system where the fuzzer monitors device responses (crashes, parameter changes, state transitions) and uses this feedback to dynamically generate subsequent test signals. The response model analyzes detected responses to determine whether they represent interesting test cases, creating a closed-loop system that continuously improves testing effectiveness based on actual device behavior.

Inventive Principle:
Principle #23Feedback

2Loss of information

If traditional fuzzing is used, then crash events can be detected, but feedback beyond crash events cannot be identified

Engineering Contradiction:
Improvefeedback information completenessVSAvoidresponse monitoring complexity
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent extends feedback detection beyond the traditional crash dimension by monitoring additional response dimensions including parameter changes, state transitions, and operational behavior modifications. The response model evaluates multiple response types simultaneously, transforming the single-dimensional crash detection approach into a multi-dimensional feedback analysis system that captures comprehensive device responses.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Adaptability or versatility

If fuzzing is applied to devices with limited measurable parameters, then testing coverage is restricted, but the system remains compatible with constrained environments

Engineering Contradiction:
Improvedevice compatibilityVSAvoidtest signal effectiveness
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent adapts the fuzzing approach to devices with limited measurable parameters by dynamically adjusting the response model based on available feedback types. The system identifies which parameters can be measured in the constrained environment and modifies its feedback analysis accordingly, using alternative response indicators such as operational state changes or timing variations when traditional measurement parameters are unavailable, thereby maintaining testing effectiveness across diverse device configurations.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3764112A1Systems and methods for fuzzing with feedback
Publication Date: 2021.01.13 ROCKWELL COLLINS INC
  • EP3764112A1 patent drawingFigure 1
  • EP3764112A1 patent drawingFigure 2
  • EP3764112A1 patent drawingFigure 3

AI summary

A system can include one or more processors (112) and computer-readable instructions that when executed by the one or more processors (112), cause the one or more processors (112) to provide a first test signal (216) to an electronic device (128), monitor at least one parameter of the electronic device (128) during a time period subsequent to the test signal (216) being provided to the electronic device (128), determine, based on the at least one parameter, a detected response of the electronic device (128) to the first test signal (216), determine, using a response model (204), an expected response of the electronic device (128) to the first test signal (216), and provide a second test signal (216) based on the detected response and the expected response to the electronic device (128). The system can include a communications circuit (120) that provides the test signal (216) and receives at least some feedback (220) indicating the parameters, and sensors (124) that receive at least some feedback (220) indicating the parameters.