Partial Pattern Recognition in Symbol Streams

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional authentication systems requiring exact matches are vulnerable to eavesdropping attacks, as noise introduced by communication channels can lead to failed authentications and expose user credentials to eavesdroppers, while systems using specified noise symbols are less effective due to predictable noise positions.

Innovation Solution

Implementing an authentication method where users introduce user-specified noise symbols at unspecified positions within credentials, allowing the authenticator to decompose submitted symbols into vectors and use a distance metric to determine a match without knowing the noise positions or symbols, enhancing security against eavesdroppers and man-in-the-middle attacks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If exact match authentication is used, then security is improved, but reliability deteriorates due to noise in communication channels

Engineering Contradiction:
Improveauthentication success rateVSAvoidvulnerability to eavesdropping
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent converts the harmful effect of noise symbols into a beneficial security feature. By requiring users to intentionally insert noise symbols at random positions, the system transforms what would normally be authentication-failing errors into a mechanism that protects credentials from eavesdroppers. The authenticator is designed to tolerate these noise symbols through vector decomposition and distance metric comparison, thereby converting the harm into improved security while maintaining reliability.

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

2Reliability

If specified noise symbols are used, then reliability is improved, but security deteriorates due to predictable noise positions

Engineering Contradiction:
Improveauthentication toleranceVSAvoidpredictability of noise positions
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent implements dynamics by making the noise symbol positions variable and user-controlled rather than fixed or predetermined. Users dynamically select random positions for inserting noise symbols during each authentication attempt, and the authenticator dynamically adapts by decomposing the submitted credential into vectors and comparing them using distance metrics. This dynamic approach ensures that noise positions are unpredictable to eavesdroppers while maintaining authentication reliability.

Inventive Principle:
Principle #15Dynamics

3Reliability

If noise tolerance is implemented, then reliability is improved, but device complexity increases

Engineering Contradiction:
Improvenoise toleranceVSAvoidauthentication processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the submitted credential into multiple vectors, where each vector represents a possible arrangement of the true credential symbols among the noise symbols. The authenticator then compares these segmented vectors against the stored credential using distance metrics, allowing it to tolerate noise while managing complexity through structured decomposition rather than brute-force comparison of all possible noise arrangements.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3967011B1Partial pattern recognition in a stream of symbols
Publication Date: 2024.03.13 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3967011B1 patent drawingFigure 1
  • EP3967011B1 patent drawingFigure 2
  • EP3967011B1 patent drawingFigure 3

AI summary

Systems and methods may be used for providing more secure authentication attempts by implementing authentication systems with credentials that include interspersed noise symbols in positions selected, for example by a user. These systems and methods secure against eavesdroppers such as shoulder-surfers or man-in-the middle attacks as it is difficult for an eavesdropper to separate the noise symbols from legitimate credential symbols. Some systems and methods may use a subset of a credential with the interspersed noise symbols.