Context Segmentation for Dynamic Response and Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current context awareness systems in communication devices face challenges in accurately determining contexts due to noisy sensor data and dynamic environments, leading to unreliable instantaneous context inference and degraded performance when transitioning between contexts.

Innovation Solution

The use of multiple streams of context data and context segmentation boundaries to identify changes in context, combining information from sensors like microphones, calendars, and location services to provide confidence in inferred contexts, allowing for smoother transitions and improved accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If instantaneous context inference is used, then dynamic response is improved, but reliability deteriorates due to noisy sensor data

Engineering Contradiction:
Improvedynamic responseVSAvoidcontext determination accuracy
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system performs preliminary actions by collecting and storing context data segments over time before making context determinations. Instead of relying solely on instantaneous sensor readings, the system accumulates multiple data segments that represent different time windows, allowing it to prepare context information in advance and make more reliable determinations while maintaining dynamic response capability.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If data averaging is used to filter noise, then reliability is improved, but dynamic response deteriorates

Engineering Contradiction:
Improvecontext determination accuracyVSAvoiddynamic response
Core Design Contradiction:
ReliabilityVSSpeed

Solution Approach 1:

The system segments context data into multiple distinct time windows or segments, each representing a different temporal perspective. Instead of averaging all data uniformly, the system divides context information into segmented data sets that can be individually evaluated. This allows the system to maintain reliability through multiple data points while preserving dynamic response by selecting or weighting segments based on their temporal relevance to current conditions.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If multiple data sources are integrated, then context determination accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvecontext determination accuracyVSAvoidsystem architecture complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements a universal context determination framework that can handle multiple data sources through a common processing architecture. The context determination module is designed to be multi-functional, capable of processing various types of sensor data, application data, and context information through unified algorithms. This universal approach allows accurate context determination across diverse data sources without proportionally increasing system complexity, as the same core mechanisms handle different data types.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP2847986B1Configuring a terminal device according to a context determined by correlating different data sources
Publication Date: 2019.07.10 QUALCOMM INC
  • EP2847986B1 patent drawingFigure 1A
  • EP2847986B1 patent drawingFigure 1B
  • EP2847986B1 patent drawingFigure 2

AI summary

Methods, systems, computer-readable media, and apparatuses for inferring context are provided. In one potential implementation, first context information associated with a first duration is identified, second context information is accessed to determine a context segmentation boundary; and the first context information and the second context information is then aggregated to generate an inferred segmented aggregated context. In a further implementation, the first context information is used to average inferred contexts, and the context segmentation boundary is used to reset a start time for averaging the first context information.