Context Server QoC Parameter Correlation for Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing context-aware systems lack efficient methods to quantify and ensure the quality of context information from multiple sources, often failing to resolve ambiguities or inconsistencies, and cannot provide context information at a requested quality level.
Innovation Solution
A method that correlates and modifies Quality-of-Context (QoC) parameters such as accuracy and probability of correctness to meet a requested quality level, using trade-offs between these parameters and Dempster-Shafer theory to handle conflicting and reinforcing information from multiple context sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If context information is provided directly from multiple context sources without modification, then the quantity of context information is sufficient, but the quality level (probability of correctness) cannot be guaranteed to meet requested standards
Solution Approach 1:
The system performs preliminary determination of the second QoC parameter (probability of correctness) based on the first QoC parameter before providing context information to the application. This advance calculation allows the system to assess whether the context information meets the requested quality level and perform necessary modifications beforehand, rather than dealing with quality issues after the fact.
Solution Approach 2:
The context server acts as an intermediary between context sources and context-aware applications. It receives context information with the first QoC parameter, determines the second QoC parameter, modifies parameters as needed, and provides the processed information to applications. This intermediary role allows the system to manage quality requirements without burdening either the context sources or the applications.
2Reliability
If QoC parameters are modified to attain requested quality level, then the reliability of context information is improved, but the processing time and complexity increase
Solution Approach 1:
The system changes QoC parameters (first QoC parameter and/or second QoC parameter) to attain the requested quality level. By systematically adjusting these parameters based on the correlation between them, the system can efficiently meet quality requirements without exhaustive processing, reducing the time and computational complexity involved.
3Quantity of substance
If context information from multiple context sources is combined, then the quantity and diversity of information is improved, but ambiguities and inconsistencies arise that reduce the probability of correctness
Solution Approach 1:
The system uses the determined second QoC parameter (probability of correctness) as feedback to decide whether to modify the first QoC parameter or select different context sources. This feedback mechanism allows the system to continuously adjust its information gathering and processing to maintain the requested quality level, ensuring that combining multiple sources does not compromise reliability.
Data Source
Figure 1~3c
Figure 4~7b
Figure 8~9b
AI summary
The invention relates to a context server (2), a context-aware system (1) and a method for providing a context information (C1) to a context-aware application (3). The method comprises the steps of: receiving at least one context information (C1,C2) together with a first QoC parameter (r1,r2), ie. an accuracy measure of the context information, from at least one available context source (CS1,CS2); determining a second QoC parameter (p1,p2), i.e. a probability of correctness of the context information; and the first and the second QoC parameters are correlated. In case that a quality level derived from the second QoC parameter of the context information is lower than a quality level (p*) which is requested by the context-aware application, the first QoC parameter (r1) is modified for adjusting the second QoC parameter in order to attain or exceed the requested quality level (p*); and the context information is provided together with the possibly modified first QoC parameter (x* r1) to the context-aware application.