Colocation Context Determination Using Sensor Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the geographic location and usage context of mobile computing devices are limited in accuracy and efficiency, particularly when devices are colocated or have complementary usage contexts, as they rely on incomplete or ambiguous sensor data.
Innovation Solution
A system and method that involve receiving sensor data from multiple devices to determine colocation and complementary usage contexts, using a remote server to process data from GPS, wireless access points, accelerometers, and other sensors to estimate locations and contexts, and refining profiles based on feedback for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data from multiple devices is processed to determine colocation and complementary contexts, then location and context determination accuracy is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system segments the determination process into distinct modules: colocation determination module that analyzes sensor data to identify devices in proximity, and context determination module that infers usage contexts. This segmentation allows independent optimization of each module and reduces overall system complexity by breaking down the complex multi-device analysis into manageable components.
Solution Approach 2:
The patent introduces intermediary data structures including device profiles that store context information, colocation indicators that represent spatial relationships, and context models that encode complementary context relationships. These intermediaries simplify the processing by pre-organizing sensor data and relationship information, reducing the computational burden during actual determination operations.
2Measurement precision
If colocation determination uses sensor data from multiple devices, then determination accuracy is improved, but data processing time and energy consumption increase
Solution Approach 1:
The system performs preliminary actions by continuously collecting and pre-processing sensor data from multiple devices, maintaining up-to-date device profiles with context information and location data. Device colocation relationships are pre-calculated and stored as colocation indicators. This preliminary preparation enables rapid determination operations without requiring intensive real-time processing of raw sensor data.
Solution Approach 2:
The patent replaces complex mechanical sensor data processing with information-based substitution. Instead of directly analyzing raw sensor streams from multiple devices during determination, the system substitutes this with pre-computed colocation indicators and context profiles, significantly reducing processing time while maintaining accuracy.
3Measurement precision
If device profiles are refined based on feedback for improved accuracy, then context determination precision is improved, but information processing and storage requirements increase
Solution Approach 1:
The system refines device profiles by dynamically changing parameters based on feedback from determination operations. Context information in device profiles is updated to reflect actual usage patterns, and colocation thresholds are adjusted based on observed device relationships. This parameter refinement improves precision while maintaining storage efficiency by updating only necessary profile attributes rather than storing complete historical datasets.
Solution Approach 2:
The patent implements selective data retention by discarding redundant or outdated sensor data while recovering and preserving only the essential context information needed for accurate determination. Device profiles maintain only the most relevant context attributes, and historical data is discarded unless it provides unique information for context refinement, optimizing the balance between precision and storage requirements.
Data Source
AI summary
Methods and systems for grouping computing devices together based on the devices being colocated with one another or being associated with complementary usage contexts, and then using the location or usage context of one device in the group to estimate the location or usage context of another device in the group are described. An example method may include receiving first sensor data from sensors of a first computing device; receiving second sensor data from sensors of a second computing device; determining, based on the received sensor data, that the first and second computing devices are colocated with one another; identifying, based on the first sensor data, a context associated with the first computing device; and determining, based at least in part on the context associated with the first computing device, a context associated with the second computing device.