iLoc Framework for Mobile Location Context Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining location context on mobile devices are computationally intensive, power-consuming, and inefficient in scaling to large user bases, making it challenging to continuously monitor and predict user location states effectively.
Innovation Solution
The Incremental Location-State Acquisition and Prediction (iLoc) framework continuously monitors signal environments to learn and predict location states, using a tightly integrated learning and prediction module that updates location states incrementally and predicts future states, reducing the need for precise GPS coordinates and minimizing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS receivers are used to continuously monitor device location, then location accuracy is improved, but power consumption increases
Solution Approach 1:
The patent combines multiple location determination techniques (GPS, WiFi, cellular signals) into a unified location context determination system. The server integrates data from various sources to derive location context, allowing the mobile device to use lower-power methods while maintaining accuracy through server-side processing.
Solution Approach 2:
The patent introduces a server as an intermediary between the mobile device and the location determination process. The server receives location data from the device and performs the computationally intensive context derivation remotely, reducing the power consumption and processing burden on the mobile device itself.
2Quantity of substance
If GPS traces are continuously stored, then location history is improved, but storage cost increases
Solution Approach 1:
The patent extracts only the essential location context information (e.g., user is at work, at home, traveling) from the raw GPS trace data. Instead of storing continuous high-precision location coordinates, the system stores derived context labels and aggregated stationary state information, significantly reducing storage requirements while preserving the essential location history.
Solution Approach 2:
The patent transforms location data from continuous coordinate pairs to discrete context categories (stationary states). By changing the parameter representation from precise GPS coordinates to contextual labels, the system reduces storage costs while maintaining the ability to derive meaningful location history.
3Measurement precision
If similar GPS traces are analyzed to derive location context, then context accuracy is improved, but computational intensity increases
Solution Approach 1:
The patent performs preliminary processing of location data by identifying stationary states and deriving context information in advance. The system pre-processes GPS traces to detect when the device remains stationary and derives context labels during these periods, reducing the computational burden during real-time operation.
Solution Approach 2:
The patent applies partial action by deriving location context only during stationary periods rather than continuously. The system determines that context derivation is most valuable when the device is stationary, and uses incremental updates during movement, reducing overall computational intensity while maintaining context accuracy.
4Adaptability or versatility
If location context is derived for thousands of users, then service coverage is improved, but scalability deteriorates
Solution Approach 1:
The patent uses a server as an intermediary to handle location context derivation for multiple users. The server centralizes the computationally intensive context derivation process, allowing multiple mobile devices to send their location data to the server for processing, thereby improving scalability and service coverage across thousands of users.
Solution Approach 2:
The patent creates a universal location context determination system that serves multiple users through a single server infrastructure. The server implements general algorithms for deriving location context that can be applied to any user's location data, enabling the system to scale to thousands of users without requiring dedicated processing resources for each user.
Data Source
Figure 1A~1B
Figure 2A~2B
Figure 2C~2D
AI summary
Techniques to derive or predict location context for a user of a mobile device include receiving signal data that indicates a set of one or more distinct signal sources from which signals are received at the mobile device for each of multiple different times. The method further comprises determining whether the mobile device is moving outside a specified area at a current time based on the signal data. The method further comprises incrementing a count for a stationary state associated with the set of one or more distinct signal sources at the current time, if the mobile device is determined to be not moving outside the specified area. The method also comprises delivering a service to the mobile device based on the stationary state.