Parametric Density Map Storage for Time-Varying Device Activity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional electronic maps fail to efficiently capture periodic patterns of device density over time, such as user device activity in specific geographic locations, which limits their effectiveness in representing time-varying raster densities.
Innovation Solution
An online system that processes telemetry records from user devices, determines raster densities within unit areas of an electronic map, and applies regression techniques to establish a polynomial relationship between time and raster density, generating a temporal function for each unit area to create a density map that illustrates how raster densities change over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional electronic maps store static map data without time-varying information, then the map structure remains simple and easy to manage, but the ability to represent device density patterns over time is lost
Solution Approach 1:
The patent segments the electronic map into discrete unit areas (grid cells) and further segments time into time windows. Each unit area has its own density function with coefficients stored separately. This segmentation allows time-varying information to be organized in a structured, manageable way without creating a monolithic complex data structure.
Solution Approach 2:
The patent represents time-varying device density using parametric functions (polynomials, sinusoids, etc.) where coefficients are stored instead of raw time-series data. This parameterization compresses the time-varying information into compact numerical parameters, enabling efficient storage and retrieval while preserving the ability to reconstruct density patterns at any time point.
2Loss of information
If detailed telemetry records are stored for every time point, then complete device activity information is preserved, but storage requirements and data processing complexity increase significantly
Solution Approach 1:
Instead of storing complete telemetry records for every time point, the patent fits the time-series density data to parametric functions and stores only the function coefficients. This transformation reduces the quantity of stored data from potentially millions of individual telemetry points to just a few parameters per unit area, while preserving the essential device activity patterns through the mathematical functions.
Solution Approach 2:
The patent creates a simplified functional representation (a copy in mathematical form) of the complex telemetry data. Rather than storing the original detailed records, it stores a compact parametric model that can generate density values at any time point, effectively copying the essential information in a compressed format.
Data Source
AI summary
Disclosed is an online system that generates a density map of a particular location based records received from one or more reporting devices. To this end, the online system determines a relationship between time and a raster density for each unit area in a density map. The determined relationship between time and a raster density may be used to store, and transmit records as parameters of a parametrized function.