On-Device Context Subsampling for Battery Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile devices inefficiently sample and upload location information, leading to redundant battery usage and suboptimal data quality due to simplistic reporting heuristics.
Innovation Solution
A machine learning-based system optimizes information sampling and uploading by generating policies that adjust scanning and upload rates based on device conditions, such as activities detected by mobile devices, thereby reducing battery consumption while maintaining data quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If mobile devices periodically sample and upload location information using simple heuristics, then data can be collected for navigation and statistics applications, but battery power is wasted due to redundant sampling and reporting frequency is suboptimal
Solution Approach 1:
The patent implements dynamic adjustment of sampling and reporting frequencies based on real-time device state analysis. The system transitions from static periodic sampling to adaptive sampling where the reporting interval dynamically changes according to detected activities, location changes, and contextual factors, thereby optimizing battery usage while maintaining data quality
Solution Approach 2:
The system changes key parameters including sampling frequency, reporting interval, and data collection intensity based on detected conditions. When the device is stationary or in low-activity states, the sampling frequency is reduced to save battery power, while during active movement or significant location changes, the system increases sampling rates to maintain data quality
2Measurement precision
If mobile devices increase reporting frequency to improve data quality, then more accurate navigation and statistics can be achieved, but battery consumption increases
Solution Approach 1:
The patent applies different sampling and reporting strategies to different contexts and locations. Instead of uniform high-frequency sampling, the system identifies specific locations and activity contexts where high-precision data is critical (such as navigation turns or significant location changes) and applies enhanced sampling only in those specific cases, while using reduced sampling in other contexts
Solution Approach 2:
The system dynamically adjusts measurement precision and sampling intensity based on real-time conditions. The reporting frequency adapts to the current state of the device, increasing precision only when necessary for accurate navigation or when significant location changes are detected, thereby avoiding unnecessary battery consumption during periods when high precision is not required
3Ease of manufacture
If mobile devices use simple heuristic-based reporting schedules, then implementation is straightforward, but reporting efficiency is suboptimal and battery is wasted on redundant sampling
Solution Approach 1:
The patent implements a self-adaptive system that automatically analyzes device state, detects activities, and determines optimal reporting schedules without requiring manual configuration or complex external control. The system services itself by making intelligent decisions about when and how to sample and report data based on its own operational context, thereby improving efficiency while maintaining implementation simplicity
Solution Approach 2:
The system incorporates feedback loops where reporting data and device state information are continuously analyzed to adjust future sampling and reporting decisions. This feedback mechanism enables the system to learn from past performance and optimize reporting efficiency over time, improving productivity while keeping the implementation approachable through iterative refinement
Data Source
AI summary
The present disclosure provides a system for intelligently sampling information, such as location, activities, etc. on device. Sampling and uploading of background context is optimized using machine learning, such that battery usage is reduced, and quality of metrics based on the reported information is maintained or improved. A policy is generated based on the machine learning, the policy dictating how scanning and upload rates should change in response to conditions on the device.