Handheld User Identification via Embedded Sensor Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current media service and device providers face limitations in providing tailored advertisements and personalized services due to ineffective user identification methods, particularly in uncontrolled lighting and cluttered environments, and privacy concerns related to camera-based solutions.
Innovation Solution
A system utilizing handheld devices with embedded data sensors and algorithms to collect and process real-time data, such as multi-axial accelerometers, thermal, pressure, and capacitive sensors, to distinguish users based on their profiles, without the need for wireless capabilities, ensuring privacy through local data storage and encryption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera-based face recognition is used to identify users, then user identification capability is improved, but privacy concerns increase and the system becomes invasive
Solution Approach 1:
The patent extracts the identification capability from camera-based systems and relocates it to sensor-based systems embedded in handheld devices. The sensors (accelerometers, gyroscopes, microphones) capture user interaction data locally without requiring camera hardware, thereby maintaining identification precision while eliminating privacy concerns associated with visual surveillance
Solution Approach 2:
The patent replaces the optical/mechanical camera-based recognition system with an electronic sensor-based system. Instead of using cameras to capture visual data, the system uses accelerometers, gyroscopes, and microphones to capture mechanical and acoustic data from user interactions with the handheld device, achieving the same identification goal through a different physical domain
2Measurement precision
If camera-based face recognition is used to identify users, then user identification capability is improved, but the system fails in uncontrolled lighting and cluttered environments
Solution Approach 1:
The patent replaces the light-dependent camera system with inertial sensors (accelerometers and gyroscopes) that are insensitive to environmental lighting conditions. These sensors detect physical interactions such as shaking patterns, tilting gestures, and device handling mechanics, enabling reliable user identification in any environmental condition without being affected by lighting or background clutter
Solution Approach 2:
The patent changes the measurement parameters from optical parameters (light reflection, color, shape) to mechanical parameters (acceleration, angular velocity, vibration frequency). This parameter transformation allows the system to operate independently of environmental visual conditions, as mechanical interactions with the device remain consistent regardless of lighting or background
3Productivity
If sensor data is collected and processed to identify users, then personalized content delivery is improved, but data security requirements increase
Solution Approach 1:
The patent implements self-service through local processing of sensor data on the handheld device itself. The device autonomously analyzes interaction patterns and identifies users without transmitting raw data externally, thereby maintaining productivity for personalized content delivery while inherently securing data through localized processing and eliminating the need for external data transmission
Solution Approach 2:
The patent extracts only the essential identification features from sensor data and transmits minimal encrypted information to the server, rather than transmitting all raw sensor data. This extraction approach enables personalized content delivery while reducing data security risks by minimizing the amount of sensitive data that needs to be stored and transmitted externally
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments for automatically distinguishing between users of a handheld device are described. An embodiment includes collecting sensor data from a user interacting with a handheld device, where the sensor data is collected via embedded sensors in the handheld device. The embodiment further includes distinguishing the user from other users of the handheld device via the collected sensor data, at least one embedded machine learning algorithm and a profile for the user. Other embodiments are described and claimed.