Wireless Sensing User Identification Using CSI and Lifelog Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless sensing-based user identification technologies face challenges in achieving accurate user recognition in varying environments due to the uniqueness of signal patterns influenced by user movement, making it difficult to generate a general model for commercialization and practical application.
Innovation Solution
The proposed method enhances user identification accuracy by utilizing lifelogs, which include environmental and personal information, in conjunction with Channel State Information (CSI) data to improve learning and prediction performance through reinforcement learning, allowing for the creation of AI devices that can recognize and identify individuals in smart home settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If wireless sensing is used for user identification based on signal patterns, then non-intrusive detection is achieved, but identification accuracy is insufficient in varying environments
Solution Approach 1:
The patent combines wireless signal-based CSI data with lifelog data (environmental information, temporal patterns, contextual metadata) to create a hybrid sensing system. This merging allows the system to leverage both the non-intrusive signal pattern detection and the contextual enrichment from lifelogs, thereby improving identification accuracy while maintaining environmental adaptability without requiring physical contact or intrusive sensors.
Solution Approach 2:
The system creates a universal user identification approach that works across multiple environments by integrating lifelog data with wireless sensing. The lifelog component provides contextual information that enables the same wireless sensing system to adapt to different settings (home, office, public spaces), making the system universally applicable rather than environment-specific.
2Measurement precision
If only signal pattern analysis is used, then system simplicity is maintained, but user recognition accuracy is insufficient
Solution Approach 1:
The patent segments the sensing system into two independent but complementary components: wireless sensing for signal pattern extraction and lifelog for contextual information. This segmentation allows each component to perform its specific function optimally while the integration layer combines them to achieve high-accuracy user recognition without overwhelming complexity in any single component.
Solution Approach 2:
The patent introduces an intermediary processing layer that integrates CSI data with lifelog data. This intermediary layer acts as a mediator between the simple wireless sensing component and the complex integration requirements, enabling accurate user recognition by harmoniously combining signal patterns with contextual information from multiple sources.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables sufficient accuracy in user recognition and identification within home environments, creating a new paradigm for IoT smart home devices by combining CSI and lifelog data to overcome the limitations of relying solely on signal patterns, thus improving commercial viability and application accuracy.
Implementation Method 1
Human movement affects wireless signal propagation (e.g., reflection, diffraction, and scattering), providing an excellent opportunity to capture human movement by analyzing the received wireless signal
Implementation Method 2
Because of its frequency-shifting, low-cost, and non-intrusive detection properties, whether researchers extract ready-to-use signal measurements or adopt frequency-modulated signals, wireless-based human activity detection has attracted considerable interest
Data Source
AI summary
Proposed are a method and device for performing wireless sensing in a wireless LAN system based on wireless sensing. Specifically, a wireless apparatus collects CSI data and lifelog. The wireless apparatus performs learning and prediction on the basis of the CSI data to acquire a first user estimation result. When the first user estimation result is at most a threshold value, the wireless apparatus predicts a second user estimation result on the basis of the lifelog.