Unregistered Device Identification via Wireless Behavior Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current enterprise management tools face challenges in accurately detecting and distinguishing on-site individuals for occupancy monitoring in office spaces, especially in bring-your-own-device (BYOD) environments, leading to high costs and inefficiencies in managing vast office spaces.
Innovation Solution
The method leverages existing wireless infrastructure and applied machine learning to identify unregistered devices by analyzing their wireless behavior, determining their category (e.g., visitor vs. employee), and automatically enrolling them into an enterprise directory, optimizing office space usage and reducing double counting through primary device determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If existing wireless infrastructure is used for device detection, then cost is reduced and ease of operation is improved, but measurement precision for identifying unregistered devices deteriorates
Solution Approach 1:
The patent introduces wireless behavior as an intermediary characteristic to bridge the gap between simple wireless infrastructure detection and accurate device identification. By analyzing behavioral patterns (connection timing, signal strength variations, interaction sequences) rather than relying solely on traditional device identifiers, the system achieves precise identification of unregistered devices using existing wireless networks, resolving the contradiction between ease of operation and measurement precision
2Device complexity
If traditional occupancy monitoring tools are used, then device identification is simplified, but productivity and accuracy of occupancy monitoring deteriorate
Solution Approach 1:
The system enables unregistered devices to self-identify and self-classify by analyzing their inherent wireless behavior patterns. The machine learning model automatically determines device categories (employee, visitor, contractor) based on observed behaviors such as connection duration, mobility patterns, and interaction with registered devices, eliminating the need for complex manual registration processes while improving occupancy monitoring accuracy
Solution Approach 2:
The patent implements continuous feedback loops where wireless behavior data is collected, analyzed by machine learning models, and used to update device classifications in real-time. This feedback mechanism allows the system to adapt to changing occupancy patterns and improve monitoring accuracy dynamically, resolving the contradiction between system complexity and monitoring productivity
3Device complexity
If unregistered devices are not properly identified, then management processes remain simple, but loss of information regarding occupancy and space usage increases
Solution Approach 1:
The patent replaces manual device registration and classification processes with automated machine learning analysis of wireless behavior. The system substitutes mechanical/administrative processes (manual device onboarding, manual occupancy tracking) with automated computational processes that analyze behavioral patterns to identify device categories and occupants, thereby eliminating information loss while maintaining simple management processes
Data Source
AI summary
One or more computer processors detect an unregistered device associated with an identified location, wherein the unregistered device is associated with wireless behavior. The one or more computer processors identify one or more registered devices in a proximity to the identified location and the detected unregistered device. The one or more computer processors identify an occupant associated with the detected unregistered device utilizing a trained device identification model, the identified location, and respective wireless behavior associated with the detected unregistered device and the identified one or more registered devices.