WiFi Latency Detection Using Sensor Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
WiFi devices often connect to or remain connected with WiFi access points outside of their intended environment due to signal range, leading to inaccurate arrival and departure latency detection, which affects presence and absence determinations in environments like homes or offices.
Innovation Solution
A computer-implemented method and system that uses machine learning to determine arrival and departure latency by analyzing reports from WiFi access points and sensor data, including connect and disconnect times, to adjust presence and absence determinations based on the latency between device connections and actual entry or exit events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If WiFi access points extend signal range to allow connections from outside the environment, then devices can connect before entry and after departure, but this causes inaccurate arrival and departure latency detection
Solution Approach 1:
The patent introduces sensor data (motion sensors, door sensors, cameras) as intermediary indicators to mediate between WiFi connection events and actual physical presence. These sensors detect when a person truly enters or leaves the environment, serving as a reference to correct the latency between connection and physical arrival/departure.
Solution Approach 2:
The system uses sensor data as feedback to continuously adjust and refine latency calculations. By comparing WiFi connection timestamps with sensor-detected presence events, the system learns and adapts the actual arrival and departure latency patterns, improving measurement accuracy over time.
2Reliability
If WiFi devices connect outside the environment due to signal range, then connectivity is maintained, but presence and absence determinations become inaccurate
Solution Approach 1:
Multiple sensor types (motion sensors, door sensors, cameras) serve as intermediaries to verify actual presence status. These sensors provide independent confirmation of whether a person is physically present, allowing the system to distinguish between device connectivity and actual human presence.
Solution Approach 2:
The patent merges WiFi connection data with sensor data from multiple sources to create a more accurate presence determination. By combining these different data streams and analyzing them together, the system overcomes the limitations of using either source alone.
3Ease of operation
If WiFi signal range extends beyond environment boundaries, then devices can connect early, but this creates latency between connection and actual entry
Solution Approach 1:
The system uses sensor feedback to measure and quantify the latency between WiFi connection and actual physical entry. This measured latency is then used to adjust presence determination timing, compensating for the early connection event and reducing the effective time loss.
Solution Approach 2:
The system performs preliminary latency calibration using sensor data before making presence determinations. By pre-learning the typical latency patterns from sensor-correlated connection events, the system can more accurately adjust for early connections without waiting for actual presence confirmation.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Systems and techniques are provided for determining arrival and departure latency for WiFi devices. Reports may be received from WiFi access points in an environment. The reports include an indication of a connection to or disconnection from one of the WiFi access points, a time of the connection or disconnection, and an identifier of the one of the WiFi access points. Data including connect times and disconnect times may be generated from the reports. Sensor and device data may be received from one or more sensors or devices in the environment. Data indicating a length of an arrival latency and a length of a departure latency for the environment may be generated with a machine learning system, where the data including connect times and disconnect times and the sensor and device data is input to the machine learning system.