Passive Sensor Event Tagging for Complete Facility Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Security information management systems struggle to provide complete situational awareness due to the inability of passive sensors to identify individuals or objects causing untagged events, leading to incomplete tracking of location and movement within facilities.
Innovation Solution
A system that uses machine learning to train a model on tagged and untagged sensor data, predicting the identity of individuals or objects causing untagged events by correlating sensor layouts and event timings, and applying the model to tag untagged events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If passive sensors are used to monitor facility conditions, then energy consumption is reduced and sensor deployment is simplified, but the ability to identify individuals or objects triggering events is lost
Solution Approach 1:
The patent introduces an event prediction model as an intermediary component that processes data from passive sensors. This model uses causal modeling to infer the identity of individuals or objects based on sensor event patterns, thereby recovering the lost identification capability without requiring active identification sensors throughout the facility.
Solution Approach 2:
The patent replaces the need for complex active identification sensor systems with a data-driven machine learning approach. Instead of using sensors that actively identify individuals through recognition technology, the system uses passive sensors combined with causal modeling to infer identity from event patterns, substituting mechanical/recognition-based identification with information-processing-based inference.
2Loss of information
If active identification sensors are deployed throughout the facility, then individual and object identification capability is improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts the identification function from the sensor hardware itself and relocates it to a software-based event prediction model. By separating the sensing function (performed by simple passive sensors) from the identification function (performed by the machine learning model), the system achieves identification capability without the complexity of active identification sensors.
Solution Approach 2:
The patent substitutes complex active identification sensor systems with a simplified passive sensor system combined with machine learning. Instead of relying on sophisticated hardware to perform identification, the system uses information processing and causal modeling to achieve the same functional goal with much simpler hardware requirements.
3Measurement precision
If machine learning models are trained on historical sensor data, then prediction accuracy for untagged events is improved, but processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by training the event prediction model offline using historical sensor data before it needs to make real-time predictions. The causal modeling component processes and learns from past event patterns in advance, so that when new sensor data arrives, the model can quickly generate predictions without requiring computationally intensive real-time training, thus reducing processing time while maintaining accuracy.
Data Source
AI summary
A tagging system gathers all events (tagged and untagged) generated by remote sensors at a location or facility over time. Based on the gathered events the tagging system uses machine learning to train a model to learn the sensor layout of a facility or location and the timing between the triggering of sensors. Once trained, the model can predict the movement and location of individuals and objects throughout the facility based on a starting tagged event. Given a series of tagged and untagged events, the system can use the movement predictions of the model to tag the untagged events in the series with the identification of an individual or object that triggered the generation of the untagged event.