Apparatus and method for event classification based on barometric pressure sensor data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for detecting intruder events in indoor environments using barometric pressure sensors struggle to distinguish between different types of events, such as window opening or breaking, and are prone to false alarms due to their simplicity and unspecific signal patterns.
Innovation Solution
An apparatus and method that employ a two-stage approach combining a predictor and a classification block, utilizing a neural network to generate a predicted signal portion and an error signal determiner to classify events by combining the predicted and measured pressure signal portions, incorporating environmental factors like weather conditions, and employing machine learning algorithms like Linear Discriminant Analysis for accurate event classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a simple threshold-based detection approach is used, then the system is easy to implement and low cost, but it cannot distinguish between different types of events and generates high false alarm rates
Solution Approach 1:
The patent segments the pressure signal into multiple portions (first signal portion and second signal portion) and processes each portion separately through different computational steps. The signal is divided into time segments that are analyzed independently to extract specific features, enabling differentiation between event types while maintaining a structured approach that balances complexity and effectiveness
Solution Approach 2:
The patent transforms the one-dimensional pressure signal into a multi-dimensional feature space by extracting multiple characteristics (mean value, standard deviation, maximum value, minimum value, and slope) from different signal portions. This dimensional transformation allows the classifier to distinguish between different event types that would be indistinguishable in the original signal domain
2Reliability
If a Hidden Markov Model is applied to classify events, then event differentiation capability is improved, but the system complexity increases significantly due to state identification and transition probability requirements
Solution Approach 1:
The patent extracts only the essential features needed for classification from the pressure signal, rather than implementing a full Hidden Markov Model. By taking out and analyzing specific signal characteristics (statistical moments and slope) from segmented portions, the system achieves event differentiation with significantly reduced computational complexity and easier implementation
Solution Approach 2:
The patent replaces the complex, resource-intensive Hidden Markov Model with a simpler, more computationally efficient classifier that uses basic statistical calculations. This substitution maintains adequate classification accuracy while dramatically reducing the computational burden, making the system more suitable for resource-constrained environments
Data Source
AI summary
An apparatus for event detection is provided. The apparatus comprises an error signal determiner for determining an error signal portion depending on a pressure signal. The error signal determiner determines a predicted signal portion depending on a first signal portion of the pressure signal. The error signal determiner also determines the error signal portion depending on the predicted signal portion and depending on a second signal portion of the pressure signal, wherein the second signal portion of the pressure signal succeeds the first signal portion of the pressure signal in time. The apparatus also comprises a classifier for determining, depending on the error signal portion, whether an event of a group of one or more events has occurred or whether no event of the group has occurred.