Sensor Measurement Pre-classification for Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the number of sensor measurements increases, determining a state associated with a device becomes increasingly complex, negatively impacting device performance and energy efficiency, necessitating a method to reduce processing costs and enhance energy efficiency while maintaining accuracy.
Innovation Solution
The method involves pre-classifying sensor measurements to select relevant feature types for extraction, thereby avoiding the processing of unnecessary features, using a processing apparatus with processors and memory to execute instructions for receiving, pre-classifying, and selecting feature types from sensor measurements, and determining the device's state based on these classifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of sensor measurements increases, then the accuracy of state determination improves, but the processing complexity and energy consumption increase
Solution Approach 1:
The patent segments the feature extraction process by pre-classifying sensor measurements into distinct categories (e.g., motion features, orientation features, position features) and selectively extracting only the relevant features for each classification. This segmentation allows the system to maintain high measurement precision while reducing processing complexity by focusing on specific feature subsets rather than processing all sensor data uniformly.
Solution Approach 2:
The patent applies local quality by tailoring the feature extraction process to the specific characteristics of different sensor measurement types. Different feature extraction algorithms are applied to different measurement categories based on their local properties and relevance to the current state determination task, optimizing both accuracy and processing efficiency for each local data type.
2Measurement precision
If the number of sensor measurements increases, then the accuracy of state determination improves, but the energy efficiency deteriorates
Solution Approach 1:
The patent segments the energy consumption by classifying sensor measurements and extracting features selectively based on their relevance. This segmentation enables the system to process only the necessary features for accurate state determination, reducing unnecessary computational energy consumption while maintaining measurement precision.
Solution Approach 2:
The patent applies partial action by extracting only the necessary features from sensor measurements rather than processing all available data. The pre-classification system identifies which features are essential for the current state determination task, allowing the system to use partial processing that achieves sufficient accuracy without the excessive energy consumption of processing all sensor measurements.
3Loss of information
If all feature types are extracted from sensor measurements, then the completeness of state information improves, but the processing cost increases
Solution Approach 1:
The patent segments feature extraction by organizing sensor measurements into predefined classification categories and extracting only the features relevant to each category. This segmentation ensures that no important state information is lost while significantly reducing processing cost by avoiding the extraction of irrelevant features.
Solution Approach 2:
The patent applies preliminary action through pre-classification of sensor measurements before feature extraction. This preliminary organization of data into meaningful categories allows the system to subsequently extract only the necessary features, ensuring information completeness while minimizing processing cost through targeted extraction rather than comprehensive extraction of all possible features.
Data Source
AI summary
A processing apparatus including one or more processors and memory receives sensor measurements generated by one or more sensors of one or more devices, pre-classifies the sensor measurements as belonging to one of a plurality of pre-classifications, and selects one or more feature types to extract from the sensor measurements based at least in part on the pre-classification of the sensor measurements. The processing apparatus also extracts features of the one or more selected feature types from the sensor measurements and determines a state of a respective device of the one or more devices in accordance with a classification of the sensor measurements determined based on the one or more features extracted from the sensor measurements.