Sensor Network Anomaly Detection Using SAX Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the IoT sensor network, anomaly detection is challenging due to subtle changes in sensor data patterns over time, leading to false positives and false negatives, as existing pattern recognition methods require exact matches and are not tolerant to deviations in wavelength and frequency.
Innovation Solution
The proposed solution involves using the Symbolic Aggregate approximation (SAX) algorithm to generate and compare patterns, allowing for tolerance by calculating a Hamming distance and average distance score between SAX words, enabling detection of anomalies even with slight deviations from known patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If exact pattern matching is used for anomaly detection, then measurement precision is improved, but reliability deteriorates due to false positives and false negatives from subtle pattern changes
Solution Approach 1:
The patent transforms the time series data into SAX (Symbolic Aggregate approXimation) words by changing the parameter representation from raw numerical values to discrete symbolic sequences. This parameter transformation enables the system to capture pattern similarities while being tolerant to subtle variations in wavelength and frequency, thus resolving the contradiction between precise anomaly detection and reliability against false positives/negatives
Solution Approach 2:
The patent introduces Hamming distance as an intermediary metric to compare SAX words. Instead of directly comparing raw sensor data patterns, the system uses SAX words as intermediaries and measures their similarity through Hamming distance. This intermediary approach allows the system to detect anomalies while accommodating natural variations in sensor data patterns over time
2Device complexity
If traditional pattern recognition is used, then device complexity is reduced, but adaptability deteriorates due to inability to handle wavelength and frequency variations
Solution Approach 1:
The patent applies parameter changes by converting time series data into SAX words, which represent the data in a discrete symbolic format. This transformation maintains relatively simple system architecture while dramatically improving adaptability to pattern variations in wavelength and frequency, as SAX words naturally capture the essential pattern characteristics without being sensitive to exact timing or value variations
Data Source
AI summary
Embodiments match sensor data output by a sensor to a trained pattern. Embodiments form a plurality of windows of an identified pattern from the sensor data, each of the plurality of windows having a substantially equal window length to a length of the trained pattern. For each of the windows, embodiments generate a corresponding first Symbolic Aggregate approximation (“SAX”) word, determine a Hamming distance between the first SAX word and a second SAX word corresponding to the trained pattern, and determine a final distance score based on coefficients between the first SAX word and the second SAX word. For each of the windows, embodiments determine a number of positions in the first SAX word that do not contribute to the final distance score, update the Hamming distance after eliminating the number of positions and determine an average distance based on the final distance score and the updated Hamming distance.