Model Based Event Detection in Sensor Networks Jayant

Model Based Event Detection in Sensor Networks Jayant Gupchup, Andreas Terzis, Randal Burns, Alex Szalay The Johns Hopkins University

Outline • Motivation • Data & Model • Experiments and Results • Discussion The Johns Hopkins University

Motivation The Johns Hopkins University

Importance of detecting events - Fixed Sampling: High Freq => too much data Low Freq => miss temporal transients - Detect Events: Adaptive Sampling (increase % of usable data) - Conserve Energy “Event starts” Detect Event Increase Sampling Frequency/Trigger Alarms “Event ends” - Alarm Triggers - Correlate events and observed phenomena in large databases The Johns Hopkins University Return to steady behavior

Sample Event Rain Event Non-Event Days The Johns Hopkins University

Solution: Rough Sketch - Model observed quantities using Principal Component Analysis (PCA). - Project original data on a “feature space” (reduce dimensionality) - Look for observations “deviating” from Average/Expected behavior in the feature space The Johns Hopkins University

Principal Component Analysis (PCA) Variable #2 PCA : - Finds axes of maximum variance X : Points original space O : Projection on PC 1 - Reduces original dimensionality (In e. g. from 2 variables => 1 variable) First Principal Component Variable #1 The Johns Hopkins University

Motivation for Using PCA Typical day: “Fits model well” Event day: “Large residuals” The Johns Hopkins University

Why Not Soil Moisture ? Reaction to event The Johns Hopkins University

Life. Under. Your. Feet Data & Model Preparation The Johns Hopkins University

Life. Under. Your. Feet data • 10 MICAz Sensors – – Air Temperature (AT) Soil Temperature (ST) Soil Moisture Photo Sensor The Johns Hopkins University

Air Temp vs. Soil Temp Notice the phase lag for Soil Temperature The Johns Hopkins University

Data Preparation • Model built on Air temperature and Soil Temperature. t=10 1 day, 10 sensors t=20 … t=1440 AT 1_1 AT 1_2 …. … …. AT 1_144 AT 2_1 AT 2_2 …. … …. AT 2_144 . . …. … …. . AT 10_1 AT 10_2 …. … …. AT 10_144 . . …. … …. . Size of matrix : [(# of days x 10) X 144] The Johns Hopkins University

The Johns Hopkins University

PCA Bases (AT & ST) Eigenvector 1 Is the Diurnal cycle similarity eigenvector 1 for ST & eigenvector 2 for AT The Johns Hopkins University

Methods and Results The Johns Hopkins University

Methods Three methods 1) Basic Method – Projections on the first principal component for AT 2) Highpass Method – Removes seasonal drift by looking at sharp changes in the local neighborhood. 3) Delta method – Makes use of the inertia of the soil and seasonal drift The Johns Hopkins University

Test Data • Test Period : 225 days between September, 2005 – July, 2006 • 48 major events were known to occur (taken from the BWI weather station, http: //www. wunderground. com/US/MD/Bwi_Airport. html) • Offline Analysis The Johns Hopkins University

Method 1 : Basic Method • • Considers only Air Temperature. First Basis Vector (PC 1) First Basis Vector covers 55% of variation in the data 1 day AT 1_1 AT 1_2 …. … …. AT 1_144 AT 2_1 AT 2_2 …. … …. AT 2_144 . . … …. . AT 10_1 AT 10_2 …. … …. AT 10_144 e 1_1 V 1_1 X = V 1_2 e 2_1 . . V 1_144 e 10_1 Average Day 1 E 1 Day n Day 2 E 2 … …. . ……………. . The Johns Hopkins University En-1 En

Method 1: Basic Method (cont. ) Results : Method Precision Recall False Negatives Basic 52. 459% 64% 18 Drawback: - Does not consider seasonal drift - Does not make use of the inertia information of the soil. The Johns Hopkins University

Method 2 : Highpass Method • Again, Considers only Air Temperature • Highpass filter on ‘E 1’ series. Call this series ‘S 1’ • Highpass filters detects sharp changes by considering the local neighborhood only => Removing seasonal drift • Threshold on ‘S 1’, values below the threshold are tagged as events. The Johns Hopkins University

Method 2: Highpass Method (cont. ) Results : Method Precision Recall False Negatives Basic Highpass 52. 459% 51. 28% 64% 80% 18 10 Drawback: - Does not make use of the inertia information of the soil. The Johns Hopkins University

Method 3 : Delta Method • Considers Air Temperature and Soil Temperature • Create E 1 series for AT and E 1 series for ST separately as discussed before • Highpass filter on AT_E 1 & ST_E 1 => AT_S 1 & ST_S 1 • Delta = AT_S 1 – ST_S 1 for all days. • Set a threshold on the Delta series. The Johns Hopkins University

Method 3: Delta Method (cont. ) Results : Method Precision Recall False Negatives Basic Highpass Delta 52. 459% 51. 28% 54. 79% 64% 80% 85. 106% 18 10 7 The Johns Hopkins University

Event detection for 12/13/2005 – 01/02/2006 Due to the inertia of the soil, ‘Delta method’ shows sharper negative peaks for event days. The Johns Hopkins University

Discussion The Johns Hopkins University

Future work • Implement “Online event detection” – Compute Basis vectors from historic data. – Load the ‘basis vectors’ and ‘threshold’ values on the motes. • Apply technique for faulty sensor detection • Detect localized events by forming clusters of motes with similar eigencoefficients. • Consider variants of PCA (Gappy-PCA, online-PCA). The Johns Hopkins University

Acknowledgements • Ching-Wa Yip 1 - PCA C# library and Discussions. • Katalin Szlavecz 2 & Razvan Musaloui-E 3 – Domain expertise and data collection. • Jim Gray 4 & Stuart Ozer 4 – Online database 1 : JHU, Dept of Physics & Astronomy 2 : JHU, Dept of Earth and Planetary science 3 : JHU, Dept of Computer Science. 4 : Microsoft Research The Johns Hopkins University

Future work • Online event detection on the motes • Apply this method for faulty sensor detection • Detect localized events by forming clusters of motes with similar eigencoefficients. • Consider incomplete days using Gappy-PCA. • Explore incremental & robust PCA techniques. The Johns Hopkins University

Training Set (Air Temp) • Seasons exhibit “Diurnal Cycles” around their daily mean (DC component) • Construct Zero-Mean Vectors for each Sensori for each day (remove DC Component) • Remove outliers using a simple median filter to build the training set X The Johns Hopkins University