Rob Purser
Development Manager for IoT and Hardware Interfacing for MATLAB at MathWorks. Leads the creation of software that connects MATLAB with the real world
Live Version at http://slides.com/rpurser/integrating_analytics#/
Most IoT data are not used currently. For example, only 1 percent of data from an oil rig with 30,000 sensors is examined.
-- McKinsey & Company
MATLAB: High-level data analytic language and interactive environment for data analysis, visualization, and algorithm development
Simulink: Block diagram environment for simulation and design. Simulate, generate code, and verify embedded systems
ThingSpeak: Analytic IoT Platform for storing sensor data and developing IoT applications
Tide Prediction - Online Analytics
Nighttime Noise - Historical Analytics
Weather - IoT Analytic Workflow
Counting Cars - Edge Node Analytics
Tide Prediction - Online Analytics
Nighttime Noise - Historical Analytics
Weather - IoT Analytic Workflow
Counting Cars - Edge Node Analytics
MathWorks Weather Station -- Revisited
Classic IoT Maker Project with Machine Learning
Deep Dive: Google "mathworks weather revisit"
1
2
3
Tide Prediction - Online Analytics
Nighttime Noise - Historical Analytics
Weather - IoT Analytic Workflow
Counting Cars - Edge Node Analytics
Counting Cars
Traffic monitor using a Raspberry Pi & Webcam
Deep Dive: Google "count cars thingspeak"
Counting Cars
Traffic monitor using a Raspberry Pi & Webcam
Deep Dive: Google "count cars thingspeak"
Embedded devices have always used:
Powerful mobile processors enable:
Foreground Detection, 2-D Median Filter, Blob Analysis
Feeds custom counting block and Transmit to ThingSpeak
Foreground Detection, 2-D Median Filter, Blob Analysis
Feeds custom counting block and Transmit to ThingSpeak
Foreground Detection, 2-D Median Filter, Blob Analysis
Feeds custom counting block and Transmit to ThingSpeak
Foreground Detection, 2-D Median Filter, Blob Analysis
Feeds custom counting block and Transmit to ThingSpeak
Foreground Detection, 2-D Median Filter, Blob Analysis
Feeds custom counting block and Transmit to ThingSpeak
Embedded devices now have enough compute power to use algorithms that previously were once only practical on desktop class machines or FPGAs
Tide Prediction - Online Analytics
Nighttime Noise - Historical Analytics
Weather - IoT Analytic Workflow
Counting Cars - Edge Node Analytics
Night Noise Analysis
Identify unusual changes in ambient noise level at night
Median Filter
Overlaid with Expected Levels
Determined expected based on median at a given time
Normalized based on first 3 hours
Polynomial fit of expected noise levels
Set Expected Levels for a given night
Move polynomial up or down to compensate for variation
Identify alarm conditions
> 1 std deviation above expected
To do: Investigate what happened this night...
Things to keep in mind:
Tide Prediction - Online Analytics
Nighttime Noise - Historical Analytics
Weather - IoT Analytic Workflow
Counting Cars - Edge Node Analytics
On Line Tide Alerts
Tide Measurement and Prediction
Deep Dive: Google "tide matlab thingspeak"
Source: SUNY StoneyBrook
23 astronomical components in tidal harmonics...
plus geography and weather -- it's always an approximation!
Tidal variation between bays 6 miles apart
Edge Node
On-Line Analysis
On-Line Analysis
http://slides.com/rpurser/environmental_analysis
By Rob Purser
In this presentation, we will discuss three kinds of analytics common in IoT systems: sensor analytics at the edge node for data reduction, cloud-based online analytics for situational awareness, and historical data analytics for developing predictive and classification algorithms. You’ll see practical techniques for integrating all these types of analytics into your IoT system. We will demonstrate these techniques using various environmental data sources, including data from weather stations, noise monitors, tide gauges, and traffic sensors. These examples utilize embedded programming, image and signal processing, and machine learning techniques to demonstrate how analytics can be integrated at all phases in your IoT system.
Development Manager for IoT and Hardware Interfacing for MATLAB at MathWorks. Leads the creation of software that connects MATLAB with the real world