Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
1© 2015 The MathWorks, Inc.
Developing Signal Processing
Applications for the Internet of Things
Jegan R Mani
Pilot Engineer
2
Agenda
IoT and its components
Developing and deploying algorithms for Edge Nodes
Connecting to Aggregation services
Conclusion
Q&A
3
Internet of Things
Internet
Thing
Thing
Thing
ThingThing
Thing
Thing
?
4
Sensors & Actuators
Consumer
Transportation
Infrastructure
Healthcare
Industrial
Internet of Things
Internet
• Application
Integration
• Data Collection
• Analytics
• Visualization
• Security
!
5
Technologies Enabling IoT
Low-cost low-power sensors
and actuators
– Data validation and clean-up
– Smart control techniques
Ubiquitous connectivity
– Wired / wireless communication
Data aggregation and analysis
– “Big data”
Signal & Image Processing
Control Systems
Communications System
Design
Cloud computing, Statistics
& Machine Learning
6
Components of IoT
7
Types of Edge nodes
Classification based on Node types
– Intelligent Nodes
– Dumb nodes
8
Intelligent Node
Capable of making local decisions
May have node level aggregation
May have local analytics capabilities
9
Developing signal processing algorithms
Access to data
– Android sensors
10
Developing signal processing algorithms
Algorithm developmentExplore and Discover
Data Analysis
& Modeling
Algorithm
Development
Application
Development
11
Developing signal processing algorithms
Deployment of algorithm as an Android App
– Android Support package Reporting and
Documentation
Outputs for Design
Deployment
Share
12
Machine learning for activity detection
13
Types of Edge nodes
Classification based on Node types
– Intelligent Nodes
– Dumb nodes
14
Dumb Node
Nodes act only as sensors
Low frequency of update
No local aggregation
15
Architecture
Weather
StationArduino
Internet
Data
aggregator
(ThingSpeak)
MATLAB
16
Connecting to Data Aggregator
17
Thingspeak connectivity
Write to a thingspeak channel from MATLAB
Read from a thingspeak channel from MATLAB
Write to a thingspeak channel from a embedded
hardware
18
Exploratory Analysis
19
Exploratory Analysis
20
Available Online On File Exchange
21
Conclusion
22
Q & A
23
Thank you !