Lecture 5 - IoT Flashcards
What is Internet of Things?
A system involving: connected devices that gather data connect with the Internet, generate analytics, and adapt behaviour based on the analytics
IoT - Layers
- Sensors - Collecting data
- Connectivity - Sending data to cloud
- Data processing - Making data useful
- User interface - Delivering information to user
How do devices gather data?
Through sensors and actuators
Detect the feature quantity of a measurement object and convert this quantity into a readable signal
Sensors on smart phones
Accelerometer, gyroscope, magnetometer, GPS, light sensor, barometer, thermometer, microphone, pedometer
Communication protocols
Options for connectivity are various
Cellular, satellite, WiFi, Bluetooth, RFID, NFC, LPWAN, zigbee
Four models for connectivity
- Device to Device
- Device to Cloud
- Device to Gateway
- Backend Data sharing
Connectivity: Device to Device
Direct communication with each other
- Via IP network, hardwire or bluetooth
- E.g. Smart watch and pacemaker
Lower power consumption
- Ideal for products to have al one battery life
Connectivity: Device to Cloud
Via ethernet, WiFi or cellular
- E.g.: Webcams to watch home while on vacation
- Tag on an animal to find where it is
Difficulty for inter product compatibility
- Due to the differences in manufacturer design
Connectivity: Device to Gateway
Intermediary between IoT device and a cloud service
- Fitness device connects to the cloud through Nike+ app
- Home appliance connects to a hub like Samsung SmartThings
Connectivity: Backend Data Sharing
Extends single device to cloud model
- Sensor data can be accessed by authorised third parties
Export, analyse smart object data fro ma cloud service
- Combine with data from other sources
- Map My Fitness: Data from Fitbit, Adidas, miCoach, etc
Trade off Communication Protocols
Tradeoff between:
- Power consumption
- Range
- Bandwith
IoT Network Topology
- Point-to-point
- Star
- Mesh
- Hybrid
IoT Architecture - Data Processing
Once the sensor data gets to the cloud: software performs processing on data
Numerous algorithms and data processing elements: ultimately become information
Consideration: storage, frequency of updates, desired output type
Edge Computing
• The round-trip can take too long
- Sending data, processing, analysing, returning instructions
• Edge computing:
- A smart edge device
- Aggregate data, analyse it and fashion responses if necessary
- All within relatively close physical distance
- Reducing delay
•Edge devices also have upstream connectivity for sending data to be further processed and stored
IoT Arhictecture - UI ways to interact
- Alert (email, text)
- Automatic notifications
- Monitoring information proactively
- Controlling system remotely