L13 - IoT Flashcards
IoT definition
system of interrelated computing devices, mechanical and digital machines, objects, animals or people that are provided with unique identifiers (UIDs) and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction.
starting point of IoT
when the number of devices connected to the internet exceeded the number of people on earth. (~2008)
actuator
An actuator is a device that produces a motion by converting energy and signals going into the system.
3 IoT Architectures
- IoT Cloud Architecture
- IoT Edge Architecture
- IoT Fog Architecture
IoT Cloud Architecture
IoT devices (microcontrollers that can communicate to an actuator) are connected via a gateway to the Wide Area Network (WAN) that connects to the cloud.
- connection to gateway can be wired or wireless
- ## sensor data is forwarded to the IoT platform in the cloud
WAN
A wide area network (WAN) is a telecommunications network that extends over a large geographic area.
Which data storage is used in the IoT Cloud Architecture?
NoSQL data storage with eventual consistency.
Iot Edge Architecture
- you have an edge computer connected to the gateway
- less latency compared to IoT Cloud Architecture
- bandwidth requirements are reduced because the edge computer which is near to the IoT device can already pre-process the data and only forwards important data (not all of it)
How are edge computers cooled?
Without a fan. Passively cooled.
3 characteristics of edge computing
- computation and data shifted to computers at the edge of the network
- extends the centralized cloud model into a distributed model
- intention to bring computation and storage closer to the location where it is needed, improve response time and save bandwidth
IoT Fog Architecture
- extension of the edge –> set of coupled edge systems for adaptivity (more compute power in a distributed fashion )
- replacement of cloud with edge: distributed location-aware shared infrastructure
Fog Computing definition -> do not really have to know
Fog computing is a layered model for enabling ubiquitous access to a shared continuum of scalable computing resources. Fog nodes can be physical or virtual. Fog nodes operate in a centralized or decentralized manner and can be federated to form clusters that provide horizontal scalability over disperse geolocations.
6 essential characteristics of fog computing
- contextual location awareness and low latency
- geographical distribution
- heterogeneity
- interoperability and federation
- real-time interaction
- scalability and agility of federated, fog node clusters
contextual location awareness and low latency of fog computing
- fog nodes are aware of their location in the context of the entire system
- they are also aware of latency costs
Heterogeneity of fog computing
- supports collection and processing of data of different form factors
Interoperability and federation in fog computing
- services like real-time streaming require the cooperation of different providers
- hence, fog computing components must be able to interoperate to enable federated services across domains.
real-time interaction in fog computing
- fog computing has support for real-time interactions instead of batch processing
Scalability and agility of federated, fog node clusters in fog computing
- fog computing is adaptive in nature
- support for elastic compute and resource pooling
- adaptation to data-load changes and network condition variations
What are possible Fog Computing Service Models?
- Software as a Service (fog provider’s applications run)
- Platform as a Service (provider provides a platform for the customer’s applications)
- Infrastructure as a Service (provision of fundamental compute resources to the customer)
What is the most used fog node deployment model?
private fog node
What are the 4 fog node deployment models?
Private (provisioned for exclusive use by a single organization), community , public (open use), hybrid (composition of two or more distinct (private, community, or public) fog nodes
When would you use the cloud architecture?
- ## power hungry ecosystem but have to run from battery power –> edge computers use too much energy
IoT Network
- wireless network
- connecting sensors and actuators to a gateway
- gateway forwards data to the internet/ Cloud
Sensor Network
- when not all the sensors have a direct connection to the gateway (e.g. in rural areas)
- sensors communicate with each other so that one of them can connect to the gateway
3 categories of IoT Networks
10-100m: Wireless Personal Area Network (WPAN) (e.g., Bluetooth)
100m-1km: Wireless Local Area Network (WLAN)
to 30km: Wireless Wide Area Network (WWAN) (e.g., LTE)
Low Power Wide Area Networks (LPWAN)
- connect sensors over long distances
- enables a long lifetime of battery-powered sensor devices
- low bandwidth
Long Range Wide Area Network (LoRaWAN)
I think this is one example of LPWAN
- low-power wireless communication technique for long distances > 10 km in rural areas; excellent connection in buildings
–> trading communication bandwidth for low power and long-range (suitable for transferring of message of a few bytes, but not for streaming videos etc.)
Is LoRa a physical layer?
Yes. The physical layer is the LoRa Modulation with the regional ISM band (radio frequency).
What are the 3 classes of device types in LoRa?
The device types depend on the bi-directional communication support.
Class A: Transfer to gateway followed by two receive windows –> best power efficiency
Class B: Download window at a predetermined point in time
Class C: Permanent receive state
What does the spreading factor determine?
Determines the #bits and time per symbol.
–> Higher –> more reliable transmission.
LoRaWAN secure?
Yes messages are encrypted on two levels
- payload is encrypted by application session key
- additionally the entire packet is encrypted by network session key
Infrastructure of LoRaWAN
The Things Network
- about enabling low power devices to use long range gateways and connect to an open-source decentralized network to exchange data with applications
- intermediate between low power devices and (IoT infrastructure, mobile applications)
- takes care of receiving messages from the device, ensuring integrity and security, and forwarding the payload
The Things Network Architecture
Gateway send messages to router –> broker > handler (knows where to deliver message to, which application) –> application
- discovery service (so that router, broker, and handler can find each other)
- network server (keeps track of information of keys and the devices that are connected via the gateway to the infrastructure)
Data Flow in TTN
G (receives LoRa message)-> R -> B (de-duplication so that the same message is not sent multiple times)|NS –> H (decryption) –> A (application) –> then the whole process the other way around back to the gateway.
5 functions of an IoT platform
- device management
- data collection
- data analysis
- device control
- user management
What is AWS Greengrass
- extension of AWS IoT to the edge
- local preprocessing of data with ML
- can work without connection to the cloud
- integration with AWS security functions
- software running on own hardware
- Cloud level programming model FAAS (install functions on the edge device that are triggered when necessary)
AWS IoT Greengrass Core
Provides local services and communicates locally with devices that run the AWS IoT Device
AWS IoT Device SDK
Allows devices to interact locally with AWS IoT Greengrass Cores
AWS IoT Greengrass SDK
Allows Lambda functions to interact with local services inside an AWS IoT Greengrass Core
