Internet of Things Platforms

Parham Alvani

parham.alvani@gmail.com

1995parham.github.io

Who I am?

school BSc. Software Engineering @ Amirkabir University of Technology

school MSc. Computer Networks @ Amirkabir University of Technology

school PhD. Computer Networks @ Amirkabir University of Technology

domain Have been in IoT/Platforms since June 2015 (Ordibehesht 1394)

Let's Light the Lamp

IoT?!

The Internet of things (IoT) is a system of interrelated computing devices, mechanical and digital machines provided with unique identifiers (UIDs) and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction.

History

• So far, the Internet of Things has been most closely associated with machine-to-machine (M2M) communication.
• Although the concept wasn’t named until 1999, the Internet of Things has been in development for decades.

Future

• The definition of the Internet of things has evolved due to the convergence of multiple technologies:
• real-time analytics
• machine learning
• commodity sensors
• and embedded systems

IoT Big Picture

Devices

• Sensors, Actuators, etc.
• Design from scratch
• Build with ready-made hardware like Arduio, Raspberry Pi, etc.
• You need to be familar with Electerical and Hardware engineering

Connection

• Zigbee, ZWave, nRF, etc.
• MQTT, HTTP and etc. over IP
• You need to know about Networking and Protocol Stack

Platform

• Kaa, Thingsboard, etc.
• You need to be familar with Backend even Frontend development technologies

Application

• Machine Learning, Data Mining, etc.
• Provide intelligence in IoT Stack

Why are we using smart things alone?

The IoT technologies in sync up with cloudinfrastructures are to result in people-centric smarter environments.

IoT Maturity Model

IoT Maturity Model

What is an IoT platform ?

What is thier job ?

• Developers can develop their applications with ease
• End-users can see his/her dashboard and customize it
• The government can provide regulation on data

Platforms in bigger scope

As we mentioned earlier, platforms can be used in a governmental scope.

• Data Regulation and Governance
• IoT Operators
• Rules and regulations applies on IoT Operators instead of Customers

Foucs on Service

Platforms must provide service to customers, and customers use these services to implement their unique applications.

Platforms Components/Services

There is no specific standard for platform architecture. Here we use a de-facto standard for having a better arrangement of materials.

• Layer 1:
• Connectivity
• Layer 2:
• Data Management
• Things Management
• Layer 3:
• Accounting
• User/Project Management
• Application Programming Interface
• Analytics

How to connect?

Many ways 😱

• LoRa, nRF, Ethernet, ...
• Custom, IP, TCP, HTTP, MQTT, CoAP, ...

One Solution 🤓

Gateways

Gateways

• By using gateways every non-TCP protocol can be changed to TCP
• By using gateways every non-(HTTP MQTT ...) protocol can be changed to (HTTP MQTT ...)
• Gateways can act in may OSI layers

How to communicate?

Many ways 😱

• JSON
• CBOR
• ASN.1
• ...

JSON?

{
    "timestamp": "Mon Feb 20 21:35:00 1995",
    "states": {
        "temperature": 24,
        "humidity": 10
    }
}

{
    "temperature": 24,
    "humidity": 10,
    "status": "online"
}

{
    "101": 10,
    "102": 10,
    "103": 1024,
    "201": 1.24
}

CBOR?

{
    "101": 10,
    "102": 10,
    "103": 1024,
    "201": 1.24
}

A4 # map(4) 63 # text(3) 313031 # "101" 0A # unsigned(10) 63 # text(3) 313032 # "102" 0A # unsigned(10) 63 # text(3) 313033 # "103" 19 0400 # unsigned(1024) 63 # text(3) 323031 # "201" FB 3FF3D70A3D70A3D7 # primitive(4608263282710586327)

• Things-Platform Interoperability
• Generated SDK
• Library
• Custom Encoder/Decoder
• Thing Data Validation
• Models

IoT Platforms contains IoT Middlewars

Middleware is a connectivity layer

IoT Platforms Features

• (Data) Access Control
• API
• ReST (Representational State Transfer)
• RPC (Remote Procedure Call)
• Big Data Streaming

IoT Platforms Features

• Open Source
• Deployment Type:
  • PaaS (Platform-as-a-Service)
  • SaaS (Software-as-a-Service)
  • Local Installation

Big Data Streaming

Kafka

A streaming platform has three key capabilities:

• Publish and subscribe to streams of records, similar to a message queue or enterprise messaging system.
• Store streams of records in a fault-tolerant durable way.
• Process streams of records as they occur.

Batch Processing

• High Latency
• Static Files

Stream Processing

• Low Latency
• Event Streams

Apache Spark

• Apache Spark achieves high performance for both batch and streaming data

Apache Flink

Programs and Dataflows

• The basic building blocks of Flink programs are streams and transformations.
• A stream is a (potentially never-ending) flow of data records.
• A transformation is an operation that takes one or more streams as input, and produces one or more output streams as a result.

Windows

• Aggregating events (e.g., counts, sums) works differently on streams than in batch processing.
• Aggregates on streams (counts, sums, etc), are scoped by windows
• count over the last 5 minutes
• sum of the last 100 elements

Data Processing Pipeline

Batch as a Special case of Streaming, processing bounded and unbounded data streams in Apache Flink

Platforms Components

Conectivity and Normalization

Agents and libraries that ensure constant object connectivity and harmonized data formats

Device Management

Backend tool for the management of device status, remote software deployment and updates

Platforms Components

Processing and action mamagement

Rule engine that allows for (real-time) actions based on incoming sensor and device data

Data visualization

Graphical depiction of (real-time) sensor data

Platforms Components

Analytics

Algorithms for advanced calculations and machine learning

Additional tools

Further development tools (e.g., app prototyping, access management, reporting)

Platforms Components

External interfaces

APIs, SDKs and gateways that acts as interfaces for 3rd party systems

Database

Repository that stores the important data sets

Kaa

Kaa [Wikipedia]

Kaa is a fictional character that Rudyard Kipling describes him as an exceptionally long, yellowish Indian rock python.

Kaa [Kaa]

100% open-source Internet of Things middleware platform for everyone.

What is Kaa?

• Reliable foundation for developing your connected products.
• Transport-agnostic link between the hardware and application worlds.
• Much more than just a message bus.
• Customizable middleware that implements necessary functional patterns for the IoT.
• Cloud enablement software for your hardware products.

Kaa Architecture

Kaa Server

The Kaa server acts as a foundation for your custom-tailored IoT back-end system. It offers a number of mechanisms for internal customization and integration with external systems:

• REST interfaces for integration with product-specific services
• Log appenders for integration with analytics
• pluggable implementations of custom transports

Kaa Server

The Kaa server was designed to handle millions of connected devices under peak pressure without ever getting tired.

Kaa Data Model

• Works with structured and unstructured data
• Well-structured data provides excellent material for any manipulation and analytics
• This kind of data also helps abstract out any unimportant details of the low-level implementation and your hardware.
• Kaa ensures that even gadgets based on very different technologies can be managed and operated identically, as long as they share the same set of data schemas.

Kaa Generated SDK

• Kaa endpoint SDKs are available in Java, C++, and C, and are designed to be embedded into your client application (or firmware).
• They handle client-server communication, authentication, data marshalling, encryption, persistence, and other functions provided by the platform.

What does kaa do for us?

• Messaging across endpoints
• Notification
• Collecting data from endpoints

Messaging across endpoints

Messaging across endpoints

• Kaa allows for delivery of events, which are structured messages, across endpoints
• The event className family can be used by one or multiple applications, thus the event can be shared between applications

Notification

Notification

• Kaa uses notifications to distribute structured messages, posted within notification topics, from the server to endpoints

Collecting data from endpoints

Collecting data from endpoints

• Kaa provides rich capabilities for collecting and storing structured data from endpoints
• The data collection subsystem in Kaa is designed to collect records (logs) of pre-configured structure
• it store them in the client, transfer from the client to Operations service, persist on server for further processing, or submit to the immediate stream analysis

Distribution Method

• Open Source
• PaaS
• The ground-breaking Kaa 1.0.0 will be available soon for the pre-release trial among early subscribers.

Let's learn it in code

2 Main Pieces

• Kaa SDK (Agent)
• Kaa Server

Thingsboard

What is ThingsBoard?

Open-source IoT Platform - Device management, data collection, processing and visualization.

What does thingsboard do for us?

• Provision and manage devices and assets
• Provision, monitor and control your IoT entities in secure way using rich server-side APIs.
• Define relations between your devices, assets, customers or any other entities.

What does thingsboard do for us?

• Collect and visualize data
• • Collect and store telemetry data in scalable and fault-tolerant way.
  • Visualize your data with built-in or custom widgets and flexible dashboards.
  • Share dashboards with your customers.

What does thingsboard do for us?

• Process and React
• Define data processing rules and trigger reactions using plugins.
• Raise alarms on incoming telemetry events or attribute updates.
• Enable use-case specific features using customizable rules and plugins.

What does thingsboard do for us?

• IoT Gateway
• Integrate devices connected to legacy and third-party systems using existing protocols.

Supported Protocols

• MQTT
• CoAP
• HTTP

IoT data analytics using Apache Spark, Kafka and ThingsBoard

Home-Assistant

What is Home-Assistant?

Open-source home automation platform running on Python 3

What does Home-Assistant do for us?

• Home Assistant will track the state of all the devices in your home, so you don't have to
• Control all your devices from a single, mobile-friendly, interface
• Set up advanced rules to control devices and bring your home alive.

Automating Home Assistant

(trigger) When Paulus arrives home (condition) and it is after sunset: (action) Turn the lights in the living room on

Microsoft Azure

IoT Hub

Connect, monitor, and manage billions of IoT assets

• Establish bi-directional communication with billions of IoT devices
• Work with familiar platforms and protocols
• Open-source device SDKs for multiple platforms
• HTTP, Advanced Message Queuing Protocol (AMQP), and MQ Telemetry Transport (MQTT)
• Authenticate per device for security-enhanced IoT solutions
• Manage your IoT devices at scale with device management
• Extend the power of the cloud to your edge device

Azure IoT Hub Integration