CSE5PRP

IoT Protocols
and Platforms
• IoT Platforms - 5
How to turn the IoT data into intelligent, actionable information
that makes an IoT solution a truly smart solution?
IoT Platform : Three layers

IoT Reference Framework - Overview. Adapted from " IoT Reference Framework," by Nguen,N and Tett,M, 2018, retrieved from https://www.iot.org.au/wp/wp-content/uploads/2016/12/IoT-Reference-Framework-v1.0.pdf.
IoT Analytics

• IoT analytics are, put, analytics that relies on data collected by internet-connected IoT
devices
• usually large and configured to provide a variety of data types.

• data sets that are both comprehensive and up to date.

• may be updated frequently as every minute or second.

• predictive analytics with IoT data - analyze that data with IoT platform to identify trends and make
predictions.

• can employ big data analytics techniques

IoT Analytics

• Analyzing large amount of accurate data provides opportunities

• predictive analytics to make better decisions.

• allowing businesses to significantly improve their ability to forecast the future

• carry out automated tasks based on a certain set of conditions being met.

• save money or time

• improve safety
Artificial Intelligence and Machine Learning

• Avoiding Downtime

• Increased Operational Efficiency

• Emotional analysis and facial recognition

• New and Improved Products and Services

• Cybersecurity and risk

Blockchain

• According to IBM (Hooper M, 2018) , there are mainly five benefits of using blockchain:

1. Greater transparency

2. Enhanced security

3. Improved traceability

4. Increased efficiency and speed

5. Reduced costs
Intelligence Enablement

• Intelligence, or ‘smart enablement’ refers to the use of smart technologies across data to
analyse, cross reference, observe behaviour and patterns, learn and predict, in order to
produce insightful outcomes, and to drive smart actions (Tett, M, IoT Reference
Framework, 2018).

• Intelligent analysis of massive amount of data collected from the IoT devices
• Streaming Analytics

• Artificial Intelligence (AI)

• Machine Learning (Ml)

• Deep Learning

• Blockchain
Application Enablement

• IoT application development,

• Operate and Manage
• device connection management,

• data storage and processing,

• event monitoring and handling,

• analytics,

• visualization,

• analytics

• adapters to connect to enterprise systems

IoT User Interface

• Users need a way to view and understand the data captured by IoT.

• An IoT user interface (UI for short) is intended to highlight how users of the IoT solution
will access it

• A user interface could be anything from a smartwatch to voice-controlled Amazon Echo to

the buttons on a smart tractor dashboard.

• Smart phones, PC or laptop, Headset Mounted Displays (HMDs) used in VR/AR

applications, or home devices such as Google Home, Amazon Alexa,

• Sometimes a UI might have to be completely customised due to specific requirements for

the solution.
IoT (Internet of Things) dataflow
• The IoT data flow refers to the process of collecting, transmitting, storing, and analyzing
data generated by interconnected devices within the IoT ecosystem.
• The data generated by IoT devices can be in the form of text, images, audio, or video
1. Data Generation: IoT devices, such as sensors, actuators, and embedded systems, collect data from their
environment. This data can include temperature, humidity, pressure, location, motion, or any other relevant
information based on the device's purpose.
2. Data Aggregation: The collected data is aggregated either locally on the device or at an intermediate
gateway. In some cases, edge computing techniques are employed to perform initial data processing and
filtering at the edge of the network.
3. Data Transmission: Aggregated data is transmitted over a network to a centralized data processing system or
a cloud-based IoT platform. This communication can occur through various wireless protocols, such as Wi-Fi,
Bluetooth, Zigbee, or cellular networks like 4G or 5G.
4. Data Storage: Once the data reaches the central system or cloud platform, it is stored in a database or a
data repository. The choice of the storage solution depends on factors like data volume, velocity, and the
required analytics capabilities.
5. Data Cleansing: Data quality issues such as missing, incomplete, or incorrect data must be addressed before
analysis can begin.
IoT (Internet of Things) dataflow
6. Data Processing: The stored data undergoes further processing, which can involve real-time analysis,
batch processing, or a combination of both. Data processing techniques like filtering, aggregation,
normalization, and transformation are applied to extract valuable insights from the raw data.Data
storage: The processed data is then stored in a database, data warehouse, or data lake for future use.
7. Data Analytics: Advanced analytics techniques, such as machine learning, artificial intelligence, and
statistical analysis, are applied to the processed data. This analysis helps derive meaningful patterns,
trends, predictions, or actionable insights.
8. Data visualization: The analyzed data can be visualized in the form of charts, graphs, and dashboards to
help stakeholders understand and interpret the data.
9. Decision Making and Action: Based on the insights gained from data analytics and visualisations,
decisions can be made. It can also to trigger specific actions and these actions can be automated
responses within the IoT system or sending notifications/alerts sent to end-users or other systems.
10. Feedback Loop: Feedback from actions taken or user interactions may be collected and incorporated
back into the IoT system. This feedback loop allows for continuous improvement and optimization of the
IoT solution.

• The exact flow of data in an IoT system can vary depending on the specific use case and the type of IoT devices involved,
deployment architecture, and the complexity of the IoT solution.
 Life cycle of big data analytics for MIoT

Image Source: Big Data Analytics for Manufacturing Internet of Things: Opportunities, Challenges and Enabling Technologies - Scientific Figure on ResearchGate. Available from: https://www.researchgate.net/figure/Life-
cycle-of-Big-Data-Analytics-for-MIoT_fig1_333814964 [accessed 8 May, 2023]
 Data pre-processing techniques

• Data preprocessing approaches on MIoT data include data cleaning,

data integration and data compression

Image Source: Data preprocessing techniques Big Data Analytics for Manufacturing Internet of Things: Opportunities, Challenges and Enabling Technologies - Scientific Figure on ResearchGate. Available from:
https://www.researchgate.net/publication/333814964_Big_Data_Analytics_for_Manufacturing_Internet_of_Things_Opportunities_Challenges_and_Enabling_Technologies
Data Analytic Approaches

Image source; Big Data Analytics for Manufacturing Internet of Things: Opportunities, Challenges and Enabling Technologies - Scientific Figure on ResearchGate. Available from: https://www.researchgate.net/figure/Data-
analytics_fig3_333814964 [accessed 8 May, 2023]
Classification of Data Analytics Approaches

Image source: https://www.researchgate.net/publication/333814964_Big_Data_Analytics_for_Manufacturing_Internet_of_Things_Opportunities_Challenges_and_Enabling_Technologies

Classification of Data Analytics Approaches

• Descriptive analytics – an exploratory analysis of historical data to tell what happened

• Diagnostic analytics – a deeper look at data to attempt to understand the causes of

events and behaviors.

• Predictive analytics - historical data to anticipate the trends of data

• Prescriptive analytics - extends the results of descriptive, diagnostic, and predictive

analytics to make the right decisions in order to achieve predicted outcomes
Data Visualization

• Visualization is a critical component of any big data analytics solution.

• It converts raw data to a form that is viewable and understandable to humans
• Visualization provides business decision-makers with graphical representations of
analysis, which can make it easier to see the implications of the data.

• Communicate information clearly and

effectively through graphical means.
• Give Meaning to otherwise raw and hard to
understand data
• For exploration, discovery, insight.

Image Source : https://upload.wikimedia.org/wikipedia/commons/9/96/Sample_JReport_Dashboards.png

Data Visualization
The power of data visualization

• A picture is “worth a thousand words”.

• Use the eye for pattern recognition; people are good at scanning, recognizing and remembering
images
• Benefits of data visualization include the ability to better observe trends and patterns, and to
identify correlations and clusters.
Data visualization
• It can help to extract and interpret the informative values from complex and high-
dimensional IoT data.
• Data visualization methods :
• Information visualization
• Exploratory data analysis
• Statistic plots
• Quantitative messages: time-series
• Ranking
• Frequency distribution
• Deviation
• Correlation
• Basic data visualization techniques :
• Statistic plots (e.g., bar chart, histogram, pie diagram, scatter plots),
• Word clouds
• Correlation coefficient matrices/functions,
• Network/graph diagrams
• Heat map
Cholera Outbreak of 1854
• The London cholera outbreak of 1854 killed more
than 500 people in just ten days in the centre of
London.

• The Miasma theory at the time concluded that

disease was spread through foul air.

• Dr. John Snow plotted every death on a

map with mapped bar charts and was
able to show that the closer to the
Broad St water pump he plotted, the
greater the number of deaths.
Why build IoT dashboards?

• Real-Time Monitoring
• Data Visualization
• Performance Analysis
• Predictive Maintenance
• Operational Efficiency
• Stakeholder Communication
• Regulatory Compliance and Reporting
Power BI

• Power BI is a business intelligence and data visualization tool developed by Microsoft.

• It converts data from different data sources to interactive dashboards and BI reports.
• It is a collection of software services, apps, and connectors that work together to turn
unrelated sources of data into coherent, visually immersive, and interactive insights.
• Power BI provides a user-friendly interface with drag-and-drop functionality, making it
accessible to business users with little or no programming knowledge.
• It empowers users to gain insights from data, make data-driven decisions, and
communicate information effectively through compelling visualizations.
• Power BI can be a powerful tool for visualizing and analyzing data generated from IoT
devices
 Power BI
• Power BI consists of several elements that all work together, starting with these three basics:
• A Windows desktop application called Power BI Desktop.
• An online software as a service (SaaS) service called the Power BI service.
• Power BI Mobile apps for Windows, iOS, and Android devices.

Image Source: Power BI , retrieved from https://learn.microsoft.com/en-us/power-bi/fundamentals/power-bi-overview retrieved on May 2023

Power BI Features

1. Data Integration:
• Power BI can connect to IoT platforms and databases to ingest and integrate data from various IoT devices.
• It supports connectors for popular IoT platforms such as Azure IoT Hub, AWS IoT, and Google Cloud IoT Core.
• This allows us to bring real-time or historical IoT data into Power BI for analysis.

2. Real-Time Dashboards:
• Power BI helps to create real-time dashboards that display live data from IoT devices.
• It helps to visualize sensor readings, device statuses, and other relevant metrics in real time.
• Power BI streaming datasets and APIs can be utilized to monitor and analyse live IoT data as new data arrives.

3. Data Transformation and Modelling:

• Power BI's data transformation capabilities, such as Power Query, can be used to clean and shape the IoT data for
analysis.
• It helps to perform data cleansing, filtering, and aggregation to prepare the data for visualization.
Power BI Features

4. Advanced Analytics:
• Power BI supports advanced analytics features that can be applied to IoT data.

• The built-in functions and writing custom calculations using DAX helps to perform calculations, create measures, and
derive insights from the IoT data.

• Machine learning algorithms or statistical models can be applied to uncover patterns, anomalies, or predictive insights.

5. Custom Visualizations:
• Power BI offers a wide range of visualizations, including custom visuals from the Power BI marketplace.

• Option to choose visualizations such as line charts, scatter plots, heat maps, or geographical maps.

6. Geospatial Visualization:
• Power BI has built-in support for geospatial visualizations, which allows for plotting IoT data on maps.

• It helps to visualise the geographical distribution of devices, track movements, or display sensor readings in different
locations.
Power BI Features
7. Alerts and Notifications:
• Power BI allows to set up alerts and notifications based on predefined thresholds or conditions.

• For example, we can configure alerts to trigger when sensor readings exceed certain limits or when specific events occur.
These notifications can be sent via email, mobile app notifications, or integrated with other collaboration tools like Microsoft
Teams.

8. Anomaly Detection:
• Power BI's analytics features can be utilized to detect anomalies in IoT data.

• By applying statistical models or machine learning algorithms, we can identify unusual patterns or outliers in sensor
readings that may indicate equipment malfunctions, security breaches, or other abnormal behavior.

9. Reports and Storytelling:

• Power BI allows to create reports that combine multiple visualizations, textual descriptions, and narratives explaining the
insights, trends, and implications.

10. Sharing and Collaboration:

• Power BI enables to share IoT dashboards and reports with others in the organization or external stakeholders.

• Power BI also facilitates collaboration by allowing users to annotate and comment on specific data points or visuals within
the reports.
 AWS Data Visualization

• Amazon SageMaker
• Fully managed service for data scientists and developers.
• It covers the machine learning workflow to label and prepare your data, choose an algorithm, train a model, and
tune and optimize the model for deployment.
• Amazon SageMaker models can also make predictions and take action.

• Jupyter notebooks
• Jupyter notebooks can be hosted with Amazon SageMaker.
• Jupyter notebooks are an open-source tool you can use to create narratives that combine text with
visualizations.
 AWS Data Visualization
• AWS IoT SiteWise

• AWS IoT SiteWise is a managed service that simplifies collecting, organizing, and analyzing industrial equipment data.

Image Source: AWS IoT SiteWise retrieved from https://aws.amazon.com/iot-sitewise/

 AWS Data Visualization
• AWS IoT Analytics
• AWS IoT Analytics service can be used to run real-time analytics directly on data sent by IoT devices.

• Can also run queries to send filtered data to a permanent data store or to visualization tools.

• It is a fully managed and highly scalable service that can be natively integrated with other services

Image Source: AWS IoT Analytics, retrieved from https://aws.amazon.com/iot-analytics/ retrieved on May 2023
 AWS Data Visualization
• Amazon Kinesis Data Analytics

• Used to perform real-time analytics on data in Amazon Kinesis streams.

• It is a fully managed, highly available, and scalable service that can handle petabytes of data in a cost-effective, pay-as-you-go model.

• Structured query language (SQL) users can run queries on it, and developers can use Java libraries for sophisticated transformations and
analysis.

Image Source: Amazon Kinesis Data Analytics retrieved from https://aws.amazon.com/kinesis/data-analytics/

 AWS Data Visualization
• Amazon QuickSight

• It is a data analysis and visualization service that you can use for a wide variety of data types and sources.

• Can use it to automatically discover data sources in your AWS solution.

• Can integrate it with other AWS services, like Amazon Redshift, Amazon S3, Amazon Kinesis, or AWS IoT Analytics to create visualizations
quickly in a cost-effective way.

• It is easy to share these visualizations with important stakeholders, which can help their decision-making process..

Image Source: Amazon QuickSighT retrieved from https://aws.amazon.com/quicksight

IoT data ingestion and visualization - Architectural Patterns

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 1: AWS Stream Manager

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 2: AWS IoT SiteWise (+ AWS IoT SiteWise
Monitor)

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 3: AWS IoT Core + AWS IoT Analytics + Amazon
QuickSight

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 4: Amazon Timestream

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 5: AWS IoT Core + Amazon Kinesis + Amazon
QuickSight

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 6: Amazon OpenSearch Service + OpenSearch
Service Dashboards/Amazon Managed Grafana

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
Pattern 7: AWS IoT Core + AWS Lambda + Amazon
DynamoDB + Amazon QuickSight / Custom Dashboards

Image Source IoT data ingestion and visualization Retrieved from : https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/ retrieved on May 2023
IoT and Big Data Application

Image Souce: IoT and Big Data Application , Retrieved from https://github.com/AWS-Big-Data-Projects/Iot-and-Big-Data-Application-using-aws-and-apache-kafka retrieved on May 2023
Can you think of some IoT solutions?
IoT Application Domains

• IoT Agriculture and Farming

• Advanced yield monitoring

• Optimal seeding

• Optimal water usage

• Livestock monitoring

• IoT Energy Solutions

• IoT smart meters

• Smart homes (connected home)

 IoT Application Domains
• IoT Smart Building Solutions
• Safety monitoring and alerting

• Smart lighting

• Flooding, water leakage, and pollution monitoring

• Detection of hazardous gases and radiation levels:

• Detection of garbage levels in containers

• IoT Oil and Gas Solutions

• Connected oil and gas fields

• Downstream applications
IoT Application Domains

• IoT Finance
• IoT usage-based auto insurance

• IoT solution to reduce fraud and liability

• IoT safety solutions

• IoT Healthcare

• Fall detection:

• Tracking of medical devices

• Medical fridges for hospitals

IoT Application Domains

• IoT Industrial
• Predictive maintenance:
• Connected factory:
• Connected mine
• Supply chain control

• IoT Retail
• Full tracking of products in stores:
• Full automation of product delivery
• Flexible shopping and loyalty programs:
• Interactive consumer engagement and operations
IoT Application Domains

• IoT Transportation
• Smart and connected parking

• Smart roads and traffic congestion

• Connected rail

• Continuous quality of shipment monitoring

Summary

• Selecting the right IoT Platform

• IoT Analytics

• Artificial Intelligence and Machine Learning

• Blockchain

• Application Enablement

• IoT User Interface

• IoT Application Domains

References

• Rayes, A., & Salam, S. (2017). Internet of things from hype to reality. The Road to Digitization; River Publisher Series in Communications; Springer:
Basel, Switzerland, 49.
• Nguen,N & Tett,M. (2018). IoT Reference Framework : Application Guide, IOT Alliance Australia retrieved from https://www.iot.org.au/wp/wp-
content/uploads/2016/12/IoT-Reference-Framework-v1.0.pdf.
• Tett,M. (2018). IoT Reference Framework : Application Guide, IOT Alliance Australia retrieved from https://www.iot.org.au/wp/wp-
content/uploads/2016/12/IoT-Reference-Framework-Application-Guide-v1.0.pdf
• Ullah, M., Nardelli, P. H., Wolff, A., & Smolander, K. (2020). Twenty-one key factors to choose an IoT platform: Theoretical framework and its
applications. arXiv preprint arXiv:2004.04924.
• Tawalbeh, L. a., Muheidat, F., Tawalbeh, M., & Quwaider, M. (2020). IoT Privacy and security: Challenges and solutions. Applied Sciences, 10(12),
4102.
• Hellbe, A., & Bohlin, G. (2018). Evaluating IoT cloud platforms in the context of smart buildings.
• Power BI documentation Retrieved from https://learn.microsoft.com/en-us/power-bi/ retrieved on May 2023
• Hakme,A,. Gupta, R,. & Kalaspurkar, U. (2022) 7 patterns for IoT data ingestion and visualization- How to decide what works best for your use case
retrieved from https://aws.amazon.com/blogs/iot/7-patterns-for-iot-data-ingestion-and-visualization-how-to-decide-what-works-best-for-your-use-case/
• Dai, H. N., Wang, H., Xu, G., Wan, J., & Imran, M. (2020). Big data analytics for manufacturing internet of things: opportunities, challenges and
enabling technologies. Enterprise Information Systems, 14(9-10), 1279-1303.