MODULE 1: INTRODUCTION - Smart watch

- Smart car
What are things? - Smart kegs
- Thing - Not a computer What is Smart?
- Phone, watches, thermostats, cars, Electric Meters, - IoT = Instrument, Interconnect, Intelligently process
sensors, clothing, TV,… (3 I’s)
- But rather… Anything, Anywhere, Anytime, Anyway, - Old: Smart = Can think → Can compute
Anyhow (5 A’s) - Now: Smart = Can find quickly, Can Delegate →
Communicate = Networking
Internet of Things - Examples: Smart Grid, Smart Meters, Smart Cars,
- Less than 1% of things around us is connected. Smart homes, Smart Cities, Smart Factories, Smart
o Refrigerator, car, washing machine, heater, a/c, Smoke Detectors, …
garage door, should all be connected but are not.
- From 10 Billion today to 50 Billion in 2020 should Internet of Brains
include processes, data, things, and people. - Brain-to-Brain Interface
- For Cisco, it is Internet of Everything - A person’s brain can send signals to other person’s
- For IBM, it is Smarter Planet brain
- More and more connected devices over the - Useful for handicap people to communicate with
ubiquitous internet others
o Smart Phones, Wearables …
o Smart Meters, Home automation sensors, …
o Connected Cars, Smart Cities, …
o Digital Warehouses, Digital Shopfloor, …
- Devices collect data that can be gathered and
analyzed
- Collected data has to be integrated into other
systems, depending on the use case
- The ecosystem is usually called the Internet of
Things (IoT)

IoT Definition
- A global infrastructure for the information society, Google Trends
enabling advanced services by interconnecting - Around for 10 years
(physical and virtual) things based on existing and - IERC-European Research Cluster on the Internet of
evolving interoperable information and Things funded under 7th Framework in 2009
communication technologies. “Internet of European Things”
- US interest started in 2009 with $4B funding for smart
IoT Characteristics grid in American Recovery and Reinvestment Act of
- Interconnectivity 2009
- Things-related services
- Heterogeneity Research Funding for IoT
- Dynamic changes - 70 M € in European Research program FP7 
- Enormous scale Internet of European Things
- Networking and Information Technology Research
IoT Sample Applications and Development (NITRD)
- Smart grid o Group of 15 Federal agencies: NSF, NIH, NASA,
- Smart health DOE, DARPA, ONR, …
- Smart home o Recommends supplement to the president’s
- Smart cities annual budget
- Smart industries o CPS is one of the areas recommended by NITRD
- Smart tv starting 2012  Smart infrastructure
 o Smart Grid, Smart Bridges, Smart Cars, tele- - The data that is processed is being produced
operational surgical robots, Smart Buildings continuously
- March 2014: £45M for IoT research in UK by David o New data is produced when a state changes – this
Cameron is called an event
o The events are produced in a certain sequence
Business Opportunities - The collection of events – the event stream - has to
- Components: Sensors, wireless radios, protocols, be processed and analyzed in the same manner (real-
- Smart Objects: Smart TV, Camera, Watch, … time and consecutive)
- Systems: Buildings, Cars, Health, … - The full process of processing event data is called
- Network service providers: ISP Event Stream Processing (ESP) or Complex Event
- Application Service Providers: Monitoring, Analytics, Processing (CEP)
Apps, Complex Event Processing
- Event Stream Processing (ESP) includes the
The Business Perspective processing of streams of event data (lower level)
- The future of business is automation, and processing - Complex Event Processing (CEP) also includes
sensor data is the key to it analysis and advanced methods like pattern matching
- Not only restricted to B2B, but also relevant for B2C to gain more insights, but both terms are used
applications equivalent in many cases (with analysis)
o Customer interactions become faster and faster - Example: The sensors of an aircraft can produce 800+
o Processing sensor data on-the-fly will not be a TBytes of data during a 12-hours flight
unique feature, but a prerequisite for success - Sense
businesses o Gather events from different sources in the real
- Processing live data not only lets you automate world in real-time
processes but also detect threads and opportunities - Analyze
o Aggregate, correlate, classify, filter the data and
The Business Perspective: Use Cases find patterns in it
- Predictive Maintenance - Respond
- Situational Marketing o Initiate a response based on the results of the
- M2M: Machines are controlled by sensor data analysis of the data
- Environmental Monitoring - With CEP, data streams can be manipulated, very
- Infrastructure Management similar to operations on static data, for example:
- Automated Manufacturing / Industrie 4.0 o Aggregations
- Financial Services o Sums
- New applications o Transformations
o Pattern Matching (machine learning comes with
Data Processing in IoT Systems 2.0)
- Data processing in IoT systems differs from o Modeling hierarchies
traditional business data o Relationship detection
o Variety of sources (sensors, information systems, o …and many more
public feeds etc.)
o Data is highly interconnected Technical Setup
o Data must be processed in real-time - IoT systems process events, that‘s why they are
- Analysis must be done in real-time as well usually based on an Event Driven Architecture
- From a technical perspective, IoT is proper Big Data
o Emphasis on Volume / Velocity
o Exponential Growth
- IoT solutions usually consist of the following
technical systems:
o Sensors and Gateways
o Stream Data Processing
o Data Storage
 o Data Analysis

Sensors - Hardware that detects changes in the

Gateways - Hardware to which the sensor is connected.

Sensor + Gateway

Data Storage
- After the data has been processed, the raw or
What is driving the IoT ?
aggregated data has to be stored for further analysis
Analysis
- Replay of data
- “Bigger picture“ analysis
o Bigger windows
o Combine streaming data with information systems
- Integration of CEP output into Business Intelligence
Why now ?
Recent IoT Products - More products are adding intelligence with MCUs to
support more sophisticated control
- Adding connectivity is getting easier & cheaper
- Low-power semiconductors allow for more battery-
powered applications
- Wi-Fi and internet access broadly available
- Tablets, PCs and Smartphones broadly available – can
be leveraged as a gateway
- Connectivity brings control, sensing & ability to
update system software over the internet

IoT is an Enabling Technology – not a Market

IoT building blocks

But there are challenges

Sensing technology is a must

The IoT must be Easy

Recent Protocols for IoT

Making it Easier to Connect to the Cloud

Typical IoT cloud services

System-level approach to power is required

Micro-Clouds on Cell Cell-Towers

Security is vital at all layers

Making the IoT of 2020 happen

IoT Research Challenges

The Problem Statement