32 Module Internet of Things

Suchit Purohit

Learning Objectives

• Definition
• Factors contributing to success of IOT
• IOT trends and statistics
• Three layer architecture and layer descriptions
• Other architectures
• Challenges

Introduction

A bridge collapses casing many casualties who holds the responsibility?? Government holds contractor responsible, contractor carries forward blame to engineer, engineer to quality control, quality control to cement manufacturer the blame gave is at terminating point and the manufacturer says…

It was fault of cement??????

If this has been said decades ago, it would be insane but in today’s scenario it is SMART.

The cement should have been smart. Smart cement is equipped with sensors to monitor scans cracks and alerts to fix it. If there is ice on bridge sensor would detect and inform the vehicles in its vicinity wirelessly. The vehicle inturn would alert the driver or slow down itself.

The communication is sensor to object, object to machine, machine to machine and machine to human. This is the exact scenario nowadays.

There are sensors monitoring, tracking and collecting all forms of data; there are cloud based applications that convert data into intelligence and transmitting to machines and humans. This information can be used for real time responses for actuators or decision systems.

The bridges are getting smart so are the cars; environment is smarter so are the cities.

The fundamental shift behind this technological outbreak to make things intelligent is called “Internet of Things” or IoT.

“The Internet of Things (IoT) is a system of interrelated computing devices that can be mechanical and digital machines, objects, animals or people tagged with unique identifiers RFID, IP address, NFC barcode etc. and have the ability to observe, identify, understand and transfer data over a network and convert it into intelligence without requiring human-to- human or human-to-computer interaction.

• A thing, in the Internet of Things, can be
• A person with wearable device counting number of steps taken
• Animals tagged with RFID to track it wherever it goes
• An automobile that has built-in sensors to alert the driver when he is snoozing
• A coffee maker that automatically starts when an alarm rings
• A cell phone that automatically raises its call ring volume when it is in concealed object like purse..

Any natural or man-made object that can be tagged and provided with the ability to transfer data over a network

From the report of BI Intelligence, Business Insider’s premium research service, it is expected that there will be more than 24 billion IoT devices on Earth by 2020. That’s approximately four devices for every human being on the planet. It comes with the tagline

“Anything that can be connected will be connected”

Such a technological outbreak also demand investments and inturn expect revenues.  It is expected that $6 billion will flow into IoT solutions, including application development, device hardware, system integration, data storage, security, and connectivity but inturn $13 trillion revenue will be generated by 2025. In our own country India, Telco is expecting to have 10-15% of revenue by IoT in 2017 which is only 1% till now.

First Internet Appliance

IoT is such a hot topic today; it opens a lot of opportunities. Though not recognised formally, IOT has been in practice since decades. In early 80’s students at Carnegie Melon University developed a coke machine recognised as first internet machine. The machine could be connected over the internet; the students could check the status of machine and determine if there is any cold drink available and then only could go to the machine.

Formally, the term IoT was coined by Kevin Ashton, Co – founder and ED of Auto – ID centre at MIT. In a presentation at Procter and Gamble in 1999 he said

“If we had computers that knew everything there was to know about things — using data they gathered without any help from us — we would be able to track and count everything and greatly reduce waste, loss and cost. We would know when things needed replacing, repairing or recalling and whether they were fresh or past their best”

It was understood that IOT has potential to change the world. Idea soon started getting materialised and opened new doors towards an interconnected today where real and virtual world merge seamlessly through deployment of embedded devices.

Factors contributing to success of IOT

IoT is one of the industries which is successful even in its nascent stage. Factors contributing to success are

• Wide availability of Broadband Internet
• Decrease in cost of Connecting
• Increase in devices embedded with Wi-Fi capabilities
• Growth of tagging technologies like RFID, NFC, barcode, QR codes and water marking
• Development of sensor technology
• Introduction of big data bases like Hadoop and NoSQL
• Development of real time analytics and machine learning skills
• Ipv6’s huge success in increasing address space
• Sky rocketed penetration of smart phones in our day-to-day life
• IoT platforms  like  cloud  to  support  applications  to  convert  data  collected  into intelligence

IoT Architecture

IOT is not a single technology it is agglomeration of various technologies that work together. Sensor/actuators collect data; it is stored on processed on the network or remote server. Actions are taken based on inference from this processed data. There is no single IOT accepted universally different architecture have been proposed by different researchers. The basic architecture is a three layered architecture. It comprises of three layers

• Perception layer
• Network/Communication layer
• Application layer

Perception Layer

It is the physical layer comprising of sensors and actuators. Sensors are considered as the troops of IoT. They are on–the–ground pieces of hardware which are responsible for monitoring the processes, talking measurement and collecting data. An actuator is a device that is used to effect a change in the environment. For eg. To change the lightning conditions. Every IoT installation required specific type of sensors. Figure 1 shows sensors used in different IoT applications.

Figure 1: Different types of sensors

Mobile phone sensor

Most basic type of sensors is mobile (smart) phone itself. A smart phone has many sensors embedded in it such as GPS, accelerometer gyroscope (Movement sensor),camera, light sensor, mic, proximity sensor and magnetometer, application build on mobile phones. Application for health monitoring, water quality monitoring, air quality monitoring also serves as sensors.

Medical sensors

Monitor and measure various parameters of human body. These sensors real time scanning data which can be provided to doctors. eg. wearable devices equipped with sensors for measuring heart rate, blood pressure, respiration levels blood Glucose level. When the parameters are required to be monitored continuously, monitoring patches are tattooed on skin. Most important consideration for these sensors is context. Data should be combined with contextual information for eg. Heart rate increase during exercises cannot be inferred abnormal. There are EEG sensors to understand the neural signals in brain to infer whether brain is calm or wandering in thoughts. This is known as neurofeedback.

Environment and chemical sensors

They are used to sense the parameters of physical environment like temperature, humidity, pressers, water pollution and air pollution. Temperature sensors can be used in many applications like monitoring of temperature inside house/industries to adjust cooling parameters or measuring of temperature of soil and water in agricultural industry. Pressure sensors can be used to measure water flow or all flow in pipes or tyres of vehicles in aircraft. Air quality can be measured by the use of sensors which sense presence of gases and other particulate matter in air .Chemical sensors are used to sense the chemicals by their odour and taste and used in food quality check, agricultural products.

Proximity and level sensors

Proximity sensors are used to find objects close to the sensors. They can be used in retail, where a retailer can use a customer’s proximity to a product to send deals and coupons directly to their smartphone. They can be used at parking lots to monitor the availability of parking spaces in large venues like airports, malls and stadiums. Level sensors measures level of content like liquid, slurries, granular materials nd powders that exhibit an upper surface in a container. They find their use in smart waste management and recycling purposes. In Oil Industry level sensors are used for measuring tank levels; diesel fuel gauging; liquid assets inventory and generate high or low level alarms. In agriculture they can be used for managing water levels in irrigation.

Infrared sensors

Used to measure certain characteristics like proximity, heat of the surroundings. They are used in applications like to monitor blood flow level, identify the heat level in the environment.

Pre-processing stage

Smart things collect huge amount of data therefor storage and computational resources are required to analyse, store and process their data.

Cloud Computing

One of the architecture is applying Cloud Computing. This is centralized architecture in which cloud is at the centre, applications above it and network below it. The performance of this architecture suffers from mobility since the devices are moving it is difficult to communicate with the central data centre. The response time of the communication is also very high which is not desirable in latency centric applications. The problem aggravates even further as more and more devices add up to the network. Furthermore, it also consumes lot of power hence posing a challenge for battery operated IoT devices.

Fog Computing

The solution to the problems of cloud computing is Fog computing. The solution is to bring computing and storage resources near the edge of the network. The concept is also known as Edge Computing. The concept is realized by employing a smart gateway between netwok and cloud. The functions of smart gateway are:

• Collect , preprocess and filter sensor data
• Providing compute, storage and networking services
• to IoT devices
• Communicate with  the  cloud  and  send  only   necessary  data  monitoring  power consumption of devices
• ensure security and privacy of data

The advantages of fog computing over cloud computing are:

• Distributed services: The nodes are distributed geographically providing services to IoT device in its area. Hence more robust
• Low latency: Fog computing has advantage of low latency rate because of smart gateways used.
• Mobility: Devices communicate with the gateways in their area supporting mobility.
• Interaction with Cloud: Communicate with the cloud sending data to and between the network and the cloud.

Network layer

Network layer is responsible for connecting to other smart things, network devices and servers. The communication can be via non – IP based networks which consume less power. They can connect  to  the internet  via  smart  gateway  .Bluetooth,  RFID  and  NFC  are  some  of  the technologies used but these are limited in range hence used for PANS. At the same time, connection to internet through IP stack demands large power and memory from devices. 6LOWPAN  is  a  trade  off  amongst  them.  It  combines  IPV6  with  low  power  personal  area networks. This technology empowers PANS with the range equivalent to LANS with much less power consumption. Other communication technologies are listed in figure 2.

Middleware

Things in Internet of things are heterogeneous in nature. Well-defined standards are required for Interoperability of such heterogeneous devices. Standardization is difficult owing to the varied requirements of different applications and devices. The solution is to have a middleware platform, which provides abstraction to support interoperability. Oracle’s Fusion Middle-ware, OpenIoT, Middlewhere and Hydra FiWare by EU are some of the middlewares used in IoT environment.

Figure 2: Communication technologies in IoT environment

Application layer

This layer is responsible for delivering application specific services to the user.It comprises of various use cases in which internet of things can be deployed.

Smart Industry

Smart City

Smart Energy

Smart Environment

Smart Retail

Smart Agriculture and farming

Smart Homes

Smart Health

The three layer architecture provide idea allow IOT but do not focus on farther details. Many more architectures are proposed in the literature. One is five layers architecture. The five layers are:

• Perception
• Transport
• Processing
• Application
• Business

Perception and application layer plays the same role as in 3 – layer architecture. The function of other 3 layers are:

Transport layer

Bridge between perception layer and processing layer transfers sensor data from perception layer to processing layer and vice versa threw wireless 3g Bluetooth, RFID and NFC.

Processing layer

This layer is also known as middleware. It stores analyse and process huge amount of data from transport layer provides services to lower layers. It comprises of database, clear computing and big data models.

Business layer

Manages whole IOT system, including appointment business and profit models.

Summary

• IoT is the most promising technology today
• It is combination of sensors, data processing and communication technology Applications are available and are open for many areas like smart cities, smart health, smart environment and many more

Suggested Reading:

1. Mobile Communication 2nd edition by Jochen Schiller, Pearson education
2. Mobile Computing by Asoke Talukder, Roopa Yavagal (Tata McGraw Hill)
3. “Wireless communication and networking” by William Stallings
4. Mobile Cellular Telecommunications — W.C.Y. Lee, Mc Graw Hill
5. Wireless Communications – Theodore. S. Rapport, Pearson Education
6. Reza B’Far (Ed), “Mobile Computing Principles”, Cambridge University Press.