The term IoT was coined by Kevin Ashton in 1999. At that time, most of the data fed to computers was generated by humans; he proposed that the best way would be for computers to take data directly, without any intervention from humans. And so he proposed things such as RFID and sensors, which gather data, should be connected to the network, and feed directly to the computer.

Today IoT (also called the internet of everything and sometimes, the fog network) refers to a wide range of things such as sensors, actuators, and smartphones connected to the internet. These things can be anything: a person with a wearable device (or even mobile phone), an RFID-tagged animal, or even our day-to-day devices such as a refrigerator, washing machine, or even a coffee machine. These things can be physical things—that is, things that exist in the physical world and can be sensed, actuated, and connected—or of the information world (a virtual thing)—that is, things that aren't tangibly present but exist as information (data) and can be stored, processed, and accessed. These things necessarily have the ability to communicate directly with the internet; optionally, they might have the potentiality of sensing, actuation, data capture, data storage, and data processing.

The International Telecommunication Unit (ITU), a United Nations agency, defines IoT as:

You can learn more at https://www.itu.int/en/ITU-T/gsi/iot/Pages/default.aspx.

The wide expanse of ICT already provided us with communication at any time or any place; the IoT added the new dimension of ANY THING communication:

It's predicted that IoT as a technology will have a far-reaching impact on people and the society we live in. To give you a glimpse of its far-reaching effects, consider the following scenarios:

• You, like me, live in a high rise building and are very fond of plants. With lots of effort and care, you've made a small indoor garden of your own using potted plants. Your boss asks you to go for a week-long trip, and you're worried your plants won't survive for a week without water. The IoT solution is to add soil moisture sensors to your plants, connect them to the internet, and add actuators to remotely switch on or off the water supply and artificial sunlight. Now, you can be anywhere in the world, but your plants won't die, and you can check the individual plant's soil moisture condition and water it as needed.
• You had a very tiring day at the office; you just want to go home and have someone make you coffee, prepare your bed, and heat up water for a bath, but sadly you're home alone. Not anymore; IoT can help. Your IoT-enabled home assistant can prepare the right flavor coffee from the coffee machine, order your smart water heater to switch on and maintain the water temperature exactly the way you want, and ask your smart air conditioner to switch on and cool the room.

The choices are limited only by your imagination. The two preceding scenarios correspond to consumer IoT—the IoT with a focus on consumer-oriented applications. There also exists a large scope of Industry IoT (IIoT) where manufacturers and industries optimize processes and implement remote monitoring capabilities to increase productivity and efficiency. In this book, you'll find the hands-on experience with both IoT applications.

Just like the OSI reference model for the internet, IoT architecture is defined through six layers: four horizontal layers and two vertical layers. The two vertical layers are Management and Security and they're spread over all four horizontal layers, as seen in the following diagram:

The Device Layer: At the bottom of the stack, we have the device layer, also called the perception layer. This layer contains the physical things needed to sense or control the physical world and acquire data (that is, by perceiving the physical world). Existing hardware, such as sensors, RFID, and actuators, constitutes the perception layer.

The Network Layer: This layer provides the networking support and transfer of data over either wired or wireless network. The layer securely transmits the information from the devices in the device layer to the information processing system. Both transmission Medium and Technology are part of the networking layer. Examples include 3G, UMTS, ZigBee, Bluetooth, Wi-Fi, and so on.

The Service Layer: This layer is responsible for service management. It receives information from the network layer, stores it into the database, processes that information, and can make an automatic decision based on the results.

The Application Layer: This layer manages the applications dependent upon the information processed in the service layer. There's a wide range of applications that can be implemented by IoT: smart cities, smart farming, and smart homes, to name a few.

Information from the network layer is often managed with the help of IoT platforms. Many companies today provide IoT platform services, where they help not only with data but also enable seamless integration with different hardware. Since they function as a mediator between the hardware and application layer, IoT platforms are also referred to as IoT middleware and are part of the service layer in the IoT reference stack. IoT platforms provide the ability to connect and communicate with things from anywhere in the world. In this book, we'll briefly cover some popular IoT platforms such as the Google Cloud Platform, Azure IoT, Amazon AWS IoT, Predix, and H2O.

You can select which IoT platform is best for you based on the following criteria:

• Scalability: Addition and deletion of new devices to the existing IoT network should be possible
• Ease of use: The system should be perfectly working and delivering all its specifications with minimum intervention
• Third party integration: Heterogeneous devices and protocols should be able to inter-network with each other
• Deployment options: It should be workable on a broad variety of hardware devices and software platforms
• Data security: The security of data and devices is ensured

A vertical market is a market in which vendors offer goods and services specific to an industry, trade, profession, or other groups of customers with specialized needs. IoT enables the possibility of many such verticals, and some of the top IoT verticals are as follows:

• Smart building: Buildings with IoT technologies can help in not only reducing the consumption of resources but also improving the satisfaction of the humans living or working in them. The buildings have smart sensors that not only monitor resource consumption but can also proactively detect residents' needs. Data is collected via these smart devices and sensors to remotely monitor a building, energy, security, landscaping, HVAC, lighting, and so on. The data is then used to predict actions, which can be automated according to events and hence efficiency can be optimized, saving time, resources, and cost.
• Smart agriculture: IoT can enable local and commercial farming to be more environmentally friendly, cost-effective, and production efficient. Sensors placed through the farm can help in automating the process of irrigation. It's predicted that smart agricultural practices will enable a manifold increase in productivity, and hence food resources.
• Smart city: A smart city can be a city with smart parking, a smart mass transit system, and so on. A smart city has the capability to address traffic, public safety, energy management, and more for both its government and citizens. By using advanced IoT technologies, it can optimize the usage of the city infrastructure and quality of life for its citizens.
• Connected healthcare: IoT enables critical business and patient monitoring decisions to be made remotely and in real time. Individuals carry medical sensors to monitor body parameters such as heartbeat, body temperature, glucose level, and so on. The wearable sensors, such as accelerometers and gyroscopes, can be used to monitor a person's daily activity.

We'll be covering some of them as a case study in this book. The content of this book is focused on information processing and the applications being implemented on IoT and so we'll not be going into details of the devices, architecture, and protocols involved in IoT reference stacks any further.

IoT has connected things never previously connected to the internet, such as car engines, resulting in the generation of a large amount of continuous data streams. The following screenshot shows explorative data by IHS of the number of connected devices in billions in future years. Their estimate shows that the number of IoT devices will reach 75.44 billion by 2025:

The reduction in sensor cost, efficient power consumption techniques, a large range of connectivity (infrared, NFC, Bluetooth, Wi-Fi, and so on), and the availability of cloud platforms that support IoT deployment and development are the major reasons for this pervasion of IoT in our homes, personal lives, and industry. This has also motivated companies to think about providing new services and developing new business models. Some examples include the following:

• Airbnb: It connects people so that they can rent out spare rooms and cottages to one another, and it earns the commission.
• Uber: It connects cab drivers with travelers. The location of the traveler is used to assign them to the nearest driver.

The amount of data generated in the process is both voluminous and complex, necessitating a big data. Big data approach and IoT are almost made for each other; the two work in conjunction.

Things are continuously generating an enormous amount of data streams that provide their statuses such as temperature, pollution level, geolocation, and proximity. The data generated is in time series format and is autocorrelated. The task becomes challenging because the data is dynamic in nature. Also, the data generated can be analyzed at the edge (sensor or gateway) or cloud. Before sending the data to the cloud, some form of IoT data transformation is performed. This may involve the following:

• Temporal or spatial analysis
• Summarizing the data at the edge
• Aggregation of data
• Correlating data in multiple IoT streams
• Cleaning data
• Filling in the missing values
• Normalizing the data
• Transforming it into different formats acceptable to the cloud

At the edge, complex event processing (CEP) is used to combine data from multiple sources and infer events or patterns.

The data is analyzed using stream analytics, for example, applying analytical tools to the stream of data, but developing the insights and rules used externally in an offline mode. The model is built offline and then applied to the stream of data generated. The data may be handled in different manners:

• Atomic: Single data at a time is used
• Micro batching: Group of data per batch
• Windowing: Data within a timeframe per batch

The stream analytics can be combined with the CEP to combine events over a time frame and correlate patterns to detect special patterns (for example, anomaly or failure).

A very popular phrase among data scientists and machine learning engineers is "AI is the new electricity" said by Prof Andrew Ng in NIPS 2017, we can expand it as follows: If AI is the new electricity, data is the new coal, and IoT the new coal-mine.

IoT generates an enormous amount of data; presently, 90% of the data generated isn't even captured, and out of the 10% that is captured, most is time-dependent and loses its value within milliseconds. Manually monitoring this data continuously is both cumbersome and expensive. This necessitates a way to intelligently analyze and gain insight from this data; the tools and models of AI provide us with a way to do exactly this with minimum human intervention. The major focus of this book will be on understanding the various AI models and techniques that can be applied to IoT data. We'll be using both machine learning (ML) and DL algorithms. The following screenshot explains the relationship between Artificial Intelligence, Machine Learning, and Deep Learning:

By observing the behavior of multiple things, IoT (with the help of big data and AI) aims to gain insight into the data and optimize underlying processes. This involves multiple challenges:

• Storing real-time generated events
• Running analytical queries over stored events
• Performing analytics using AI/ML/DL techniques over the data to gain insights and make predictions

For IoT problems, the most used data management (DM) methodology is cross-industry standard process for data mining (CRISP-DM) proposed by Chapman et al. It's a process model that states the tasks that need to be carried out for successfully completing DM. It's a vendor-independent methodology divided into these six different phases:

1. Business understanding
2. Data understanding
3. Data preparation
4. Modelling
5. Evaluation
6. Deployment

Following diagram shows the different stages:

As we can see, it's a continuous process model with data science and AI playing important roles in steps 2–5.

A large number of cloud platforms with both AI and IoT capabilities are available today. These platforms provide the capability to integrate the sensors and devices and perform analytics on the cloud. There exist more than 30 cloud platforms in the global market, each targeting different IoT verticals and services. The following screenshot lists the various services that AI/IoT platforms support:

Let's briefly find out about some popular cloud platforms. In Chapter 12, Combining it all together, we'll learn how to use the most popular ones. The following is a list of some of the popular Cloud platforms:

• IBM Watson IoT Platform: Hosted by IBM, the platform provides device management; it uses the MQTT protocol to connect with IoT devices and applications. It provides real-time scalable connectivity. The data can be stored for a period and accessed in real time. IBM Watson also provides Bluemix Platform-as-a-Service (PaaS) for analytics and visualizations. We can write code to build and manage applications that interact with the data and connected devices. It supports Python along with C#, Java, and Node.js.
• Microsoft IoT-Azure IoT suite: It provides a collection of preconfigured solutions built on Azure PaaS. It enables a reliable and secure bidirectional communication between IoT devices and cloud. The preconfigured solutions include data visualization, remote monitoring, and configuring rules and alarms over live IoT telemetry. It also provides Azure Stream Analytics to process the data in real time. The Azure Stream Analytics allows us to use Visual Studio. It supports Python, Node.js, C, and Arduino, depending upon the IoT devices.
• Google Cloud IoT: The Google Cloud IoT provides a fully managed service for securely connecting and managing IoT devices. It supports both MQTT and HTTP protocols. It also provides bidirectional communication between IoT devices and the cloud. It provides support for Go, PHP, Ruby, JS, .NET, Java, Objective-C, and Python. It also has BigQuery, which allows users to perform data analytics and visualization.
• Amazon AWS IoT: The Amazon AWS IoT allows IoT devices to communicate via MQTT, HTTP, and WebSockets. It provides secure, bi-directional communication between IoT devices and the cloud. It also has a rules engine that can be used to integrate data with other AWS services and transform the data. Rules can be defined that trigger the execution of user code in Java, Python, or Node.js. AWS Lambda allows us to use our own custom trained models.

For the implementation of IoT-based services, we need to follow a bottom-up approach. For each IoT vertical, we need to find the analytics and the data and, finally, implement it in code.

Due to its availability in almost all AI and IoT platforms, Python will be used for coding in this book. Along with Python, some helping libraries such as NumPy, pandas, SciPy, Keras, and TensorFlow will be used to perform AI/ML analytics on the data. For visualization, we will be using Matplotlib and Seaborn.

TensorFlow is an open source software library developed by the Google Brain team; it has functions and APIs for implementing deep neural networks. It works with Python, C++, Java, R, and Go. It can be used to work on multiple platforms, CPU, GPU, mobile, and even distributed. TensorFlow allows for model deployment and ease of use in production. The optimizer in TensorFlow makes the task of training deep neural networks easier by automatically calculating gradients and applying them to update weights and biases.

In TensorFlow, a program has two distinct components:

• Computation graph is a network of nodes and edges. Here all of the data, variables, placeholders, and the computations to be performed are defined. TensorFlow supports three types of data objects: constants, variables, and placeholders.
• Execution graph actually computes the network using a Session object. Actual calculations and transfer of information from one layer to another takes place in the Session object.

Let's see the code to perform matrix multiplication in TensorFlow. The whole code can be accessed from the GitHub repository (https://github.com/PacktPublishing/Hands-On-Artificial-Intelligence-for-IoT) filename, matrix_multiplication.ipynb:

import tensorflow as tf
import numpy as np

This part imports the TensorFlow module. Next, we define the computation graph. mat1 and mat2 are two matrices we need to multiply:

# A random matrix of size [3,5]
mat1 = np.random.rand(3,5)  
# A random matrix of size [5,2]
mat2 = np.random.rand(5,2)  

We declare two placeholders, A and B, so that we can pass their values at runtime. In the computation graph, we declare all of the data and computation objects:

# Declare placeholders for the two matrices 
A = tf.placeholder(tf.float32, None, name='A')
B = tf.placeholder(tf.float32, None, name='B')  

This declares two placeholders with the names A and B; the arguments to the tf.placeholder method specify that the placeholders are of the float32 datatype. Since the shape specified is None, we can feed it a tensor of any shape and an optional name for the operation. Next, we define the operation to be performed using the matrix multiplication method, tf.matmul:

C = tf.matmul(A,B)

The execution graph is declared as a Session object, which is fed the two matrices, mat1 and mat2, for the placeholders, A and B, respectively:

with tf.Session() as sess:
    result = sess.run(C, feed_dict={A: mat1, B:mat2})
    print(result)

Keras is a high-level API that runs on top of TensorFlow. It allows for fast and easy prototyping. It supports both convolutional and recurrent neural networks, and even a combination of the two. It can run on both CPUs and GPUs. The following code performs matrix multiplication using Keras:

# Import the libraries
import keras.backend as K
import numpy as np

# Declare the data
A = np.random.rand(20,500)
B = np.random.rand(500,3000)

#Create Variable
x = K.variable(value=A)
y = K.variable(value=B)
z = K.dot(x,y)
print(K.eval(z))

In the coming chapters, we'll be learning different DL models and ML methods. They all work on data; while a large number of datasets are available to demonstrate how these models work, in this book, we'll use datasets available freely through wireless sensors and other IoT devices. The following are some of the datasets used in this book and their sources.

This dataset contains 9,568 data points collected from a combined cycle power plant (CCPP) in a course of six years (2006-2011). CCPP uses two turbines to generate power, the gas turbine and the steam turbine. There're three main components of the CCPP plant: gas turbine, heat recovery system, and steam turbine. The dataset, available at UCI ML (http://archive.ics.uci.edu/ml/datasets/combined+cycle+power+plant), was collected by Pinar Tufekci from Namik Kemal University and Heysem Kaya from Bogazici University. The data consists of four features determining the average ambient variables. The averages are taken from various sensors located around the plant that record ambient variables per second. The aim is to predict the net hourly electrical energy output. The data is available in both xls and ods formats.

The features in the dataset are as follows:

• The Ambient Temperature (AT) is in the range 1.81°C and 37.11°C
• The Ambient Pressure (AP) is in the range 992.89—1033.30 millibar
• Relative Humidity (RH) is in the range 25.56% to 100.16%
• Exhaust Vacuum (V) is in the range 25.36 to 81.56 cm Hg
• Net hourly electrical energy output (PE) is in the range 420.26 to 495.76 MW

Wineries around the world have to undergo wine certifications and quality assessments to safeguard human health. The wine certification is performed with the help of physicochemical analysis and sensory tests. With the advancement of technology, the physicochemical analysis can be performed routinely via in-vitro equipment.

We use this dataset for classification examples in this book. The dataset can be downloaded from the UCI-ML repository (https://archive.ics.uci.edu/ml/datasets/Wine+Quality). The wine quality dataset contains results of physicochemical tests on different samples of red and white wine. Each sample was further rated by an expert wine taster for quality on a scale of 0—10.

The dataset contains in total 4,898 instances; it has a total of 12 attributes. The 12 attributes are as follows:

• Fixed acidity
• Volatile acidity
• Citric acid
• Residual sugar
• Chlorides
• Free sulfur dioxide
• Total sulfur dioxide
• Density
• pH
• Sulfates
• Alcohol
• Quality

The dataset is available in the CSV format.

Air pollution poses a major environmental risk to human health. It's found that there exists a correlation between improved air quality and amelioration of different health problems such as respiratory infections, cardiovascular diseases, and lung cancer. The extensive sensor networks throughout the world by Meteorological Organizations of the respective country provide us with real-time air quality data. This data can be accessed through the respective web APIs of these organizations.

In this book, we'll use the historical air quality data to train our network and predict the mortality rate. The historical data for England is available freely at Kaggle (https://www.kaggle.com/c/predict-impact-of-air-quality-on-death-rates), and the air quality data consists of daily means of ozone (O3), Nitrogen dioxide (NO2), particulate matter with a diameter less than or equal to 10 micrometers (PM10) and PM25 (2.5 micrometers or less), and temperature. The mortality rate (number of deaths per 100,000 people) for England region is obtained by the data provided by the UK Office for National Statistics.

In this chapter, we learned about IoT, big data, and AI. This chapter introduced the common terminologies used in IoT. We learned about the IoT architecture for data management and data analysis. The enormous data generated by IoT devices necessitates special ways to handle it.

We learned about how data science and AI can help in both analytics and prediction generated by the many IoT devices. Various IoT platforms were briefly described in this chapter, as were some popular IoT verticals. We also learned about special DL libraries: TensorFlow and Keras. Finally, some of the datasets we'll be using throughout the book were introduced.

The next chapter will cover how to access the datasets available in varied formats.