In the recent times we are witnessing a rapid evolution of technologies besides the creation of new ones. The rapidity of the change results in several complexities, while highlighting the need to be fleet-footed in running with the evolution.
Along with the evolution of the technology, the industry too has been evolving. From Industrial revolution 1.0 we have seen the global transformation to Industrial revolution 4.0. At each stage, the complexity of managing the manufacturing process has increased, while the management itself has become more multi-faceted.
While Industry 3.0 was marked by the advent of computation and automation of processes and activities, several activities that required monitoring and taking actions based on observations could not be significantly automated. The physical world and the digital world remained separate with a manual intervention being required to connect the two.
Industry 4.0 refers to the intelligent networking of machines and processes for industry with the help of information and communication technology. The seamless connectivity of the digital world to physical world is achieved. One of the most significant developments that has aided this connectivity has been the growth of Sensor networks and Artificial Intelligence. The sensor network along with other technologies governing communication, automation and controls has resulted in the growth of a new domain, “The Internet of Things.”
The Internet of Things, popularly known as IoT, is the network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment. There are two parts to the IoT – namely, sense or interact in a manner to connect the physical world with the digital world, and the reverse, of the interaction of the digital world with the physical world.
Significant domains impacting IoT
The physical world is marked with measures such as mechanical, thermal, electrical, and so on, while the digital world is a world of 0s and 1s. The other significant difference is the location – digital world seldom exists at the location of the physical world. Together, a lot of technologies from different domains become involved in enabling the realms of IoT. So, what are some of these significant domains?
Very broadly, we can categorize the discussions into the following areas:
- The domains impacting the sensor and actuator devices
- The domains impacting the transfer of data from physical location to the digital location
Data handling and digital processing
Let us look at these areas closely and identify the impacts that different domains have on IoT.
Sensors and Actuators
As we have seen, the most significant transformation in IoT is that of the physical measure to digital form. This transformation is achieved by the group of devices called “Sensors.” These sensors operate between two domains – the physical domain on one side and the electronic domain on the other.
At the other end of the activities related to IoT, once the digital processing is completed, the instruction from the digital world is transformed into actions in the physical world. This is achieved by devices called “Actuators.” Actuators functionally are like Sensor, though playing the reverse role. In device design and construction, they are quite different.
As we can expect, every sensor is some form of an electronic device that leverages certain unique properties of materials. For example, a temperature sensor detects the change in temperature and provides an electronic output. For achieving this, certain thermal properties of materials is leveraged – materials that show a definite relationship between the temperature sensed and the electrical output in the form of voltage or current. Similarly, we have pressure sensors, that may leverage the property of “piezoelectric effect” to sense changes in pressure, for example.
In all these cases, though we deal with different physical property on one side, the design and fabrication of sensors is all about exploitation of different electronic properties of semi-conductors. Thus, there are two significant domains involved:
Material Science – Studying the materials, their properties and how the properties vary with changes in the physical conditions.
Electronics – Design and fabrication of integrated circuits with the semiconductor materials, including amplification of the observed changes and taking the generated electrical output through leads to measuring instruments.
Communication link
Having transformed the physical activity to an electrical form, it is important to communicate this information to a convenient location. In every case, the location of the physical activity will not be suitably located for performing digital processing and transformations or for storing. The information must be transmitted over communication links to the appropriate location.
To make the sensor devices communicate, they must connect themselves to a network. To achieve the connectivity, we would require a controller which can control many such devices and perform other management functions on the devices. When communicating, data from multiple such sensor devices may have to be aggregated. A microcontroller will be required at a site in the vicinity of the sensor devices to be able to control and construct the data in a communicable form.
The data is then carried from the controller to the data processing unit, which could be in a different location. This calls for appropriate communication technologies spanning over the spectrum of Local area to Wide area networks to be able to reach the required distances.
In short, we are dealing with the following domains:
Sensor networks – Communication from sensor to sensor and to microcontroller.
Microcontrollers and edge systems – To enable data cleansing, aggregating, and packaging near location of the sensors
Communication networks – Carry the packaged data to the remote site for further processing
Data handling and processing
Having successfully transmitted the data to the remote site, the data received is to be stored, processed and meaningful information extracted. Though the size of each data packet containing a reading from a sensor is small, typically any IoT application accumulates enormous number of such readings. The number of data records received will be large – typically it could be as high as several million records being received daily. Besides, the data from different sensors and sources of data could be quite different. Thus, we need a mechanism to store and process large volumes and varieties of data. Storing and processing such data takes us to the realms of Database management and Big Data.
To extract relevant information, the data processing would involve performing analytic operations. These analytic algorithms could range from quite simple ones to complex ones involving Artificial Intelligence with Deep-learning and machine-learning.
The storage and processing of data could be performed at a single remote location, or could be over a distributed network, in which case the computing facility is provided through “Cloud.”
This area will involve:
- Cloud storage and computing
- Server and storage technologies
- Database management systems and Big Data
- Visualization techniques
- Data analytics involving AI with machine-learning and deep-learning
Security
Securing and protecting the data purity, integrity and maintaining the confidentiality are important. As the complexity of the application and solution increases, so does the complexity of security of data. In IoT, with several devices, communication links and servers involved, it is imperative that every part of the network understands the security in every other part of the network.
Some parts of the IoT solution provide for strong security through appropriate levels of authentication and encryption. But, there are other parts which require external means to secure the network and the data therein.
Securing each part of the IoT solution involves its own technology, and hence, at a solution level, securing the solution would involve a group of different security solutions and technologies all of which must interact and coordinate and work together.
In Conclusion
An IoT solution is an interesting confluence of number of technologies and domains. It brings together different branches of engineering like Mechanical, Material Science, Electrical, Electronics, Communications, Computer Hardware Engineering, Data Sciences and Analytics, and so on. There is also plenty of physical and chemical concepts as under-currents running through the solution making it a domain that can captivate and challenge people of varied skills and interest in Science, Engineering and Technology.
Here, we have just had a peep into the different areas of engineering; IoT warrants a deeper dive and exploration of the various parts of the solution. An attempt in this direction will be made in subsequent articles to follow.
About the Author
Dr. Anand Lakshmanan is a Senior-Member of IEEE, a Technologist, and an organization builder. He is currently pursing advisory and consulting roles for EdTech companies, and member of curriculum committee and Senate in Institutes of National repute.