Whither Internet of Things?

In a nutshell, Internet of Things (IoT) is alive and well. I read an article a few days ago in the Korean Times which reports on the development of an IoT-based device in which an alert can be sent to you wherever you may be located, if your car is about to be stolen. Useful, certainly; though a technology that prevents the car from being stolen at all will be much more valuable. At any rate, rather than go through the details of the invention, I thought it would be better to write about some basics of IoT, with a focus on the state-of-the-art of this magical technology.

IoT, on which this column has written a lot about - please refer to three articles on this topic starting from that in the 11 May, 2015 issue of Daily Trust - has been considered by many to be the next big disruption in IT. I have crudely defined IoT as the technology associated with the embedding of a device into physical objects and communicating with the device across the Internet. The devices include those for sensing, such as GPS trackers, switches, temperature probes, cameras, and those for actuating, such as valves, bulbs, and locks - most of which communicate using low-power radios, Wi-Fi, and cellular means.

It is important to remember that the Internet is the medium through which data is transmitted. An example given in one the articles, data centers - those infrastructures that house high-performance servers in the thousands or more in one location are voracious energy consumers, to the point that almost half of the lifetime costs of data centers come from power consumption. Thus, the need to manage energy in those installations is dire. Some manufacturers have developed sensors that track the power usage by each server and deployed software for balancing computing loads and temporarily decommissioning servers and storage devices that are underutilized.

I have also previously described in this column the (IOT)-enabled Predictive Manufacturing (PM), wherein you manufacture hardware that you sell to your customers, but during the manufacturing process, you plant small gadgets (RFID, sensors, video monitoring, cameras, remote information distribution capabilities, and actuators) that will monitor the hardware and create data for you on the state of your hardware as it journeys out of your manufacturing floor into the world. (RFID stands for radio frequency identification waves.) Data collected on the hardware - such as temperature, humidity, location, some diagnostics on the state of the hardware - can be communicated instantaneously to you, the manufacturer, via some cloud interactions. Processing the collected data allows you to assess the state of the hardware, communicate with it, and remotely fix impending problems with it, even before the problems occur! By this means, you prevent a downtime at the customer’s site.

An easy-to-read description of the “state-of-the-art” of IoT is provided in a July 2016 paper by T Santhi Sri and his co-authors from Siddhartha Institute of Technology in India. This reference describes the four layers that constitute IoT: Application layer, Middleware layer, Network layer, and Perception layer. Acquiring information from the physical layer, using for example, sensors or recognition technologies as described above, occurs in the physical layer. A specific example is determining the temperature of the engine in your car. Sensors of different kinds: sound, smoke, and vibration can be used to identify the physical object, as can RFID and bar codes.

The collected data is communicated through some network to the system that processes the data. As Santhi Sri points out, “Telecommunication network acts as core host network which communicates between sensor and transmission network, such as WiFi, WiMAX technologies, and with core telecommunication network such as 2G, 3G, 4G, etc. An addressing scheme like IPV6 to uniquely identify network devices is crucial.” The protocol or the set of rules and regulations required for communicating in IoT - is different from the standard HTTP and TCP/IP, owing to the presence of devices and other constraints.

One of the issues that one has to contend with in IoT involves handling the huge volume of collected data, which will probably be communicated to you via the cloud, to take advantage of massive storage capacities available in typical data center installations. In expansive deployments, such as those that collect and analyze timeinstantaneous data, an exorbitantly huge amount of data will need to be processed (analyzed) for business or manufacturing intelligence. The bandwidth required for transmitting large data across networks will also have to be contended with.

There is a lot risk associated with IoT in the area of security. For example, transmitted data could be intercepted, while standard security mechanisms do not directly apply to IoT because of heterogeneity. Storing sensed and transmitted data at the back end should enhance data integrity management. A satisfactory resolution of the foregoing issues will certainly be necessary for you to access the benefits accorded by IoT.