Back in August we wrote an article answering the question on everyone’s mind: What is the internet of things? We spoke about some of the embedded systems used in IoT, and how it can transform a number of industries. As the internet of things has grown, it has defined itself by the distinct elements that are contained within an IoT deployment. We will now take a look at what makes up an internet of things system and provide a graspable understanding of the IoT.
In its article Green Internet of Things for Smart World, the Institute of Electrical and Electronics Engineers (IEEE) gave an overview of IoT and all of its elements. The IEEE gives five main elements of an IoT setup as shown in the image below, and gives a description for each of them:
Identification plays a crucial role in naming and matching services with their demand. Examples of identification methods used for the IoT are electronic product codes (EPC), ubiquitous codes (uCode), etc.
Sensing is for collecting various data from related objects and sending it to a database, data warehouse or data center. The gathered data is further analyzed to perform specific actions based on required services. The sensors can be humidity sensors, temperature sensors, wearable sensing devices, mobile phones, and many others.
Communication technologies connect heterogeneous objects together to offer specific services. The communication protocols available for the IoT are: Wi-Fi, Bluetooth, IEEE 802.15.4, Z-wave, LTE-Advanced, Near Field Communication (NFC), ultra-wide bandwidth (UWB), LPWAN and emerging standards.
Computation, the hardware processing units, like microcontrollers, microprocessors, system on chips (SoCs) or field programmable gate arrays (FPGAs), and software applications perform this task. Many hardware platforms like Arduino or Raspberry PI are developed and various software platforms are utilized. The cloud platform is a particularly important computational part of IoT, since it is very powerful in processing various data in real-time and extracting all kinds of valuable information from the gathered data.
The services in IoT can be categorized into four classes: identity-related services, information aggregation services, collaborative-aware services and ubiquitous services. Identity-related services lay the foundation for other types of services, since every application mapping real world objects into the virtual world needs to identify the objects first. Information aggregation services, gather and summarize the raw information which need to be processed and reported. The obtained data are further utilized by the collaborative-aware services to make decisions and react accordingly. Ubiquitous services are for offering the collaborative-aware services to anyone on demand, anytime and anywhere.
Semantic means the ability to extract knowledge intelligently so as to provide the required services. This process usually includes: discovering resources, utilizing resources, modeling information, recognizing and analyzing data. The commonly used semantic technologies are: resource description framework (RDF), web ontology language (OWL), efficient XML interchange (EXI), etc.
A simpler answer to what elements make up the IoT are three categories that have been around for a long time:
- Hardware
- Software
- Cloud
Common elements found in IoT hardware include:
- Low energy sensors
- Communication services – gateways, modems, routers
- Touch screens and battery support/power
Hardware trends include decreasing the size of devices and moving to a system that does not require battery. Once data from sensors is collected, that information needs to be filtered and sent to the end user. That may be a consumer, a commercial or an industrial user. It may also be another device in the M2M workflow. Once the information is provided, it can be presented in visualizations on mobile devices.
Software is slowly starting to take over the spotlight from hardware. Software is loaded with APIs and SDKs that let you build IoT solutions. Software products give enterprises the ability to see real-time metrics as well as data pre and post-cloud analysis.
And of course, there is cloud. Cloud solutions allow large, complex processes to be completed away from the device. Gathered data is transmitted to a cloud-based service where the information coming in from the IoT device is aggregated with other cloud based data to provide useful information for the end user. The data being consolidated can be information from other internet sources as well as from others subscribing with similar IoT devices. Most often, there will be some data processing required to provide useful information that is not necessarily obvious in the raw data, according to Intelligent Product Solutions.
Irena Bojanova, member of the IEEE member society, posted a graphic on its site mapping out IoT categories and the elements present within each:
| Sensors | Sense the physical environment |
| Actuators | Affect the physical environment |
| Virtual Objects | Electronic tickets, Agendas, Books, Wallets |
| People | Ex.: Humans can control the environment via mobile apps |
| Services | Ex.: Cloud services – can be used to:
•  Process big data and turn it into valuable information •  Build and run innovative applications •  Optimize business processes by integrating device data. |
| Platforms | Type of middleware used to connect IoT components (objects, people, services, etc.) to IoT. Provide numerous functions:
•  Access to devices •  Ensuring proper installation/behavior of device •  Data analytics •  Interoperable connection to local network, cloud or other devices. |
| Networks | IoT components are tied together by networks, using various wireless and wireline technologies, standards, and protocols to provide pervasive connectivity. |
