Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

What are the components of an IoT product?

Let’s look at the different technology components that make up an Internet of Things product.


Sensors

Illustration of dots (representing sensors) connected to each other. Underneath is a photograph of Mischa Dohler in a speech bubble, indicating that the following text are his words.

First, we need things we want to connect. These could be sensors, actuators, robots, whatever can be connected. The most developed technology field here is pertaining to sensors. There are many different sensors available today. Whether you want to measure temperature, humidity, light, noise, pollution, pressure, torsion, tension, acceleration, position, images, magnetic fields, electric fields, etc, there is a large array of sensors to choose from. They used to be large, bulky and power hungry. Today they are invisible and energy efficient, whilst maintaining a high measurement precision. The technology field of actuators and robots, whether large or small, is still evolving. But even here, miniaturization is trending and hopefully the Internet of Things by 2020 will be a healthy mix of sensors measuring things and robots acting on the insights.


Connectivity

Illustration of lines (representing a wireless network). Underneath is a photograph of Mischa Dohler in a speech bubble, indicating that the following text are his words.

Second, after you have chosen the sensor or actuator, you will need to connect it reliably to the Internet. Since wireless connectivity is central to the success of the IoT, we will discuss this in great depth in Week 2. We will establish that connecting all objects in the long-term future with cellular mobile phone technology is probably the best business proposition since it is a technology which ensures availability, reliability and viability … all ingredients which we need for a successful IoT uptake.

Low Power Wifi, a new system, is also a great contender since Wifi already enjoys global coverage today. Furthermore, an exciting class of Low Power Wide Area networking technologies is also emerging, diversifying the connectivity portfolio. On the other hand, whilst the initial technology of choice, known as Zigbee, is suitable for sporadic short-range connections, it is surprisingly unfit to meet the demands for a viable Internet of Things at global scale. So watch out when building a future-proof IoT business!


Platform

Illustration of a cloud (representing a platform). Underneath is a photograph of Mischa Dohler in a speech bubble, indicating that the following text are his words.

Third, the collected data needs to be stored and processed somewhere. Known as IoT platforms, these are typically cloud-based infrastructures which:

  1. receive and send data via standardised interfaces, known as API;
  2. store the data; and
  3. process the data.

Many commercial platforms are available today where I would recommend going for scalable cloud solutions in a software-as-a-service model. There is no need to have a fully-fledged backend when you only support 10 sensors at the beginning of your stellar IoT career, whilst you want to use the same platform to support millions of daily sensor readings some years after your IoT product has scaled massively.


Analytics

Illustration of a graph (representing analytics). Underneath is a photograph of Mischa Dohler in a speech bubble, indicating that the following text are his words.

Fourth, some data analytics needs to be applied to the data as the value is not in the raw bits and bytes, but rather in the insights gathered from them. Big data analytics tools are generally available today, which stretch from simple statistical tools to more sophisticated machine learning approaches, with deep learning being the latest trend. Think of statistical tools finding you the known knowns in the data; machine learning finding the known unknowns; whilst deep learning is able to find the unknown unknowns.


User Interface

Illustration of a graph (representing analytics). Underneath is a photograph of Mischa Dohler in a speech bubble, indicating that the following text are his words.

Finally, an important component is how the data is presented to the final users. Make sure your IoT product has a very appealing user interface, both web based as well as smart phone or tablet based. I found that it is often a very sexy front-end which convinces clients to buy into the IoT journey. Fortunately, there are many open-source as well as paying front-end products available today.



In summary, the IoT data undergoes a long and fascinating journey. There is a data up-stream from the sensors in the field, via the wireless connectivity, into the IoT platforms. Then, there are data mash-up opportunities within the platforms allowing us to leverage Big Data opportunities. Finally, there is a data down-stream from the platforms back to the actuators in the field or some beautiful frontends on your computer screen or smart phone.

Share this article:

This article is from the free online course:

The Internet of Things

King's College London

Contact FutureLearn for Support