Understanding the Building Blocks of IoT

The Internet of Things (IoT) is changing how our world works – from smart homes and wearables, to retail and banking. Everything is accessible on our smartphones and all our devices are connected to the Internet. This is urging enterprises to transform radically as customer needs are constantly evolving.

A study by Cisco estimates that there will be 50 billion Internet-connected devices by 2020. Today, there are about 20 billion physical objects or ‘things’ connected to the network, embedded with sensors, software and network connectivity. It is changing the way we do business and revolutionizing business verticals like automotive, healthcare, banking, energy, aerospace, manufacturing, retail, security and surveillance, smart cities, environment and logistics.

Collecting and exchanging data forms the foundation of the IoT. In the last decade, we managed to raise the value of ‘things’ into responsive tools by adding the ability to share data. It has elevated plain objects from its normal status into an IoT architecture and added a new layer to digitalisation.

Industrial IoT is a multi-billion-dollar opportunity for enterprises. Here are the four key aspects of an IoT network that will help you transform and solve your business needs:

  • ‘Things’

The bottom layer of IoT architecture comprises of the device layer. It is key to choose the right hardware and peripherals, along with necessary sensors to meet your business need. Devices can be of various types, but to qualify as an IoT device, it must have some communications that directly or indirectly connects to the Internet. Some devices need no operating systems. For instance, wearables, health and fitness monitoring devices, watches, and even human implanted devices have no operating system. On the other hand, industrial sensor devices may run on a 32 or 64 bit computing platforms. In many cases, it could be smartphone which is not just a phone but also an IoT object itself with sensing capability like accelerometer, proximity, gyroscope, magnetometer, barometer, light, touch, camera, thermometer, heartbeat etc. This is a classic example of optimized hardware package with board, chipset, sensors, modules and application ready APIs. The major players in the device layer space are ARM mbed, Atmel, Texas Instruments, Freescale, Qualcomm, Ericsson, Arduino, Raspberry Pi, Beagle etc.

 

  • Connectivity

Connectivity in the IoT network is often wireless, and uses many solutions like Wi-Fi, ZigBee, GPRS, 3G, 6LoWPAN and Bluetooth. Each of these solutions have pros and cons. ZigBee requires very low power though has limited range, while 3G has good coverage but is expensive both in terms of hardware and usage. You can choose the right one for your enterprise based on your business need.

  • Cloud/ Fog Platform

Connected devices generate a large volume of data, and this data gets transferred, processed and stored. If there is a need for historical data to be stored, datacenters must be scaled accordingly to handle this. The real-time nature of IoT places a huge demand on the processing capabilities of a data broker. Quantum computing can play a major role in addressing this issue. The reality is that you’ll need a highly robust datacenter to handle all the device data coming in.

In industrial IoT, some of the major challenges are faced due to large data volumes, limited network bandwidth, high network latency and poor network reliability. To address these problems, Cisco has introduced the concept of Fog Computing by creating a computing layer in the network called the ‘Fog’. It is a computing service/application which runs at the edge, operating closer to the devices. Though fog is not the replacement for cloud, it enables data filtering, faster decision making at the edge. Data is eventually also sent to the cloud for post processing and analysis.

 

  • Application

Sensor enabled devices collect both structured and unstructured data, and requires application specific analytics to make it meaningful and convert it into actionable decisions. Most of the IoT cloud solution providers provide an IoT analytics platform that makes it easy for developers to embed platform components into their applications, saving time and money for building the components themselves. To receive and analyse data pouring in from devices, the right application needs to be developed and customised. Such applications will enable businesses to analyse data in near-real time, and make automated decisions. For the system to understand your customers’ business needs, an enterprise must choose the right IOT technology stack from the above-mentioned choices — things, connectivity and cloud platform.

IoT is an evolutionary path in a new era of digital transformation. It has the potential to radically change the future and is even termed the fourth industrial revolution due to the impact it is bound to create. Designing IoT and all its aspects is a very important factor. The result of this phase will play a huge role in deciding the future of enterprises.

Comments

comments

Anupam Wahi
Anupam Wahi

Anupam Wahi, Senior Vice President, Communication Engineering at Incedo Inc.