Bench Talk

There has been plenty written about the growth of the Internet of Things (IoT) and the more specialized Industrial Internet of Things (IIoT). IIoT is a driving factor in forthcoming the so-called fourth generation of the Industrial Revolution. There are many promises being made in the vision of the so-called Industry 4.0. In a nutshell, it promises a seamless marriage of the previous industrial revolutions and the information revolution. Industry 4.0 brings the efficiencies that the Internet helped enable from the Information Technology (IT) world and into the realm of Operational Technology (OT). Put more simply, a world where customized products are delivered to consumers at mass production costs. Of course, bringing the power of the internet isn't without challenges. All the bad things associated with the Internet—botnets, ransomware, cryptocurrency mining malware, denial-of-service attacks— come with all the positives.

No doubt that there are a lot of problems to be worked out in order to achieve the promises of this latest iteration of the Industrial Revolution. Additive manufacturing equipment, Artificial Intelligence (AI) powered robotics, digital twins, and the IIoT are amongst the still nascent technologies that are driving us towards a world of on-demand product customization at mass production costs. So, what are the roadblocks and what is being done to eliminate them?

A foundational element of this revolution is seamless communications. Through seamless communications, data can easily flow to and from customers, business systems, and manufacturing equipment. While the internet of today has helped to close the gap between customers, suppliers, and internal business systems at the front end, many organizations are struggling with making similar strides behind the scenes. With the massive amount of potential IoT endpoints, wired networking just isn’t feasible from a cost or infrastructure perspective. Wireless networking—especially for facilities with large square footages such as factories—has been efficient though not always reliable method to network such production facilities.

Wireless mesh networking is an oft-praised solution to the problems encountered in networking large areas, at least in theory. Historically the reality of many multihop mesh network implementations is that they suffer from high latency and low throughput. Especially as the number of nodes in the network increases. The amount of onboard RAM memory is a key a factor in determining how large a mesh network can scale. Each device holds a sort of map of the mesh network—sometimes referred to as a routing table. More devices in the network mean more relations between devices need to be tracked and thus the memory limitation.

However, over the last decade mesh network technology has continued to mature. In fact, it might be a surprise that IEEE incorporated a wireless mesh amendment (802.11s) into its 802.11—a.k.a. Wi-Fi—wireless specification back in 2012. Furthermore, products such as the Google Wi-Fi router use 802.11s to provide it’s mesh capabilities. Later this year embedded platform developer Particle.io is set to deliver a family of mesh network development platforms to the market.

A mesh network is typically comprised of three types of devices:

1. Endpoints: These devices send data to or receive data from a repeater. They are not involved in the transport of messages between other devices on the mesh network. Machinery and sensors are examples of endpoints.
2. Repeaters: These devices dynamically communicate with multiple endpoints, repeaters, and gateways. They can be endpoint themselves as well if they have sensors or actuators themselves. Primarily they are responsible for efficiently moving data across the mesh network.
3. Gateways: Gateways are devices that serve as the backhaul between the mesh network and external networks of the internet. They backhaul can use any number of communications methods including Wi-Fi or any number of cellular data standards.

In non-mesh networked IoT implementations where IoT devices require a callback to a web server for passing data amongst devices if the backhaul connection to the internet is down, then no devices can talk as to one another. But with a mesh network, devices can continue to operate without the internet connectivity. This helps to make mesh networks a safer and more reliable option for applications that prioritize geographically localized availability, such as factories or other industrial facilities.

That’s it for now, but remember to check back for parts two and three of this blog series. In part 2, we will discuss RESTful APIs and Moving Data Across the Internet of Things. Part three will focus on Edge Computing and the Internet of Things.