Cellular technology may trigger IoT's next growth wave: Kaye

How engineers deliver connectivity to an IoT implementation can vary dramatically from project to project – even when IoT deployments involve the same use cases in seemingly identical settings.

For example, two sets of temperature sensors in two remote facilities may use similar Bluetooth sensors and have a similar RF deployment strategy because of the physical environment, but the two implementations need to use very different technology for delivering data from those sensors to the cloud. That is because wireless infrastructure varies so much from one location to the next, and that can force engineering teams to use two different connectivity strategies for what would otherwise be identical projects.

The first remote facility in the example above might have available Wi-Fi infrastructure as the means of data backhaul to on-premise servers or the cloud from the Bluetooth sensors. Meanwhile, the second facility may lack available access to that infrastructure and must rely on a different infrastructure overlay, such as longer-range technology like LoRaWAN for backhaul. Both scenarios still require the ability to bridge between the Bluetooth radio in the temperature sensors and the available backhaul connectivity infrastructure.

These differences in wireless infrastructure from one location to another – or even from spot to spot within a given implementation site – add a layer of complexity to every IoT project. Early on, the engineering team must assess what infrastructure is available, develop a connectivity strategy based on what is and isn’t available, and customize the IoT deployment to those circumstances. That requires a lot of time and effort. It would be a lot simpler if there were a single, uniform source of backhaul connectivity that was available everywhere.

As long as engineers have been working on IoT deployments, they have thought about cell networks as the data backhaul backbone for these projects. That is because cell connectivity is nearly ubiquitous, has significant geographic range, and penetrates buildings effectively to support interior deployments. IoT that leveraged cellular infrastructure could be deployed at nearly every spot on the map, and that infrastructure wouldn’t need to be built by the company deploying the IoT network. It sounded perfect!

Except that it wasn’t perfect. Two issues stood in the way. The high-performance nature of LTE connectivity driven by consumer devices (media rich phones and tablets) would drain the more limited batteries of wireless IoT devices like sensors in the blink of an eye unless they were wired for power or charged every night like a cell phone. And the cost of cellular connectivity for hundreds or thousands or tens of thousands of these IoT devices would be like paying the cell phone bill of a million teenagers. Those two drawbacks made cell infrastructure impractical as the source of connectivity for all but a very few IoT deployments, but two new versions of the cellular standard are designed to overcome those obstacles to broaden usage of cell connectivity for IoT.

The two new cellular protocols that were designed for IoT are LTE-M and NB-IoT. Here is a deep dive into both technologies in this white paper for those of you who are interested in more information about how they work. The key things to know, however, are that LTE-M and NB-IoT solve both of those drawbacks that prevented cellular from being practical in the past: energy usage and optimized connectivity for infrequent connectivity of IoT devices like sensors.

Unlike the always-on, battery-guzzling architecture of the version of cellular that is used by phones and tablets, LTE-M and NB-IoT were designed very purposefully to preserve battery life of IoT devices. Although they still use the same LTE network, they are built with an operational architecture that doesn’t seek to have a continuous connection with the nearest cell tower – one of the biggest battery drains for cellular-connected devices. LTE-M and NB-IoT combine that architecture with other battery preserving features that enable devices to have projected life spans of 10 years or more in the field. This solves the first of those major objections to cellular-based IoT.

LTE-M and NB-IoT also solve the data costs issue with an architecture that has operational costs that make them competitive with other types of connectivity by using the cell network only as-needed for delivering batches of intermittent data. This minimizes data costs compared to typical cell-connected devices. LTE-M and NB-IoT also remove the expense of building out network infrastructure from the customer’s and OEM’s cost structure. By using public cellular network infrastructure, the organization doesn’t need to deploy the network, own it or maintain it. Those costs are carried by the wireless carrier, which eliminates a source of operational costs for IoT deployments.

Another way that these technologies are optimized for IoT is that they do not require the physical radio and architecture that traditional LTE devices use – because IoT does not require the high level of processing, data throughput and processing – which keep the cost of these devices down.

By eliminating concerns about battery longevity and cost, LTE-M and NB-IoT allow engineers working on IoT projects to finally have the ubiquitous, uniform infrastructure that has been so coveted since the emergence of IoT. Anywhere that you have bars, you can have IoT that taps into the same digital infrastructure. And that is significant for a long list of use cases, including cold supply chain for food and vaccines, telehealth monitoring of remote medical devices and remote monitoring of industrial equipment – all of which have connectivity challenges that are solved by the ubiquity of cell connectivity.

For engineers who have worked on cellular projects, you are likely aware of the caveats for projects involving that technology. For instance, certifications can be arduous for any cell-connected device because of the additional certification hurdles for products with a cellular radio, including regulatory agencies (e.g. FCC/IC, etc.), network approvals (e.g. PTCRB/GCF) and carrier approvals (e.g. AT&T certifications). Working with pre-certified cellular IoT socket modems and an experienced wireless design partner can help navigate those steps in a way that accelerates the certification process.

Another important piece of advice for engineers working on cellular IoT projects is about how to successfully pair LTE-M and NB-IoT with complementary short-range technologies like Bluetooth. Bluetooth is a particularly attractive complement to LTE-M and NB-IoT because of the way Bluetooth can perform all of the short-distance device-to-device communication alongside LTE-M/NB-IoT doing the backhaul communication to the LTE network, only when needed. For IoT projects that combine cellular with other technologies like Bluetooth, care should be taken with how the multiple radios are co-located to ensure performance, avoid interference issues, and ensure certifications. This process can be easily navigated, though, with the right RF modeling, antenna selection and testing partner.

The two new versions of the cellular protocol have finally made that long-time dream about cellular IoT a reality, and I expect to see explosive growth in the number of deployments that utilize LTE networks as their data backbone. IoT can now be easily deployed anywhere that you can see bars on your phone, which is sure to open the door to an even larger wave of IoT deployments.

Jonathan Kaye is the Senior Director of Product Management at Laird Connectivity. In this role at the company, Kaye is a lead developer of Laird Connectivity’s embedded wireless connectivity solutions. He has more than 20 years of experience in the embedded wireless and product design field, including positions at EZURiO and Lever Technology before joining Laird Connectivity a decade ago.