Integrating wireless capabilities into products can be challenging for engineers who are not specialists in RF design, even when working with familiar technologies like Wi-Fi and Bluetooth. But those challenges become even bigger when the design project involves the new low-power cellular Internet of Things (IoT) protocols: LTE CAT M1 (CAT M1) and narrowband-IoT (NB-IoT). That complexity is driven by several interconnected factors:
- The complexity of cellular chipsets, which make them very difficult to work with for engineers other than true cellular experts.
- The need for a long list of niche expertise—including specific areas of embedded systems programming, expertise in time series-based architectures, and more—to build each segment of the connectivity pathways.
- The rigorous security and performance standards for the devices’ applications, including industrial and healthcare IoT use-case implementations.
- The hurdles involved in cellular certification, which are more rigorous than typically involved in products using other protocols.
- The sheer newness of the cellular IoT protocols, which means fewer resources and knowledgeable colleagues available for assistance.
Even a highly skilled engineer who has experience working with other wireless technologies may be intimidated by the degree of difficulty of cellular IoT projects. However, because the advantages of CAT M1 and NB-IoT make them very attractive technologies for IoT deployments, product engineers must have a solid strategy to successfully navigate the world of cellular IoT. This article provides six key tips to help engineers get off to a successful start with CAT M1 and NB-IoT projects, while also avoiding missteps that might cause major setbacks.
Before we get to those, let’s talk about why CAT M1 and NB-IoT are poised to be a major focus of development pipelines over the next two years. Given how much cellular infrastructure currently exists around the world, cellular networks may seem like an ideal platform for IoT deployments. But the always-on architecture of cellular technology causes it to drain batteries at an impractical rate for IoT deployments. And even if energy usage is not an issue, the high cost of hardware and connectivity would be just as much of an impediment.
Those hurdles are removed with the aforementioned new low-power versions of LTE technology—CAT M1 and NB-IoT. They have battery lives of up to 10 years and a cost structure that’s ideal for IoT deployments. One emphasizes extending battery life and minimizing cost, while the other provides more flexibility and performance when engineers want to prioritize those at the expense of a bit of battery life and cost efficiency (interested readers can get a deeper dive into the attributes of CAT M1 and NB-IoT in a recent white paper written by Laird Connectivity).
The bottom line for an engineering team is that the remarkable battery life, ubiquitous network infrastructure, and low total cost of ownership (TCO) mean that these two protocols will soon likely be the centerpiece of IoT design projects. Here are a series of tips to help you navigate your first cellular IoT projects:
1. It’s not either/or with cellular IoT: One of the questions often asked regarding CAT M1 and NB-IoT surrounds which protocol a team should choose at the beginning of a given project. That binary choice is common with other wireless protocols in which one must choose a path early in the development process and stick to it. But it’s not an either/or choice with cellular IoT. The choice will likely be both.
CAT M1 and NB-IoT are designed to coexist in IoT deployments, allowing engineers to make a choice at the device level (to suit that device’s job) rather than to make a one-size-fits-all choice for the entire implementation. This gives engineers a great deal of flexibility and does not mean that a high-stakes decision must be made early in the planning process.
2. One of your most important decisions involves the cloud: While you don’t have to make an either/or choice about CAT M1 and NB-IoT early in the planning process, you do have some important early decisions to make regarding cloud connectivity. How you use the cloud significantly impacts operational costs for your deployment.
Yes, NB-IoT and LTE-M are designed to make data costs manageable for IoT deployments, but there are still data costs, and those costs can be high if you choose to deliver large amounts of data from your device back to the cloud. This is because data-transfer rates from these two protocols are purposefully designed to not be robust. That means sending a large amount of data to the cloud may require long transfer sessions that rack up lots of data costs. To avoid that, engineering teams should carefully plan what data the device reports, how much of it truly needs to be fed back to the cloud, and how data processing and storage are handled by the device versus the cloud.
3. Your carrier matters: One of the misconceptions raised by engineers considering cellular IoT is that the carrier decision comes later in the process. It matters earlier on because carriers have networks with differing coverage areas, technical specifications, performance, and customer offerings for IoT deployments. Evaluating carriers to pinpoint the one that best fits your needs is critical because that decision drives many engineering choices made during the rest of the design process. And the right fit is dictated by the specifics of your deployment, including geographic location, proximity to nearby towers, IoT data plans offered by competing carriers, and other criteria.
4. The earlier the better regarding discussions about certifications: Cellular certifications are notoriously difficult, not only because of their rigor, but because so many organizations are involved, including the carriers, telecommunications industry organizations, and national/international governmental agencies. Unexpected hurdles with these certifications can cause significant delays that could easily derail a cellular IoT project. Thus, engineering teams should pay close attention to the pre-certification of components, modules, and antennas that later expedites the certification process. Having partners with experience in cellular testing and certification also can be invaluable in keeping projects on time and on budget.
5. Think carefully about antennas: Literally thousands of antennas are available to choose from for RF design projects, which means there are thousands of potentially ill-suited antennas for a given project. Selecting the right antenna is always hard, but it’s particularly hard for cellular projects.
One reason for the complexity is that frequency range must match the carrier you work with, while also meeting the performance needs your team mapped out. You may also want to deploy other wireless technologies like Bluetooth as a complement to CAT M1 and NB-IoT, which means you also must look at antennas that have multiple bands with performance characteristics that meet your needs. As with all antennas, pre-testing is a key step to ensure that the antenna performs as desired before you make a major commitment.
6. PSM and eDRX are the keys to ultra-long battery life: For cellular IoT devices to achieve the decade-long battery life that makes CAT M1 and NB-IoT so attractive, engineering teams must effectively manage sleep/wake cycles to optimize performance versus battery life. This includes effectively using deep-sleep modes that conserve the most energy.
Engineers should understand how to utilize the PSM and eDRX power-saving capabilities during the design process, including when to use deep-sleep modes, how to orchestrate devices that have varying sleep settings, and how to coordinate that with complementary technologies like Wi-Fi and Bluetooth, which have different sleep/wake cycles as part of their battery-management capabilities (those interested in a deeper discussion of PSM and eDRX can read the previously mentioned white paper).
Cellular IoT holds exciting possibilities for ultra-low-power wireless device networks that can take advantage of the readily available cellular connectivity. It’s a complicated logistical process with timetables and lots of moving parts, but the payoff is significant if you address the items in the checklist as part of your planning process. I hope this set of best practices and caveats proves to be a helpful tool for your first few design projects with these exciting technologies.
Matt Stergiou is the senior product manager for IoT at Laird Connectivity Solutions.