Most IoT projects use a simple formula:
- An object you want to monitor, track, or control
- Hardware or devices such as sensors and gateways to enable the collection and sending of data
- Connectivity to send that data to the right place
- Platforms and portals to centralise the data and turn in into actionable analytics
Sounds pretty simple. But it becomes a little more complex when you consider that each of area of an IoT project has no “catch all” list of criteria which is totally correct.
Just like with connectivity options, there’s a whole roster of hardware contenders that boast their own strengths and weakness. Whether it’s a sensor in a drain sending water level data to councils, or a gateway connecting to multiple networks to transfer gigabytes of data per day, the requirements for each project are extremely varied. And it’s worth putting in the extra effort to get that deployment right from the get-go—hardware can be at the heart of the project, influencing every other choice in the formula from connectivity to portals.
Here are some of the essentials to keep in mind when matching a device to your project.
Can you count on the IoT device in your time of need? Reliability is generally an important factor, but in some cases it’s absolutely critical; for example, in solutions within healthcare or emergency services.
Depending on your project’s scale and operational structure, it may be a serious hassle to keep their battery life topped-up—especially in the case of large scale deployments that you can’t just plug into the nearest outlet.
In these cases a large battery size would obviously cover all the power requirements, but at the expense of weight and physical footprint. To save precious battery life, you might therefore look at how often the device is sending data, reducing it to a regular heartbeat rhythm instead of constant streaming.
And if you do go for battery power, you’ll need to ask if the device IS easily accessed for battery replacement; and, more importantly, are there measures in place to alert you when they’re almost out of power?
When things go wrong
You’ve deployed your solution and everything’s going great. But then one of your devices stops working. If you’ve got GPS enabled, you might be able to find that device and manually find out what’s going on—but what if the GPS has gone down too?
You’ll therefore need to consider how you’re finding, diagnosing, and solving problems when they do occur—and how much that’s going to cost you from an operational perspective. Do you have your own team to support the hardware internally, or is there a tried and tested support structure to keep the device estate running?
If a solution’s expected life cycle is 10 years, you’ll need to keep on top of a few things:
Keeping things updated:
Is the firmware upgradable? And how do you update it?
IoT deployments tend to be somewhere between the hundreds, thousands, millions, or somewhere in between. Those deployments are typically spread over a large area, with some solutions designed to be constantly on the move. So going on-site to push out an update manually often isn’t feasible. In these instances devices that use over-the-air updates are a lifesaver.
Running out of storage:
Will it be able to hold enough data for the lifecycle? Or should you ensure the device can ship that data off to the cloud instead?
Keeping up with connectivity:
Does the device support the newest WiFi or bluetooth standards, which will be ubiquitous in a few years? Is it ready for technologies such as NB-IoT and LTE-M, even though these haven’t been rolled out yet in some countries?
Improper planning for connectivity has already caused problems for some devices: the 2G shutdown has already commenced in some countries, leaving 2G network-only devices unusable. Simply ensuring that those devices also had LTE capability on deployment would have saved a bunch of hassle and cash.
The connectivity conundrum
Does your solution require a particular type of connectivity? If so, you’ll need to ensure your device can run with your particular choice. The usual suspects are WiFi, bluetooth, LPWAN, or cellular. Each method has its pros and cons—you can check our connectivity roadmap for some pointers on how to choose the right connectivity. (https://pangea-group.net/2018/10/04/picking-perfect-iot-connectivity/)
Some typical considerations might include distance, location, and whether your deployment being online all the time is mission critical.
All in all, there’s a plethora of devices to pick from and and we’ve only scratched the surface in this post. You’ll need to start by examining your project specs and mapping out your requirements for each element of the IoT formula. In doing so, you’ll come closer to pinpointing which IoT device is best able to bring what your project needs to the table.