72 degrees. 72 degrees. 72 degrees. 72 degrees. 96 degrees. 72 degrees. 72 degrees. 72 degrees….
This is what most IoT data looks like — the same information over and over and over again. What is useful here, and what is not? For those that follow our articles, you have long seen us discuss deriving context from data. This is really what matters. Not that it is 72 degrees for 9,999 out of 10,000 readings, but rather, why was that one reading different? That is the real question.
Many have argued against a massive onslaught of useless sensor data, given its impact on storage, bandwidth, power consumption and, of course, money. And to a degree, they are correct. Many great IoT use cases for predictive maintenance and asset tracking have been blocked based on these concerns.
So how have successful implementations seen their way over the wall of objections and reached deployment? There are many explanations, but they almost always come down to starting with a great business case, and then finding a practical technical solution.
Recent IoT Trends
From a technical perspective, we are seeing consistent trends across successful implementations:
- Successful IoT projects are increasing intelligence and analytics capabilities at the edge, with the ability to act locally and autonomously. This means they can answer the following questions:
- What information is contextually interesting?
- What action should be taken as a result?
- What should be communicated or shared?
- Is it urgent, or can it wait?
While traditional sensors based on the WiFi and Bluetooth Low Energy (BLE) protocols continue to gain traction, we are also seeing increased adoption of longer range solutions, such as LoRa, Sigfox, LTE-M and Narrowband IoT (NB-IoT).
Let us talk about some of these emerging standards. First and foremost, they have common attributes:
- They are low-power (with battery life measured in months, or even years)
- They are relatively long-distance (think kilometers vs. meters)
- They are designed for moving small quantities of data (think bits and bytes vs. megabits and gigabytes.)
In many uses cases, a meager 50 megabyte per month data plan will suffice. Think about that in comparison to your monthly smartphone data plan consumption, where you may be blowing through 5-10 gigabytes (or at least your kids are).
By contrast, WiFi is a short-range, power-hungry protocol that is well-suited to high bandwidth applications. BLE, on the other hand, is a short-range, power-sipping protocol that is a better fit for low bandwidth use cases. Both also have low-cost licensing fees.
In an effort to further extend battery life and decrease unit costs, devices leveraging the newer protocols are typically designed to be relatively “dumb” — with limited onboard analytics or event processing capabilities — and are intended to work either independently or with a gateway.
For example, consider the “precision agriculture” use case where farmers are leveraging sensor data to maximize crop yields — and by extension, their profits. To do so, they need to better understand what and when to plant, when to water, when to fertilize and when to harvest. To make this possible, they are deploying sensors to measure temperature, moisture and nitrogen levels in the soil. These sensors may collect data continuously, or they may sleep and wake up hourly. They may communicate data as it is collected, or, for purposes of managing data costs and/or battery life, the data may be batched and sent, store-and-forward style, on a daily or weekly basis.
These sensors need to be strategically deployed around fields in order to provide statistically useful data. While the quantity of data collected is quite small per sensor, when analyzed at the aggregate level, it can become quite large, and very powerful. The greater the density of sensors, the more granular the predictions and recommendations of the analytical models. Each reading in this case is simply a set of data points that needs to be fed into machine learning and analytical models to make determinations. Those models are frequently supported by external data sources, such as the National Weather Service, which further help determine the best course of action to maximize crop yields.
Practical and Cost-Effective Data Capture
While the potential results of acting on the basis of this data-driven analysis are readily apparent in this use case, the biggest technical question remains: How can you capture the data in a practical, cost-effective manner?
Other key questions to consider: What density of sensor data is statistically significant enough to make the data analysis compelling? Does the currently available technology make the analysis both practical and cost-effective? How quickly can the information be responded to? Will the workers in the field be able to respond to and act on such a granular/micro-level to plan and execute their daily activities quickly enough to make the endeavor worthwhile? These are good questions for a team of farmers, engineers and data scientists to decide.
A big part of the answer will also depend on the technical solution required to capture the data. This brings us back to that first big question: How should you capture the data?
New IoT Standards
Traditional, low-cost communications technologies, such as WiFi and BLE, work very well indoors, where bandwidth is relatively unlimited, power is generally available and continuous data streaming is both possible and cost-effective.
WiFi and BLE are ill-suited, however, for working in outdoor, remote and open spaces, where their power requirements and range profiles make them impractical. Clearly there is a need for a different architectural design approach that can communicate outdoors and cover much more ground.
Enter the new IoT protocols and communication standards, such as LoRA, Sigfox, LTE-M and NB-IoT. All are designed to operate outdoors and cover relatively long distances. All place a priority on power management over speed, and they are all designed to securely move small quantities of data at a leisurely pace. They are well-suited to communicating daily or hourly sensor readings from outdoor, remote devices, but they are not designed to handle more bandwidth-intensive, rich-data scenarios, such as continuous video or audio streaming from those locations.
While the jury is still out as to which protocol(s) will win the day and achieve wide adoption, they all have different advantages/disadvantages, and they all have their respective proponents and detractors. If there is one degree of commonality across all of them, however, it is the recognition that all these protocols are embedded hardware solutions that address connectivity, power management, data storage, security and battery life.
If there is an area where there is some divergence of opinion, and in fact the area where real solutions intersect with reality, it is in intelligent edge solutions. This is where the communications travel a relatively short range to a gateway, rather than directly to the cloud. There are many reasons why this may prove to be just as common as direct-to-cloud communication.
Why communicate first to a gateway?
- Better battery life
- Reduced communication costs
- The ability to take local action and run analytics on beefier hardware
Clearly there will be a place for both device-to-gateway and device-direct-to-cloud solutions. And undoubtedly we will soon see hybrid scenarios as well, where, for example, most communications flow through a gateway unless there is some urgent event condition that dictates otherwise. We will have lots of fun putting these together in the months and years to come.
In conclusion, this is an exciting time for those seeking to build advanced IoT solutions for the outdoors. We have arrived at the point where edge compute, communications and battery technologies have finally caught up with the vision and desire to drive outcomes in the field. Thanks to our extensive group of partners covering the end-to-end solution — from sensor to gateway to cloud — and our deep bench of IoT architects, data scientists, AI and Machine Learning specialists, CTP and HPE are uniquely qualified to assist you in designing, building and managing your outdoor IoT solutions.