So much has been read about sensors, measurement, big data and big IoT, but much of the useful work we read is being done on a smaller scale than expected.

It's not news that the Internet of Things (IoT) is high on the hot-topic list these days. Certainly, the potential benefits and payback of IoT, including industrial IoT, are huge, but there will also be a lot of deflation as the glow of "whatever your problem, IoT will solve it" dissipates. That's fairly standard trajectory for an emerging technology, as the latest edition of the Gartner "hype cycle" shows.

 
Gartner hype cycle chart cr *Figure 1: The Gartner hype chart serves as a reality check in seeing progress of an emerging technology.  

I think that part of the issue is that there is so much talk of "big IoT" networking of huge factories and installations. This will likely happen, but it's a lot of work, takes a lot of investment, a lot of time, and there will be a lot of problems debugging it all. Perhaps the better way to look at IoT is through smaller, more manageable projects which you can get both your hands and mind around more quickly and have quicker results as well.

That's why I found a recent project done by some engineers at Analog Devices to be a good example of what the IoT can do in a directly meaningful, shorter-cycle way. The small-scale project, "IoT: The Internet of Tomatoes," focused on improving the quality, taste, and appearance of locally grown tomatoes (and thus the yield and market price) while minimising environmental impact, water use, and costs.

 
InternetOfTomatoes Schweber Blog30 366x210 (cr) Figure 2: Analog Devices' "Internet of Tomatoes" project shows that small-scale IoT applications can achieve what big-funded investments can.  

To do this, the team instrumented a tomato patch with sensors that measured and reported critical data such as humidity, temperature, and ambient sunlight. This data, in turn, was uploaded to a cloud-based application, which kept the farmer up to date on real-time conditions and trends, including disease potential and ripening status.

In some ways, this IoT project is conceptually only a little different from what farmers have been trying to do for many years in measurement: get better data using field-based sensors. However, today's technology under the broad IoT heading makes this sort of measurement objective much more practical and easy to install. The sensors are better and cheaper, local power is feasible (battery or solar), the reporting is done via wireless link, and there's access to a database and smart application which cannot only look at a single site's data and make suggestions but also integrate it with expert knowledge, thus improving that insight by looking at data and outcomes across many locations and scenarios.

Thus, while the basic objective is the same, the execution is much easier, more sophisticated, and capable of figuring out so much more — and doing it all at a far lower cost. While I am very sensitive to overhyped buzzwords, in this case, IoT is not just flash and facade but also embodies some of the advances that make it possible and that it can provide.

One thing about this tomato IoT system design truly intrigues me. A detailed story on the project in Design News, "Attack of the Tomato Engineers," mentions an interesting point, “We put the MCU into a deep sleep when it’s not being used, then use the accelerometer to wake up the rest of the system, take a measurement, and then put everything back to sleep again.” Huh? Use an accelerometer for system wake-up? I'd sure like to know more about why and how that was done. One reader asks the same question, as have several chat rooms I looked at. I am sure it was a solid engineering decision; and if it was a clever use of a product in way not originally intended, but which solved a problem and hopefully had lower cost… well, that’s the essence of engineering creativity.

Have you had any experience in implementing a real but modest IoT project — or perhaps a big one?