This is part thirteen of a series. Read part twelve here.
My first introduction to the concept of a smart sensor happened a decade or so ago. The team I headed up at the time was looking into the world of industrial control to determine if there was a place for a Digital Signal Processing device. In our initial look into the market we learned that many industrial manufacturing plants had tens of thousands of sensors. They were used to keep track and manage the manufacturing process. About the same time we ran across some work being done at UC Berkeley by Dr. Kristofer Pister. It seems he and his colleagues had created a device they were calling Smart Dust. We were intrigued with the concept, but the concept of smart dust begged several questions about the idea of a smart sensor:
- What size should it be?
- How does it get powered?
- How does it communicate?
- Why (where, how) would I use them?
Figure 1 is a picture of the Smart Dust concept Dr. Pister envisioned.
[Figure 1 | The concept of Smart Dust as envisioned by Dr. Pister. Courtesy of Kristofer Pister, c 2000.]
I am from the semiconductor industry so the idea of shrinking something has been permanently burned into my mind. But I also know that smart sensors will be of all sizes. I also am still not sure exactly what I would do with smart dust. The best idea I have is smart paint. With smart paint I could change the color of the walls in my house using a remote control. Now, wouldn’t that be interesting?
Power and communications
I’ll discuss these two aspects later in this series, but it is worth noting that power and wireless communications will be the dominant design criterion of a smart sensor. In a later blog I’ll show a chart we found that puts the concepts of performance, communications and power dissipation in perspective.
Why, where and how?
I think this is the big question to be answered for smart sensors. I told a story earlier about a camera the size of a grain of rice that the researches at USC developed. The surprise was how many opportunities there could be for such a device beyond giving sight to the blind. As an aside, when I would ask audiences where else in the body should we put a camera, one of the most popular answers was, "In the end of my finger." Another that seemed to be popular was, "In the back of my head." It seems that would allow me to keep track of things going on behind my back. As a child I was convinced that my mother had an eye in the back of her head. She seemed to be able to see me even when I was behind her making faces (or something worse).
This leads me to a question or two for you:
- What would be the killer application for smart dust?
- What would you do with a camera the size of a grain of rice? Implants? Industrial? Consumer?