This is part twelve of a series. Read part eleven here.

We’ve discussed the aggregator and the cloud in previous blogs. So, how do smart sensors fit into the IoT ecosystem? It seems the smart phone is the ultimate smart sensor. Further, the smart phone has been available to us for almost all our lives – okay, maybe not that long, but it does seem that way. Even though the smart phone seems to fit, it is not what I am thinking about when I use the term “smart sensor.”

Allow me to describe what I think the ultimate smart sensor looks like. I’ll base this description on a camera that the University of Southern California designed for an artificial vision research project. The intent was to insert the camera in the eye of a blind person. The camera would then communicate with a set of electrodes on the retina. With this system, a blind person could once again see.

An interesting result of the involvement I had with the research team was a market assessment we made based on a list of products that could be derived from the camera’s initial design. It seemed that the ability to have an ultra-low powered camera the size of a grain of rice could be used in more ways than helping the blind to see. The application ideas we put together suggested that the opportunities combined could create businesses valued in the ten-billion-dollars-per-year order of magnitude. Of that ten billion dollars about 0.1 percent (about $10M) would come from the camera used in giving sight back to the blind.

Now let’s get into the P’s and Q’s of smart sensors. A generic block diagram of the camera is shown in Figure 1. These components would make up three P’s of any smart sensor. I’ve added a few Q’s (questions) below each component to help explain each block’s possible design:

Performance: (Blue)

• Processor
  • Q: What does the processor do?
  • Q: Does it compress the data prior to sending?
  • Q: Is the processor analog or digital?
• Communications
  • Q: What method of communications?
  • Q: Is it standard or proprietary?
  • Q: Is security a concern?

Power: (Green)

• Energy scavenger
  • Q: What method?
  • Q: How much energy?
• Energy buffer
  • Q: Battery?
  • Q: Capacitor?
  • Q: Other?
• Energy management
  • Q: How to convert?
  • Q: How to distribute?

Personality: (Red)

• Image sensor (in this example, the personality is the USC imager).
  • Q: How many pixels?
  • Q: How many bits per pixel?
  • Q: Black and white or color?

In a later blog I will spend time on each of the blocks.

[Figure 1 | A block diagram of a smart sensor.]

Recall that the Aggregator can look like the Cloud in an IoT system – under our new definition of a Smart Sensor, the Aggregator can also be the Smart Sensor, or a High IQ Sensor. Consider the OSD335x System-in-Package for example, has the flexibility and performance and size for any of these roles.

Here are a few questions for you to think about:

1. Where would you use an imager the size of a grain of Rice?
2. What is your guess as to how many pixels are needed to replicate the human eye?
3. What is your guess as to how many pixels are good enough to give sight back to a blind person?
4. What is your guess on how many bits per pixel would be needed to replicate the human eye? Dynamic range? Accuracy?

Gene Frantz is Co-Founder and Chief Technology Officer at Octavo Systems.

Read part thirteen of the series here.

eletter-07-09-2018

eletter-07-10-2018