With the AR prefix, I suppose this come out of the former "Aptina" portion of ON. Austin Cho and Alex Krymski (Photobit then Micron) demonstrated a 176x144 30 fps 8b CMOS image sensor with 550uW power dissipation that was published in 2003. So, now we have 560x560 3000uW @ 1fps 17 years later? Something seems off. Is there a typo? Or is the 3 mW expended over less time than a second? I know there have been even lower power sensors demonstrated since 2003.
From the product overview: "The product has an innovative super low power mode which draws less than 2.5 mW while active, and can detect motion or changes in lighting conditions and wake the rest of the system up"
Are there really other full image systems that in active state that draw less than 2.5 mW ? I would be interested in seeing a reference. I've seen select chips but when integrated into a system they usually have 10s of mW in all modes but dormant.
What is a full image system? I thought we were talking about an image sensor chip? Here is the reference to our 2003 paper I mentioned above: K-B Cho, A. Krymski, E.R. Fossum, A 1.5V 550 uW 176x144 autonomous CMOS active pixel image sensor, IEEE Transactions on Electron Devices, vol. 50(1), pp. 96-105 (2003) You can find it here: http://ericfossum.com/Publications/Papers/A%201.5-V%20550-uW%20176x144%20autonomous%20CMOS%20active%20pixel%20image%20sensor.pdf
Many lab publications report low power. Once you go consumer, and face yield issues, standards to be followed, it is difficult to meet these experimental numbers...
With the AR prefix, I suppose this come out of the former "Aptina" portion of ON. Austin Cho and Alex Krymski (Photobit then Micron) demonstrated a 176x144 30 fps 8b CMOS image sensor with 550uW power dissipation that was published in 2003. So, now we have 560x560 3000uW @ 1fps 17 years later? Something seems off. Is there a typo? Or is the 3 mW expended over less time than a second? I know there have been even lower power sensors demonstrated since 2003.
ReplyDeleteFrom the product overview:
ReplyDelete"The product has an innovative super low power mode which draws less than 2.5 mW while active, and can detect motion or changes in lighting conditions and wake the rest of the system up"
Are there really other full image systems that in active state that draw less than 2.5 mW ? I would be interested in seeing a reference. I've seen select chips but when integrated into a system they usually have 10s of mW in all modes but dormant.
ReplyDeleteWhat is a full image system? I thought we were talking about an image sensor chip? Here is the reference to our 2003 paper I mentioned above:
DeleteK-B Cho, A. Krymski, E.R. Fossum, A 1.5V 550 uW 176x144 autonomous CMOS active pixel image sensor, IEEE Transactions on Electron Devices, vol. 50(1), pp. 96-105 (2003)
You can find it here:
http://ericfossum.com/Publications/Papers/A%201.5-V%20550-uW%20176x144%20autonomous%20CMOS%20active%20pixel%20image%20sensor.pdf
Many lab publications report low power. Once you go consumer, and face yield issues, standards to be followed, it is difficult to meet these experimental numbers...
ReplyDelete