many applications will not run on 100% harvested power but many other will.
The ones that will not operate on 100% harvested power still will benifit because it will extend the life by the proportion of power being supplied via vibrations and/or heat gradients.
it can even allow infinite life with reduced dutycycle if coupled with a capacitor even after the battery fails or degrade beyond adding value.
This is possible if the battery and capacitor can be switched in independently by the micro based on health monitoring.
So even when the battery fails and continuos operation or higher duty cycle with batteries it can still function at a much lower duty cycle.
take a bridge strain gauge monitor...these take years to fail and so a once an hour update that was stretched to once a day or week would still have value to the monitoring system and allow high level of safety.
Also as a vibration mode changes (think rusted bolt on a bridge or a bad bearing on rotating equipment the vibration waveform and amplitude would change. this could actually add extra power just when it is needed!
(assuming the bandwidth and amplitude envelope that can be capture is sufficient.
Nice article. I've found the bridge time to be a challenge (using super caps). Also keeping the average current as low as possible means a much less transmission rate when occupancy. There are also challenges with LUX level on ceilings which tend to be closer to 50-100 LUX level. Your 51uA with transmission every second for 3 minutes is impressive.
Harvesting energy from sun is not that challeging now a days due to availability of many single chip energy harvesting and power management chips but mostly the sensor networks are installed within building areas where sunlight can hardly reach. If energy harvesting devices becomes more efficient for use in indoor ambient condition harnessing energy from indoor lights or room temperature variations then definitely it will be a great leap in energy harvesting technology.
Wireless sensor node will be adopted widely if it can run on renewable energy source. Efficient energy harvesting is indeed one of the major breakthroughs. Using high efficient solar panel can help to better supply energy. High capacity energy storage is crucial to keep the sensor running while Sun is not available. The cost of ownership will include the maintenance cost. When a company deploys a thousand, there is very little chance that they can run to hunt for the units to replace battery even though the battery energy can keep the unit run for 2-3 years.
Harvesting shall not limit to renewable energy source. I have seen product which is able to harvest energy off from open a door or a window. To turn a light switch on has produced enough energy to send a signal to a remote switch to turn on an actual light. It is amazing how energy harvesting technology has gone so far.