Not an Either-Or Proposition

E-mail Tom Kevan

The debate over batteries vs. power harvesting in wireless sensor applications is fermenting just below the surface as the RF technology gains broader acceptance. The mistake some of us have made is to look for the supremacy of one technology in all instances. That will not be the case.

Battery Power
Battery developers and manufacturers are turning out strong, well-designed products that work well in many of today's wireless sensor applications, and these power sources continue to get better and better. Many RF industry experts will tell you that current battery technology can easily match the operational life of sensors used in industrial applications; indeed, they can last as long as the systems the sensors are monitoring. For example, it is believed that the life of a mini steel mill is only about seven years. After that, their technology becomes obsolete, and the mill is no longer economically competitive. In this case, why not use a battery that can last up to ten years?

And a static, undemanding application—such as a temperature sensor that takes measurements every 10 minutes—does not even tax the potential of the battery that powers it. If the sensor and battery can function in the given environment, why go to power harvesting?

The fact is that batteries will be the power source of choice in many cases. And that will remain true for the foreseeable future. "Batteries play a huge role now and will continue to play an important role, says Joerg Bertholdt, senior director of marketing for Crossbow Technology.

Meeting Future Needs
But keep in mind that the more a technology can do, the more it will be asked to do. As wireless-sensing applications become more complex, the less practical a battery may be.

What if the static temperature-sensing application I mentioned earlier begins to show data that cause the engineer monitoring the sensor to want more data, more often? What if larger amounts of real-time data would enable the engineer to learn more about the system or condition being monitored? You hit the wall of the battery's energy budget.

What if you want to do in situ processing and analysis that cannot be supported or practically sustained by battery power?

What happens when your sensor is in an impossible-to-reach location or it would be too costly to have someone replace a large number of batteries?

This is where power harvesting comes into play. "In five years, energy-harvesting technology will become very realistic," says Bertholdt. "As the cost comes down and the technologies improve, more and more applications will be enabled. The more we can become independent of batteries and have autonomously powered devices, the value proposition of wireless sensor networks becomes bigger and bigger."

"If anyone gets a reasonable power density from a power-harvesting device that can operate in an industrial temperature range, they will win big," says John Suh, senior application engineer for Crossbow Technology.

The point is that both power sources have a place in wireless sensing applications. The challenge will be to play to their strengths.