Download (IEEE Pervasive) GE EHAC (sub)

IEEE Pervasive Special Issue: Energy Harvesting and Conservation
Guest Editor Introduction
Roy Want, Keith I. Farkas, Chandra Narayanaswami
10th January 2005, DRAFT version 3.0
One the goals of pervasive computing is that computation should become integrated with our daily work practice
and, unlike traditional personal computing, blend in to such an extent that it enhances these activities without being
noticed. In other words, computing becomes truly invisible. Yet, at the heart of every pervasive computing system,
there are electronic components that consume energy. For mobile systems, or systems for which power is not
guaranteed to be reliable, perhaps because the mains electricity is intermittent, managing their energy needs can be
a significant distraction for users. As most users can attest from their experience with mobile phones, and portable
media recorders and players, the magnitude of this distraction depends on the specific application, the hardware on
which the application is based, and the characteristics of the energy source.
This problem of reducing user involvement in power management is being addressed predominantly in two ways,
the first of which is to reduce power consumption by using hardware that consumes less power, or software that
judiciously manages power consumption, or both. Advances in VLSI techniques have led to dramatic decreases in
the power consumption of electronic circuits. This decrease has been in large part due to decreases in the dynamic
power consumption of transistors. This reduction has allowed a pervasive system to deliver greater capabilities on
a fixed power budget, or similar capabilities on a smaller power budget. However, in the future, it remains unclear
to what extent this trend will continue as static power consumption is becoming a large fraction of total power
consumption, and no long term fix currently exists. In addition, over time, as users become more accustomed to the
convenience of wireless capabilities and transfer large amounts of data over a wireless interface, physical laws will
dictate the minimum amount of energy that needs to be transmitted, which cannot be circumvented by VLSI
circuitry per se. Energy-aware software is a compatible technique, wherein software is designed to conserve energy
where possible, for example, by judiciously powering down components that are not used, and engaging powersaving modes of operation when higher-power modes are not warranted to deliver a given quality of service. The
challenge for such software is to limit the degree to which users are distracted by these optimizations, and to limit
the amount of user attention and configuration that is required. This problem is further exacerbated when mobile
devices evolve from single purpose devices to multi-purpose computers with similar operational characteristics to
desktop computers.
The second way in which the distraction of energy management is being minimized is the development of improved
energy sources. Today, batteries represent the dominant energy source for systems that cannot be readily powered
from mains electricity or for which mains power does not exist, for example, at a remote weather station. While
exponential improvements have occurred in hardware components, such as dramatic improvements in processor
speed, memory density, and network bandwidths, such improvements have not occurred in battery technology nor
are any anticipated in the future. At the same time, due to the persistent integration of more features and capabilities
by the consumer electronics industry, due to shrinking component sizes, our expectations for mobile systems
continue to outpace the energy-storage capabilities of conventional batteries. This trend has given rise to the
development of alternative energy-storage solutions, such as direct methanol fuel cells. These cells, however, are no
panacea, as like batteries, they must be managed by a user, and hence distract from the application at hand.
Moreover, the energy density of batteries in some mobile devices is already very high and safety concerns arising
from rapid accidental discharge would become more serious with significant increases in energy density.
Thus, for the reasons discussed here, energy management for pervasive systems continues to be a distraction for
users. But there are techniques that given the appropriate applications and modes of use, can have a significant
positive effect on a user’s involvement in this process, namely the subject of this special issue.
Energy Harvesting and Conservation
For applications with low power requirements, a promising approach for reducing a user’s involvement in the
energy management of a pervasive computing device is energy harvesting. The objective of energy harvesting is to
collect ambient energy for the purpose of powering systems, possibly storing energy in times when it is not required
– this process is also known as energy scavenging.
A number of energy transducers exist, with perhaps the best known being solar cells. Solar cells have long been
used to power simple hardware components such as calculators, emergency telephones along a highway, and low
grade lighting for pathways at night (charged during the day). Another type of transducer converts the energy
contained in a vibrating object into electrical energy. Such transducers have been used by researchers to harvest
energy from floors, stairs, and equipment housings. A third type of transducer is one that harvests mechanical
energy, such as that produced by a person walking and the movement of an object: self winding watches, which
wind themselves from the swing of a persons arm, even predate electronic watches. Hybrid cars that transfer
energy from the engine to the battery during braking are a more modern example of this technique, which has been
used in commercial vehicles. Yet another type of transducer is one which converts the momentum generated by
radioactive reactions into electrical energy. Some transducers are also able to exploit temperature gradients, or
pressure gradients, and produce electrical energy. Each of these transducers has differing degrees of usefulness
depending on the situation in which they are used.
Energy harvesting is most applicable to applications that demand small amounts of continuous power, or which
have short periods of high-power use that can be met from previously stored harvested energy. Energy harvesting
can also be used to supplement more conventional energy sources, such as powering a mobile computer while it is
in a low-power sleep state, or charging the battery. These supplements may be a significant contribution to the
energy needs of a computer given a suitable usage pattern. For instance, if a mobile user makes extensive use of a
computer for short periods of time, an energy harvesting system might be able to ensure the battery is always
topped up during the standby period when the computer is not being used.
A complementary technique for extending battery lifetime is the use of system software to judiciously manage
power consumption. This represents a large number of techniques that can be summarized by the term energy
conservation. Most users of today’s personal computers are familiar with some of these techniques – from the
shutting down of disk drives after a time out period to slowing down the clock of a processor. Given that the energy
contributions of energy harvesting will be modest, and will also likely be serendipitous based on the location of use,
conservation techniques must be used hand-in-hand with energy harvesting to make best use of its contribution.
Commercial Uses of Energy Harvesting
We thought it would be instructive to include a brief survey of the commercial uses of energy harvesting in our
News department (see page XXX), providing insight into the types of products that are both under development and
available, the stories behind some of them, and their commercial success. This survey examines energy harvesting
in a broader context, considering not only how it can be used to replace batteries, but also how it can be used to
reduce energy costs and provide a source of energy when one is not readily available, such as in underdeveloped
regions or in disaster situations where conventional energy sources are no longer available. This survey describes
products ranging from new VLSI technology, and thermoelectrically-powered microprocessors, to shake
flashlights, wind-up solar powered radios, and solar-powered fridges.
Future Trends and Directions
Although energy harvesting techniques can have a significant benefit for some systems and the associated
applications, they are limited by the efficiency of the transducers and the raw energy available. Since the latter
places a fundamental limit on the energy that can be harvested, it will always be necessary to apply energy
conservation techniques in tandem to make best use of the energy available. In cases where energy harvesting is
sufficient for an application, conservation can still extend the processing capabilities of mobile systems and enable
new applications.
The semiconductor industry believes feature sizes for semiconductor devices will continue to decrease for some
years ahead, although the rate at which we can build them, due to significant manufacturing complexity, is slowing.
The result will be continued benefits in switching power, with leakage current now also beginning to play a
significant role. This means that power dissipation is not just related to the switching frequency, but to the area of a
chip that is actively connected to the supply lines. Consequently, the need for ever more capable electronic circuits
to match our growing user expectations may have to be increasingly balanced against the power cost of delivering
the associated increased integration.
As more VLSI designs are built with power characteristics that lend themselves to low-power operation, the
commercial opportunity for energy harvesting will become apparent. This will drive interest in improving the
existing energy transducers, perhaps with more efficient components, or in the search for fundamentally new
materials with improved energy conversion properties. For example, solar cells with greater than 20% efficiency,
perhaps based on non-silicon solutions, or a better piezo-electric material than the commonly used PZT (lead
zirconate titanate).
Better understanding of solid state physics and refined manufacturing techniques will play a role in improving
efficiency. As the demand for processing power for personal computers continue to increase, the demands on the
quality of the manufacturing process also increase to guarantee the required high-yields of silicon chips containing
tens of millions of transistors. These improved processes have a knock-on effect for other devices types, including
those used as energy transducers.
Although energy harvesting is a compelling concept, is not a solution for all energy needs, but will increasingly
replace batteries for lower-power applications with the expected advances in low-power circuit design techniques,
and improving transducer energy efficiencies. If the economics are right, in some cases it may just be a question of
applying known material characteristics in novel combinations to improve existing energy harvesting techniques.
The Papers in this Issue
The first full paper in this special issue, “Energy Scavenging for Mobile and Wireless Electronics” by Paradiso and
Starner, provides a survey of the amount of energy available from various ambient sources, and the technology that
can be used to harness it. Roundy et al, in a paper titled “Vibration-base Energy Scavenging for Pervasive
Computing: New Designs and Research that Increase Power Output”, provide us with an in depth study of how
mechanical vibration can be used as an energy source, one of the previously less studied ambient energy sources. In
the paper “The Wireless Identification and Sensing Project: Usage Model, Applications and Working Prototype”,
Philipose et al demonstrate how UHF radio frequency energy generated for the purpose of reading RFID tags, can
be put to further use to provide a remote sensing capability; and in “Inductive Telemetry of Multiple Sensor
Modules”, Chevalerias et al, present an analysis of the capabilities of inductive power in the context of powering
RF sensor modules. Looking to novel forms of power, Lal et al. provide us with a description of “Pervasive Power
form Radioisotope-power Piezoelectric Generators”. This paper can be considered energy harvesting from lowgrade radioactive materials or it can be thought of as a new type of battery technology, and shows how concepts for
a harvesting technology and a battery technology might in some cases blur together. Finally, we turn our thoughts
to energy conservation for “ECOSystem: Explicit Energy Management as a First Class OS resource”, by Zeng et
al, and learn how energy management can be effectively integrated into an Operating System without needing each
application to be individually customized for power efficiency.
Complementing the full papers, in the work-in-progress department, we present two energy harvesting projects.
Kansal and Srivastava report on their ongoing research into the challenges of building practical energy harvesting
sensor networks, while Randall et al. report on their ongoing research into extracting user-level information from
the rate of energy harvesting, such as where the user is located or the orientation of a mobile device.
To summarize, energy harvesting is a nascent technology. At the current time, modest amounts of energy can be
harvested for our environment and put to use. We believe this area will receive more research attention due to the
increasing use of pervasive electronics and the difficulty that arises in managing the power for large numbers of
devices, and with new technologies, the amount of energy that can be harvested will increase. Over time, practical
deployment of these technologies could occur in a wide variety of settings. We hope you find the articles in this
issue of Pervasive Computing interesting, and that some of you will contribute to further advances that will lead to
widespread use of this technology.