IntJouCompAppl

International Journal of Computers and Applications, Vol. 28, No. 3, 2006
A CIRCUIT MODEL FOR PASSIVE
RF AUTONOMOUS DEVICES WITH
PROTOCOL CONSIDERATIONS
M.H. Mickle,∗
M. Mi,∗
J.T. Cain,∗
and T. Minor∗∗
Abstract
This paper focuses on a lumped circuit model for autonomous
untethered wireless devices powered by remote radio frequency (RF)
sources including radio frequency identification (RFID) tags. The
technology presents major challenges for these passive (no on-board
power) devices, including the ability to provide maximum power
for data transmission and data rates under powering constraints
due primarily to governmental regulations. This paper presents the
electronic circuit model of a remote temperature-sensing device less
than 0.5 cubic inches in volume. The RF energy is first converted
to a DC voltage that is then used to power circuitry to measure
temperature and transmit the reading to a remote receiver. These
devices are typically deployed in large groups functioning as sensors
or RFID tags that communicate with a single receiver. The typical
rate at which energy can be harvested, converted, and used to power
circuitry and transmitter is about 0.025% of the rate at which the
power is used. Thus, careful consideration must be given to the
circuit model, including parameters and communication protocols.
Included in the paper are a logically derived circuit topology and
the methodology by which the parameters can be determined. The
protocol can then be analyzed with respect to power using the
resulting circuit model.
Key Words
Energy harvesting, protocol, wireless devices, communications
1. Introduction
Untethered autonomous wireless remote identification [1,
2] and sensing devices [3–5] with no on-board power (no
battery) and essentially an infinite shelf life have become
extremely important to sensors and radio frequency iden-
tification devices. These devices may be powered by light,
heat, vibration, or radio frequency (RF) energy [6]. This
paper presents the model for powering and operation of
the device using RF energy. The RF source may be a
∗ Department of Electrical and Computer Engineering, 348
Benedum Hall, University of Pittsburgh, Pittsburgh, PA 15261,
USA; e-mail: {mickle@ee., mimst42+@, cain@ee.}pitt.edu
∗∗ Gnostic communications, Inc., PO Box 663, Belle Vernon, PA
15012, USA; e-mail: timminor1@yahoo.com
(paper no. 202-1723)
locally supplied by a directed energy source [4, 7] or it
may be RF ambient energy such as that produced by
AM or FM radio broadcast bands. There is technically
no difference, but the magnitudes of energy viable at the
device typically differ by orders of magnitude. The RF
energy can be characterized in terms of watts per square
meter where the antenna of the source can be designed to
produce different power (energy) densities depending on
the application. That specification becomes a part of the
total design scenario. The input energy for the unteth-
ered device considered here is characterized in terms of
watts/meter2
.
The device contains an antenna that collects the in-
coming RF energy and directs it to electronic circuitry that
converts the RF energy (at some efficiency) into a direct
current (DC) voltage [4]. The design goal for the electronic
conversion circuitry is to maximize the harvested energy
through the efficiency of the RF energy conversion.
The device contains an antenna that is affected by
the other device circuitry contained on these small-scale
devices. The analysis of this effect requires a circuit model
of the device. Fig. 1 illustrates two devices spanning the
scale from printed circuit board (PCB) devices (inches)
to a complementary metal oxide semiconductor (CMOS)
die (millimeters). The device on the left is a temperature
Figure 1. A PCB scale device (left) and an RFID CMOS
die (right).
243

sensor fabricated under a NASA contract, and the die on
the right is an RFID device [7].
The use of traditional analysis of the antenna, match-
ing circuitry, and RF-to-DC conversion circuitry [8] is ex-
tremely limited as the device becomes smaller and smaller
due to the interaction of the various components in the
energizing RF field [9]. Thus, as the device size becomes
smaller [10], the complicated element interactions suggest
a lumped model that is empirically calibrated. The de-
vice is actually quite simple in terms of the input, output,
and state. It is assumed, without loss of generality, that
the device performs a single function such as temperature
measurement with the resulting data communicated to a
remote base station along with an identification number.
If there is a measurement, the device is a sensor. If there
is no measurement (only identification), the device is an
RFID tag [1].
The input to the device is an RF energy field measured
in power/square area, for example, watts per square meter
[8]. The state of the device is a DC voltage that is ob-
tained through conversion of the RF energy. Once the state
(voltage) of the device reaches a sufficiently high value,
the sensor measurement and identification information are
transmitted by the on-board RF transmitter, which is the
output of the device [4, 5]. Given a functioning device
topology, it is normally desired to maximize the distance
between the autonomous device and the base station that
powers the device and reads the information transmitted
by the device. Device optimization is thus a function of
increasing the efficiency of the energy conversion to pro-
vide more power; and of minimizing the nontransmitted
energy, that is, the circuitry overhead. In effect, it would
be desirable to convert virtually all of the harvested en-
ergy to transmitted energy with an appropriate modu-
lation scheme. Increasing the energy available for data
transmission also increases the operating distance between
the device and the receiver.
The goal of this paper is the development of the model
with the parameter values for the autonomous device that
directly link the RF input in terms of watts per square
meter to the output of the power supply in terms of
DC volts that produces an RF signal that is delivered
to the device communications antenna. The topology of
the model is implemented with simple electrical circuit
elements that are included based on obvious functions that
are characterized by simple logic.
2. System Overview
The cartoon of the total system, including the base station
and the device, is given in Fig. 2, where the device is within
Figure 2. The total functional system.
the enclosed block and the energy and communications
antennas of the base station are outside the enclosed block.
The discrete device example of Fig. 1 is shown in Fig. 3 in an
orientation to illustrate the correlation to the components
of the device elements in the block of Fig. 2.
Figure 3. An example sensing device.
The RF energy source and the communications receiver
may be collocated, although it is not a requirement. The
power supply is not ideal, and care must be taken in the
design of a communications protocol. When the device is
operating, energy is transmitted to it to power the on-board
circuitry. Thus, the energy source and device are operating
simultaneously [5]. However, the communication of data
to the base station receiver uses energy at a faster rate
than can be supplied by the source. Thus, the transmitter
must be controlled by the on-board microprocessor to only
transmit data for an interval of time, followed by a period
of time that will allow the energy to build up (recover) in
terms of supply voltage, VDD. This would allow the device
to function continuously with energy recovery occurring
between transmissions. A summary of the operation of the
system of Fig. 2 follows:
1. The power delivered to the device (PRD) is a function
of the power transmitted by the base station (PTB).
2. The power input to the device (PRD) is also a function
of the distance between the device and the base station
antenna due to the inverse square law [8].
3. The power actually available for operation of the device
is a function of the efficiency of conversion of the
antenna and matching and DC conversion circuitry.
4. The power used by the device (PU), other than the
power delivered to the transmitting antenna, is a func-
tion of the on-board circuitry, which is required to be
low power circuitry.
5. The actual power transmitted by the device (PTD)
is a function of the communications antenna and the
matching to the transmitter.
6. The external receiver (at the base station) specifies a
minimum amount of power that must be received by
the base station (PRB).
Thus, the maximum range of the device is essentially a
function of the power received (PRD) when the device
has been optimized by minimizing the power used (PU)
by the on-board electronics. In the present example,
PU is more than an order of magnitude less than the
transmitter. Thus, as a first-order approximation, the
primary concern is characterized by the power received by
the device. However, there is a second consideration. The
device electronics require a certain minimum voltage level
in order to operate. Thus, maintaining this minimum value
of voltage for VDD is an additional factor. The energy
harvesting input will be shown to be equivalent to charging
244

a capacitor. Once the device circuitry begins to function,
including transmitting, the voltage will be shown to drop
as a capacitor discharge [11].
Given an initial voltage value, the rate of using the
energy exceeds the rate of energy supply, and eventually
cannot provide the necessary voltage for the electronics to
continue to function indefinitely. The power to operate the
device comes from a remote RF source that is harvested
by an antenna/matching circuit/charge pump combination
[4, 5], which does not act as an ideal voltage source.
3. Device Circuit Topology
The device lumped circuit model is given in Fig. 4. The RF
powered device converts incoming RF power (energy) that
is received by the device (PRD) to drive a microprocessor
and the sensing circuitry that uses some of the converted
power (PU). The measured temperature data are then
transmitted at some power level (PTD) to the base station
receiver that receives the data at some power level (PRB).
Figure 4. Device equivalent circuit.
The topology of the model will be presented in a
straightforward discussion based on function, testing, and
verification. Operation begins with an inert (zero energy
state) and proceeds to an operational state with suffi-
cient voltage, VDD, with a charging curve of the form
v(t) = V(1 − e−αt
).
The elements RC and C follow directly from the ob-
served charging (recovery) curve of Fig. 5. This is a clas-
sical charging curve for an RC circuit [11]. The capacitor
shown in the model is not a specific circuit element, but
rather the lumped effect of a number of circuit elements.
Figure 5. Voltage during a data transmission (9 bit) and
voltage recovery.
Once the capacitor element is charged to a sufficient
VDD, the static (no transmitting) discharge (on-board
electronics) is very slow. This implies an RC decay [11]
with a very large resistor, RS, in parallel with the capacitor.
The long delay of the discharge is the result of optimizing
the on-board circuitry. The value of the resistor can simply
be determined by applying a fixed voltage source for VDD
and measuring the resulting current.
The on-board electronics (exclusive of the transmitter)
automatically turns on when VDD reaches a sufficiently
high level. The transmitter is turned on by an on-board
microprocessor according to a predetermined protocol (im-
plemented in firmware) based on the calculated system
power requirements and the empirical testing of the power
consumption. The transmitter circuitry is simply a sink for
power, which has been represented classically as a resistor
whose value can be determined independent of the device
interconnections.
All of the elements are given as lumped representations
[11] of the resistive and capacitive effects of the device
circuitry. The input RC equivalent circuitry is given by
X Ω and α F with the nontransmitting load given by RS.
Once the device begins to transmit, indicated by closing
the switch, S, current will flow through RL. The load can
be represented by a resistor, RL, that is much less than
the nontransmitting electronic circuit model, RS.
The energy source, VIN(t), is the simple power input
to the circuitry that comes from the base station RF
transmitter (PTB in Fig. 2).
When the switch, S, is closed, this implies that the
load is the parallel combination of RS and RL. However,
based on empirical measurements, the significant difference
in power drawn allows the parallel combination to be
considered as simply RL.
4. Calibration of the Device Model
Testing for the model calibration was performed by placing
the device (see Figs. 2 and 3) in an electro-magnetic
field generated by the base station transmitter providing
the power (PTB). Thus, the testing mode duplicates the
normal mode of operation of the device.
The power density [8] of the field was known and
certain static measurements were made with the device
having on-board instrumentation used only in the testing
phase. With the test instrumentation, the measured value
of the device on-board DC voltage can then be wirelessly
transmitted using Infra Red to a test instrument [9].
The initial parameter determination is the input RC
combination where the charging (recovery) curve of Fig. 5
determines the values of X Ω and αF of Fig. 4. Devices of
this type have essentially no power regulation and show a
sharp drop in voltage once the transmission begins. Fig. 5
illustrates this voltage drop when the device transmits a
byte of information preceded by a START bit for a total of
9 bits.
In this research, the device is placed in an RF field of
a given power density specified in terms of watts/meter2
.
The device contains an antenna and energy-harvesting
circuitry such that for a given power density there is a
resulting amount of power available (harvested) as direct
current (DC) at the output of the harvesting circuitry. This
harvesting measure of this circuit will be lumped into the
term, AH, specifying an area and having the dimensions
245

of meter2
such that the following equation holds true in a
field of X watts/meter2
:
X watts
meter2 ∗ AH = Y watts (1)
In Fig. 4, the Y watts would be the amount of power that
could be supplied to the resistor combination, RS and
RL (when S is closed). For example, assume that under
steady-state operating conditions, it is possible to supply
a voltage, VDD = 2 volts and a total power of 1 mW, to
the resistive load consisting of the specified resistors. The
parallel resistor combination can be determined from (2):
Power =
22
Volts
RS||RLOhms
= 1 mW (2)
RS||RL =
4 Volts
1 × 10−3watts
= 4 K Ohms (3)
From a static measurement with S open and no energy-
harvesting energy applied, the value of RS can be deter-
mined to be 3 M Ohms.
RS||RL = RS
RL
RS + RL
= 3 × 106 RL
3 × 106 + RL
(4)
3 × 106 RL
3 × 106 + RL
= 4 × 103
Ohms (5)
(3 × 106
− 4 × 103
)RL = (3 × 106
)(4 × 103
) Ohms (6)
RL ≈ 4 × 103
Ohms (7)
The model of the RF energy source in Fig. 4 thus consists
of a battery and a resistor because the source is non-ideal
when measured as VDD. The constant energy (power
source) is thus the battery, VRF, which is turned on or off
by the switch, T, in Fig. 7, based on the availability of the
RF wave having X watts per square meter.
The transmission (discharge) curve of Fig. 5 can now
be used to determine the value of the capacitor in Fig.
4. From close analysis of the recovery curve, the RC time
constant is determined to be 6 × 103
. The resistor RS of
the circuit in Fig. 4 can be neglected as contributing very
little to the RC time constant due to its very large value.
Thus:
RL ∗ C = 8 × 104
= 4 × 103
∗ C, (8)
and C is 2 × 10−8
F = 2 µF
By a similar method, the RC charging (recovery) curve
of Fig. 5 has an RC time constant of 0.016 = 16 × 10−3
.
Thus:
RX ∗ C = 0.016 = RX ∗ 2 × 10−3
, and RX = 8 ∗ 144
Ω
(9)
This completes the determination of all circuit ele-
ments of Fig. 4. The figure with the component values is
repeated in Fig. 6.
Figure 6. Equivalent circuit with all component values.
5. The Model for the Base Station as a Source for
VIN(t)
Although the circuit model is complete as shown in Fig. 6, it
is still a useful exercise to establish the value of the voltage
source, VIN(t). Considering Fig. 6, a nodal equation can be
written at the node, VDD, and algebraically manipulated
to the form of (10).
−VRF − VDD
RX
+ C
dVDD
dt
+
VDD
RS
= 0 (10)
C
dVDD
dt
+
VDD
RX
+
VDD
RS
=
VRF
RX
(11)
dVDD
dt
+
VDD
RXC
+
VDD
RSC
=
VRF
RXC
(12)
dVDD
dt
+
1
RXC
+
1
RSC
VDD =
VRF
RXC
(13)
Now, from the curve fit for the charging curve of Fig. 5:
VDD = A(1 − e−αt
) (14)
Substituting (14) into (10) gives:
αA e−αt
+
1
RXC
+
1
RSC
[A(1 − e−αt
)] =
VRF
RXC
(15)
αA e−αt
− A
1
RXC
+
1
RSC
e−αt
=
VRF
RXC
− A
1
RXC
+
1
RSC
(16)
The steady-state solution for the particular example illus-
trate in Fig. 6 is:
VRF
RXC
− A
1
RXC
+
1
RSC
= 0 (17)
The value of A for the specific empirical test is 2.6763 volts.
VRF
0.016
− 2.6763
1
0.016
+
1
6 × 10−1
= 0 (18)
VRF ≈ 2.748 (19)
246

The result shown in (19) is extremely important for the
design of these autonomous devices in that the efficiency of
the device is primarily determined by the ratio of RC/RS
where the losses due to the harvesting circuitry are given
by the I2
R loss across RC during harvesting and the static
power (I2
R) loss across RS during operation.
6. Communications Protocol Considerations
As indicated previously, remote RF powered devices must
be optimized for power loss. The power used (PU) by the
device is critical for circuit overhead optimization. How-
ever, in general the largest amount of power is consumed
by the RF transmitter. Fig. 7 is a magnification of the
oscilloscope trace of the transmitting period for a specific
device under test.
Figure 7. Final model equivalent circuit with all elements.
The device under consideration transmits data to the
base station receiver through ON/OFF modulation, with
the receiver forwarding the information to a personal com-
puter through a serial port similar to the RS 232 bit format
and baud rates. Due to the byte format, the nominal mes-
sage length is considered to be one byte. Thus, data limits
extend from 1 to 8 data bits. For messages less than 8
bits, an intermediate receiver device collects the bits. With
ON/OFF modulation, a START bit is used to provide the
synchronization for the receiver to look for the single bit as
well as the larger number of bits. Thus, data transmission
from the device to the base station can be from 2 to 9 bits.
On the right hand side of Fig. 8, it is obvious that
during transmission the rate of consumption of power
is much greater than the recovery (harvesting) rate can
sustain. Thus, all transmission protocols must take into
account the rate of energy use, as somewhere the voltage
level of the power supply will decrease below the VDD
Figure 8. Magnification of transmission and initial transmission period of Fig. 5.
limit. Each application must take into account the amount
of data to be transmitted and the VDD limit.
The intent of this section of the paper is not to present
a simple answer or mathematical technique as to how this
problem is solved. As remote devices become smaller and
smaller, there is increased difficulty in being able even to
characterize the device equivalent circuit other than by
empirical means due to the proximity of the individual
circuit elements.
7. The RF Model of the Base Station Powering
Source (PTB)
The results of the model parameters will now be related
to the powering source and harvesting area. The device
of Fig. 1 has an energy harvesting antenna that can be
characterized as having a specific parameter defining how
much of the power (energy) density per unit volume can
be harvested for a given density of watts/meter2
. This
parameter is specified in terms of area and is defined as
the harvesting area, AH. The harvesting area is essentially
the same in concept as the effective area, Ae, of traditional
antenna theory [8]. The method described here is simplified
in concept, and it is not necessary to get into a theoretical
presentation of traditional antenna theory [8] in this paper.
The primary energy measure at the device is the volt-
age, VDD, across the capacitor, C. Based on the relation-
ship among charge, capacitance and voltage, Q = CV [12],
it is possible to solve for the charge on C:
Q = 2 × 10−7
∗ 2.748 = 5.496 ∗ 10−7
Coulombs (20)
Now from the relation among energy, voltage, and charge,
Volt = Joule/Coulomb, it is possible to determine the en-
ergy:
2.748 ∗ 5.496 ∗ 10−7
= 1.51 ∗ 10−6
Joules (21)
From the experimental results illustrated in Fig. 5, the
voltage, VDD = 2.748, occurs at approximately 0.050 sec-
onds. Assume that the power density of the RF field in
which the device is placed is 1 watt/4.18 meter2
(the test-
ing field), which represents a transmitted power of 1 watt
from an isotropic antenna at a distance of 1 meter. The
energy from this field with the harvesting area, AH, taken
247

into account [12] is:
Energy =
0.050
0
(1 watt/4.18 meters2
) ∗ AH dt
= 1.51 ∗ 10−6
Joules (22)
Energy =
1 watt
4.18 meters2
∗ AHt
0.050
0
AH = 0.126 × 10−3
meters2
(23)
As indicated earlier in this paper, the classical effec-
tive area [8], Ae, and the energy harvesting area, AH, of
this paper are similar in concept but are not necessarily
equivalent. The effective area has a function that can be
determined mathematically from the physics of the system.
The energy harvesting area is empirically determined
by a series of effects that are represented as a single
factor for a particular implementation of the antenna and
charge pump implementation measured using the harvested
energy available for the device to use for the on-board
circuitry and the communications transmission.
8. Model Application
8.1 Directed Source Requirements
The model developed has a number of uses. For example,
given the model with the parameters of Fig. 6, it is
necessary to maintain a minimum voltage of 2.5 volts to
operate the electronics represented by RS. The voltage
recovery time can be determined using the time constant
of RC, and C to determine the time to recover the energy
used for the data transmission. The necessary voltage,
VDD, to ensure enough time for the transmission can be
determined from VDD and RC. Once the voltage VRF is
determined, it is then possible to determine the amount of
power that must be provided to the antenna represented
by PRD.
(1) Given VRF, the required charge is calculated using
(17)–(19) for the specific device under analysis.
(2) Using the necessary charge, the total required energy
for recovery is calculated using the technique presented
in Section 7.
(3) The energy harvesting area, AH, is known and is used
with the energy to determine the upper limit (time)
desired, which allows the determination of required
power from (20). Note that the value of 1 watt/4.18
meter2
was a given in the initial development of (20).
It is not the value to be calculated:
Power Density =
Required Energy
Required Time ∗ AH
(24)
8.2 Receiver Power Requirements (Sensitivity)
Again, given the model of Fig. 7, the power for transmis-
sion in the model can be characterized by RL. By mea-
suring the power received at some known distance, we can
simply calculate the gain (efficiency) of the antenna based
on the power to the transmitter and the power received
in the direction of the receiver. Thus, given the distance
from the device to the receiver, the power delivered to the
receiver (PRB) could be determined from classical electro-
magnetic theory. If the power is not sufficient, one or
more of the following will be required: the device antenna
will need to be improved; the amount of power supplied
to the transmitter will need to be increased; a more sensi-
tive receiver will be required (less PRB); the distance may
be decreased (assuming it is not a design goal); or some
combination of the previous.
From the above considerations, the importance of the
elements of the model of Fig. 7 can be seen in terms of the
overall system design and parameters such as transmitted
power (PTB), power received at the base station communi-
cations receiver (PRB), the distance of the device from the
energy source having the direct affect on the power density,
and the distance between the device and the communica-
tions receiver. The antennas for energy harvesting and
communications in these devices of high-density circuitry
are best evaluated empirically due to the complicated mu-
tual interactions of the associated electronics making up
the power used (PU) component in Fig. 2 and the minimal
overall size of the device.
9. Conclusion
In summary, for a wireless autonomous device, this paper
has demonstrated the ability to develop and calibrate a
model for relating RF energy transmitted to the device
being powered through to the power that is transmitted
by the device to the base station receiver. The trans-
mitter power will determine the sensitivity of the receiver
necessary to receive the transmitted message based on the
device and receiver separation distance. The parameters of
the model can easily be related to the main considerations
of the device function. These include: charging time for
sufficient energy to obtain VDD as a function of C; the
power for the on-board electronics by RS; the transmitted
power in RL; and the required supply voltage, VRF, in
relation to VDD through RC. From the above, the available
parameters link all of the major functions of the device and
can be determined from empirical data using traditional
analytic techniques.
The device input and output power can in turn be
linked to the powering transmitter and the communica-
tions receiver in terms of traditional physical relationships
as illustrated in Fig. 2. Thus, the analysis presented using
the device model makes it possible to account for the high
drain of energy compared to the rate of energy recovery,
thereby accounting for charging and discharging dynamics
to provide the analytical basis for appropriate communi-
cation protocols. The model can also be correlated to the
classical backscatter radio frequency identification (RFID)
tag [1]. The relative parameter values will be different, but
the same model and parameter characterization are possi-
ble. This paper has developed a model for the power sup-
ply that directly links the RF input in terms of watts per
square meter to the output of the power supply in terms
of DC volts. This model coupled with the communication
248

model thus links the entire energy train from RF source to
device to receiver in terms of variables that can be directly
measured and used in the device and total system design.
Acknowledgements
This research was funded in part by the National
Aeronautics and Space Administration under contract
no. NNK04OA29C; the Pittsburgh Digital Greenhouse
through a grant from the Commonwealth of Pennsylvania,
Department of Community and Economic Development,
project no. 2264323; the Defense Research Projects
Agency under contract no. GS-35F-5556H; the NCIIA
grant nos. 41-96, 101-97, and 199-98; the U.S. Defense Lo-
gistics Agency; and the McGowan Center of the University
of Pittsburgh.
References
[1] Total asset visibility using intelligent labels, in J.D. Porter,
M.H. Mickle, & R.E. Billo, Warehouse of the Future Conference,
Atlanta, GA, May 14–16, 2000.
[2] Klaus Finkenzeller, RFID Handbook: Radio-frequency identi-
fication fundamentals and applications (Chichester, UK: John
Wiley and Sons, 1999).
[3] K.S.J. Pister, J.M. Kahn, & B.E. Boser, Smart dust: Wireless
networks of millimeter-scale sensor nodes, Highlight Article
in 1999 Electronics Research Laboratory Research Summary,
University of California, Berkeley.
[4] M.H. Mickle, K.W. Wells, & R.G. Hoelzeman, Apparatus for
energizing a remote station and related method, U.S. Patent
No. 6,289,237, University of Pittsburgh, September 11, 2001
(40 claims).
[5] M.H. Mickle, M. Lovell, L. Mats, L. Neureuter, & D. Gorodet-
sky, Energy harvesting: Profiles and potential sources, In-
ternational Journal of Parallel and Distributed Systems and
Networks, 4(3), 2001, 150–160.
[6] J.T. Cain, W.C. Clark, L. Schaefer, W. Moldeki, D. Ulinski, &
M.H. Mickle, Energy harvesting for DNA gene sifting and sort-
ing, International Journal of Parallel and Distributed Systems
and Networks, 4(3), 2001, 140–149.
[7] M.H. Mickle, M. Mi, D. Gorodetsky, L. Mats, & L. Neureuter,
Apparatus for energizing a remote station and related method,
U.S. Patent No. 6,615,074, 2003.
[8] C.A. Balanis, Antenna theory: Analysis and design (Hoboken,
NJ: John Wiley & Sons, Hoboken, 1997).
[9] M. Mi, Analysis, design and optimization of antennas on
CMOS integrated circuits for energy harvesting applications,
doctoral diss., University of Pittsburgh, 2003.
[10] M.H. Mickle, R.R. Hoare, R.E. Billo, J.T. Cain, I. Kourtev, &
R.G. Hoelzeman, Autonomous wireless information systems on
a chip with seamless integration, 4th World Multiconference
on Systemics, Cybernetics and Informatics, Orlando, FL, July
2000, 172–176.
[11] D.K. Cheng, Analysis of linear systems (Reading, MA:
Addison-Wesley, 1959).
[12] F.W. Sears & M.W. Zermansky, University physics (Reading,
MA: Addison Wesley, 1957).
Biographies
Marlin H. Mickle received his
Ph.D. from the University of Pitts-
burgh in 1967. He is currently
the Nickolas A. DeCecco Profes-
sor and is the executive director of
the Swanson Center for Product
Innovation. He is active in the ar-
eas of energy harvesting and high-
technology applications, and is the
co-author or co-editor of over 20
books and over 125 refereed pub-
lications. He has held engineering
positions with IBM and Westinghouse and has also served as
program director of the Systems Theory and Applications
Program of the National Science Foundation. He is a Life
Fellow of the IEEE, 1988 recipient of the Systems Research
and Cybernetics Award of the International Institute for
Advanced Studies in Systems Research and Cybernetics,
and received the Carnegie Science Center 2005 Award for
Excellence in Corporate Innovation.
Minhong Mi received his B.Sc.
in electronics from Peking Univer-
sity in 1996 and M.Sc. in opto-
electronics from Shanghai Insti-
tute of Technical Physics, Chinese
Academy of Sciences, in 1999. In
2003 he received his Ph.D. in elec-
trical engineering from the Uni-
versity of Pittsburgh, with a dis-
sertation topic on antenna inte-
gration on silicon wafers. After
working as a research associate at
the University of Pittsburgh for one year, he recently joined
Ansoft Corporation in Pittsburgh as an application engi-
neer to support its high-frequency structure and circuit
simulation tools.
James T. Cain joined the De-
partment of Electrical Engineer-
ing, University of Pittsburgh, in
1966 and is currently a professor
of computer engineering and elec-
trical engineering. He received his
B.Sc., M.Sc., and Ph.D. degrees
from the University of Pittsburgh
in 1964, 1966, and 1970 respec-
tively. He has been a visiting pro-
fessor at the University of Karl-
sruhe in Germany. His industrial
experience includes full-time and consulting positions with
multiple divisions within the Bell Telephone Company of
Pennsylvania, Bell Telephone Laboratories, Westinghouse
Electric Corporation, General Electric Company, RPS In-
corporated, and Westinghouse Savannah River Company.
He has been an active contributor to and office holder in
a variety of professional organizations, including the IEEE
(1995 president), the Computing Sciences Accreditation
249

Board (CSAB; co-founder and 1988-89 president), the Ac-
creditation Board for Engineering and Technology (ABET),
the Federation on Computing in the United States (FO-
CUS), and the IEEE Foundation. His current research
interests are in the embedded systems area with emphasis
on systems utilizing energy harvesting. He is a fellow of the
IEEE.
Timothy Minor of Gnostic Com-
munications, Inc., is the com-
pany’s president and chief exec-
utive oﬃcer. He has been en-
gaged in electronic R&D, design,
project management, program
management, high-tech business
consolidations, and startups for
radio communications, control,
networking, and RF identiﬁca-
tion (RFID) for 29 years. His
background includes both sys-
tems and hardware development incorporating advanced
signal processing for military, aerospace, and commercial
applications development in the defense electronics sector.
Mr. Minor holds an MSEE from Southern Methodist Uni-
versity in Dallas, Texas, and a BSEE from the University
of Pittsburgh.
250

IntJouCompAppl

More Related Content

What's hot (19)

Viewers also liked (17)

Similar to IntJouCompAppl (20)

IntJouCompAppl