See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/327185347

Radio frequency energy harvesting and management for

wireless sensor networks

Chapter · April 2016

CITATIONS READS

58 574

4 authors, including:

Murtala Zungeru Adamu S. R. S. Prabaharan

Botswana International University of Science and Techn… SRM Institute of Science and Technology
96 PUBLICATIONS   994 CITATIONS    101 PUBLICATIONS   2,121 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Special Collection on Wireless Sensor Networks for Pests and Diseases Monitoring: Detection and Control View project

Network Traffic Analysis in an Industrial Environment: A case of BIUST network View project

All content following this page was uploaded by Murtala Zungeru Adamu on 25 August 2018.

The user has requested enhancement of the downloaded file.

Chapter 13

Radio Frequency
Energy Harvesting and
Management for Wireless
Sensor Networks
Adamu M. Zungeru, Li-Minn Ang,
S. R. S. Prabaharan, and Kah Phooi Seng

Contents
13.1 Introduction............................................................................................. 342
13.2 RF Energy Harvesting............................................................................. 342
13.3 Review of Energy Harvesting Systems and Power Consumption in
WSNs....................................................................................................... 344
13.3.1 Ambient RF Sources and Available Power.................................... 344
13.4 RF Energy Harvesting and the Use of the Powercast Harvester................345
13.4.1 Intentional Energy Harvesting...................................................... 346
13.4.2 The Powercast TX91501 Powercaster Transmitter......................... 348
13.4.3 Powercast Power Harvester Receiver............................................. 348
13.4.4 Measurement of RF Power Received and Gains.............................349
13.4.4.1 Friis Transmission Equation............................................349
13.4.4.2 Power Density.................................................................350
13.4.5 Energy Storage............................................................................... 351

341
342  ◾  Adamu M. Zungeru, et al.

13.5 Energy Management in WSNs..................................................................352

13.5.1 Algorithm Operations....................................................................356
13.6 Experiment and Simulation Results..........................................................358
13.6.1 Experimental Results.....................................................................358
13.6.2 Simulation Results.........................................................................359
13.6.3 Real-Time Implementation of RF Powercast Energy Harvester.....362
13.7 Conclusions and Future Work.................................................................. 364
References..........................................................................................................365

13.1  Introduction
Radio frequency (RF) energy harvesting holds a promising future for generating
a small amount of electrical power to drive partial circuits in wireless commu-
nicating electronics devices. Reducing power consumption has become a major
challenge in wireless sensor networks. As a vital factor affecting system cost and
lifetime, energy consumption in wireless sensor networks (WSNs) is an emerging
and active research area. This chapter presents a practical approach for RF energy
harvesting, and management of the harvested and available energy for WSNs using
the improved energy efficient ant-based routing algorithm (IEEABR) as our pro-
posed algorithm. The chapter looks at measurement of the RF power density, cal-
culation of the received power, storage of the harvested power, and management of
the power in WSNs. The routing uses IEEABR technique for energy management.
Practical and real-time implementations of the RF energy using Powercast™ har-
vesters and simulations using the energy model of our Libelium Waspmote to verify
the approach were performed.
The chapter is organized in the following format: The first part of the chap-
ter, provided in section 13.2, covers a general perspective and the objective of the
chapter. Section 13.3 reviews energy harvesting systems and power consumption
in WSNs. Section 13.4 gives a detailed explanation of our RF energy harvesting
method using the Powercast harvesters. Section 13.5 looks into the management of
the harvested energy in our WSNs. Section 13.6 presents experimental setup and
results, while also looking at the simulation results and its environment. Finally,
section 13.7 concludes the chapter with an open research problem and future work
to be done, and a comparative summary of our results with the EEABR algorithm,
and ad hoc on-demand distance vector (AODV), which form strong energy man-
agement protocols.

13.2  RF Energy Harvesting

Finite electrical battery life is encouraging companies and researchers to come up
with new ideas and technologies to drive wireless mobile devices for an enhanced
period of time. Batteries add to size and their disposal adds to environmental
 Radio Frequency Energy Harvesting and Management  ◾  343

pollution. For mobile and miniature electronic devices, a promising solution is

available in capturing and storing the energy from external ambient sources, a
technology known as energy harvesting. Other names for this type of technology
are power harvesting, energy scavenging, and free energy, which are derived from
renewable energy [1]. In recent years, the use of wireless devices is growing in many
applications like mobile phones and sensor networks [2]. This increase in wireless
applications has generated an increasing use of batteries. Many research teams are
working on extending the battery life by reducing the consumption of the devices.
Other teams have chosen to recycle ambient energy like in microelectromechani-
cal systems (MEMS) [3]. The charging of mobile devices is convenient because the
user can do it easily, as for mobile phones. But for other applications, like wireless
sensor nodes that are located in difficult to access environments, the charging of
the batteries remains a major problem. This problem increases when the number
of devices is large and they are distributed in a wide area or located in inaccessible
places. The research on RF energy harvesting provides reasonable techniques of
overcoming these problems.
The rectification of microwave signals to DC power has been proposed and
researched in the context of high-power beaming. It has been proposed for heli-
copter powering [4], solar power satellites [5], and the SHARP system [6]. The DC
power depends on the available RF power, the choice of antenna, and frequency
band. An energy harvesting technique using electromagnetic energy, specifically
radio frequency, is the focus of this chapter. Communication devices generally
have omnidirectional antennas that propagate RF energy in most directions, which
maximizes connectivity for mobile applications. The energy transmitted from the
wireless sources is much higher, up to 30 W for 10 GHz frequency [7], but only
a small amount can be scavenged in the real environment. The rest is dissipated
as heat or absorbed by other materials. The RF power harvesting technique also
is used in radio frequency identification (RFID) tags and implantable electron-
ics devices. Most commonly used wireless sensor nodes consume few µW in sleep
mode and hundreds of µW in active mode. A great factor contributing to energy
harvesting research and development is ultralow-power components.
The management of power available for sensor nodes has been dealt with to
an extent using ant-based routing [8]–[16], which utilizes the behavior of real ants
searching for food through pheromone deposition while dealing with problems
that need to find paths to goals. The simulating behavior of an ant colony leads
to optimization of network parameters for the WSN routing process to provide a
maximum network lifetime.
The main goal of this chapter is to propose practical harvesting of RF energy
using Powercast harvesters while managing the harvested and available energy of
the sensor networks using our proposed algorithm (IEEABR), which helps in the
optimization of the available power. The objective is to efficiently power sensor
networks with or without batteries to maintain network lifetime at a maximum
without performance degradation.
344  ◾  Adamu M. Zungeru, et al.

13.3 Review of Energy Harvesting Systems

and Power Consumption in WSNs
For proper operation of sensor networks, a reliable energy harvesting technique
is needed. Over the years, much work has been done on the research from both
academic and industrial researchers on large-scale energy from various renewable
energy sources. Less attention has been paid to small-scale energy harvesting tech-
niques, although, quite a number of work has been carried out on energy scaveng-
ing for WSNs. The efficient far-field energy harvesting [17] uses a passively powered
RF-DC conversion circuit operating at 906 MHz to achieve power of up to 5.5
µW. In related work [18]–[21], all consider the little available RF energy while uti-
lizing it to power the sensor networks. Bouchouicha, et al. [2] studied ambient RF
energy harvesting in which two systems, the broad band without matching and
narrow band, were used to recover the RF energy. Among but not all of the avail-
able energy harvesting system for wireless sensors are: solar power, electromagnetic
energy, thermal energy, wind energy, salinity gradients, kinetic energy, biomedical,
piezoelectric, pyroelectric, thermoelectric, electrostatic, blood sugar, and tree meta-
bolic energy. These could be further classified into three [22]: thermal, radiant, and
mechanical energy. Based on these, Table 13.1 and Table 13.2 show the comparison
of the different and common energy scavenging techniques.
Beside the harvested energy for the sensor network, the consumption of the
harvested power for the different mode of the network has to be looked upon before
choosing a power harvesting source. A review of some power consumption in some
selected sensor nodes can be found in Gilbert and Balouchi [23]. For some com-
mercial sensor network nodes, the consumption differs, as shown in Table 13.3;
power consumption of the nodes differs among manufacturers.

13.3.1 Ambient RF Sources and Available Power

A possible source of energy comes from ubiquitous radio transmitters. Radio waves,
a part of the electromagnetic spectrum, consist of magnetic and electrical compo-
nents. Radio waves carry information by varying a combination of the amplitude,
frequency, and phase of the wave within a frequency band. On contact with a
conductor, such as an antenna, the electromagnetic (EM) radiation induces electri-
cal current on the conductor’s surface, known as skin effect. The communication
devices use the antenna for transmission and/or reception of data by utilizing the
different frequencies spectrum from 10 to 30 Kz. The maximum theoretical power
available for RF energy harvesting is 7.0 µW and 1.0 µW for 2.4 GHz and 900
MHz frequency, respectively, for free space distance of 40 m. The path loss of sig-
nals will be different in an environment other than free space [29], although, for
our work using the Powercast harvester, the power available for P2110, which oper-
ates at 915 MHz, is 3.5 mW before conversion and 1.93 mW after conversion at a
distance of 0.6 m, and 1 µW at a distance of 11 m [30].
 Radio Frequency Energy Harvesting and Management  ◾  345

Table 13.1  Comparison of Energy Harvesting Sources for WSNs

Energy Performance
Source Classification (Power Density) Weakness Strength

Solar Power Radiant 100 mW/cm3 Requires Can use

energy exposure to without limit
light, and low
efficiency if
device is in
building

RF Waves Radiant 0.02 µW/cm2 at Low efficiency Can use

energy 5 Km from AM inside a without limit
radio building

RF Energy Radiant 40 µW/cm2 Low efficiency Can use

energy at 10 m if out of line without limit
of sight

Body Heat Thermal 60 µW/cm2 Available only Easy to build

energy at 5oC when using
temperature thermocouple
difference is
high

External Thermal 135 µW/cm2 Available only Easy to build

Heat energy at 10oC when using
temperature thermocouple
difference is
high

13.4 RF Energy Harvesting and the Use

of the Powercast Harvester
RF power harvesting is a process whereby RF energy emitted by sources that gen-
erate high electromagnetic fields, such as TV signals, wireless radio networks, and
cell phone towers, but through power generating circuits linked to a receiving
antenna, are captured and converted into usable DC voltage. Most commonly
used is an application for RFID tags in which the sensing device wirelessly sends
a radio frequency to a harvesting device that supplies just enough power to send
back identification information specific to the item of interest. The circuit sys-
tems that receive the detected RF from the antenna are made on a fraction of a
micrometer scale, but can convert the propagated electromagnetic waves to low
voltage DC power at distances up to 100 m. Depending on concentration levels
that can differ through the day, the power conversion circuit may be attached to
346  ◾  Adamu M. Zungeru, et al.

Table 13.2  Comparison of Energy Harvesting Sources for WSNs (Continued)

Body Motion Mechanical 800 µW/ Dependent High power
energy cm3 on motion density not
limited on interior
and exterior

Blood Flow Mechanical 0.93 W at Energy High power

energy 100 mmHg conversion density not
efficiency is limited on interior
low and exterior

Air Flow Mechanical 177 µW/cm3 Efficiency is High power

energy low inside a density
building

Vibrations Mechanical 4 µW/cm3 Has to exist High power

energy at density not
surroundings limited on interior
and exterior

Piezoelectric Mechanical 50 µJ/N Has to exist High power

energy at density not
surroundings limited on interior
and exterior

a capacitor that can disperse a constant required voltage for the sensor and circuit
when there is not a sufficient supply of incoming energy. Most circuits use a float-
ing gate transistor as the diode, which converts the signal into generated power,
but is linked to the drain of the transistor, and a second floating gate transistor
linked to a second capacitor can enable a higher output voltage once the capacitors
reach full potential [31].
The effectiveness of energy harvesting depends largely on the amount and pre-
dictable availability of energy sources, whether from radio waves, thermal differ-
entials, solar or light sources, or even vibration sources. There are three categories
for ambient energy availability: intentional, anticipated, and unknown, as shown
in Figure 13.1.
Our research relies basically on the intentional use of the Powercast harvester.

13.4.1  Intentional Energy Harvesting

The designs rely on an active component in the system, such as an RF transmitter
that can explicitly provide the desired type of energy into the environment when
the device needs it. Powercast supports this approach with an energy source of
3 W, 915 MHz RF transmitters; the P1110 and P2110 also are used along with it as
 Radio Frequency Energy Harvesting and Management  ◾  347

Table 13.3  Comparison of Power Consumption of Some Selected Sensor

Network Nodes
Manufacturers
Operating Intel
Conditions Crossbow Waspmote IMote2 Jennic JN5139
MICAz [24] [25-26] [27] [28]

Radio Standard IEEE 802.15.4/ IEEE IEEE IEEE

Zigbee 802.15.4/Zigbee 802.15.4 802.15.4/Zigbee

Typical Range 100 m (outdoor), 500 m 30 m 1 km

30 m (indoor)

Data Rate 250 kbps 250 kbps 250 kbps 250 kbps

Sleep Mode 15 µA 62 µA 390 µA 2.8 µA

(deep sleep)

Processor 8 mA active 9 mA 31–53 mA 2.7+0.325 mA/

Consumption mode MHz

Transmission 17.4 mA (+0 dBm) 50.26 mA 44 mA 34 mA (+3 dBm)

Reception 19.7 mA 49.56 mA 44 mA 34 mA

Supply Voltage 2.7 V 3.3 V 3.2 V 2.7 V

(min)

Average Power 2.8 mW 1 mW 12 mW 3 mW

receivers. The intentional energy approach is appropriate for other types of energy
as well, such as placing an energy harvester on a piece of industrial equipment that
vibrates when it is operating. Using an intentional energy source allows designers
to engineer a consistent energy solution. A quick look at the basic operation of the
transmitter and receiving circuit is discussed below.

Intentional Anticipated Unknown

Figure 13.1  Pictorial view of intentional, anticipated, and unknown energy sources.
348  ◾  Adamu M. Zungeru, et al.

13.4.2  The Powercast TX91501 Powercaster Transmitter

The Powercast TX91501 is an RF power transmitter specifically designed to pro-
vide both power and data to end devices containing the Powercast P2110 or P1110
power harvester receivers [30]. The transmitter is housed in a durable plastic case
with mounting holes. It is powered by a regulated 5 V DC voltage, mostly from a
power source of 240 V AC, rectified and regulated to its accommodated voltage of
5 V DC from its built-in internal circuitry. The transmitter has a factory set, fixed
power output, and no user adjustable settings. Also, a beautiful control feature is
the status LED that provides a feedback on a functional state. It provides a maxi-
mum of 3 W EIRP (equivalent or effective isotropic radiated power). A side view,
real view, and its transmission state are shown in Figure 13.2.
The Powercast transmitter transmits power in the form of direct sequence
spread spectrum (DSSS) and data in the amplitude shift keying (ASK) modulation
and at a center frequency of 915 MHz. The power output is 3 W EIRP and verti-
cally polarizes for optimal transmission. For data communication, it has an 8-bit
factory set, TX91501 identification (ID) number broadcast with random intervals
up to 10 ms using ASK modulation. Its operating temperature is within the range
of -20°C to 50°C at the power input from mains of 5 VDC/1A.

13.4.3  Powercast Power Harvester Receiver

The Powercast receivers can harvest directly a directed or ambient RF energy and
convert it to DC power for remotely recharging batteries or battery-free devices.
The two modules available for our research are P1110 and P2110 and both have
similarities and differences in their area of applications, as shown in Table 13.4.
The Powercast P2110 power harvester receiver is an RF energy harvesting device
that converts RF energy to DC voltage. It has a wide RF operating range, and provides

Figure 13.2  (a) A side view and (b) a real view of a TX91501 Powercaster trans-
mitter in its transmission state.
 Radio Frequency Energy Harvesting and Management  ◾  349

Table 13.4  Comparison of the Two RF Energy Powercast Eeceivers

Receivers Differences Similarities

P2110 1.  Design for battery charging and 1.  Harvesting range
direct power applications from 850–950 MHz
2.  Provide intermittent/pulsed power 2.  Works with standard
output 50-ohm antennas
3.  Configurable, regulated output
voltage up to 5.25 V
4.  Power management and control
I/O for system optimization

P1110 1.  Design for battery charging and 1.  Harvesting range
direct power applications from 850–950 MHz
2.  Configurable over voltage protection 2.  Works with standard
up to 4.2 V 50-ohm antennas
3.  Connect directly to rechargeable
batteries including alkaline, lithium
ion, and Ni-MH.

RF energy harvesting, and power management for battery-free micropower devices.

It converts the RF energy to DC and stores it in a capacitor as well as boosting the
voltage to the set output voltage level and enables the voltage output.

13.4.4  Measurement of RF Power Received and Gains

Utilized in our experiment were power meters that provide the most accurate mea-
surement of RF power of any of the types of RF measurement equipment, and
the simplified Friis equation that provides a reasonable estimate of the amount of
power that is received and available for use.

13.4.4.1  Friis Transmission Equation

The Friis transmission formula is used solely for studying RF communication links
[32]. The formula can be used in situations where the distance between two anten-
nas is known, and a suitable antenna needs to be found. Using the Friis transmis-
sion equation, one can solve for the antenna gains needed at either the transmitter
or receiver in order to meet certain design specifications.

2
Pr ⎛ λ ⎞
= Gt Gr ⎜ (13.1)
Pt ⎝ 4 πR ⎟⎠

350  ◾  Adamu M. Zungeru, et al.

where Pr is the received power in watts (W), Pt is the transmitted power, Gt is the
transmitting antenna’s gain, Gr is the receiving antenna’s gain, λ is the wavelength
of the transmitted and received signal in meters, and R is the distance between the
antennas in meters. The gain of the antennas, usually measured in decibels, can be
converted to a power ratio using;
GDB

G = 10 10 (13.2)

c
λ= (13.3)
f

where C is the speed of light in meters per second, and f is the frequency in Hz.
Hence, C is equal to 3 × 108 m/sec.
A simplified version of the Friis equation [33] is provided by the Powercast com-
pany for quick and easy calculation on a spreadsheet, where a reasonable estimate
of the amount of power generated, received, and available for use are calculated.

13.4.4.2  Power Density

RF propagation is defined as the travel of electromagnetic waves through or along a
medium. For RF propagation between approximately 100 MHz and 10 GHz, radio
waves travel very much as they do in free space and travel in a direct line of sight
and a slight difference in the dielectric constants of air and space [34]. For air, the
dielectric is one and 1.000536 at sea level. In antennas theory, an isotropic radiator
is a theoretical, lossless, omnidirectional (spherical) antenna [34] [35]. That is, it
radiates uniformly in all directions. The power of a transmitter that is radiated from
an isotropic antenna will have a uniform power density (power per unit area) in all
directions. Power density at any distance from an isotropic antenna is the ratio of
the transmitted power by the surface area of a sphere (4 ≠R 2 ) at that distance. The
surface area of the sphere increases by the square of the radius, therefore, the power
density, PD, (watts/square meter) decreases by the square of the radius.

Pt
PD = (13.4)
4 πR 2
where Pt is the peak or average power, PD the power density, and R the distance
between the transmitter and the receiving antenna. Radars use directional anten-
nas to channel most of the radiated power in a particular direction. The gain (Gt)
of an antenna is the ratio of power radiated in the desired direction as compared to
the power radiated from an isotropic antenna, or:

Maximum radiation intensity of actual antennaa

Gt = (13.5)
Radiation intensity of isotropic antenna with same power input
 Radio Frequency Energy Harvesting and Management  ◾  351

The power density at a distant point from the radar with an antenna gain of
Gt is the power density from an isotropic antenna multiplied by the radar antenna
gain. Power density from the radar:

PG
t t
PD = (13.6)
4 πR 2

13.4.5  Energy Storage

The most common energy storage device used in a sensor node is a battery, either
nonrechargeable or rechargeable. A nonrechargeable battery (e.g., alkaline) is suit-
able for a microsensor with very low power consumption (e.g., 50 µW). Alternatively,
a rechargeable battery (e.g., lithium ion) is used widely in sensor nodes with energy
harvesting technology [36]. A battery is used not only for storage of energy gener-
ated by the harvesting device, but also to regulate the supply of energy to a sensor
node. A wireless sensor node is powered by exhaustible batteries [37]. Several fac-
tors affect the quality of these batteries, but the main factor is cost. In a large-scale
deployment, the cost of hundreds and thousands of batteries is a serious deploy-
ment constraint. Batteries are specified by a rated current capacity, C, expressed
in ampere-hours. This quantity describes the rate at which a battery discharges
without significantly affecting the prescribed supply voltage (or potential differ-
ence). Practically, as the discharge rate increases, the rated capacity decreases. Most
portable batteries are rated at 1 C. This means a 1,000 mAh battery provides 1,000
mA for 1 hour, if it is discharged at a rate of 1 C. Ideally, the same battery can dis-
charge at a rate of 0.5 C, providing 500 mA for 2 hours, and at 2 C, 2,000 mA for
30 minutes and so on. 1 C is often referred to as a 1-hour discharge. Likewise, a 0.5
C would be a 2-hour and a 0.1 C a 10-hour discharge. In reality, batteries perform
at less than the prescribed rate. Often, the Peukert equation is applied to quantify-
ing the capacity offset (i.e., how long a battery lasts in reality):

c
T = (13.7)
In

where C is the theoretical capacity of the battery expressed in ampere-hours, I is the

current drawn in ampere (A), T is the time of discharge in seconds, and n is the
Peukert number, a constant that directly relates to the internal resistance of the battery.
The value of the Peukert number indicates how well a battery performs under
continuous heavy current. A value close to 1 indicates that the battery performs
well; the higher the number, the more capacity is lost when the battery is dis-
charged at high current. The Peukert number of a battery is determined empiri-
cally. For example, for lead acid batteries, the number is typically between 1.3 and
1.4. Drawing current at a rate greater than the discharge rate results in a current
352  ◾  Adamu M. Zungeru, et al.

consumption rate higher than the rate of diffusion of the active elements in the
electrolyte. If this process continues for a long time, the electrodes run out of active
material even though the electrolyte has not yet exhausted its active materials. This
situation can be overcome by intermittently drawing current from the battery and
also proper power management techniques.

13.5  Energy Management in WSNs

Despite the fact that energy scavenging mechanisms can be adopted to recharge
batteries, e.g., through Powercast harvesters [30], solar panels [2], or piezoelectric or
acoustic transducers [21], energy is a limited resource and must be used judiciously.
Hence, efficient energy management strategies must be devised at the sensor nodes
to prolong the network lifetime as much as possible. Many routing, power manage-
ment, and data dissemination protocols have been specially designed for WSNs
[38]. The EAGRP (energy-aware geographic routing protocol) [39], an enhanced
AODV (ad hoc on-demand distance vector) [40], and an EEABRA for WSNs [41]
all have developed different protocols in order to manage the available energy in
WSNs. In a related work, Alippi, et al. [42] use energy-hungry sensors in trying to
manage the available energy in WSNs. Reducing power consumption has become
a major challenge in wireless sensor networks. As a vital factor affecting system
cost and lifetime, energy consumption in WSNs is an emerging and active research
area. The energy consumption of WSNs is of crucial concern due to the limited
availability of energy. Whereas energy is a scarce resource in every wireless device,
the problem in WSNs is more severe for the following reasons [37]:

1. Compared to the complexity of the task they carry out, sensing, processing,
self-managing, and communication, the nodes have been very small in size to
accommodate high-capacity power supplies.
2. While the research community is investigating the contribution of renewable
energy and self recharging mechanisms, the size of nodes is still a constrain-
ing factor.
3. Ideally, a WSN consists of a large number of nodes. This makes manually
changing, replacing, or recharging batteries almost impossible.
4. The failure of a few nodes may cause the entire network to fragment
prematurely.

The problem of power consumption can be approached from two angles. One
is to develop energy efficient communication protocols (self-organization, medium
access, and routing protocols) that take the peculiarities of WSNs into account. The
other is to identify activities in the networks that are both wasteful and unnecessary
and mitigate their impact. Wasteful and unnecessary activities can be described as
local (limited to a node) or global (having a scope network-wide). In either case, these
 Radio Frequency Energy Harvesting and Management  ◾  353

activities can be further considered as accidental side effects or results of nonoptimal

software and hardware implementations (configurations). For example, observations
based on field deployment reveal that some nodes exhausted their batteries prema-
turely because of unexpected overhearing of traffic that caused the communication
subsystem to become operational for a longer time than originally intended [43].
Similarly, some nodes exhausted their batteries prematurely because they
aimlessly attempted to establish links with a network that had become no longer
accessible to them. Most inefficient activities are, however, results of nonoptimal
configurations in hardware and software components. For example, a considerable
amount of energy is wasted by an idle processing or a communication subsystem.
A radio that aimlessly senses the media or overhears while neighboring nodes com-
municate with each other consumes a significant amount of power. A dynamic
power management (DPM) control strategy is aimed at adapting the power/per-
formance of a system to its workload. The DPM, having a local or global scope,
or both, aims at minimizing power consumption of individual nodes by providing
each subsystem with the amount of power that is sufficient to carry out a task at
hand [37]. Hence, it does not consider the residual energy of neighboring nodes.
IEEABR as the proposed algorithm considers the available power of nodes and the
energy consumption of each path as the reliance of routing selection. It improves
memory usage and utilizes the self organization, self-adaptability, and dynamic
optimization capability of the ant colony system to find the optimal path and mul-
tiple candidate paths from source nodes to sink node. The protocol avoiding using
up the energy of nodes on the optimal path and prolongs the network lifetime while
preserving network connectivity. This is necessary because, for any WSN protocol
design, the important issue is the energy efficiency of the underlying algorithm due
to the fact that the network under investigation has strict power requirements.
It has been proposed by Kalpakis, Dasgupta, and Namjoshi [44] that forward
ants be sent directly to the sink node; the routing tables only need to save the neigh-
bor nodes that are in the direction of the sink node. This considerably reduces the
size of the routing tables and, in essence, the memory needed by the nodes. Since
one of the main concerns in WSN is to maximize the lifetime of the network,
which means saving as much energy as possible, it would be preferable that the
routing algorithm could perform as much processing as possible in the network
nodes, rather than transmitting all data through the ants to the sink node to be
processed there. In fact, in huge sensor networks where the number of nodes can
easily reach more than thousands of units, the memory of the ants would be so large
that it would be unfeasible to send the ants through the network. To implement
these ideas, the memory Mk of each ant is reduced to just two records, the last two
visited nodes [41]. Since the path followed by the ants is no longer in their memo-
ries, a memory must be created at each node that keeps records of each ant that was
received and sent. Each memory record saves the previous node, the forward node,
the ant identification, and a timeout value. Whenever a forward ant is received, the
node looks into its memory and searches the ant identification for a possible loop.
354  ◾  Adamu M. Zungeru, et al.

If no record is found, the node saves the required information, restarts a timer, and
forwards the ant to the next node. If a record containing the ant identification is
found, the ant is eliminated. When a node receives a backward ant, it searches its
memory to find the next node to where the ant must be sent. In this section, we
modify the EEABR to improve the energy consumption in the nodes of WSNs and
also to improve the performance and efficiency of the networks. The main focus of
this chapter is on IEEABR power management strategies in WSNs.
The algorithm of our proposed power management techniques is as follows.

1. Initialize the routing tables with a uniform probability distribution:

1
Pld = (13.8)
Nk

where PId is the probability of jumping from node l to node d (destination),

and Nk is the number of nodes.
2. At regular intervals, from every network node, a forward ant k is launched
with the aim of finding a path to the destination. The identifier of every vis-
ited node is saved onto a memory Mk and carried by the ant.
Let k be any network node; its routing table will have N entries, one for each
possible destination.
Let d be one entry of k routing table (a possible destination).
Let Nk be the set of neighboring nodes of node k.
Let Pkl be the probability with which an ant or data packet in k, jumps to a
node l, l∈Nk, when the destination is d (d ≠ k). Then, for each of the N entries
in the node k routing table, it will be nk values of Pld subject to the condition:

∑ l ∈N k
Pld = 1; d = 1,  , N (13.9)

3. At every visited node, a forward ant assigns a greater probability to a destina-

tion node d for which falls to be the destination among the neighbor nodes,
d ∈ Nk.. Hence, initial probability in the routing table of k is then:

9N k − 5
Pdd = (13.10)
4 N k2
Also, for the rest of the neighboring nodes among the neighbors for which
m ∈ Nk will then be:

⎧ 4N k − 5
⎪ , if N k > 1
Pdm = ⎨ 4 N k2 (13.11)
⎪ 0, if N k = 1
⎩
 Radio Frequency Energy Harvesting and Management  ◾  355

Of course, equation (13.10) and equation (13.11) satisfy equation (13.9). But,
if none of the neighbors is a destination, equation (13.8) applies to all the
neighboring nodes. Or
4. The forward ant selects the next hop node using the same probabilistic rule
proposed in the ACO (ant colony optimization) metaheuristic:

⎧
⎪
⎪
⎪⎪ [τ(r , s )]α.[ E ( s )]β
Pk (r , s ) = ⎨ , s ∉ Mk
∑
⎪
[τ(r , u )]α.[ E ( s )]β (13.12)
⎪ u∉Mk
⎪
⎪⎩ 0, else

where pk(r, s) is the probability with which ant k chooses to move from node r
to node s, τ is the routing table at each node that stores the amount of phero-
mone trail on connection (r, s), Ε is the visibility function given by (C −1es ) (c is
the initial energy level of the nodes and es is the actual energy level of node s),
and α and β are parameters that control the relative importance of trail versus
visibility. The selection probability is a trade-off between visibility (which says
that nodes with more energy should be chosen with high probability) and
actual trail intensity (which says that if on connection (r, s) there has been a
lot of traffic, then it is highly desirable to use that connection).
5. When a forward ant reaches the destination node, it is transformed into a
backward ant whose mission is now to update the pheromone trail of the path
it used to reach the destination and that is stored in its memory.
6. Before backward ant k starts its return journey, the destination node com-
putes the amount of pheromone trail that the ant will drop during its journey:
1
Δτ = E Mink − Fd k ⎤ (13.13)
C − ⎡⎢
⎣ E Avg k − Fdk ⎦⎥

And, the equation used to update the routing tables at each node is

⎡ Δτ ⎤
τ(r , s ) = (1 − ρ) ∗ τ(r , s ) + ⎢ ⎥ (13.14)
⎣ φ.Bdk ⎦

where ϕ is a coefficient and Bdk is the distance travelled (the number of visited
nodes) by the backward ant k until node r, in which the two parameters will
force the ant to lose part of the pheromone strength during its way to the
source node. The idea behind the behavior is to build a better pheromone
distribution (nodes near the sink node will have greater pheromone levels)
356  ◾  Adamu M. Zungeru, et al.

and will force remote nodes to find better paths. Such behavior is important
when the sink node is able to move, because pheromone adaptation will be
much quicker [41].
7. When the backward ant reaches the node where it was created, its mission is
finished and the ant is eliminated.

By performing this algorithm in several iterations, each node will be able to know
which are the best neighbors (in terms of the optimal function represented by equa-
tion (13.14)) to send a packet toward a specific destination. The flow chart describing
the action of movement of a forward ant for our proposed algorithm is shown in
Figure 13.3. The backward ant takes the opposite direction of the flow chart.

13.5.1  Algorithm Operations

After the initialization of the routing table and setting up a forward ant for hopping
from node to node in search of the sink, at every point in time, a node becomes a
source holding, in its stack or memory, information about an event around itself
(neighbors). The information gathered in its memory is transferred or dissemi-
nated toward the sink node with the help of neighbor nodes behaving as repeaters.
Associated raw data generated at each source (nodes) are divided into M pieces,
known as data parts. An integer value M also represents the number of ant agents
involved in each routing task. This raw data provided by the source node about an
event contains information, such as source node identification, event identification,
time, and the data about the event. The data size is chosen based on the sensor
nodes deployed and the size of the buffer. After the splitting of the raw data, each
part is associated with routing parameters to build a data packet ready to transfer.
These parameters are code identification, describing the code following as data,
error, or acknowledge: CID. Next is the node identification to which the packet
is transferred: NID. The packet number also represents the ant agent k; SN is the
sequence number, Nk, which contains the number of visited nodes so far and the
kth data part shown in Figure 13.4. In this figure, the group of the first four fields
is named the data header. When delivery of all data packages is accomplished, the
base combines them into raw data.
When a node participating in a routing receives a data packet whose agent num-
ber is given, it makes decisions about the next destination for that packet of data.
The decision on the next node or destination for which the packet of data should be
transferred will depend on equation (13.10), otherwise equation (13.12) with the
highest probability. The pheromone level of the neighbors that is the first determin-
ing factor follows the energy levels of the neighbor nodes, which are most important
on the decision rule. For any of the neighbors chosen, the NID field of the node is
updated and the packet is then broadcast. The remaining neighbors among the cho-
sen node also hear the broadcast; they check the NID field and understand that the
message is not made for them. They then quickly discard the packet immediately
 Radio Frequency Energy Harvesting and Management  ◾  357

Start

Initialize the
routing table and
get list of
neighboring nodes

Delete Yes Nodes

Memory visited = 2?

No

Select node No Yes

according to Loop? Die
Eqn 12

Node Yes Remove node from

previously
list
visited?

No

Move to the Next

Node

Add Node to No Destination

Memory and (Sink) reached?
repeat step 1

Yes

Die

End

Figure 13.3  An IEEABR forward ant flow chart.

CID NID SN Nk Data part(k th)

Data Header

Figure 13.4  Data packet content.

358  ◾  Adamu M. Zungeru, et al.

after only listening to the NID field of the packet. The Nk is updated with an incre-
ment by one after ensuring that SN is not in the list of the tabus (routing table) of
the chosen node. The next node is determined to update the NID field by performing
the same operation as performed earlier by the first node and the sequence continues
until the packet gets to the sink node. The reversed operation is done for the back-
ward ant as for the acknowledgment, which gets to the source: now the last bus stop
for the backward ant, and it dies off after reaching the source.

13.6  Experiment and Simulation Results

Different experiments are conducted to measure the circuit’s parameter and
the influence of the RF power source. Simulation results based on the per-
formance of the circuit with differences in distance of the harvester from the
power sources, the energy usage, and energy management using our proposed
IEEABR, are all analyzed below while showing the harvesting setup and the
simulation environment.

13.6.1  Experimental Results

Using the Powercast calculator and setting components P2110 at 1.2 V to 915
MHz, battery capacity at 1150 mAh for P2110, and P1110 at 4.0 V to 915 MHz
for the same battery capacity, while varying the distance between the transmitters,
the readings will be as shown in Table 13.5 to Table 13.8, with differences in the
receiver and antenna used in the experiment. The behavior of the packets received
with time is shown in Figure 13.5a,b , while the packets received with distance for
different harvesters and antennas are compared in Figure 13.6a,b.

Table 13.5  Amount of Power Harvested by P2110

Harvester Using a Dipole Antenna
Distance (ft) P (µW) I (µA) Recharge Time (hrs)

2 3687 3073 22.08

5 523 436 155.04

10 135 112 602.64

12 85 71 952.32

15 37 31 2169.12

18 11 9 7360.56

20 1 1 68339.28
 Radio Frequency Energy Harvesting and Management  ◾  359

Table 13.6  Amount of Power Harvested by P2110

Harvester Using a Patch Antenna
Distance (ft) P (µW) I (µA) Recharge Time (hrs)

5 1925 1604 42.24

10 386 322 210.50

15 189 158 429.40

18 131 109 618.5

20 102 85 797.50

25 50 41 1639.00

30 19 16 4353.00

35 5 4 15517.00

36 1 1 70019.00

Table 13.7  Amount of Power Harvested by P1110

Harvester Using a Dipole Antenna
P
Distance (ft) (µW) I (µA) Recharge Time (hrs)

2 3688 922 62.40

4 1085 271 211.92

6 259 65 888.72

7 86 22 2659.92

13.6.2  Simulation Results

We use event-driven network simulator-2 (NS-2) [45] based on the network topol-
ogy to be able to evaluate the implementation of the proposed energy management
protocol. This software provides a high simulation environment for wireless com-
munication with detailed propagation, MAC (media access control), and radio
layers. AntSense (an NS-2 module for Ant Colony Optimization) [46] was used
for the EEABR. The simulation parameters are shown in Table 13.9. We assume
that all nodes have no mobility since the nodes are fixed in the application of most
wireless sensor networks. Simulations were run for 60 minutes (3,600 seconds)
each time the simulation started, and the remaining energy of all nodes were taken
and recorded at the end of each simulation. The average energy was calculated
while also noting the minimum energy of the nodes. This helps in recording tracks
360  ◾  Adamu M. Zungeru, et al.

Table 13.8  Amount of Power Harvested by P1110

Harvester Using a Patch Antenna
Distance (ft) P (µW) I (µA) Recharge Time (hrs)

2 16115 4029 14.16

4 3070 768 74.88

6 1551 388 148.30

8 810 203 283.90

10 366 92 627.60

12 93 23 2475.00

13 26 7 8750.00

of the performance of the management protocols in term of the network’s energy

consumption.
As energy is the key parameter to be considered when designing protocol for
power management to enhance the maximum lifetime of sensor networks, we use:
1. The minimum energy that gives the lowest energy amount of all nodes at the
end of simulations
2. The average energy that represents the average of energy of all nodes at the
end of simulation
3. The simulation was done on a static WSN where sensor nodes were randomly
deployed with the objective of monitoring a static environment
Nodes were responsible for monitoring and sending the relevant sensor data to the
sink node where nodes near the phenomenon will easily depreciate in energy because

9 15
P2110, Dipole antenna
8
Time between Packets (sec)

Time between Packets (sec)

P2110, Patch antenna

7
6 10
5
4
3 5
2
1
0 0
4 6 8 10 12 14 16 18 20 22 5 10 15 20 25 30 35 40 45
Distance (feet) Distance (feet)
(a) TX91501-3W EIRP, 915MHz power transmitter (b) TX91501-3W EIRP, 915MHz power transmitter

Figure 13.5  Variation in time between Packets received and distance of harvesting.
 Radio Frequency Energy Harvesting and Management  ◾  361

9 8000
P2110, Dipole antenna P2110, Dipole antenna
8 P2110, Patch antenna 7000 P2110, Patch antenna

Recharge Time ×10000 (hrs)

Time between Packets (sec)

7
6000
6
5000
5
4000
4
3000
3
2 2000

1 1000
0 0
4 6 8 10 12 14 16 18 20 22 4 6 8 10 12 14 16 18
Distance (feet) Distance (feet)
(a) Comparison of packet received with (b) Comparison of battery recharge time,
distance using different antenna using dipole and patch antennas with P2110

Figure 13.6  Comparison of power harvesting using dipole and patch antennas
with P2110.

Table 13.9  Simulation Parameters

Parameters Values

Routing Protocols AODV, EEABR, IEEABR

MAC Layer IEEE 802.15.4

Frequency 2.4 GHz

Packet Size 1 Mb

Area of Deployment 200 × 200 m2 (10 nodes), 300 × 300 m2

(20 nodes), 400 × 400 m2 (30 nodes),
500 × 500 m2 (40 nodes), 600 × 600 m2
(50–100 nodes),

Data Traffic Constant Bit Rate (CBR)

Simulation Time 3,600 sec.

Battery power 1,150 mAH, 3.7 V

Propagation Model Two-ray ground reflection

Data Rate 250 Kbps

Current Draw in Sleep Mode 62 µA

Current Draw in Transmitting Mode 50.26 mA

Current Draw in Receiving Mode 49.56 mA

Current Draw in Idle Mode (Processor) 9 mA

362  ◾  Adamu M. Zungeru, et al.

they will be forced to periodically transmit data. Simulations were run for 60 min-
utes (3,600 seconds) each time the simulation started, and the remaining energy of
all nodes was taken and recorded at the end of each simulation. The average energy
was calculated while also noting the minimum energy of the nodes. Figure  13.8
presents the results of the simulation for the studied parameters: the average energy,
and minimum energy of AODV, EEABR, and IEEABR. As can be seen from the
results presented in Figure 13.8, the IEEABR protocol had better results in both
the average energy of the nodes and the minimum energy of nodes experienced at
the end of simulation. The AODV as compared to the EEABR performs worst in all
cases. In term of average energy levels of the network, IEEABR as compared with
EEABR average energy values varies between 2 and 8 percent, while AODV is in
the range of 15 to 22 percent also to the minimum energy of the nodes. Figure 13.7
shows a screenshot of a NAM window of the simulation environment for 10 nodes
randomly deployed, while the results of the simulations are as shown in Figure 13.8.

13.6.3 Real-Time Implementation of RF
Powercast Energy Harvester
The real-time implementation in harvesting RF energy from Powercast harvest-
ers and the management of the power that is harvested, using the IEEABRA, is

Figure 13.7  Graphical representation of the simulation environment in NS-2.34

with 10 nodes.
 Radio Frequency Energy Harvesting and Management  ◾  363

1
1
0.95
0.9 0.9

Minimum Energy (%)

Average Energy (%)

0.85
0.8 0.8
0.75
0.7 0.7
0.65
AODV AODV
0.6 EEABR 0.6 EEABR
0.55 IEEABR IEEABR
0.5 0.5
0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100
No. of Nodes No. of Nodes
(a) Average energy of network nodes (b) Energy of node with minimum energy

Figure 13.8  Performance analysis of AODV, EEABR, and IEEABR energy man-
agement protocols.

presented in Figure 13.9. The experiment was set up and available power mea-
sured while also varying the distance between the harvester and transmitter. The
time between each packet delivery, i.e., the harvesting period, were noticed and
recorded. The power consumption of the Waspmote under consideration can be
found in Waspmote: Technical Guide–Libelium [25] and Zigbee Protocol [26].

(a) (b) (c)

(d) (e)

Figure 13.9  Hardware setup of the real-time implementation.

364  ◾  Adamu M. Zungeru, et al.

The battery powering of the Waspmote is 1,150 mAh at 3.7 V, which can suffi-
ciently power each of the nodes separately under constant transmission or recep-
tion for 19.39 h. For our management protocol that was applied, the maximum
energy consumed was found to be 23 percent of the supply energy amounting
to total current drawn to be 264.5 mA for 1 h. It then means that the battery
can sufficiently power the node with the minimum energy for 4.35 h without
recharging. For the recharging of the battery at 15 ft, as shown in Table 13.6, it
takes 429.4 h to fully recharge when empty, and 91.9 h to replenish the drawn
current of 264.5 mA. However, with constant harvesting, it then means that the
total energy of the battery remains without reduction, which can then sustain
the network for the required number of years needed for sensing. A quick look at
the receiver in its receiving and conversion state with dipole and patch antenna
connected for the application of harvesting from the Powercast transmitter and
the harvesting mode of the receiver 3 ft (0.914 m) away from the transmitter,
respectively, are shown in Figure 13.9a-c. Shown are the P2110 Powercast har-
vester receiver with (a) dipole (omnidirectional) antenna, (b) patch (directional)
antenna, (c) TX95101 Powercaster transmitter in its harvesting mode, (d) with
Waspmote, and (e) Gateway connected to the sink. The results of the measure-
ments of the harvested RF energy are presented in Table 13.5 to Table 13.8 and
Figure 13.5a,b.

13.7  Conclusions and Future Work

In this chapter, research based on the application of RF energy harvesting, using
Powercast harvesters to support the limited available energy of wireless sensor
networks, and its management using Ant Colony Optimization metaheuristic
was adopted. In this work, we proposed an Improved Energy Efficient Ant Based
routing Algorithm energy management technique, which improves the lifetime
of sensor networks. The IEEABR utilizes initialization of uniform probabilities
distribution in the routing table while giving special consideration to neighboring
nodes that fail to be the destination (sinks) in other than to save time in searching
for the sink, leading to reduced energy consumption by the nodes. The experi-
mental results showed that the algorithm leads to very good results in different
WSNs. Also looking at the harvested energy, the time of charging the battery
powering the sensor nodes is drastically reduced, while requiring time intervals of
91.9 hours to recharge the battery. The protocol considers the residual energy of
nodes in the network after each simulation period. Based on NS-2 simulation, the
IEEABR approach has effectively balanced the WSN node power consumption
and increased the network lifetime. Consequently, our proposed algorithm can
efficiently extend the network lifetime without performance degradation. This
algorithm focused mainly on energy management and the lifetime of wireless
sensor networks.
 Radio Frequency Energy Harvesting and Management  ◾  365

As for future work, we intend to build a linking circuit so as to directly charge

the Waspmote battery from the Powercast harvesters, harvest the useless energy
from the Waspmote, study a dual approach in the selection of sink and the self
destruction of the backward ants should there exist a link failure, and find an alter-
nate means of retrieving the information carried by the backward ant to avoid loss
of information. We also intend to design a maximum power point tracker (MPPT),
so as to dual power the Waspmote, and model both sources for perpetual operation
of the sensor networks.

REFERENCES
[1] F. E. Little, J. O. McSpadden, K. Chang, and N. Kaya, “Toward Space Solar Power:
Wireless Energy Transmission Experiments Past, Present and Future.” AIP Conference
Proceedings of Space Technology and Applications International Forum, vol. 420, pp. 1225–
1233, 1998.
[2] D. Bouchouicha, F. Dupont, M. Latrach, and L. Ventura, “Ambient RF Energy
Harvesting,” International Conference on Renewable Energies and Power Quality
(ICREPQ’10), Granada (Spain), March, 2010.
[3] S. P. Beeby, M. J. Tudor, and N. M. White, “Energy Harvesting Vibration Sources for
Microsystems Applications,” Measurements Science and Technology, vol. 17, pp. 175–
195, 2006.
[4] R. M. Dickinson, “Evaluation of a microwave high-power reception-conversion array
for wireless power transmission,” Tech. Memo 33-741. Jet Propulsion Laboratory,
California Institute of Technology, Pasadena, CA, Sept. 1975.
[5] H. Hayami, M. Nakamura, and K. Yoshioka, “The Life Cycle CO2 Emission
Performance of the DOE/NASA Solar Power Satellite System: A Comparison of
Alternative Power Generation Systems in Japan,” IEEE Transactions on Systems, Man,
and Cybernetics Part C: Applications and Reviews, vol. 35, no. 3, August 2005.
[6] T. W. R. East, “Self-Steering, Self-Focusing Phased Array for SHARP,” Antennas and
Propagation Society International Symposium, 1991, AP-S, Digest 24–28, pp. 1732–
1735, vol. 3, June 1991.
[7] RF HAMDESIGN Microwave equipments and parts. Online at: http://www.rfham-
design.com/products/parabolicdishkit/1682909a390cc1b03/index.html
[8] Y. Zhang, L. D. Kuhn, and M. P. J. Fromherz, “Improvements on Ant Routing for
Sensor Networks,” Ant Colony, in Optimization and Swarm Intelligence, Lecture Notes
Computer Science, 2004, 3172, pp. 289–313.
[9] P. X. Liu, “Data Gathering Communication in Wireless Sensor Networks Using
Ant Colony Optimization,” 2004 IEEE International Conference on Robotics and
Biomimetics, 2004, pp. 822–827.
[10] Y.-Feng Wen, Y.-Quan Chen, and M. Pan, “Adaptive ant-based routing in wireless sen-
sor networks using Energy*Delay metrics,” Journal of Zhejiang University SCIENCE A,
vol. 9, Mar. 2008, pp. 531–538.
[11] R. GhasemAghaei, M. A. Rahman, W. Gueaieb, and A. El Saddik, “Ant Colony-
Based Reinforcement Learning Algorithm for Routing in Wireless Sensor Networks,”
2007 IEEE Instrumentation & Measurement Technology Conference (IMTC), May
2007, pp. 1–6.
366  ◾  Adamu M. Zungeru, et al.

[12] X. Wang, L. Qiaoliang, X. Naixue, and P. Yi, “Ant Colony Optimization-Based

Location-Aware Routing for Wireless Sensor Networks,” in Proceedings of the Third
International Conference on Wireless Algorithms, Systems, and Applications (WASA’08),
Springer-Verlag, Berlin, Heidelberg, vol. 5258, 2008, pp. 109–120.
[13] M. Paone, L. Paladina, M. Scarpa, and A. Puliafito, “A multi-sink swarm-based rout-
ing protocol for Wireless Sensor Networks,” in IEEE Symposium on Computers and
Communications, July 2009, pp. 28–33.
[14] G. De-Min, Q. Huan-Yan, Y. Xiao-Yong, and W. Xiao-Nan, “Based on ant colony multi-
cast trees of wireless sensor network routing research,” Journal of iet-wsn.org, vol. 2, 2008,
pp. 1–7. Online at http://www.iet-wsn.org/webeditor/UploadFile/201062611395470472.
pdf.
[15] S. Xia and S. Wu, “Ant Colony-Based Energy-Aware Multipath Routing Algorithm
for Wireless Sensor Networks,” in 2009 Second International Symposium on Knowledge
Acquisition and Modeling, Nov. 2009, pp. 198–201.
[16] G. Wang, Y. Wang, and X. Tao, “An Ant Colony Clustering Routing Algorithm for
Wireless Sensor Networks,” in 2009 Third International Conference on Genetic and
Evolutionary Computing, Oct. 2009, pp. 670–673.
[17] T. Le, K. Mayaram, and T. Fiez, “Efficient Far-Field Radio Frequency Energy
Harvesting for Passively Powered Sensor Networks,” IEE Journal of Solid-State Circuits,
vol. 43, no. 5, pp. 1287–1302, May 2008.
[18] H. Jabbar, Y. S. Song, and T. D. Jeong, “RF Energy Harvesting System and Circuits for
Charging of Mobile Devices,” IEEE Transaction on Consumer Electronics, Jan. 2010.
[19] C. Lu, V. Raghunathan, and K. Roy, “Micro-Scale Harvesting: A System Design
Perspective,” in Proceedings of the 2010 Asia and South Pacific Design Automation
Conference (ASPDAC’10), IEEE Press: Piscataway, NJ, 2010.
[20] L. Tang and C. Guy, “Radio Frequency Energy Harvesting in Wireless Sensor Networks,”
in Proceedings of the 2009 International Conference on Wireless Communications and
Mobile Computing (ICWCMC’09): Connecting the World Wirelessly, ACM: New
York, 2009.
[21] J. A. Hagerty, T. Zhao, R. Zane, and Z. Popovic, “Efficient Broadband RF Energy
Harvesting for Wireless Sensors,” Colorado Power Electronics Center (COPEC),
Boulder, CO, 2003.
[22] W. Seah and Y. K. Tan, “Review of Energy Harvesting Technologies for Sustainable
Wireless Sensor Network,” in Sustainable Wireless Sensor Networks, Y. K. Tan (ed.), pp.
15–43. Tech Publishing, Rijeka, Croatia, Dec. 2010.
[23] J. M. Gilbert and F. Balouchi, “Comparison of Energy Harvesting Systems for Wireless
Sensor Networks,” International Journal of Automation and Computing, vol. 5, no. 4,
pp. 334–347, Oct. 2008.
[24] MICAz  Datasheet.  Online at: http://courses.ece.ubc.ca/494/files/MICAz_Datasheet.
pdf
[25] Waspmote: Technical Guide-Libelium. Online at: http://www.libelium.com/documen-
tation/waspmote/waspmote-technical_guide_eng.pdf
[26] Zigbee Protocol. Online at: http://wiki.kdubiq.org/kdubiqFinalSymposium/uploads/
Main/04_libelium_smfkdubiq.pdf
[27] Imote2  Documents—WSN.  Online at: http://www.cse.wustl.edu/wsn/images/e/e3/
Imote2_Datasheet.pdf
[28] JN5139—Jennic  Wireless  Microcontrollers.  Online at: http://www.jennic.com/
products/wireless_microcontrollers/jn5139
 Radio Frequency Energy Harvesting and Management  ◾  367

[29] T. T. Le, “Efficient Power Conversion Interface Circuits for Energy Harvesting
Applications,” PhD disser., Oregon State University, Corvallis, 2008.
[30] Powercast Documentation. Online at: http://Powercastco.com/
[31] B. Dixon, “Radio Frequency Energy Harvesting.” Online at: http://rfenergyharvesting.
com/
[32] J. A. Shaw, “Radiometry and the Friis Transmission Equation.” Online at: http://www.
coe.montana.edu/ee/rwolff/EE548/sring05%20papers/Friis_Radiometric_2005Feb9.
pdf
[33] Wireless Power Calculator. Online at: http://Powercastco.com/wireless-power-calculator.
xls
[34] Power Density. Online at: http://www.phys.hawaii.edu/~anita/new/papers/military-
Handbook/pwr-dens.pdf
[35] R. Struzak, “Basic Antenna Theory.” Online at: http://wirelessu.org/uploads/
units/2008/08/12/39/5Anten_theor_basics.pdf
[36] N. Dusit, H. Ekram, M. R. Mohammad, and K. B. Vijay, “Wireless Sensor Networks
with Energy Harvesting Technologies: A Game-Theoretic Approach to Optimal Energy
Management,” IEEE Wireless Communications, vol. 14, no. 4, pp. 90–96. September 2007.
[37] D. Waltenegus and P. Christian, Fundamental of Wireless Sensor Networks: Theory and
Practice, New York: Wiley Series on Wireless Communication and Mobile Computing,
pp. 207–213, 2010.
[38] A. Kansal, J. Hsu, S. Zahedi, and M. B. Srivastava, “Power management in energy har-
vesting sensor networks,” in Proceedings of ACM Transactions on Embedded Computing
Systems, vol. 6, September 2007, pp. 32–66.
[39] A. G. A. Elrahim, H. A. Elsayed, S. El Rahly, and M. M. Ibrahim, “An Energy Aware
WSN Geographic Routing Protocol,” Universal Journal of Computer Science and
Engineering Technology, vol. 1, no. 2, Nov. 2010, pp. 105–111.
[40] W. Li, M. Chen, and M-M. Li, “An Enhanced AODV Route Protocol Applying in the
Wireless Sensor Networks,” Fuzzy Information and Engineering, vol. 2, AISC 62, pp.
1591–1600, 2009.
[41] T. C. Camilo, Carreto, J. S. Silva, and F. Boavida, “An Energy-Efficient Ant Based
Routing Algorithm for Wireless Sensor Networks,” In Proceedings of 5th International
Workshop on Ant Colony Optimization and Swarm Intelligence, Brussels, Belgium, pp.
49–59, 2006.
[42] C. Alippi, G. Anastasi, M. D. Francesco, and M. Roveri, “Energy management in
Wireless Sensor Networks with Energy-Hungry Sensors,” IEEE Instrumentation and
Measurements Magazine, vol. 12, April 2009, pp. 16–23.
[43] X. Jiang, J. Taneja, J. Ortiz, A. Tavakoli, P. Dutta, J. Jeong, D. Culler, P. Levis, and
S. Shenker, “Architecture for Energy Management in Wireless Sensor Networks,” ACM
SIGBED Review, Special issue on the workshop on wireless sensor network architec-
ture, vol. 4, no. 3, pp. 31–36, July 2007.
[44] K. Kalpakis, K. Dasgupta, and P. Namjoshi, “Maximum Lifetime Data Gathering
and Aggregation in Wireless Sensor Networks,” in Proceedings of IEEE International
Conference on Networking, vol. 42, no. 6, August 2003.
[45] NS2 installation on Linux. Online at: http://paulson.in/? p=29
[46] NS-2 Module for Ant Colony Optimization (AntSense). Online at: http://eden.dei.
uc.pt/~tandre/antsense/
View publication stats