A battery-free smart sensor powered with RF Energy

R. La Rosa, G. Zoppi L. Di Donato, G. Sorbello

STMicroelectronics Dipartimento di Ingegneria Elettrica Elettronica ed
Catania, Italy Informatica, Universita’ di Catania

C. A. Di Carlo, P. Livreri
[email protected] Dipartimento di Energia, Ingegneria dell’Informazione e
Modelli Matematici, Università di Palermo, Palermo, Italy

Abstract — The development of Internet of Things (IoT) limitation of system lifetime due to the presence of a battery,
infrastructure and applications is stimulating advanced and achieving the advantage of being non-disposable, more cost
innovative ideas and solutions, some of which are pushing the efficient and functional as long as power is delivered. This
limits of state-of-the-art technology. The increasing demand system architecture is based on RF Energy Transfer where an
for Wireless Sensor Network (WSN) that must be capable of external RF power source, which is assumed to have unlimited
collecting and sharing data wirelessly while often positioned energy, is used to feed the battery-free sensor for the time
in places hard to reach and service, motivates engineers to needed to dispatch both the sensing and communication
look for innovative energy harvesting and wireless power functions. Although far from efficient, RF energy transfer is
expected to be used as a power source in sensor nodes for
transfer solutions to implement battery-free sensor nodes. Due
wireless sensor networks since it is a very convenient way to
to the pervasiveness of RF (Radio Frequency) energy, RF
provide power to low energy devices such as IoT and WSN
harvesting that can reach out-of-sight places could be a key nodes, eliminating the need for periodic battery replacements.
technology to wirelessly power IoT sensor devices, which in RF energy can be used to either directly power ultra-low power
order to be ubiquitous, need to be wireless, maintenance-free, battery-free devices, which inherently work in discontinuous
battery-free and low cost enough to be used almost anywhere. mode, or to charge a storage element such as an ultra-capacitor
In this article the implementation of an RF power transfer when a continuous source of power is needed to enable the
system for battery-free wireless sensor nodes will be shown. node to operate for considerable durations in the absence of an
The system comprises an RF power transmitter which delivers RF power source. The primary benefit of RF energy is its
a power of 0.5 W at 900 MHz and an RF energy harvesting pervasiveness, which makes it possible to reach and feed
which is used as an RF power receiver to power a battery-free out-of-sight sensors. On the other hand, due to the poor
sensor node. Experimental results will show how the system efficiency of energy transfer, this technology is restricted
can feed a 10 mA load for a 50 ms acquisition time with the primarily to small energy devices. According to the Friis
transmitted power delivered from several distances. transmission formula [6], in a typical power transmission in the
range of 900 MHz in free space, the received power is about
Keywords—Radio Frequency Harvesting, Wireless Power −30 dB (1/1000) of the transmitted power after only 1 meter of
Transfer, Wireless Battery Charger, Litium Ion Battery, distance, and decays
Wireless Sensor Networks, Internet of Things. −20 dB every 10 meters. This is the main reason why this
technology is recommended mostly for short-range
I. INTRODUCTION applications.
The main challenge to be addressed in Wireless Sensor Today, RF energy transfer in battery-free systems is
Networks is the maintenance of the nodes due to limited principally used in passive RFID tags where the required
battery capacity and lifetime. In order to solve this issue, some power for communication and sensing is very low due to the
studies so far have focused on improving the communication small size of the data (identification information) to deal with,
and computing algorithms in order to make WSNs more energy the reduced distance, the absence of a sensing activity, and
efficient [1,2], while others have focused on the possibility to active communication, since the communication is based on
null the standby power consumption of the sensor nodes [3,4] backscattered power. On the contrary, in a wireless sensor node
and others have shown how to charge a battery from a distance the data acquired by the sensor must be transmitted, and this
[5]. Even if those research efforts have achieved relevant involves a much higher power consumption for both the
results, their purpose is only to prolong the lifetime of the sensing and communication activities.
sensor nodes in WSNs, assuming that they still rely on battery
power. However, a battery-powered sensor node is a disposable
item, the use of which is strictly limited by the life span of the
II. SYSTEM DESCRIPTION
battery. The main consequence of this is a high maintenance
cost of WSNs especially when, as usually occurs, the sensors The system described in this section comprises two main
are placed out of reach or in hazardous places. In this paper, an units: the power unit, the function of which is to provide power
innovative power architecture for a totally battery-free wireless to the system, and the sensor unit, which, in the most typical
sensor node is proposed with the intent to eliminate the

978-1-5386-6282-3/18/$31.00 ©2018 IEEE

case, includes several sensors, a microcontroller and a case the capacitor supplies a current to the load in the
communication element (Bluetooth, Wi-Fi, UHF, etc.). order of milli amperes, much higher than the harvested current
The power unit described in this paper for a battery-free which is instead in the order of micro amperes. As soon as the
sensor node is the unique power source of the system. It wireless sensor node has finished its activity, the sensor unit
primarily consists of an RF energy harvester capable of shuts down all the circuitry of the power unit through the
transferring the received power to the sensor unit of the system. digital signal . In this case, the LDO is turned off and
Specifically, the IC [7], provided by STMicroelectronics and the voltage goes down, causing the entire system to shut
described in Fig.1 is a self-powered RF IC for energy down. The value of the charging current, converted by the
harvesting applications. This device fully integrates a high RF-to-DC from the radio frequency energy, depends primarily
performance wide-bandwidth energy harvester (350 MHz – 2.4 on the received power at the RF-to-DC, which in free space can
GHz), with a sensitivity of −18 dBm at 900 MHz, a power be derived using the Friis transmission formula [5] below:
efficiency of 45% at 900 MHz with an input power of −10
dBm and provides an average output voltage of 2.3 V (Fig.2).
4 (1)

where is the received power, is the transmitted power,

the transmitting antenna gain, the receiver antenna gain,
λ the wavelength of the RF signal used, and the distance
between receiver and transmitter antenna. Thus, the power
delivered to the capacitor is given by the product of
the received power ( ) at the antenna and the efficiency of
the RF harvester  .

 (2)

Considering a distance between transmitter and receiver of 2

meters, free space condition, a transmitted power of 0.5 W

Fig. 1. The IC in a battery-free wireless sensor node

Fig. 3. Energy efficiency values versus input power levels for

the RF-energy harvester IC.

(27 dBm) at a frequency of 900 MHz and the gain of 1 for

Fig. 2. Power Sequence both antennas ( 1 ), the received power is
= 88 µW (~ −10.5 dBm). The efficiency of the RF
With reference to Fig. 1 and Fig. 2, the received radio
harvester (Fig. 3) is ~ 42% when = −10.5 dBm, and the
frequency energy is converted by the RF-to-DC energy
transducer into a current that charges the capacitor , power delivered to the capacitor ( ) is
causing the build-up of the voltage . When the voltage 37 µW. In steady state the average voltage at the ,
reaches the maximum value of 2.4 V, all other internal is about 2V, so the charging current is
circuits of the IC turn on and the digital signal is / ≈ 20 µA. Fig. 4 shows the current
asserted, enabling the Low Drop Output (LDO) voltage delivered to the battery vs the input power.
regulator. The LDO biases all the other units of the sensor node
(microcontroller, sensors and communication unit) with a
regulated voltage of 1.8 V. At this point, the system is
completely on with a higher current consumption, which
causes the voltage to drop because of the imbalance
between the harvested and required energy. In fact, in a typical
 voltage delivered to bias the battery-free sensor unit with
emphasis on the first acquisition time and cycle time at three
different distances, 60 cm, 120 cm and 180 cm respectively.

Fig. 4. Output current values versus input power levels for the
RF-energy harvester IC.

Regarding the load current, we targeted a sensor unit able to

carry out its sensing and communication activity with an
average current of 10 mA within a time window of 50 ms with
a maximum voltage drop of 0.4 V. This choice Fig. 6. Battery-Free bias voltage with RF source at 60 cm
guarantees enough voltage headroom to supply low power
electronic devices requiring a typical bias voltage of 1.8 V.
These requirements can be achieved by means of a low cost
1 mF capacitor. As an example of a feasible
battery-free sensor node, assuming 2 meters of free space
transmission, which means a charging current of about 20 µA,
and the capacitor completely discharged, the sensor
unit is able to perform its first acquisition 120 s after the
beginning of the power transmission. All acquisitions after the
first one, as long as the RF power transmitter stays on, will be
performed much more quickly since the capacitor
will keep a charge floor. Considering that the max allowed
voltage drop is 0.4 V, the maximum time needed to get all Fig. 7. Battery-Free bias voltage with RF source at 120 cm
subsequent acquisitions is 20 s in the absolute worst case.

Fig. 8. Battery-Free bias voltage with RF source at 180 cm

Fig. 5. Test Setup
III. CONCLUSION
EXPERIMENTAL RESULT A commercially available sensor unit has been wirelessly and
The test setup (Fig. 5) is made up of two units, a generic power successfully supplied in a battery-free configuration. Three
transmitter board and the sensor node consisting of the IC different cases in which the power transmitter and the RF
evaluation board as power unit, a sensor unit adopting a low harvester have been placed at three distances: 60 cm, 120 cm
power Bluetooth system as communication unit based on the and 180 cm have been studied. The study has also been
STMicroelectronics IC BlueNRG-1 and a set of sensors such as focused on the acquisition time in order to understand if the
accelerometer, pressure and temperature sensor. performance of the system allows this technology to be useful
Both duty cycle and active time are influenced by distance,
since the charging current always on reduces both of them.The in real use cases. The experimental results have shown the
power transmitter is programmed to deliver 27 dBm (0.5 W) at possibility to get data every 5 s by transmitting 0.5 W from a
900 MHz. The following pictures (Fig. 6, 7 and 8) show the distance of 60 cm, and roughly every 20 s from a distance of
180 cm. Overall, the measured values for receive and [3] R. La Rosa N. Aiello, G. Zoppi, “An innovative system capable to turn
on any turned off electrical appliance by means of an efficient optical
acquisition times are in line with theoretical values and the energy transfer “, PCIM Europe Conference Proceedings, pp. 1559-
system is quite easy to scale for different conditions such as 1566, 2014.
load currents, transmitted power, etc. The main outcome of [4] R. La Rosa, N. Aiello, G. Zoppi, “RF remotely-powered integrated
this work has been to show that RF wireless power transfer system to nullify standby power consumption in electrical appliances”
allows the implementation of battery-free wireless sensor IECON 2016 – 42nd Annual Conference of the IEEE Industrial
Elctronics Society.
nodes in which the lifetime is no longer limited by battery
[5] R. La Rosa, G. Zoppi, A. Finocchiaro, G. Papotto, L. Di Donato, G.
lifespan, but rather by the lifespans of the electronic Sorbello, F. Bellomo, C. A. Di Carlo, P. Livreri, “An over the distance
components. Wireless Battery Charger based on RF Energy Harvesting” SMACD
2017 – International Conference on Synthesis Modeling Analysis and
REFERENCES simulation methods and Applications to Circuit Design.
[1] I.F. Akyildiz et al., “A survey on Sensor Networks” IEEE Commun. [6] Harald Friis “A Note on a Simple Transmission Formula” Proceedings
Mag., vol. 40, no. 8, Aug. 2002, pp. 102-14. of I.R.E. and Waves and Electrons May, 1946.
[2] T. Soyata, L. Copeland ; W. Heinzelman, “RF Energy Harvesting for [7] R. Guerra, A. Finochiaro, G. Papotto, B. Messina, L. Grasso, R. La
Embedded Systems: A Survey of Tradeoffs and Methodology” IEEE Rosa, G. Zoppi, G. Notarangelo, G. Palmisano “ An RF-powered
Circuits and Systems Magazine, vol. 16, Feb. 2016, pp. 22-57. FSK/ASK receiver for remotely controlled systems”. 2016 IEEE Radio
Frequency Integrated Circuits Symposium.