Content uploaded by Andrey S. Andrenko
Author content
All content in this area was uploaded by Andrey S. Andrenko on Feb 18, 2016
Content may be subject to copyright.
978-1-4799-8641-5/15$31.00©2015 IEEE
Outdoor RF Spectral Survey: a Roadmap for Ambient
RF Energy Harvesting
Andrey S. Andrenko, Xianyang Lin and Miaowang Zeng
SYSU-CMU Shunde International Joint Research Institute
No. 9 Eastern Nanguo Road,
Shunde, Guangdong, China 528300
andrey_andrenko@sdjri.com
Abstract- In this paper, the results of outdoor RF spectral
survey carried out in an urban area in Southern China are
presented. The motivation of this work has been an attempt to
record the dataset of ambient RF power radiated by the mobile
base stations and Wi-Fi hotspots in the urban environment so as
to select the frequency bands most suitable for the future
ambient energy harvesting prototype design. RF spectrum survey
was undertaken in the city center of Shunde, China, which is one
of the most developed cities in Guangdong province. Typical
places including shopping center square and a residential area
have been considered in this survey. The ambient power level
within the RF spectrum of 0.7 to 3 GHz has been measured to
cover all the mobile frequency bands (2G, 3G and 4G) together
with the 2.4 GHz Wi-Fi band. The survey results show that the
available RF power levels of CDMA800, GSM900 and GSM1800
bands are the most promising for the ambient energy harvesting
design and outdoor applications.
Keywords—spectral survey; ambient RF energy harvesting;
spectrum analyzer; antenna gain
I. INTRODUCTION
During the last decade, wireless power transmission and
harvesting have become the high-impact technologies which
will have a profound influence on the wireless sensor design
and accelerate the development of internet of things. Several
applications have been reported, such as small scaled smart-
dust sensors [1], implantable biomedical sensors [2] and
human body device circuits [3]. Recently, energy harvesting
assisted UHF RFID sensor tag design has attracted much
attention for increasing the read range and advancing the
performance of next-generation RFID networks [4].
In the design of the smart RFID tags, the most attractive
way to collect energy without dedicated power supply is an
ambient RF energy harvest. In urban environments, signals of
mobile and Wi-Fi services are generally everywhere and the
energy is “free” to use. Furthermore, the minimum startup
power for the RFID chip is low, which makes it possible to
collect an ambient energy and to amplify the back scatter
signal of a passive RFID tag.
In order to select the most suitable frequency bands to
deploy the outdoor ambient RF energy harvester, energy level
of the available RF frequency bands needs to be measured in
different locations in the city. In the previous works, several
spectrum surveys of RF power level at different frequency
bands such as mobile phone and TV broadcasting have been
reported. Some energy surveys are performed in general
scenarios (e.g. indoor, outdoor, etc.) rather than have more
detail of location and time [5]. In [6], energy measurements
were conducted in 16 places of Tokyo city center. However,
this survey separates the frequency bands into 3 general
categories (TV, mobile and FM radio) and little details of
energy level at different mobile bands (GSM, CDMA, 3G, etc.)
and Wi-Fi band are presented. Some results of spectral survey
in central London have been reported in [7]. It should be noted
that the ambient RF conditions are generally country specific
depending on the regulations for the wireless services.
In this paper, a survey of ambient RF energy levels in the
shopping square and residential area of Shunde, China are
presented. The main purpose was to define the feasible
frequency bands for the future energy harvesting circuit design.
Energy levels at different places are analyzed and the most
promising frequency bands for ambient harvesting design in
the Chinese urban conditions are selected.
II. RF MEASUREMENT SCENARIO
In obtaining the feasible energy levels in the city, the
measurement places were selected based on a typical urban
environment in Shunde. The first one has been an open
parking lot outside the large shopping center. The second one
has been selected in a residential area being close to the
mobile base stations of China Mobile and where Wi-Fi signals
from resident apartments are considered to be stronger. Fig. 1
shows the approximate measurement positions in the
aforementioned places.
Fig. 1. Measurement places outside the shopping center and in a residential
area.
As mobile phones are used more frequently in the daytime,
the energy level was expected to be stronger than in the
nighttime. Therefore the series of measurements have been
conducted from 2:30pm to 4:30pm. The spectral survey has
been carried out to cover all the mobile frequency bands and
2.4GHz Wi-Fi band with the measured frequency range being
defined from 700MHz to 3GHz based on the China radio
spectrum resource illustrated in Fig. 3.
In this spectral survey, Agilent 9343C portable spectrum
analyzer (SA) combined with a double-ridge high-gain horn
antenna has been used. The measured antenna gain versus
frequency is presented in Fig. 2. In order to produce the
measurement results close to the ones measured with
omnidirectional antenna, the signal energy has been recorded
with the antenna facing 4 different directions at each location
as shown in Fig. 1. To account for both horizontal and vertical
wave polarization in rich multipath environment, horn antenna
has been rotated by 90-deg measuring the RF power level in
both E- and H-planes. Thus 8 measurements in total have been
conducted at each place and average power has been
calculated from 8 measured datasets. The spectrum analyzer
was set to “max-hold” in order to record the maximum power
level. In each measurement, about 30 seconds are allocated
and the resolution has been 461 points. Finally, the actual
ambient RF power level at each measurement point has been
obtained by subtracting all the measured data from the antenna
gain at each frequency within the 0.7 to 3 GHz band so as to
account for the gain frequency dependence of the antenna.
III. RF POWER MEASUREMENT RESULTS
A. China’s Frequency Band Allocation
Three main wireless telecom providers, China Mobile
(CM), China Unicom (CU) and China Telecom (CT) are
operating in Mainland China. Fig. 3 shows the frequency
bands allocation for mobile communication in Mainland
China. Digital cellular mobile communication (DCMC)
frequency bands are located from 825MHz to 2.655GHz,
covering the 2G, 3G and 4G wireless technologies. 2G mobile
communication frequency band consists of GSM900,
GSM1800 and CDMA800. As a matter of fact, CDMA2000
EVDO of CT service is operated in CDMA800 frequency
band. Thanks to the benefits of low frequency CDMA800
band, CDMA2000 mobile communications of CT become the
Fig. 2. Horn antenna gain versus frequency in 700MHz to 3GHz band.
3G network with the broadest coverage in Mainland China. In
addition, CT also operates 1.92 and 2.11GHz CDMA2000
frequency bands. 3G network of CU is based on WCDMA
technology. It works over the 1.94GHz and 2.13GHz bands
and only covers the rural places. 3G network of CM is based
on TD-SCDMA technology and spreads over 1.88GHz and
2.01GHz frequency bands. As for 4G networks, CM adopts
TD-LTE technology and its frequencies are on 2.32 and
2.575GHz. By 2015, CM has constructed a largest 4G network
in the world. Both CT and CU have utilized TD-LTE and
FDD-LTE as their 4G network technologies, but the coverage
area is much smaller than that of CM.
Fig. 3. Mobile frequency band allocation in Mainland China.
B. Measurement Results Outside the Shopping Mall
Signal power was measured in parking lot of a shopping
mall in Shunde. In all the measurements, the antenna has been
placed 1m above the ground. Fig. 4 and Fig. 5 show the
received RF power with the horn antenna facing the shopping
mall building (as indicated in Fig. 1) and the average received
power, respectively. It has been shown that the signal power
levels of 3G and 4G (TD-SCDMA, TD-LTE and FDD-LTE)
are quite sensitive to the direction. The maximum signal
power has been -60.9dBm, -61.5dBm and -62.7dBm in the
direction 1, respectively, while the average power in all the
directions is obtained as -70.8dBm, -71.4dBm and -78.3dBm.
This indicates that the 3G and 4G signal power is unstable and
relatively low so that these bands are less suitable for the
ambient RF energy harvesting. According to Fig. 5, the
frequency band providing the highest power outside the
shopping mall is 870 to 880MHz CDMA800 downlink band
of CT with the average power reaching about -38.99dBm in
the measurement. The second strongest signal frequency band
is 935 to 960MHz GSM900 downlink band of CM and CU,
where the average power level is about -50dBm. The SA has
detected a range of 1.805 to 1.85GHz with power maximum
being -54.9dBm that is the GSM1800 of CT. The 2.11 to
2.15GHz is the WCDMA downlink band of CU and
CDMA2000 of CT, and the measured power here is at the
level of -59dBm. As this measurement point was located
outside the shopping mall and far away from the residential
area, the power of 2.4GHz Wi-Fi signal was at the noise level
of less than -80dBm.
Fig. 4. RF power received outside shopping mall with antenna in direction 1.
Fig. 5. Average RF power received outside shopping mall.
C. Measurement Results in the Residential Area
Figs. 6 and 7 display the ambient RF powers measured in
the direction 1 and 2 in the residential area, respectively. The
average RF power at this location is presented in Fig. 8. The
antenna was again 1m above the ground. The measurement
results were similar to the ones obtained in the shopping
square but two additional bands were detected. They are 2.4 to
2.45GHz Wi-Fi band and 2.32 to 2.335GHz TD-LTE band of
CM. The Wi-Fi signal has been detected mainly because there
were many WLAN hot spots around residential area as well as
in the office building nearby. However, the signal power level
is relatively low. Despite the fact that the measurement point
has also been close to a few mobile base stations located on
top of the office building, the signal power of 2.32GHz and
2.575GHz bands of TD-LTE is still low with the maximum
average power being equal to -72.2dBm. The reason of such a
low TD-LTE signal level is that the measurement spot has
been located just 3m from the wall of this high office building.
So, no line-of-sight propagation from the base station has been
observed here. Instead, the measurements are considered to be
carried out in a typical for urban conditions rich multipath
environment.
Considering the spectral survey results obtained for the
antenna facing two directions, it can be concluded that the 810
to 870MHz CDMA800 is the most promising frequency band
for ambient RF energy harvesting. CDMA800 signal has the
highest power level varying between -38 to -30dBm and is
less sensitive with respect to the direction, location, and wave
polarization. The second feasible group of RF energy sources
is the 935 to 960MHz GSM900, 1.805 to 1.85GHz GSM1800
and 2.11 to 2.145GHz CDMA2000 and WCDMA
communications with a relatively high average power level
varying between -56 and -40dBm depending on the antenna
orientation. It should be noted though that the GSM1800 and
CDMA2000/WCDMA signal powers are more sensitive with
respect to the measurement direction as shown in Figs. 6 and 7.
Fig. 6. RF power received in residential area with antenna in direction 1.
Fig. 7. RF power received in residential area with antenna in direction 2.
Analyzing the data presented in Figs. 6-8, we can define
them as an encouraging result towards the rectenna-circuit
design of ambient RF energy harvesters for sensor
applications where the arbitrary oriented wireless nodes could
be moving in urban areas. On the other hand, the average
radiated power of 3G TD-SCDMA, 4G (TD-LTE, FDD-LTE)
and 2.4G Wi-Fi bands show much lower signal levels of -65 to
-55dBm. Therefore, these bands are much less suitable for
ambient wireless RF energy harvesting design for outdoor
applications.
Fig. 8. Average RF power received in residential area.
IV. CONCLUSIONS
A methodology of measuring an ambient RF signal power
is described in this paper, and the frequency band allocation in
Mainland China is discussed. RF power level has been
measured and analyzed at two different urban locations.
Considering the results, it has been found that 810 to 870MHz
CDMA800 together with the GSM900 and GSM1800 are the
most reliable ambient RF resources for wireless energy
harvesting sensor design and applications. The results
obtained will be used for defining the design frequencies of
the rectifying circuits in the future works.
REFERENCES
[1] T. Salter, G. Metze, and N. Goldsman, “Parasitic aware optimization of
an RF power scavenging circuit with applications to smartdust sensor
networks,” Proc. of IEEE Radio Wireless Symp., 2009, pp. 332-335.
[2] C. C. Lo, Y. L. Yang, C. L. Tsai, C. S. Lee, and C. L. Yang, “Novel
wireless impulsive power transmission to enhance the conversion
efficiency for low input power,” Microwave Workshop Series on
Innovative Wireless Power Transmission: Technologies, Systems and
Applications (IMWS), 2011 IEEE MTT-S International, 2011, pp. 55-58.
[3] N. Barroca, H. M. Saraiva, P. T. Gouveia, J. Tavares, L. M. Borges, F. J.
Velez, C. Loss, R. Salvado, P. Pinho, R. Goncalves, N. BorgesCarvalho,
R. Chavez-Santiago and I. Balasingham, “Antennas and circuits for
ambient RF energy harvesting in wireless body area networks,” Proc. of
IEEE International Symposium on Personal Indoor and Mobile Radio
Communications (PIMRC), pp. 532-537, London, UK, Sept. 2013.
[4] Y. Zuo, “Survivable RFID systems: issues, challenges, and techniques,”
IEEE Trans on Systems, Man, and Cybernetics, Part C: Applications and
Reviews, vol. 40, no. 4, pp. 406-418, July 2010.
[5] G. Thuroczy, F. Molnar, J. Szabo, G. Janossy, N. Nagy, G. Kubinyi, and
J. Bakos, “Public exposure to RF from installed sources: Site
measurements and personal exposimetry,” Proc. of 1st European Conf.
on Antennas and Propagation, 2006, pp. 1–4.
[6] Y. Kawahara, K. Tsukada, and T. Asami, “Feasibility and potential
application of power scavenging from environmental RF signals,” in
IEEE AP-S Int. Symp., 2009, pp. 1–4.
[7] M. Pinuela, P. D. Mitcheson, and S. Lucyszyn, “Ambient RF energy
harvesting in urban and semi-urban environments,” IEEE Trans on
Microwave Theory and Tech, vol. 61, no. 7. pp. 2715-2726, July 2013.