Buletinul Științific al Universității Politehnica Timișoara TRANSACTIONS on ELECTRONICS and COMMUNICATIONS Volume 60(74), Issue 2, 2015 Ionospheric… [608002]

Buletinul Științific al Universității Politehnica Timișoara

TRANSACTIONS on ELECTRONICS and COMMUNICATIONS

Volume 60(74), Issue 2, 2015

Ionospheric Propagation Investigation in Western
Romania – An Experimental Approach

Cornel Balint1 Aldo De Sabata2 Septimiu Mischie3

1 Faculty of Electronics and Telecommunications, Communications Dept.
Bd. V. Parvan 2, 300223 Timisoara, Romania, cornel.balint @upt.ro
2 Faculty of Electronics and Telecommunications, Measurement and Optical Electronics Dept.
Bd. V. Parvan 2, 300223 Timisoara, Romania, aldo.de -sabata @upt.ro
3 Faculty of Electronics and Telecommunications, Measurement and Optical Electronics Dept.
Bd. V. Parvan 2, 300223 Timisoara, Romania, septimiu.mischie @upt.ro Abstract – We present a receiving system in the HF
frequency band that has been devised for testing the
possibility of communication through the ionospheric
channel in emergency situations in the Western part of
Romania. Several signals that have been received,
acquisitioned and processed are reported in order to
demonstrate the functionality of the receiver and to
illustrate the problems associated with ionospheric
propagation in the considered geographical region.
Keywords: ionosphere propagation, ionosphere
sounding, chirp s ounding , time -frequency analysis , SDR

I. INTRODUCTION

The ionosphere is an upper atmospheric region
containing several layers of ionized gas, denoted D, E
and F [1]. Some of the layers may split in several sub –
layers. This region make possible point -to-point
transmissions for Earth radio stations in the HF
frequency band by one or several refractions on the E
and F layers and no or several reflections on the
ground. A greater number of hops results in a larger
connectio n distance and a greater attenuation. The D
layer is mainly responsible for attenuation of the
transmitted signal. Additional attenuation is
introduced by reflections and propagation through
auroral regions [2, 3].
The ionosphere is affected by a high vari ability of
parameters, as indicated by many measurement
campaigns performed in various parts of the world
throughout the years [4, 5, 6]. Consequently, the
ionospheric communication channel has to be
considered as randomly time -variant and
characterized ac cordingly [7]. The variability is a
drawback of the ionosphere considered as a
communication channel. However, since establishing
a radio connection between two remote points situated
on the ground requires no infrastructure other than
transmitting and rec eiving antennas, a solution based
on the ionospheric channel becomes interesting for
broadcasting into remote regions, for point -to-point
communications in remote, under -developed areas
and for communications in emergency situations
when built infrastructu re becomes unavailable. In order to implement a communication service on the
ionospheric channel, one has to rely on a prediction
model or to frequently sound the signal path for
matching the transmitter parameters to the channel.
Several ionosphere models are available [8, 9, 10, 11]
and can be used. However, these models are in
general global and applying such a model for a
particular place requires some interpolation or input
of some local parameters. Furthermore, the
information provided by the model ma y miss the
variability of the channel since it is based on averaged
data.
An HF receiving system and two antennas have been
mounted at the location of Timișoara, Romania
(45°45′13″ N, 21°13′32″ E) in order to test the
possibility of ionosphere communication with the rest
of the country in emergency situations.
The present paper is the first report on this work. We
outline the conception of the system and we present
the measurement results on "opportunistic signals" [2]
received by our antennas, demonstrating in this way
the functionality of the receiver. The last section of
the paper is reserved for conclusions.

II. PRESENTATION OF THE RECEIVER

HF signals are received on location by means of two
antennas. The Harris RF 1936 is an omnidirectional
antenna working in the ra nge 1.6 -30 MHz. According
to producer's data, the antenna is mainly horizontally
polarized, having a nominal input impedance of 50 ,
a gain of -16 dBi@2 MHz and -2 dBi@30 MHz, and a
VSWR in the range 1 -2.8. The antenna occupies a
square of maximum 60 m si de on the ground. The
Harris antenna has been placed on the roof of the
building where experiments have taken place and it
has been connected to the receiver with a 30 m long
coaxial cable. The input parameters of the antenna
have been measured with an Agi lent N5230A network
analyzer, carefully calibrated. The VSWR and Smith
Chart are reported in Figs. 1 and 2 for the range 1 –
15 MHz .
14

The Diamond WD 330 J antenna is used in the 2 –
30 MHz frequency range for reception and emission
up to 150 W, is 25 m long, w eights 3.1 kg and it can
be mounted in a straight line for a dipole -type
radiation pattern or as a reversed Vee for an
approximately omnidirectional radiation pattern.
According to producer's data, the 50  VSWR is
tipically S=2 in the 2 -18 MHz frequency r ange and
S=3 at higher frequencies. These values may be
modified by the working conditions.
The Diamond antenna has been mounted on the
ground, in front of the building where experiments
have been performed.
The input parameters have been measured in the same
conditions as for the preceding case in the 3 -12 MHz
frequency range and are reported in Figs. 3 and 4.
By comparing the measurement results concerning the
two antennas, one can notice that the Diamond
antenna is better matched for a 50  load in alm ost all
the tested frequency range.
However, in some practical situations, the Harris
antenna performed better than the Diamond antenna, a
fact that could be explained by one of the following
reasons: either the Harris antenna had a better VSWR
at the wor king frequency (see Figs. 1 and 3), or it
received a stronger signal due to its better position
that compesated the poorer matching or due to the
fading conditions of the moment (the two antennas are
separated by more than 40 m).

The receiver consists of a NI USRP -2950R, 50 MHz –
2.2 GHz Dual RF Transceiver SDR (Kintex – 7, 410T
FPGA) and GPS Clock, an extension LFRX USRP
Daughterboard (DC – 30 MHz) and a PCIe – MXI
Express Interface Kit for USRP R IO. This SDR
equipment allows for the acquisition of the I and Q
components of the input signal in a specified
frequency band at a 60 MHz sampling rate. The
necessary software has been developed under Linux
and GNURadio environment.

III RECEPTION OF SIGNA LS

The receiving equipment has been tested on several
signals captured by the antennas. We report here
results obtained with a FSK signal and a chirp signal
used for ionosphere sounding.

A. FSK Signal

A FSK signal is transmitted permanently by a station
near Hamburg, Germany. This signal has been
acquired with the Harris antenna during tests
performed for several days in October 2015 in the
hourly interval 10 -13, local time. We consider here a
recording consisting of 172012 I and Q samples that
have been us ed for generating the corresponding
complex signal in a 200 kHz wide frequency band
centered on 10 MHz. The magnitude of the discrete
Fourier transform of the raw signal is presented in
Fig. 5. The FSK signal is distinguishable in the Fig. 3. VSWR on the cable of the Diamond antenna.
Fig. 4. Smith Chart at the input of the Harris antenna.
Fig. 1. VSWR on the cable of the Harris antenna.
Fig. 2. Smith Chart at the input of the Harris antenna.
15

vicinity of the cente r frequency (10 MHz). Other HF
signals are present in the received spectrum. For a
clearer rendering of the signals, allowing for a more
precise frequency assessment, the following
processing has been performed on the raw signal: a
1024 – sample wide Hammi ng window has been
displaced along the signal with a step of 500 samples.
The Discrete Fourier Transform has been performed
at each step and the results have been averaged. The
magnitude of the result is displayed in Fig. 6.
The two maxima of the FSK signa l are now clearly
visible at 9.99 MHz and 10.02 MHz respectively.
Other signals are present at 9.365, 9.766, 9.955,
10.17, 10.31, 10.68, 10.73 and 10.86 MHz.
Averaging is not a good option if (in)stability in time
determined by the variation in electron density of the
ionosphere, fading and the Doppler shift caused by
displacement of layers have to be investigated.
Therefore, a time -frequency analysis of the si gnal has
been performed, and reported in Fig. 7. A 1024 –
sample rectangular window and a 32 – sample step
have been used for the representation. The experiment
demonstrates the variability of reception in the HF
range, clearly visible on the 10.86 MHz signal in Fig.
7, but also visible on the FSK signal.

B. Chirp Si gnal

A modern solution to obtain real -time information
about the ionosphere is the ionosonde , a special type
of radar system , consist ing of a transmitter equipment
that sweeps all or part of the HF frequen cy range, and
it transmit s short pulses or continu ous signals. These
signals, reflected at various layers of the ionosphere,
are collected by the receiver and analysed by an
appropriate control system.
Ionospheric sounding can be vertically or oblique.
Vertical sounder s transmit the wave vertically and the
reflected wave is received by the receiver placed in
the vicinity of the transmitter. Oblique ionospheric
sounding allow s for monitoring the radio channel over
large and very large distances, impossible to achieve
by other techniques. This technique also allows a
single receiver to receive signals from several
transmitters located in different geographical places .

In both cases, the signal round trip delay allows to
calculate the height at which the reflection occurs and
to estimate the critical frequency and the
communication distance for a given frequency.
In this paper we focus only on the particular
ionosonde type sounder known as the Chirp Sounder,
that transmits a low power Frequency Modulated
Continuous Wave (FMCW) or chirp, i.e. a continuous
signal in which the instantaneous frequency increases
linearly with time. Chirp sounders usually transmit in
range 2 – 30 MHz and some of th e sounders skip
some particular frequencies used for critical
communications.
Software defined radio is a very powerful and
versatile tool that can be used both for generating and
receiving the chirp sounding signal [12]. In our
experiments we have used the USRP 2950R based
receiving system and the t wo antennas described
above .
There are many ionosondes that transmit chirp signals
but the information about the ir location and time
parameters are either incomplete or difficult to find .
The transmitting stations are hard to identify as chirp
signal does not contain any information in addition to
the FM-CW wave.
For testing our receiving system, we have found a
reliable chirp sounder located near Akrotini, Cyprus,
approximately 1600 km from reception point. Fig. 7. Time -frequency representation of the received signal.
Fig. 6. Averaged spectrum of the received signal.
Fig. 5. Spectrum of the received signal.
16

This chirpsounder sweeps the HF band at the
frequency rate of 100 KHz/s. The transmission repeats
every 300 s, thus during every hour the sequence
repeat s itself 12 times. The chirp time, i.e. the
moment when the chirp starts at 0 MHz is 235 s.
Taking the midnight time as a reference point the first
transmission starts at 00:03:55 and the second at
00:08:55. The spectrum of the received chirp signal of
Cyprus1 ionosonde is presented in Fig. 8. The
receiver bandwidth was set to 10 kHz around the
central frequency of 6775 kHz. The chirp signal level
has been approximately +20 dB above the noise floor .
There were no other signals except the chirp and the
noise in the receiving frequency band. The peak hold
option for FFT has been activated, and the chirp
signal spectrum has been plotted in successive
positions corresponding to the FFT refresh rate (5ms ).
Fig. 9 shows the received chirp signal from Cyprus
ionosonde on 20/10/2015 at 8:53:55 GMT (10:53:55
local time) plotted in a time -frequency representation
around 10500 kHz. At 100 kHz/s the c hirp signal need
2 s to travel along the received bandwidth of 200 kHz.
A strong amplitude modulated signal is present at
10500 kHz and o ther modul ated signals are visible in
the received frequency band .

Fig. 9. Time -frequency representation of chirp signal from
Cyprus ionosonde.

CONCLUSIONS

A HF receiving equipment has been mounted in order
to investigate ionospheric propagation in the Western
part of Romania . The mounted equipment has been
tested on "opportunistic" signals captured by
antennas, namely a FSK signal and a chirp signal used
for sounding the ionosphere. Several signal
processing techniques have been applied in order to
assess the correct function ing of the equipment and to
test ionospheric propagation conditions. Variability of
propagation conditions has been clearly demonstrated.
Therefore, a need for a systematic assessment of
propagation conditions has become evident. The goal
of this research is to test the possibility of establishing
an emergency communication network on the
Romanian territory.

ACKNOWLEDGMENT

This research was funded by the Ministry of
Education and Research of Romania through
UEFISCDI, project code PN -II-PT-PCCA -2013 -4-
0627 , contract no. 292/2014.

REFERENCES

[1] R. E. Collin, Antennas and Radiowave Propagation, Ch. 6,
New York: McGraw -Hill, 1985.
[2] K. Khoder, R. Fleury, P. Pagani, „Monitoring of Ionosphere
Propagation Conditions using Opportunistic HF Signals”, The 8th
European Conference on Antennas and Propagation (EuCAP 2014),
6-11 Apr., The Hague, pp. 2697 -2701, 2014.
[3] D. V. Blagoveshchensky, M.A. Sergeeva, " Impact of the
auroral ionosphere on HF radio propagation ", General Assembly
and Scientific Symposium, 2011 XXXth URSI , 13 -20 Aug.,
Istanbul, pp. 1 -4, 2011.
[4] C. Williams, "Low -frequency radio -wave propagation by the
ionosphere, with particular reference to long -distance navigation ",
Proc. IEE – Part III: Radio and Communication Engineering vol. 98
, no. 52, pp. 81 -99, 1 951.
[5] C.C. Watterson, J.R. Juroshek, W.D. Bensema, "Experimental
Confirmation of an HF Channel Model", IEEE Trans. Comm.
Technol., vol. 18, no. 6, pp. 792 -803, 1970.
[6] S. Shi, G. Chen, G. Yang, T. Li, Z. Zhao, J. Liu, “Wuhan
Ionospheric Oblique -Incidence Sou nding System and Its New
Application in Localization of Ionospheric Irregularities”, IEEE
Trans. Geoscience Remote Sens., vol. 53, no. 4, pp. 2185 -2194,
2015 .
[7] P.A. Bello, "Characterization of Randomly Time -Variant
Linear Channels", IEEE Trans. Comm. Syst., vol. 11, no. 4, pp.
360-393, 1963.
[8] F.T. Koide, “A Computer Method of HF Ionospheric
Propagation Prediction and Analysis”, IRE Trans. Antennas
Propag., pp. 540 -558, September, 1963.
[9] C.M. Rush, “Ionospheric Radio Propagation Models and
Predictions -A Mini -Review”, IEEE Trans. Antennas Propag., vol.
34. no. 9, pp. 1163 -1170, 1986.
[10] S.A. Cummer, “Modeling Electromagnetic Propagation in the
Earth –Ionosphere Waveguide”, IEEE Trans. Antennas Propag., vol.
48, no. 9, pp. 1420 -1429, 2000.
[11] Z. Yan, G. Wang, G. Tian, W. Li, D. Su, T. Rahman, “The HF
Channel EM Parameters Estimation Under a Complex Environment
Using the Modified IRI and IGRF Model”, IEEE Trans. Antennas
Propag., vol. 59, no. 5, pp. 1778 -1783, 2011.
[12] P. B. Nagaraju, E. Koski, T. Melodia, "A Software -defined
Ionospheric Chirpsounder for HF Propagation Analysis," in Proc.
of HF Nordic Conference, Faro, Sweden, August 2010. Fig. 8. Spectrum of chirp signal around 6775 kHz .
17