67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.

Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

IAC-16-B.2.8-GTS.3.7.x33593
A LOW-COST MOBILE GROUND STATION FOR SATELLITE COMMUNICATION IN VHF BAND
Sneha Velayudhana, Miguel Bazdrescha*
a

Department of Electrical, Computer & Telecom Engineering Technology, Rochester Institute Of Technology,
Rochester, NY 14623, United States, mxbiee@rit.edu
* Corresponding Author
Abstract
In this paper, we present the architecture and performance analysis of a low-cost, small, mobile and easily
deployable ground station to track and receive signals from amateur band satellites which operate on the VHF-band
(144 MHz to 147 MHz). The proposed ground station uses a handheld 5-dB gain Yagi-Uda antenna, and a small lownoise amplifier with 23 dB gain. The rest of the analog front-end is a software-defined radio receiver with a USB
interface to a laptop computer. The software-defined nature of the front-end gives the station the flexibility to target
satellites with diverse power, modulation and error-correcting schemes. The system is completed by open-source and
freely-available software for tracking satellite positions and for signal decoding and processing. We report
performance evaluation results for indoor and outdoor operation. The proposed ground station can be adopted in
locations with limited access to fixed outdoor antennas, whether because of financial, regulatory or other restrictions.
Given its low cost, it can be easily adopted for classroom and laboratory activities related to satellite signal processing.
Keywords: Ground station, link analysis, low cost, portability
1. Introduction
A ground station provides the communication
window to and from a satellite [1]. It is usually
designed to track, communicate with, and process the
data received from satellites. In this paper, we propose
an architecture for a satellite ground station that is
inexpensive, small, and portable, while providing
enough performance to successfully detect and receive
satellite signals.
The main reason for requiring low cost, small size
and high portability is that we wish to allow interested
amateurs, students, and researchers to have easy
access to satellite signals in the absence of more
traditional infrastructure, such as a roof-mounted or
parabolic antenna, or dedicated laboratory space. We
envisage applications as varied as apartment-dwelling
amateurs being able to record measurements during a
field trip, students working on tracking and detecting
satellites even during laboratory off-hours, and signal
processing researchers having ready access to a
diverse set of signals.
The proposed ground station is designed to operate
on the band used by many amateur and cube satellites,
the VHF band (30 300 MHz, in particular
frequencies around 145 MHz). It is worth noting that
the ground station can also be used in the UHF band
(around 450 MHz) with minor modifications.
The ground station presents a stereo sound-card
interface to a computer. This is advantageous because
it allows the use of any signal processing software that
understands an audio signal. The stations maximum
sampling rate is 192,000 complex samples per second,

IAC-16-B.2.8-GTS.3.7.x33593

for a maximum bandwidth of 192 kHz. Since many

amateur satellites transmit signals with bandwidth no
larger than a few kilohertz, this sampling rate is
appropriate. The in-phase and quadrature components
of the satellite signals complex envelope are
presented in the left and right stereo audio channels.
The fact that the satellite signals are ultimately
digitized offers significant flexibility and opens up
many potential applications. The signals can be
processed either in real-time or stored for later
processing. They can be used as input to any of the
large variety of open-source and free of cost software
packages available that are capable of processing and
demodulating them.
In a more technical or educational environment,
any of the signals properties can be analyzed, as well
as their data decoded and recovered. Research and
experiments can be done on Doppler shift
measurement and correction, up- and downconversion, frame synchronization, error correction
codes, demodulation, etc.
2. Ground station Architecture
We propose the architecture shown in generic form
in Figure 1. This architecture may be considered a
simplification of other, more sophisticated and
costlier, proposals [2, 7]. There are only five elements.
To meet our cost and portability requirements, the
elements should meet the following constraints.
The antenna must be small and light enough
to hold and point with bare hands, for at least
a few minutes at a time. It must also be as

Page 1 of 6

67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.
Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

Fig 2. Proposed ground station configuration.

Fig 1: Architecture of the proposed ground station.

directional and
possible.

frequency selective as

The LNA should cover all frequencies of

interest, with a noise figure less than 2 or 3
dB. It must be small and inexpensive. Ideally,
it should have a bias-T connection to get
power from the front-end. Optionally, it can
be battery-operated.

The radio-frequency front-end must provide

sufficient filtering and amplification to obtain
a good signal-to-noise ratio. Ideally, it should
be able to provide power to the LNA via a
bias-T set-up. It must be able to receive a
quadrature signal and produce its complex
envelope. It should be configurable from the
computer it is attached to, from which it must
also be able to draw power.

The dual-channel ADC should have at least

14 bits of resolution, with a sampling rate of
at least 96,000 complex samples per second.
It should offer a standard interface to the
computer its attached to, such as a sound
card or USB device.

The digital signal processor may be a laptop

computer or a DSP or FPGA board with an
appropriate interface to the ADC. As a
minimum, it should have enough processing
power to record the samples in a standard
format.

It should be pointed out that two additional

elements may be required. Depending of the amount
and power of out-of-band emissions, a band-pass filter
may be necessary between the antenna and the LNA.
A battery may also be needed to power the LNA. The
use of the LNA itself may be considered optional,
since in many circumstances the amplification
provided by the front-end may suffice.
IAC-16-B.2.8-GTS.3.7.x33593

Finally, all the elements of the ground station

would ideally be available off-the-shelf, without
requiring complex assembly.
2.1 Current proposal configuration
Our actual proposed configuration is shown in Fig.
2. A description of each element follows:
The Yagi antenna is tuned to 145 MHz and
designed to be hand-held. It is manufactured
and sold by Arrow Antenna. A picture is
shown in Fig. 3.
The LNA is based on the Mini-Circuits
PSA4-5043+ MMIC amplifier and offers a
gain of 23.5 dB at 145 MHz, with a
theoretical noise figure of less than 1 dB. It
can be powered by a 6 to 9-volt battery or by
a bias-T power connection from the RF frontend.
The RF front-end is the FUNcube dongle
Pro+, a software-defined radio designed
specifically for VHF satellite signal
reception. This is a quadrature front-end with
several high-quality band-pass filters and an
integrated LNA. The noise figure at 145 MHz
is 3.5 dB. It integrates an ADC at 192,000
complex samples per second at 16 bits per
sample and appears as a USB sound card to a
computer attached to it. Finally, it is capable
of powering the LNA using a bias-T
connection.
The digital signal processor we have used is
simply a general-purpose laptop computer.
Any modern computer is able to handle the
sampling rate produced by the front end with
considerable ease, and a fully-charged
battery can last for hours in the field.
Software installed on the laptop may be used
for all tasks involved, from tracking and
locating a specific satellite, to recording and
processing the received signals.
2.2 Cost and Portability
One of the highlights of this project is its cost
(compare with e.g. [2]). The components needed to set
up a ground station are generally considered to be very
affordable. The detailed cost list is shown in Table 1.
Page 2 of 6

67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.
Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

4. Link analysis
Link budget is calculated to know the margin for
the communication [3]. Link analysis/budget is the
detailed calculation of the transmitted power, the
losses, received power and other attenuation and
distortions in the signal due to interference,
atmospheric losses etc. The link analysis for FUNcube
was calculated as follows. For this calculation, we
assume that only the FUNcubes internal LNA is used.
The carrier-to-noise ratio for the downlink is
defined as
Fig 4: A handheld Yagi antenna for the VHF band.
We have omitted the cost of the laptop, since they vary
greatly in price, and in many cases one will be already
available for this use.
Regarding portability, our experience shows that
the equipment is light-weighted and easy to handle and
it can be taken anywhere and set up very quickly. The
antenna is collapsible and takes five minutes to
assemble or disassemble. The dimensions and weight
of each component is listed in Table 2, where we have
included the weight of a large 9-V alkaline battery. We
have again omitted the weight of the laptop, since they
tend to present large variation.
Table 1: Cost of each ground station component, in
US dollars (including shipping of European
components to the US).
Item

Cost

Antenna

$90

Low Noise Amplifier

$80

Receiver

$200

Connectors

$10

Cables

and its value can be calculated as

[EIRP] = [ ] + [ ]
= 10 log(0.3) + 10 log(1)
= 5.2 dB.

$390

Table 2: Dimensions and weight of the ground station

components, including the optional battery.
Item

Length
(in)
37.5

Width
(in)
20.25

Height
(in)
--

Weight
(g)
250

FUNcube
dongle

3.3

0.7

0.4

50

LNA

1.6

0.6

75

Battery

2.6

2.6

3.7

680

Total:

1055

Antenna

where EIRP is the effective isotropic radiated power,

is the total loss in the link including free-space and
equipment losses, is the receiver gain, is the
equivalent system noise temperature, and is
Boltzmanns constant.
We assume the following system parameters [4]:
Power transmitted from satellite antenna:
= 300 mW.
Transmitted
carrier
frequency:
=
145.935 MHz.
Transmitter antenna gain: = 0 dB.
Receiver antenna gain: = 5 dB.
Receiver noise figure: = 3 dB.
We also assume that [] denotes that the quantity
inside the brackets is expressed in decibels. The
satellite EIRP is defined as
EIRP =

$10
Total:

1
= EIRP
0

The wavelength is calculated as

=

= 2.07 m.

The free space loss, then, is given by

4 2
= (
)

where is the height of satellite from Earth. Then,

[ ] = 20 log(4600 103 ) 20 log(2.07)

IAC-16-B.2.8-GTS.3.7.x33593

Page 3 of 6

67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.
Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

= 131.2 dB.
Since our antenna is not circularly polarized, we
assume a polarization loss of 3 dB and further
atmospheric losses of 0.8 dB. The total propagation
loss is then given by
[] = 131.2 + 3.0 + 0.8
= 135 dB.
The Arrow antenna manufacturers do not publish
specifications for the antennas gain, but we know that
it is directional and has a relatively high gain.
Assuming that the antenna gain is = 5 dB, the
receiver noise figure is = 3 dB and the environment
temperature is 300 K, we can estimate the equivalent
temperature to be = (2 1)(300 K) = 300 K.
Then, the receiver figure of merit (or receiver
sensitivity) is given by
[

] = 19.8 dB.

Now we can calculate the carrier-to-noise ratio in

decibels as

[ ] = 5.2 135 19.8 + 228.6

0
= 68.6 dB.
Now we calculate the signal-to-noise ratio /0 .
The FUNcube transmits at a rate of 1,200 bits per
second using BPSK modulation [4]. The bit energy is
then given by
[ ] = [

] = 166 dB

leading to a signal-to-noise ratio

ignored other noise sources such as galactic noise.

However, this analysis provides a strong indication
that the proposed station should be capable of
receiving the satellites signal. Even if our margin
calculations are off by almost 30 dB, the ground
station would still perform adequately.
5. Performance results
We have performed an experimental analysis of the
proposed ground station in both indoor and outdoor
environments.
5.1 Indoor performance results
We are interested in two different ground station
tests in an indoor environment.
The first test is simply observing a satellite from a
laboratory. At the telecommunications engineering
technology department at RIT, we have access to a
laboratory room with a large window facing north.
Detecting a satellite from inside the lab would make it
easier to assign students or assistants the task of
recording specific satellite passes. However, despite
line-of-sight conditions, we have been unable to
record any signal. More work needs to be performed
in this area, which has known difficulties [5, 6].
The second kind of test is a functional test to verify
that all components are operational. In this test, a
separate software-defined radio capable of
transmitting at the desired frequency is set up to
emulate the satellite signal, while analysis software is
used to verify that signal samples are properly
received and processed at the ground stations
computer. While, strictly speaking, our tests have been
successful and useful, our transmitter is optimized for
transmitting signals of tens to hundreds of milliwatts,
resulting in complete saturation of the ground stations
receiver. An example of the behavior observed is
shown in Fig. 4. In the figure, a popular signal analysis
package called HDRSharp is showing a waterfall plot

= 37 dB.
0
The satellites BPSK signal is encoded using a
Reed-Solomon forward error control code. Setting our
desired bit-error rate (BER) to 105 , we obtain a
desired signal-to-noise ratio equal to 10 dB. The
ground stations link margin is then
= 37 10 = 27 dB.
This result may be considered very optimistic,
since we havent included all loss sources, such as
efficiency, feeder, and pointing losses, and we have
IAC-16-B.2.8-GTS.3.7.x33593

Fig. 4: Strong saturation in and distortion are evident

in this test of the ground station using a laboratory
test signal.
Page 4 of 6

67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.
Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

of the received signal; the saturation and strong

distortion are evident (top third of the figure).
5.2 Outdoor performance results
The FUNcube satellite was successfully tracked
outdoors, in an open field near a residential area. Its
telemetry signal was received and recorded. The
telemetry signal consists of a continuous repetition of
0.7 seconds of an unmodulated carrier, followed by 4.7
seconds of a 1.2 kbps BPSK signal. The procedure to
detect the signal is as follows. In order to avoid the
receivers strong DC offset, we downconvert the
satellites signal to a small intermediate frequency, 39
kHz in this case. This allows us to avoid the DC offset
while still fitting the signal within the receivers
bandwidth. We recorded the signals to audio files;
final downconversion to baseband is performed in
software after capture is finished.
We are interested in proving that the proposed
ground station is capable of receiving FUNcubes
telemetry data with adequate signal-to-noise ratio. To
do this, we show the estimated power spectral density
of the recorded intermediate frequency signal. We
show the signal as an unmodulated carrier in Fig. 5,
and as a modulated BPSK signal in Fig. 6.
The periodograms correspond to the quadrature
signals. They were calculated by windowing the signal
with a Blackman window of 211 elements, and running
a Welch periodogram estimation [8] with 64 samples
of overlap and FFT calculations with 212 elements.

Fig. 5: Estimation of the power spectral density of the

unmodulated carrier transmitted by FUNcube in its
telemetry signal.

IAC-16-B.2.8-GTS.3.7.x33593

Fig. 6: Estimation of the power spectral density of

the FUNcubes BPSK telemetry signal.
The peak amplitude to noise floor ratio can be
estimated to be around 13 dB for the unmodulated
carrier and 11 dB for the BPSK signal.
While these measurements are not as strong as
those suggested by our (optimistic) link margin
calculations, they are still useful and prove that the
proposed station is capable of receiving VHF satellite
signals.
We also report successful reception of signals from
the International Space Station and from weather
satellites from the US National Oceanic and
Atmospheric Administration.
6. Conclusions and discussion
We have shown that with a relatively small
investment and using widely available off-the-shelf
components, almost any individual and certainly any
research or educational institution may have access to
space communications. We have calculated the link
margin showing the ground stations potential
theoretical performance, and demonstrated its actual
performance in fairly typical operating conditions.
The proposed ground station is small and portable.
It is no longer necessary to purchase and maintain
delicate and expensive equipment or to devote large
amounts of space to this purpose. Furthermore, the
ground station we have proposed is flexible enough
that those interested in the technical details can have
full access to the signals for further study; at the same
time, it is possible to simply use existing software as a
black-box if the end result is given priority.
We believe that these developments will promote
public and student interest in space, engineering,
signal processing and related fields.

Page 5 of 6

67th International Astronautical Congress (IAC), Guadalajara, Mexico, 26-30 September 2016.
Copyright 2016 by the International Astronautical Federation (IAF). All rights reserved.

Much work remains to be done in improving the

signal-to-noise ratio and in documenting the stations
performance and operation. As well, alternative
configurations using different components need to be
explored. Many areas of space signal processing
remain locked behind proprietary software that is
difficult to use.
References
[1] V.Dascal, P. Dolea, O. Critea, T. Palade, Aspect of
a low cost ground station for GENSO network, Acta
Technica Napocensis, Vol. 52, Number 4, 2011.
[2] V. Dascal, P. Dolea, O. Critea, T. Palade,
Advanced VHF ground station for NOAA weather
satellite APT image reception, Acta Technica
Napocensis, Vol 53, Number 3, 2012.
[3] B. Sklar, Digital Communications: Fundamentals
and Applications, Prentice Hall, US, 1988.
[4] AMSAT-UK, The FUNcube Handbook, AMSATUK, Dorset, UK, November 2013.

IAC-16-B.2.8-GTS.3.7.x33593

[5] S. Cakaj and K. Malari, Rigorous analysis on

performance of LEO satellite ground station in urban
environment, International Journal of Satellite
Communications and Networking, Vol. 25, Issue 6,
pp. 619643, 2007.
[6] S. Cakaj et al, Elevation impact on signal to
spectral noise density ratio for low earth orbiting
satellite ground station at S-band, Science and
Information (SAI) Conference, 2014.
[7] M. Fischer and A. L. Scholtz, Design of a Multimission Satellite Ground Station for Education and
Research, Second International Conference on
Advances in Satellite and Space Communications,
Athens, 2010, pp. 58-63.
[8] P. D. Welch, The use of the fast Fourier transform
for the estimation of power spectra, IEEE Transactions
on Audio Electroacoustics, vol. AU-15, pp. 70-73,
June 1967.

L
N

Page 6 of 6