Detecting Drones and Human Beings With DVB-S Based COTS Passive Radar For Short-Range Surveillance
Detecting Drones and Human Beings With DVB-S Based COTS Passive Radar For Short-Range Surveillance
Detecting Drones and Human Beings With DVB-S Based COTS Passive Radar For Short-Range Surveillance
Abstract—This work addresses the exploitation of Digital Video others. For this reason, the purpose of this paper is to analyze
Broadcasting - Satellite (DVB-S) signals as sources of opportunity the feasibility of polarimetric DVB-S based PBR for small
for the surveillance of small targets in the proximity of critical target surveillance. Passive coherent location technology is
infrastructures. A performance analysis has been carried out
and it is reported to demonstrate the feasibility of the proposed particularly suitable for this task as it does not emit any elec-
system. Then, a preliminary experimental validation is performed tromagnetic radiation and thus it also allows covert operations.
using data collected by means of a low cost COTS based multi- The main contribution of this work is the preliminary
polarimetric receiver to detect small and close targets, as drones demonstration of the possibility to perform the monitoring of
and human beings. We present promising experimental results
small targets, drones and humans, in critical areas. To this aim,
that show the capability of the proposed system to perform
the desired task and, potentially, the opportunity to gain more we develop a DVB-S based PBR demonstrator and we use it
information on the target motion based on its Doppler signature. for an experimental validation.
The reminder of the paper is organized as follows. We
Index Terms—Passive bistatic radar, DVB-S signal, COTS introduce the system and its feasibility for the considered
based, small target detections, polarimetric diversity
application in Section II. In Section III, we describe the acqui-
sition campaign and the employed signal processing scheme.
I. I NTRODUCTION
The experimental validation reported in Section IV shows the
Passive bistatic radar (PBR) sensors [1] can detect and local- capability of the employed system to detect small targets.
ize targets using signals emitted by non-cooperative sources, Finally, some conclusion remarks are reported in Section V.
usually referred to as illuminators of opportunity (IOs). They
are based on the bistatic radar technology, which consists of
II. DVB-S BASED PASSIVE B ISTATIC R ADAR
an emitter and a receiver that are not co-located.
In recent years, significant progress has been made in A typical PBR geometry is shown in Fig. 1. It includes
PBR technology and, depending on the considered applica- two directional antennas, namely the reference antenna (R)
tion, different signals have been considered as sources of pointed straight at the IO to collect the reference signal, and
opportunity. Among them, a recently renewed interest has the surveillance antenna (S) steered toward the coverage area
been received by the exploitation of satellite signals, such to collect target echoes.
as GNSS (Global Navigation Satellite System) and DVB-
S (Digital Video Broadcasting - Satellite), see e.g. [2]-[10]
and the references therein. Among the main advantages of
using this kind of signals, we recall their permanent world-
wide availability as well as their wide transmitted bandwidth,
which results in a good range resolution. Recently, some
contributions have been published and have demonstrated
the potential of satellite signals based PBR for a variety of
applications. With particular reference to DVB-S signals, a
waveform analysis has been reported in [11]; the authors in [4]
showed the capability of such systems to detect ships, while
the system feasibility for car detection was demonstrated in
[10].
Nowadays, a challenging task is represented by the surveil-
lance of critical infrastructures, that are potentially threatened
by the intrusion of drones and unauthorized people, among Fig. 1. Principal rays and antennas disposition.
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.
Several geostationary communication satellites are currently
working and transmitting to Europe, specifically to Italy,
with different equivalent isotropically radiated power (EIRP)
values. Therefore, it is necessary to carry out a thorough
investigation in order to find the best satellite with a good EIRP
and a correct position, among other desirable parameters.
Once the satellite to be used has been selected, a per-
formance prediction analysis must be carried out. In order
to analyze the possibility to exploit DVB-S transmissions to
detect targets, we use the well known bistatic radar equation to
have a first approximation of the signal-to-noise ratio (SN R)
level. The bistatic radar equation is given by
EIRP G c2 σ Ti
SN R = (1)
(4π)3 f02 d2E d2T F kB T Fig. 2. The SN R as a function of dT for different values of σ.
being the employed parameters defined in Table I for the
consider Astra 1M satellite and the receiving RF front-end surveillance antenna. In order to have two cross-polarizations
developed for this project. in the reference parabolic reflector we install a small feedhorn
antenna with an orthomode transducer (OMT) in the focus of
TABLE I
M AIN PARAMETERS OF EQ .(1)
the reflector (see Fig. 4). In the same way an OMT is installed
in the surveillance feedhorn (see Fig. 5).
PARAMETER SYMBOL VALUE
transmitted Power EIRP 53 dBw
carrier frequency f0 10.7 GHz
emitter distance dE 36000 Km
receiver antenna S gain G 5 dB
noise figure F 1 dB
coherent processing interval Ti 0.5s
Boltzmann constant kB 1.38 10−23 J/K
light velocity c 3 · 108 m/s
Temperature T 290 K
38
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.
Fig. 4. Reference antenna with a parabolic reflector, a feedhorn, an OMT
and two LNBs.
39
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.
In particular, Fig.10 (a) and (b) show the results obtained
using the H and V polarization, respectively, while Fig.10
(c) shows the results obtained using the polarimetric NCI
approach. Fig.10 confirms the considerations made on Figs.
8 and 9 and shows that, using the proposed solution, we are
able to detect the drone with a good continuity for its entire
trajectory.
40
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.
(a) (a)
(b) (b)
(c) (c)
Fig. 9. Results against drone for datafile #2: Minimum nominal Pf a values Fig. 10. Results against drone: Raw detections for 40 consecutive data
in the range-Doppler domain to be set in order to detect each bin, using (a) windows with Pf a = 10−4 , using (a) single-pol H (b) single-pol V (c)
single-pol H (b) single-pol V (c) polarimetric NCI. polarimetric NCI. Different gray shades map the temporal information, starting
from the darkest one.
0.5s, extracted at the time when the human was starting its CFAR detection stage must be carried out. In Fig. 11 (c), we
run (a) and at the time he was running quickly (b). In both show the detection results obtained for the human target for a
cases, the target peak is approx. 20dB above the disturbance set of 40 consecutive datafiles with Pf a = 10−6 , using training
power level. Incidentally, note that this result matches with the data selected only in the range dimension in order to avoid
SNR analysis reported in Fig. 2 for a RCS of 1 m2 , which including micro-Doppler contributions in the secondary cells.
is likely to correspond to the monostatic RCS of a human Furthermore, to face the problem of having multiple detections
[17]. Moreover, it is evident that along with the mass center in the Doppler domain at each range bin we only considered
that was moving with an approx. constant velocity, the limbs one detection per range cell, using the detections center of
were also fluctuating, resulting in the insurgence of the micro- gravity. Using these strategies, the human trajectory can be
Doppler phenomena. This is particularly evident in Fig. 11 (b) easily identified in Fig. 11 (c).
and could potentially allow us to infer on the type of target
movement, via a classification based on the different micro- V. C ONCLUSION
Doppler signatures. Clearly, Fig. 11 (b) also suggests that a The feasibility of passive radar systems based on DVB-S
proper selection of the training data to be used in the CA- opportunity transmissions for the detection of small targets
41
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.
surveillance applications as well as the employment of the
developed RF front-end architecture to localize the detected
targets. Moreover, advance signal processing strategies to ex-
ploit polarization diversity will be considered with the purpose
of further improving the performance while different solution
to extract the micro-Doppler signatures will be investigated.
VI. ACNOWLEDGEMENT
The authors gratefully acknowledge Genex RF s.r.l. for the
valuable advices regarding the RF components selection.
R EFERENCES
(a) [1] H. D. Griffiths and C. J. Baker, Introduction to Passive Radar, Norwood,
MA, USA: Artech, Mar. 2017.
[2] D. Pastina et al., ”Maritime Moving Target Long Time Integration for
GNSS-Based Passive Bistatic Radar,” in IEEE Transactions on Aerospace
and Electronic Systems, vol. 54, no. 6, pp. 3060-3083, Dec. 2018.
[3] J. -L. Bárcena-Humanes, N. del-Rey-Maestre, M. P. Jarabo-Amores, D.
Mata-Moya and P. Gomez-del-Hoyo, ”Passive radar imaging capabilities
using space-borne commercial illuminators in surveillance applications,”
2015 Signal Processing Symposium (SPSympo), Debe, 2015, pp. 1-5.
[4] L. Daniel, S. Hristov, X. Lyu, A. G. Stove, M. Cherniakov and M.
Gashinova, ”Design and Validation of a Passive Radar Concept for Ship
Detection Using Communication Satellite Signals,” in IEEE Transactions
on Aerospace and Electronic Systems, vol. 53, no. 6, pp. 3115-3134, Dec.
2017.
[5] H. Nies, F. Behner, S. Reuter and O. Loffeld, ”First Results of Passive
Radar Imaging and Tracking Using Geostationary Satellites,” 2018 15th
(b) European Radar Conference (EuRAD), Madrid, 2018, pp. 214-217.
[6] I. Pisciottano, D. Cristallini and D. Pastina, ”Maritime target imaging via
simultaneous DVB-T and DVB-S passive ISAR,” in IET Radar, Sonar
and Navigation, vol. 13, no. 9, pp. 1479-1487, 9 2019.
[7] A. G. Stove, M. S. Gashinova, S. Hristov and M. Cherniakov, ”Passive
Maritime Surveillance Using Satellite Communication Signals,” in IEEE
Transactions on Aerospace and Electronic Systems, vol. 53, no. 6, pp.
2987-2997, Dec. 2017.
[8] D. Cristallini, M. Caruso, P. Falcone, D. Langellotti, C. Bongioanni, F.
Colone, S. Scafè, P. Lombardo, “Space-Based Passive Radar Enabled
By The New Generation Of Geostationary Satellites”, IEEE Aerospace
Conference, Big Sky (MT), USA, 6-13 March 2010.
[9] A. Arcangeli, C. Bongioanni, N. Ustalli, D. Pastina, P. Lombardo, “Pas-
sive Forward Scatter Radar based on satellite TV broadcast for air target
detection: preliminary experimental results”, IEEE Radar Conference
2017, 9-11 May 2017, Seattle, WA (USA).
[10] P. Gomez-del-Hoyo, M. Jarabo-Amores, D. Mata-Moya, N. del-Rey-
(c) Maestre and J. Rosado-Sanz, ”First Approach on Ground Target Detection
with GPS based Passive Radar: Experimental Results,” 2019 Signal
Fig. 11. Results against human target : (a)-(b) ψN CI (fD , RB ) obtained for
Processing Symposium (SPSympo), Krakow, Poland, 2019, pp. 71-75.
two different datafiles (c) human path detections for 40 consecutive datafiles
[11] Z. Sun, T. Wang, T. Jiang, C.Chen and W. Chen, ”Analysis of the
with Pf a = 10−6 . Different gray shades map the temporal information,
properties of DVB-S signal for passive radar application,” 2013 Interna-
starting from the darkest one.
tional Conference on Wireless Communications and Signal Processing,
Hangzhou, 2013, pp. 1-5.
[12] W. J. Alberhseim, ”A closed-form approximation to Robertson’s detec-
tion characteristics,” in in Proceedings of the IEEE, vol. 69, no. 7, pp.
(drones and humans) has been preliminarly demonstrated. 839-839, July 1981.
[13] F. Colone, D. W. O’Hagan, P. Lombardo and C. J. Baker, ”A Multistage
To this aim, an experimental analysis has been carried out. Processing Algorithm for Disturbance Removal and Target Detection
Furthermore, it has been proven that a big parabolic reflector in Passive Bistatic Radar,” in IEEE Transactions on Aerospace and
is not necessary to detect small targets at very short ranges. Electronic Systems, vol. 45, no. 2, pp. 698-722, April 2009.
[14] P. Lombardo and F. Colone, “ Advanced processing methods for passive
In this work, a simple feedhorn with approximately 5 dB gain bistatic radar’, in Melvin, W. L., and Scheer, J. A. (Eds.): ’Principles
has been used. This approach also allowed to achieve wide of Modern Radar: Advanced Radar Techniques’, Raleigh, NC: SciTech
angular coverage. This work also demonstrates that the joint Publishing, 2012, pp. 739–821.
[15] F. Colone and P. Lombardo, ”Polarimetric passive coherent location,” in
exploitation of different polarimetric channels according to IEEE Transactions on Aerospace and Electronic Systems, vol. 51, no. 2,
the polarimetric NCI can effectively improve the performance pp. 1079-1097, April 2015.
of the system with respect to the use of single polarimetric [16] F. Filippini, F. Colone, D. Cristallini and G. Bournaka, ”Experimental
results of polarimetric detection schemes for DVB-T-based passive radar,”
channels. It is expected that higher gain antennas will enable in IET Radar, Sonar and Navigation, vol. 11, no. 6, pp. 883-891, 6 2017.
the detection of small targets at higher distances. Future [17] M.A. Richards, Fundamentals of Radar Signal Processing, McGraw-Hill
works will include a performance assessment in longer range Companies, New York (2005).
42
Authorized licensed use limited to: AMITY University. Downloaded on November 30,2022 at 04:56:37 UTC from IEEE Xplore. Restrictions apply.