Aerospace and Electronic Systems Magazine November 2017 - 50

ADS-B Jamming Mitigation
iii. the implementation has not legacy problems and doesn't
by the EPA algorithm. Note that (d) is quite similar to the original
need to use new system approaches (as in the case of messignal (a) and can be decoded. The two found sources are then
sage authentication/encryption), but only a "software upsent to the decoder. Only the second one is recognized as genuine.
date" in the case of an existing multichannel station.
Comparing the decoded bit stream of the unjammed message and
the jammed one with or without the application of the EPA it is
The one drawback we have to mention is the fact that the propossible to evaluate the performance of the algorithm.
posed method requires a multichannel antenna and receiver, thus
All the results, for all the types of jammer and for different SIR
increasing costs of ADS-B stations. Note that this may be acceptare summarized in Figure 5 and Figure 6. In particular, Figure 5
able considering that a multichannel receiver with signal separashows the rate of "lost replies" ("lost replies" are replies not recogtion algorithms also improves the Mode S channel capacity, i.e. the
nized from the decoder) for various SIR and various type of jamcapability to manage many simultaneous aircraft with respect to
mer, with and without EPA processing. Figure 6 shows the rate of
the classical single antenna receiver.
"decoded replies" ("decoded replies" are replies that are decoded
Last, it must be remarked that EPA doesn't need calibrated
without errors from the decoder) for various SIR and various type
channels (as in the case of the classical adaptive beamforming or
of jammer, with and without EPA processing. The rates are calAOA estimation), therefore, its hardware is "more" but not "more
culated with respect to all considered replies and all considered
complicated" than single channel receiver. The cost of future inreplies were jammed; therefore, the rate is computed with respect
to the total number of interfered signals. Table 1 shows
a summary of the results for any type of jammer considered. The rate of signals decoded with zero error and the
rate of lost replies is reported.
These results show that for any kind of tested jammer,
with the proposed processing method, the number of the
decoded replies without error is drastically increased and
the number of lost replies is drastically reduced. Note,
for example, that in the worst case (SIR = -26 dB) the
lost replies are almost the totality for any kind of jammer
in the case of the classical (i.e. without EPA capability)
receiver. This number is reduced (in the worst case) to
22.7%. Furthermore, also considering the number of replies received without errors, for any kind of jammer we
have no reply with zero error, for the classical receiver.
This figure becomes, at least, 63.3% of the total jammed
replies without errors, in the case of EPA application, a
significant improvement in ADS-B message receiving Figure 5.
EPA-Jammer mitigation results. Rate of lost replies for different type of jammer and
and decoding.
different SIR values. (*) without EPA processing, (+) with EPA processing.

CONCLUSION
In this article, the effects of jamming on ADS-B receivers
were analyzed and discussed using real data, resulting in
a strong reduction of the coverage, a reduced detection
probability, and a possible overload of the processor are
caused by jamming. A physical level solution based on a
signal separation method in a multichannel receiver was
also proposed, evaluated with real ADS-B signals, and
discussed.
The signal separation is implemented before the detection procedures and allows the remaining receiver
chain not to be changed. The benefits of this solution are:

the ADS-B station has the capability to work also
in jamming conditions if the receiver is not saturated;

ii.

the presence of any kind of malicious signal can
be monitored and an alarm can be raised (integrity monitoring of the receiver);

Figure 6.

EPA-Jammer mitigation results. Rate of decoded replies without error for different
type of jammer and different SIR values. (*) without EPA processing, (+) with EPA
processing.

IEEE A&E SYSTEMS MAGAZINE

NOVEMBER 2017

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine November 2017