aerospace

Article
Chirp Signals and Noisy Waveforms for Solid-State
Surveillance Radars
Gaspare Galati *, Gabriele Pavan † and Francesco De Palo †
Department of Electronic Engineering, University of Rome “Tor Vergata”, via del Politecnico 1,
00133 Rome, Italy; gabriele.pavan@uniroma2.it (G.P.); francesco.de.palo@uniroma2.it (F.D.P.)
* Correspondence: gaspare.galati@uniroma2.it; Tel.: +39-06-7259-7417
† These authors contributed equally to this work.

Academic Editors: Yan (Rockee) Zhang and Mark E. Weber

Received: 1 December 2016; Accepted: 8 March 2017; Published: 14 March 2017

Abstract: Since the advent of “pulse compression” radar, the “chirp” signal (Linear Frequency
Modulation, LFM) has been one of the most widely used radar waveforms. It is well known that,
by changing its modulation into a Non-Linear Frequency Modulation (NLFM), better performance
in terms of Peak-to-Sidelobes Ratio (PSLR) can be achieved to mitigate the masking effect of nearby
targets and to increase the useful dynamic range. Adding an appropriate amplitude modulation,
as occurs in Hybrid-NLFM (HNLFM), the PSLR can reach very low values (e.g., PSLR < −60 dB),
comparable to the two-way antenna sidelobes in azimuth. On the other hand, modern solid-state
power amplifier technology, using low-power modules, requires them to be combined at the Radio
Frequency (RF) stage in order to achieve the desired transmitted power. Noise Radar Technology
(NRT) represents a valid alternative to deterministic waveforms. It makes use of pseudo-random
waveforms—realizations of a noise process. The higher its time-bandwidth (or BT) product, the higher
the (statistical) PSLR. With practical BT values, the achievable PSLR using pure random noise is
generally not sufficient. Therefore, the generated pseudorandom waveforms can be “tailored”
(TPW: Tailored Pseudorandom Waveforms) at will through suitable algorithms in order to achieve
the desired sidelobe level, even only in a limited range interval, as shown in this work. Moreover,
the needed high BT, i.e., the higher time duration T having fixed the bandwidth B, matches well
with the low power solid-state amplifiers of Noise Radar. Focusing the interest on (civil) surveillance
radar applications, such as ATC (Air Traffic Control) and marine radar, this paper proposes a general
review of the two classes of waveforms, i.e., HNLFM and TPW.

Keywords: radar waveforms; chirp; pulse compression; non-linear FM; low PSLR; solid-state
amplifiers; Noise Radar Technology

1. Introduction
For any new radar, the waveforms design is a key element, in which conflicting requirements have
to be satisfied, i.e., (to mention the main ones), range resolution and accuracy (defined as the degree
of conformance between the estimated or measured position and the true position), dynamic range,
Multiple-Input Multiple-Output (MIMO) or polarimetric channels orthogonality, Low Probability of
Intercept (LPI) [1–3], low interference to other apparatuses [4], and best exploitation of the transmitter
power with a low (or unitary) crest factor. To mitigate the masking effect of nearby targets [5],
the sidelobe level of the autocorrelation function of the transmitted pulse has to reach very low values
comparable to the two-way antenna sidelobes in azimuth (typical of marine radars). Waveform design
is of paramount interest in weather radar community, as solid-state and phased-array technology is
being increasingly used also there. However, in that community the Integrated Sidelobe Level (ISL)
is of interest. In this paper, mostly related to man-made targets (planes, ships) the peak sidelobe

Aerospace 2017, 4, 15; doi:10.3390/aerospace4010015 www.mdpi.com/journal/aerospace

Aerospace 2017, 4, 15 2 of 14

level is considered. This approach leads to specify the Peak-to-Sidelobes Ratio (PSLR) typically better
than −60 dB [6,7]. One way to meet this requirement is the use of sidelobe-suppression filters in
reception. However, these sidelobe-suppression filters [8,9] add Signal-to-Noise-Ratio (SNR) losses
and an increase of the Doppler mismatch effect.
Alternatively, the optimal design of “chirp” radar waveforms [10–12], may be considered. Since
the advent of “pulse compression” radar, the “chirp” signal (Linear Frequency Modulation, LFM) has
been one of the most widely used radar waveforms. It is well known that, by changing its modulation
into a Non-Linear Frequency Modulation (NLFM) [13–15], better performance in terms of PSLR can
be achieved.
Historically, the design of chirp radar in the Western was made public in the 1960s [16] and, in the
Eastern world, in the works (1950s) by Yakov Shirman [17], who received the IEEE Pioneer Award in
2009 “For the independent discovery of matched filtering, adaptive filtering, and high-resolution pulse
compression for an entire generation of Russian and Ukrainian radars” [18,19].
The “chirp” theory is well established and is related to the Stationary Phase principle [10,12,20]:
for a large enough number of independent samples, or product (also called Compression Ratio, or BT)
between time duration T and bandwidth B, the group delay of a LFM signal is proportional to the
instantaneous frequency. Moreover, the spectrum of the LFM signal after the matched filtering is
quasi-rectangular, with constant amplitude and linear phase in the bandwidth B (and ideally, zero
amplitude outside it). Its autocorrelation has a time duration 1/B and an unacceptable PSLR about
13.2 dB below the main peak. The sidelobes can be lowered by using an NLFM law, which can be
easily designed by means of the well-known spectral windows [13,20]. However, this method, which
relies on the Stationary Phase method, works well when BT is “large” (in practice for BT greater than
a few thousands) but becomes less and less effective when BT decreases, as occurs in various civil
applications like ATC (Air Traffic Control) radar and marine (or navigation) radar.
A new design method to cope with the sidelobe problem has been presented in [21,22], leading to
a kind of Hybrid NLFM (HNLFM) whose amplitude, however, is not constant during the duration T
of the signal. Such a requirement creates implementation problems with the widely-used saturated
(C-class) power amplifiers. This family of waveforms is enhanced and analyzed in the following.
The power solid-state technology is increasingly used for radar transmitters using microwave
transistor amplifiers leading, in most cases, to production/generation power issues. In fact, to achieve
the typical medium-range for air or sea surveillance of civil targets, i.e., about 90 to 110 km (50 to
60 NM, Nautical Mile), a peak power of a few tens of kW has to be transmitted with a duty cycle
around 10%. This power requirement is met by combining about 16 solid-state power modules at
the Radio Frequency stage, typically emitting 50 to 100 W each. The pertaining implementation is
not easy [23], since it should introduce low losses to minimize the attenuation of the coming through
signal. It should also present good isolation between input/output ports so that each module is not
influenced by the other, and finally, it should provide a well-matched circuitry to maximize the power
transfer without occurring in load pulling. Apparatus size and weight can also increase if a large
number of solid-state modules are needed. The high power requirement becomes more problematic if
the before mentioned HNLFM signal is used because of its very precise amplitude modulation.
Another method, technically unrelated to the HNLFM, is the Noise Radar Technology (NRT).
It represents a valid candidate to overcome the high-power stage issue. It makes use of pseudo-random
waveforms, realizations of a Gaussian noise process [24]. The higher their time-bandwidth (or BT)
product, the higher their (statistical) PSLR. In fact, since the PSLR is related to how much the waveform
is equal to itself, the number of independent samples (with their phases) represents the degree
of freedom available to make the correlation sidelobes as low as possible. Using pure random
noise, practical BT values are generally not large enough to achieve PSLR. Therefore, the generated
pseudorandom waveforms should be “tailored” (TPW: Tailored Pseudorandom Waveforms) at will
through suitable algorithms [25] in order to achieve the desired sidelobe level. The suppression can
also be applied over a limited range interval, as will be shown throughout this paper. Since the PSLR
Aerospace 2017, 4, 15 3 of 14

increases with the signal time duration T (assuming a fixed bandwidth B), we consider a 100% duty
cycle operation for NRT with separated transmitting and receiving antennas. Then the architecture
considered is a bistatic radar with co-located antennas, with low electromagnetic coupling between
them. This architecture allows us to lengthen up the signal to the Waveform Repetition Time (WRT)
which is the equivalent of the Pulse Repetition Time (PRT) for pulsed radar. This framework provides
some advantages of the TPW approach over the pulsed HNLFM, including:

• the longer time duration allows to lower the transmitted peak power (with respect to an equivalent
pulse compression radar) by about a figure of approximately 30 times, keeping unchanged the
overall system requirements (e.g., maximum range, range resolution, bandwidth allocation) for
the same application;
• the lower transmitted power calls for less solid-state modules to be combined, easing the radar
transmitter architecture;
• a large number of quasi-orthogonal waveforms can be transmitted, alleviating mutual interference
problems and paving the way for an MIMO operation [26,27].

The drawbacks of TPW include the dual antenna configuration (typical of Continuous Wave (CW)
radars) as well as the burden of the real-time computation of the cross-ambiguity function. However,
the matched filter approach can be successfully used. Design criteria and ensuing comparison of
deterministic and pseudo-random waveforms are shown hereafter.
This paper focuses the attention on two waveform classes: HNLFM and TPW, suitable to be used
in (civil) surveillance radar applications, such as ATC and marine radar. Most of this paper is a review
of (many) works, some unpublished, by the authors.
The paper is organized as follows: Section 2 describes the deterministic waveforms design and
compares classical pulse compression waveforms with the novel ones; Section 3 shows a different
approach to pulse compression based on random waveforms (noisy signals) making use of CW
operation; finally, Section 4 reports conclusions and future perspectives.

2. Deterministic Waveforms Design

The complex envelope of a generic signal, modulated both in Amplitude and in Phase (AM-PM), is:

s(t) = a(t)e jφ(t) , (1)

where a(t) and φ(t) denote respectively the AM and PM modulation. The stationary phase
principle [10,16] establishes that, at the instantaneous frequency ωt the amplitude spectrum |S(ωt )|2
of a given signal s(t), having the amplitude a(t), can be approximated as [10]:

a2 ( t )
|S(ωt )|2 ∼
= 2π 00 , (2)
|φ (t)|

i.e., the energy spectral density at the frequency ωt = φ0 (t) (in radians) is larger when the rate of
change of ωt is smaller. From Equation (2) the amplitude modulation function a(t) can be evaluated as:
r
1 
2
a(t) ∼
= S φ0 (t)  |φ00 (t)| . (3)
2π

However, the validity of the stationary phase approximation depends on the compression ratio
BT, and it should be considered reliable only for large BT (in practice, greater than a few thousands).
It is less and less effective when a low BT is required, as it happens in various radar applications.
Considering the Millett waveform, i.e., a “cosine squared on a pedestal” weighting [13], the output of
the Matched Filter (MF) reaches the theoretical PSLR of −42 dB only when BT is greater than about one
thousand. For any lower compression ratio, a convenient frequency modulation function (in radians)
Aerospace 2017, 4, 15 4 of 14

can be obtained
Aerospace
Aerospace 2017,
2017, 4, 15 as a weighted sum of the Non-Linear tangent FM term and the LFM one, hence
4, 15 the
44 of
of 14
14
name Hybrid-NLFM [21,22]:

11
  2γ  2
0 ( )=π
ϕ 2γt + (1 − )2t ,, (4)
φ (t) = πB α tg + (1 − α ) (4)
tg(γ(γ)
) T T
where ∈ (0, 1) is the relative weight between the linear and the tangential terms, B is the swept
where α ∈ (0, 1) is the relative weight between the linear and the tangential terms, B is the swept
frequency (nearly equal to the signal’s bandwidth), γ is the Non-Linear tangent FM rate and, being
frequency (nearly equal to the signal’s bandwidth), γ is the Non-Linear tangent FM rate and, being
∈ h− T, + T ,iT denotes the pulse-width.
t ∈ − 2 , + 2 , T denotes the pulse-width.
If ( ) is a signal with a Gaussian spectrum: | (ω )| = exp −βω2  with β = 1, using the
If s(t) is a signal with a Gaussian spectrum: |S(ωt )|2 = exp −β B2t with β = 1, using the
optimized values of α and γ as evaluated in [21,22] (i.e., those values that reach the optimum PSLR
optimized values of α and γ as evaluated in [21,22]⁄⁄ (i.e., those values that reach the optimum PSLR
maximizing the transmission efficiency: = R +T/2( ) ), the energy loss results as low as 0.58
maximizing the transmission efficiency: η = T1 ⁄⁄ −T/2 a2 (t)dt), the energy loss results as low as
dB with
0.58 dB respect to the to
with respect case
theofcase
unityofamplitude. InsertingInserting
unity amplitude. EquationEquation
(4) into Equation
(4) into (3), the amplitude
Equation (3), the
modulationmodulation
amplitude has been evaluated
has beenasevaluated
shown inasFigure
shown 1. in
Zooming around
Figure 1. Zooming ( )around
= 1, i.e.,a(in
t) the
= 1,interval
i.e., in
(−0.3, 0.3), some ripples of the relative amplitude order of 10 are −observed.
3 In
the interval (−0.3, 0.3), some ripples of the relative amplitude order of 10 are observed. In Figure Figure 2, the2,
corresponding instantaneous frequency modulation is
the corresponding instantaneous frequency modulation is shown. shown.

Amplitude
Amplitude Modulation
Modulation
1.2
1.2

11

11
0.8
0.8
0.999
0.999
a(t)
a(t)

0.6
0.6
0.998
0.998
-0.3
-0.3 -0.2
-0.2 -0.1
-0.1 00 0.1
0.1 0.2
0.2 0.3
0.3
0.4
0.4

0.2
0.2

00
-0.5
-0.5 -0.4
-0.4 -0.3
-0.3 -0.2
-0.2 -0.1
-0.1 00 0.1
0.1 0.2
0.2 0.3
0.3 0.4
0.4 0.5
0.5
t/T
t/T

Figure
Figure 1. Normalized amplitude
1. Normalized amplitude weighting
weighting for
for Hybrid
Hybrid Non-Linear
Non-Linear Frequency
Frequency Modulation
Modulation (HNLFM)
(HNLFM)
(red
(red line:
line: optimum
optimum weighting,
weighting, blue
blue line:
line: sub-optimum
sub-optimum weighting).
weighting).
f/B

Figure 2. Instantaneous normalized frequency for HNLFM.

Figure 2. Instantaneous normalized frequency for HNLFM.

Figure 3 shows the PSLR of the matched filter output for the optimized waveform (solid line-
circles). With BT decreasing, the approximation due to the principle of the stationary phase becomes
worse and worse causing an increase in the PSLR. However, with BT = 64 the PSLR (−51 dB) is still
Aerospace 2017, 4, 15 5 of 14

Figure 3 shows the PSLR of the matched filter output for the optimized waveform
(solid line-circles).
Aerospace 2017, 4, 15 With BT decreasing, the approximation due to the principle of the stationary 5 of 14
phase becomes worse and worse causing an increase in the PSLR. However, with BT = 64 the PSLR
(compatible withcompatible
−51 dB) is still many applications,
with many and for BT = 128
applications, andthefor
PSLR
BT =is128
−75the
dB,PSLR
which is75
is − suited to most
dB, which is
applications.
suited to most applications.
These excellent
These excellenttheoretical
theoretical results
results (circles
(circles in Figure
in Figure 3) are
3) are only only possible
possible if the amplitude
if the amplitude weighting
is the one shown in Figure 1 (red line) with the instantaneous frequency of Figure 2.of Figure 2.
weighting is the one shown in Figure 1 (red line) with the instantaneous frequency
PSLR (dB)

Figure 3.
3.Effect of the
Effect compression
of the ratio ratio
compression BT onBT the on
Peak-to-Sidelobes Ratio (PSLR),
the Peak-to-Sidelobes RatioHNLFM
(PSLR),(optimum
HNLFM
(optimum and sub-optimum)
and sub-optimum) and Milletand Millet
signals, thesignals, the lastonly
last is shown is shown only as a reference.
as a historical historical reference.

In practice,
In practice, it it may
may be be hard
hardto toimplement
implementthe therequirement
requirementon on a((t),),with
withripples
ripplesofofthetheorder
orderofof1
over 1000. In general, if a PSLR level of −60 dB is required, a simple
1 over 1000. In general, if a PSLR level of −60 dB is required, a simple rule of thumb tells us that rule of thumb tells us that the
the
fidelity of the waveform, in terms of (I, Q) or (Amplitude, Phase) shall
fidelity of the waveform, in terms of (I, Q) or (Amplitude, Phase) shall be better than 1:1000, or 10 bits,be better than 1:1000, or 10
bits,matter
no no matter
howhow the waveform
the waveform is designed.
is designed. It It
is is widelyknown—from
widely known—frommany manypractical
practical experiences
experiences
(unfortunately, not available in the open literature)—that hardware
(unfortunately, not available in the open literature)—that hardware imperfections increase the imperfections increase the PSLR
PSLR
with respect
with respect to to the
the “ideal”
“ideal” compressed
compressed waveform.
waveform. This This means
means thatthat if
if the
the required
required PSLR
PSLR is is for
for example
example
−60 dB, the intrinsic level of the waveform (which is our focus)
−60 dB, the intrinsic level of the waveform (which is our focus) has to be −67 dB, the difference has to be −67 dB, the difference
between these
between these figures
figures depending,
depending, of of course,
course, on on the
thestability/linearity
stability/linearity of the hardware.
of the hardware. In In aa recent
recent
private communication
private communication [28] [28] it
it is
is explained
explained that:that: “In
“In one
one case,
case, related
related to to aa NAVICO
NAVICO civil civil marine
marine radar
radar
using GaN for the solid state amplifier, and a non-linear FM, amplitude-weighted pulse with BT =BT
using GaN for the solid state amplifier, and a non-linear FM, amplitude-weighted pulse with =
150,
150, the measured PSLR is −55 dB, versus the “ideal” −60 dB of the
the measured PSLR is −55 dB, versus the “ideal” −60 dB of the waveform”. So, the effect of hardware waveform”. So, the effect of
hardware
is significantis significant but notand
but not dramatic, dramatic,
can beand can be with
controlled controlled
a carefulwith a careful design.
design.
If the amplitude of the transmitted signal is kept
If the amplitude of the transmitted signal is kept constant (the power constant (the power amplifier
amplifier working
working in in
saturation) with
saturation) with the theinstantaneous
instantaneousfrequencyfrequencyofofFigure Figure 2, 2,
thethetheoretical
theoretical PSLRPSLRincreases
increasesby about
by about25–
30 dB. An improvement can be obtained using a sub-optimum
25–30 dB. An improvement can be obtained using a sub-optimum waveform where the ripples shown waveform where the ripples shown
in the
in the zoom
zoomof ofFigure
Figure1 1are areremoved
removed imposing(a()t=
imposing ) =1 in this
1 in interval
this interval (dashed
(dashedlineline
in Figure 1). The
in Figure 1).
PSLR of this “sub-optimum” waveform results only 10–20 dB worse
The PSLR of this “sub-optimum” waveform results only 10–20 dB worse than the optimized signal than the optimized signal when
when ≥ 128
BT ≥(diamonds-line
128 (diamonds-line in Figure 3). Figure
in Figure 4 shows
3). Figure the half-power
4 shows the half-power pulse width
pulse (τ (τ) versus
width the
) versus
3dB
compression
the ratio
compression BT. BT.
ratio TheThe
values are are
values compared
comparedwith the the
with ones of the
ones LFM,
of the τ τ3dB
i.e.,i.e.,
LFM, = =. B1 .
For BT =
= 128
128 (frequency
(frequency swept of B ==1 MHz
swept of 1 MHz andand duration T
duration = 128 μs),
= 128 µs), Figure
Figure 5a shows
5a shows the
the compressed
compressed pulse, pulse,
whilewhile
FigureFigure 5b reports
5b reports its spectrum.
its spectrum. The PSLR
The PSLR reachesreaches −75 dB−for
about about 75 the
dB
for the optimum
optimum waveform waveform
case andcase andfor−the
−66 dB 66 dB for the sub-optimum
sub-optimum one, still for
one, still compatible compatible for most
most applications,
applications, e.g.,the
e.g., lower than lower thantwo-way
typical the typical two-way
antenna antennainsidelobes
sidelobes azimuth.inTheazimuth. i.e., ∆ =
The range resolution,
range resolution,
i.e., ∆R(= 2B c representing
(B−3dB representing
the −3 dB −3 dB bandwidth),
thebandwidth), is aboutis 550
aboutm.550 m.worth
It is It is worth clarifying
clarifying that that
the
−3dB
the
bandwidth of a waveform is very important when spectral limitations apply, which is the case of
bandwidth of a waveform is very important when spectral limitations apply, which is the case of
most civil radar. Of course, bandwidth values only give a rough indication of the range resolution,
whose correct numerical value is of course the −3 dB width of the output of the reception filter, either
matched or possibly mismatched for some reason.
Aerospace 2017, 4, 15 6 of 14

most civil radar. Of course, bandwidth values only give a rough indication of the range resolution,
whose correct numerical
Aerospace 2017,
2017, 4,
4, 15
15
value is of course the −3 dB width of the output of the reception filter, either
Aerospace 66 of
of 14
14
matched or possibly mismatched for some reason.
To improve
To improve
To improve the theresolution,
the maintaining
resolution,
resolution, maintainingthe
maintaining the same
the pulseduration,
same pulse
same pulse duration,
duration, thethe
the swept
swept
swept frequency
frequency
frequency has to
has has
to beto be
be
increased,
increased, with with
the the
onlyonly practical
practical limitation
limitation
increased, with the only practical limitation being being
being the
the restricted
restricted use
useofofthe spectrum.
the spectrum.
restricted use of the spectrum.

T=
T = 100
100 microsec
microsec -- Compressed
Compressed pulse
pulse width
width
88
Sub-optimum HNLFM
Sub-optimum HNLFM
77 LFM ((
LFM = 1/B)
3dB = 1/B)
3dB
66
55
44
33
22
11
00
32
32 64
64 96
96 128
128 160
160 192
192 224
224 256
256
BT
BT

Figure 4. Effect
Figure
Figure 4. of of
4. Effect
Effect thethe
of compression
the compressionratio
compression BT on
ratio BT
ratio onthe
on thecompressed
the compressed
compressed pulse
pulse
pulse width
width
width (τ (τ).).3dB ).
(τ

(a)
(a)

(b)
(b)
Figure 5.
Figure 5. HNLFM
HNLFM compressed
compressed pulse:
pulse: = 128
= 128 ,, = 128 μs
= 128 μs ,, = 1 MHz
= 1 MHz .. (a)
(a) Normalized
Normalized
Figure 5. HNLFM(b)
autocorrelation; compressed pulse:
Power spectrum
spectrum BT = 128, T = 128 µs, B = 1 MHz. (a) Normalized
density.
autocorrelation; (b) Power density.
autocorrelation; (b) Power spectrum density.
3. Noise
3. Noise Radar
Radar Technology
Technology
3. Noise Radar Technology
The HNLFM
HNLFM capability
capability to
to get
get very
very low
low sidelobes
sidelobes of
of the
the compressed
compressed pulse
pulse is
is strictly
strictly dependent
dependent
The
on the
The
on the ability of
HNLFM
ability ofcapability
the signal
the signal to
generation and
get veryand
generation amplification
lowamplification chain,
sidelobes ofchain, including the
the compressed
including thepulse
RF power
RF power amplifier,
is strictly to
dependent
amplifier, to
on thefaithfully
faithfully reproduce
reproduce
ability of thegeneration
the
the signal amplitude modulation
amplitude modulation
and of Figure
of Figure
amplification 1, calling
1,
chain, calling for highly
for
including highly linear
the RFlinear
powerA-class
A-class
amplifier,
to faithfully reproduce the amplitude modulation of Figure 1, calling for highly linear A-class amplifiers.
Aerospace 2017, 4, 15 7 of 14

Moreover, the available bandwidth is not fully exploited because of the particular frequency law of
Equation (4), which is the main law responsible for the low sidelobe level.
Noise Radar Technology (NRT) [29–35] may be a valid alternative to overcome the high-power
amplitude modulation issue. NRT makes use of pseudo-random waveforms that are realizations
of a Gaussian band-limited random process [24]. These “pure noise” realizations, once generated
and stored, are not strictly random anymore as they act as deterministic signals with known PSLR,
range resolution and ambiguity function. The number of different possible realizations to be used is
theoretically unlimited (modern pseudorandom numbers generator can reach a period of 21492 [36],
as implemented in Matlab generator—practically infinity), so that each radar can operate with its own
radar signal, possibly different from the others. Being noisy, the waveforms themselves possess a low
enough degree of mutual cross-correlation to pave the way for distributed MIMO operation with a
low level of mutual interferences [27] even operating at the same, or nearby, carrier frequency.
For a pure noise waveform, the PLSR value does not strongly depend on the amplitude
modulation [24] but, rather, on the Time Bandwidth product BT. The bandwidth being limited by the
application context (e.g., about 50 MHz, and 200 MHz as a maximum, for marine radar), BT may be
increased at will by selecting a continuous wave (CW) architecture instead of the pulsed one, keeping
unchanged the compression processing (computation of the correlation or the ambiguity function) at
the receiver side. The power can be significantly lowered with respect to an equal-performance pulse
radar architecture.

3.1. Power Budget

Let us consider a typical maximum range of 150 km (80 NM) with a range resolution of 150 m.
A comparison between pulsed HNLFM and CW noise radar can be done while keeping the transmitted
energy constant. For this purpose, let us consider the saturated HNLFM since the power loss with
respect to the optimal amplitude modulation is only 0.58 dB. In a CW architecture, the maximum delay
due to range is generally set as one-fourth of the Waveform Repetition Time (WRT), which corresponds
to the signal time duration if we neglect the data processing time between consecutive sweeps:

WRT 2 · Rmax
= . (5)
4 c
2·150000 (m)
For a maximum range of 150 km, it is required that: WRT 4 = 3·108 (m/s) = 1 ms,
i.e., WRT = 4000 µs, i.e., 4096 µs in order to work with power-of-two values. Knowing the
time duration of the noise signal (WRTnoise ) together with the HNLFM pulse width (THNLFM ),
the relationship between the needed peak powers can be evaluated as:

THNLFM
Pnoise = PHNLFM · , (6)
WRTnoise

where Pnoise and PHNLFM are respectively the peak power of CW noise and pulsed radar.
With WRTnoise = 4096 µs and THNLFM = 128 µs, it results: Pnoise = 0.03125 · PHNLFM . The equality
holds also for the optimal amplitude modulated HNLFM if the 0.58 dB power loss is considered.
Equation (6) states that the required power for a CW noise radar is about 15 dB lower than the peak
power required for a pulsed radar, keeping unchanged the maximum range. Lowering the transmitted
power means less solid-state modules and a straightforward RF power production/generation design.
The generation of M independent pure noise band-limited signals is shown in Figure 6. The (I, Q)
samples are generated through a white Gaussian random process. The signal to be filtered by the
frequency window H ( f ) is { xi }iN=1 where N is the number of generated samples and xi the ith complex
( I, Q) sample.
Aerospace2017,
Aerospace 2017,4,4,15
15 88 of
of14
14
Aerospace 2017, 4, 15 8 of 14

Figure 6. Pure noise generation block-diagram (ZMNL: Zero-Memory-Non-Linear transformation,

6. Pure
Figure 6.
FFT: Fast
noise
noise
Fourier
generation
generation
Transform,
block-diagram
block-diagram
iFFT:
(ZMNL: Zero-Memory-Non-Linear transformation,
inverse-FFT).(ZMNL: Zero-Memory-Non-Linear
FFT: Fast Fourier Transform, iFFT: inverse-FFT).
Fourier Transform, iFFT: inverse-FFT).
3.2. Unimodular Pure Noise
3.2. Unimodular Pure Noise
Being a white process, the signal has to be band-limited with a proper frequency window ( )
paper, we consider ((the
Being a white with aa proper window H f ))
frequency window
which can be any, process, the signal has to[37]
e.g., Blackman–Nuttall be or
band-limited
simply rectangular. In thisfrequency
which can be be any,
any, e.g.,
e.g.,Blackman–Nuttall
Blackman–Nuttall[37] [37]ororsimply
simply rectangular. In this paper, we consider
rectangular frequency window to fully exploit the available bandwidth; other windowsconsider
rectangular. In this paper, we can be usedthe
the rectangular
rectangular frequency
frequency window window to fully exploit
to fully the available bandwidth; other windows can
to gain a slight improvement in PSLRexploit
value. theAfteravailable bandwidth;
the frequency otherwindowing,
domain windows can
thebesignal
used
be
to used
gain ato gain
slight a slight
improvement improvement
in PSLR in PSLR
value. value.
After the After
frequencythe frequency
domain
amplitude is saturated to the maximum (i.e., a unimodular signal is obtained) value through a Zero- domain
windowing, windowing,
the signal
the signal is
amplitude amplitude
Memory-Non-Linear istosaturated
saturated(ZMNL) the maximumto the maximum
(i.e.,
transformation, while (i.e.,
a unimodular asignal
the phase unimodular
isiskept signal
obtained) value
unchanged. is obtained)
through
Since the a value
Zero-
( , )
through a Zero-Memory-Non-Linear
Memory-Non-Linear (ZMNL) (ZMNL)
transformation, transformation,
while the phase while
is kept the phase
unchanged. is kept ( , )
unchanged.
Since the
samples come from a random process, at each run, the algorithm provides different realizations
Since
samples
having ( I,same
thethe
come Q)from
samples
average come
a random from a random
process,
performances inattermsprocess,
each run,
of PSLR at and
the each run, the
algorithm algorithm
provides
cross-correlation provides
different
level, different
realizations
while the range
realizations
having the having
same the
average same average
performances performances
in terms of in terms
PSLR and of PSLR and cross-correlation
cross-correlation
resolution only depends on the used ( ). Figure 7 shows the spectrum of the unimodular band- level, whilelevel,
the while
range
the rangepure
resolution
limited resolution
only onlythe
depends
noise and depends
onspectrum
the used of (aused
on the ). Figure
LFM Hsignal
( f ).7Figure
having7the
shows shows theBT.
thespectrum
same spectrum
of
Thethe of thebandwidth
unimodular
unimodular
1 MHz band- is
band-limited
limited pure
fully exploited. pure
noise noise
and and
the the spectrum
spectrum of a of
LFMa LFM
signalsignal having
having the the
same sameBT.BT.
The The
1 1
MHz MHz bandwidth
bandwidth is
is fully
fully exploited.
exploited.
|X(f)|
|X(f)| [dB]
[dB]

Figure 7. Spectra
Figure 7. Spectra of
of Unimodular
Unimodular Pure
Pure Noise,
Noise, and
and LFM. With B ==11 MHz,
LFM. With MHz, T =
= 4096 µs.
4096 μs.
Figure 7. Spectra of Unimodular Pure Noise, and LFM. With = 1 MHz, = 4096 μs.
Once compressed
Once compressed by by the
the matched
matched filter,
filter, the
the pure
pure noise
noise and
and thethe LFM
LFM give
give aa range
range resolution
resolution of of
Δ =
∆R =Once
150 m compressed
150 m (1 (1 μs) by the matched filter, the pure noiseμs,
µs) as shown in Figure 8. In the first 5–10 µs,
and
thethe
the
LFM give
secondary
secondary
a range
lobes
lobes of pure
of
resolution
pure noise are
noise
of
are
Δslightly
= 150 m (1 μs) as
fluctuating shown
(the PSLRin Figure 8. In to
thethefirst μs,far
5–10For the secondary lobes of pure noise are
slightly fluctuating (the PSLR isiscomparable
comparable to the LFM).
LFM). For far lobes,
lobes, the PSLR
the PSLR is empirically
is empirically related
related
slightly
to the
the BTfluctuating
BT [24]
[24] by (the
by Equation PSLR
Equation (7):(7):is comparable to the LFM). For far lobes, the PSLR is empirically related
to
to the BT [24] by Equation (7):
= −10 ⋅ ( ) + [dB] , (7)
PSLR = −10 ·⋅ log10 ((BT )) +
= −10 + k [[dB]
dB] ,, (7)
(7)
with of the order of 10–13 dB, which corresponds, for BT = 4096, to −23 dB (see Figure 8). The cross-
with
with of the
k of
correlation the order
order
level ofof10–13
between 10–13
twodB, which
dB, corresponds,
which
independently corresponds,
generatedfor BT
for= 4096,
pure BT
noise= to4096,
−23 dB
signals tois(see Figure
−of23thedBsame 8).level
(see The cross-
Figure
as the8).
correlation
The level between
cross-correlation level two independently
between two generated
independently pure noise
generated signals
pure is
noise
autocorrelation sidelobes, excluding the central zone. The advantage of the pure noise signal with of the same
signals is level
of the as the
same
autocorrelation
level
respectas to
thethe sidelobes,
autocorrelation
HNLFM excluding
signalsidelobes, the central
excluding
is two-fold: a narrowerzone. The advantage
the central
bandwidthzone. getof
to The the pure noise
advantage
the required thesignal
ofrange pure with
noise
resolution
respect to the HNLFM signal
and a cross-correlation level atis the
two-fold: a narrower
same value bandwidth to get
as its autocorrelation the required range resolution
sidelobes.
and a cross-correlation level at the same value as its autocorrelation sidelobes.
Aerospace 2017, 4, 15 9 of 14

Aerospace 2017,respect
signal with 4, 15 9 of 14
to the HNLFM signal is two-fold: a narrower bandwidth to get the required range
resolution and a cross-correlation level at the same value as its autocorrelation sidelobes.
The main reason suggesting the use of noisy signals, beyond the MIMO application, is their
The main reason suggesting the use of noisy signals, beyond the MIMO application, is their
capability to be tailored at will to reach better PSLR values, e.g., −60 dB as discussed above.
capability to be tailored at will to reach better PSLR values, e.g., −60 dB as discussed above.
Normalized Output [dB]

Normalizedautocorrelations
Figure8.8.Normalized
Figure autocorrelationsof
ofUnimodular
Unimodular Pure
Pure Noise
Noise and
and LFM: = 11 MHz,
LFM: B = = 4096 μs.
MHz, T = 4096 µs.

3.3. Range
3.3. Range Sidelobe
Sidelobe Suppression
Suppression Algorithms
Algorithms
Many
Many approaches
approaches have have been been used
used in in the
the past
past years
years to to cope
cope withwith the the range
range sidelobe
sidelobe problem,
problem,
starting
starting fromfrom the the time
time (or (or frequency)
frequency) weighting
weighting of of thethe received
received signalsignal [10,11],
[10,11], to to the
the modern
modern
algorithms
algorithms able ableto togenerate
generatesignalssignalswithwithsuitable
suitableautocorrelation
autocorrelation characteristics
characteristics without
without thetheneedneedof
sidelobe
of sidelobe suppression
suppression in in
reception.
reception. TheThemain
main difference
differencebetween
betweenthese thesetwo twoapproaches
approachesisisthat, that, using
using
the
the generation
generation algorithms,
algorithms, significant
significant complexity
complexity isis demanded
demanded to to the
the generation
generation side side inin terms
terms of of
computational
computational burden burdentotoachieve achieve a “useful”
a “useful” waveform.
waveform. In anyIncase,
any this
case, thiscan
issue issue can beovercome
be trivially trivially
overcome by offline generating
by offline generating a large enough a largesetenough
of noisyset of noisy waveforms
waveforms ready to beready to be transmitted,
transmitted, and storedand in a
stored in a mass
mass memory memory [31].
[31].
Generally,
Generally, aa noisenoise generation
generation algorithm
algorithm startsstarts from
from aa random
random process process realization
realization (e.g.,(e.g., the
the
unimodular, band-limitedone,
unimodular, band-limited one,seesee Figure
Figure 6) and
6) and runsruns in order
in order to minimize
to minimize a certain a objective
certain objective
function
function
with defined with constraints.
defined constraints. In this
In this case, thecase, the objective
objective functionfunction
is the PSLRis theand PSLR theand the constraints
constraints are the
are the limited
limited bandwidth bandwidth
and theand theamplitude
unity unity amplitudeneededneededto fullytoexploit
fully exploit the solid-state
the solid-state amplifier.amplifier.
Often,
Often,
due todue to convergence
convergence considerations,
considerations, the Integrated-Side-Lobe
the Integrated-Side-Lobe (ISL) is(ISL) is minimized
minimized instead instead
of theof the
PSLR
PSLR
because because the former
the former is an integrated
is an integrated valuevalue over
over all theallsidelobe
the sidelobe
region region
whilewhile the latter
the latter is onlyis only
a locala
local
valuevalue thatrapidly
that can can rapidlychange change
pointpoint by point.
by point.
AA well-known
well-known sidelobe
sidelobe suppression
suppression algorithm
algorithm family
family is is the
the Cyclic
Cyclic Algorithm
Algorithm New New (CAN)
(CAN) [25][25]
that
that provides
providesseveral
severalinteresting
interestingways waystotoapproach
approachthe the suppression
suppression problem.
problem. ToTosuit thethe
suit algorithm
algorithm to
particular
to particular needs,
needs, thetheauthors
authors of of
this paper
this paperhave havedeveloped
developeda anew newalgorithm
algorithmto to generate
generate noisy
noisy
waveforms
waveforms having having aa limited
limitedbandwidth
bandwidthand andananunimodular
unimodular amplitude,
amplitude, with
with thethe possibility
possibility to tune
to tune the
the suppressed
suppressed zone zone length
length depending
depending on the
on the particular
particular application.
application. In theIn the following,
following, some some results
results are
are
shownshown andand discussed,
discussed, highlighting
highlighting advantages
advantages andand drawbacks.
drawbacks.
In
In some
some applications,
applications, the the matched
matched filter
filter output
output is is performed
performed by by aa periodic
periodic correlation
correlation between
between
the
the radar
radar return
return andand the
the reference,
reference, which
which can can significantly
significantly differ
differ from
from thethe aperiodic
aperiodic case. case. The
The periodic
periodic
approach
approachcan canbe beconsidered
consideredininthe thecase
caseof of
FFT FFTimplementation,
implementation, with it being
with it beinga spectrum
a spectrum product, i.e.,
product,
ai.e.,
circular correlation.
a circular correlation.In thisIn paper, onlyonly
this paper, the aperiodic
the aperiodic correlation is discussed.
correlation is discussed.
CAN algorithms [25]
CAN algorithms [25]havehavebeenbeendeveloped
developed both
both forfor aperiodic
aperiodic andand periodic
periodic cases;cases; in general,
in general, there
there
are noare nobasic
large largedifferences
basic differences
betweenbetween
them. The them.
mainThe ideamain idea is tothe
is to minimize minimize
difference thebetween
differencethe
between
obtainedthe andobtained
the desired and autocorrelation
the desired autocorrelation
functions throughfunctions through
a process thata process that runsuntil
runs cyclically cyclically
a stop
until a stop criterion is satisfied, e.g., the difference between two consecutive steps is less than a given
threshold. The constraints to be satisfied within this minimization lead to different algorithms
Aerospace 2017, 4, 15 10 of 14

criterion is satisfied, e.g., the difference between two consecutive steps is less than a given threshold.
The constraints to be satisfied within this minimization lead to different algorithms belonging to the
CAN family. These constraints can be the unit amplitude, the number of suppressed sidelobes as well
as the mainlobe width (i.e., the needed bandwidth) or others.
Especially
Aerospacein MIMO
2017, 4, 15 radar configuration, a number of M orthogonal waveforms 10 are required to
of 14
th
discriminate, after the matched filter, the m signal among the others. For this purpose, the CAN
belonging to the CAN family. These constraints can be the unit amplitude, the number of suppressed
family also provides
sidelobes as wella “MIMO” version
as the mainlobe widthfor
(i.e.,the algorithms
the needed in which
bandwidth) the quantity to be minimized
or others.
is the difference between
Especially the obtained
in MIMO and theadesired
radar configuration, number ofcovariance
M orthogonal matrix [25],are
waveforms until a stop
required to criterion
is reached.discriminate, after the matched filter, the mth signal among the others. For this purpose, the CAN
family also provides a “MIMO” version for the algorithms in which the quantity to be minimized is
The main drawback of the CAN algorithms is their inability to manage the bandwidth increase,
the difference between the obtained and the desired covariance matrix [25], until a stop criterion is
i.e., the mainlobe
reached. of the signal generated by using the CAN algorithm is very narrow. In fact, these
algorithms prefer a deep
The main drawbacksidelobe suppression
of the CAN algorithms is attheir
theinability
expense of having
to manage a full-Nyquist
the bandwidth increase, occupied
bandwidth, i.e.,which
the mainlobe of the signal
unfortunately generated
is not by using
suitable the CAN algorithm
for applications is very
in which narrow. Inregulations
spectrum fact, these must be
algorithms prefer a deep sidelobe suppression at the expense of having a full-Nyquist occupied
met. Only one CAN algorithm (Stopband CAN (SCAN) [25]) is able to manage the spectrum constraint.
bandwidth, which unfortunately is not suitable for applications in which spectrum regulations must
In this study,
be met.theOnly
SCAN one has
CANbeen applied
algorithm to generate
(Stopband noise[25])
CAN (SCAN) unimodular signalsthe
is able to manage with BT = 4096 and
spectrum
B = 1 MHz. The related
constraint. aperiodic
In this study, the SCANautocorrelation
has been appliedistoshown
generate in Figure
noise 9a. Keeping
unimodular signals withBT BTunchanged,
=
the SCAN4096 and B = improves
algorithm 1 MHz. Thethe related
PSLR aperiodic
by about autocorrelation
20 dB with is respect
shown into Figure 9a. Keeping BTnoise from
the unimodular
unchanged, the SCAN algorithm improves the PSLR by about 20 dB with respect to the unimodular
which the algorithm starts. The mainlobe is kept wide since the SCAN spectrum, shown in Figure 10,
noise from which the algorithm starts. The mainlobe is kept wide since the SCAN spectrum, shown
is well-shaped
in Figurewithin the required
10, is well-shaped 1 MHz
within bandwidth.
the required 1 MHz bandwidth.

Figure 9. Compressed pulse comparison between Unimodular Noise: (a) Stopband Cyclic Algorithm
Figure 9. Compressed pulse comparison between Unimodular Noise: (a) Stopband Cyclic Algorithm
New (SCAN); (b) Band Limited Algorithm for Sidelobes Attenuation (BLASA) Single Input, Single
New (SCAN); (b) Band Limited Algorithm for Sidelobes Attenuation (BLASA) Single Input, Single
Output (SISO): B = 1 MHz, T = 4096 μs.
Output (SISO): B = 1 MHz, T = 4096 µs.
The drawback of the SCAN algorithm is that the sidelobes close to the mainlobe are still quite
high. To overcome the issue, the authors have developed the BLASA (Band Limited Algorithm for
The drawback of the SCAN
Sidelobes Attenuation) algorithm
algorithm whoseisresult
thatisthe sidelobes
shown in Figureclose
9b. Ittocomes
the mainlobe are CAN
from the same still quite high.
To overcome
ideathe
andissue, the
provides authors
very have developed
low sidelobes theclose
even in the area BLASA (Band Limited
to the mainlobe. Algorithm
It is called for Sidelobes
“SISO” (Single
Input,
Attenuation) Single Output)
algorithm because
whose a single
result waveform
is shown inisFigure
generated at each
9b. run. The
It comes BLASA
from theSISO spectrum
same CAN idea and
is shown in Figure 10 and it is still well shaped within the allowed 1 MHz bandwidth, providing a
provides very low sidelobes even in the area close to the mainlobe. It is called “SISO” (Single Input,
range resolution of 150 m.
Single Output) because a single waveform is generated at each run. The BLASA SISO spectrum is
shown in Figure 10 and it is still well shaped within the allowed 1 MHz bandwidth, providing a range
resolution of 150 m.
Aerospace 2017, 4, 15 11 of 14

Being a “SISO” algorithm, BLASA does not lower the cross-correlation level between two
independently generated waveforms. Hence, their cross-correlation is at the same level as the initial
unimodular pure2017,
Aerospace noise.
4, 15 11 of 14
As in Aerospace
the CAN 2017, 4, 15
family, even for the BLASA SISO, a “MIMO” version exists which is11able of 14
to jointly
Being a “SISO” algorithm, BLASA does not lower the cross-correlation level between two
generate a independently
number M
Being a “SISO” of waveforms
generatedalgorithm, at each
BLASA
waveforms.
run, i.e.,
doestheir
Hence,
M signals
not cross-correlationbelonging
lower the cross-correlation
is at the same
to the
level same
levelbetween
set.two BLASA
as the initial
The
MIMO hasindependently
been developed
unimodular pure generatedto manage
noise. waveforms. the suppression
Hence, of both auto
their cross-correlation is at and
the samecross-correlation
level as the initialfunctions,
unimodular
still keeping theAsbandwidth
in the pure
CAN noise.limited.
family, even for the BLASA SISO, a “MIMO” version exists which is able to jointly
As inathe
generate numberCAN M family, even for the
of waveforms BLASA
at each run, SISO,
i.e., Ma “MIMO”
signals version exists
belonging to thewhich
same is able
set. The toBLASA
jointly
Due to the limited
generate a number
number of samples
M of waveforms at each
that BLASA
run, i.e., M signals
MIMO can manipulate, the joint suppression
MIMO has been developed to manage the suppression of bothbelonging
auto and to the same set. The
cross-correlation BLASA
functions,
region in cross
MIMO and auto
has been
still keeping
correlations
developed to
the bandwidth manage the suppression of both auto and cross-correlation functions, length of
cannot
limited.
be as long as the whole length of the pulse. The
the suppressed Due to the limited number ofnumber
zone
still keeping depends
the bandwidth on the
limited.samples that M of the waveforms
BLASA in the generated
MIMO can manipulate, set: the lower M,
the joint suppression
the longer region Due to
the suppressed the limited
in cross and auto number
length. of samples
Moreover,
correlations that
cannotthe BLASA
be assuppressed MIMO can
long as the whole manipulate,
zonelength
length the
of depends joint suppression
the pulse. The on length
the amplitude
region in cross andzone
ofittheincreases
suppressed autodepends
correlations
on cannot
the number be as long
Mis as waveforms
ofallowed
the the whole length
in the of the pulse.
generated The
set:if the length
lower
constraint:of if amplitude modulation while it decreases the unimodular
M,the thesuppressed
longer thezone depends on
suppressed the number
length. Moreover, M ofthethe suppressed
waveforms in the generated
zone length depends set: theon lower
the
constraint M,
is applied.
the longer
amplitude
Hereafter,
the suppressed
constraint:
only
it increases
the
length.unimodular
Moreover,
if amplitude
case will
the suppressed
modulation
be considered.
zone while
is allowed lengthit depends
decreasesonif the the
Figureamplitude
11 shows
unimodular the BLASA
constraint:
constraint it MIMO
increases
is applied. ifcompressed
amplitude
Hereafter, thepulse
only modulation with
unimodular M
case=will
is allowed 2 and
while the unimodular
it decreases
be considered. if theamplitude
constraint.unimodular
It isFigure
generated constraint
11 showsfor isBT
the applied.
BLASA= 256 Hereafter,
MIMOwhich only
is
compressed thethemaximum
unimodular
pulse case
with Mvalue
= 2 andwillatbewhich
the considered.
unimodular the current
amplitudealgorithm
Figure
constraint. 11
It shows
is the
generated BLASA
for BT MIMO
= 256 compressed
which is the pulse
maximum with M =
value2 and
at the
which unimodular
the current amplitude
algorithm
version canconstraint.
work. Theis occupied bandwidth is always B = 1 MHz but, as it thehappens algorithm with the HNLFM,
version canItwork. generated for BTbandwidth
The occupied = 256 which is is the maximum
always B = 1 MHzvalue at it
but, as which
happens current
with the HNLFM,
the −3 dB version
bandwidth
the −3 dB can is a quarter
work. Theisoccupied
bandwidth
of the
a quarterbandwidth
total occupied
of the totalisoccupied
bandwidth.
always Bbandwidth.
= 1 MHz but, as it happens with the HNLFM,
the −3 dB bandwidth is a quarter of the total occupied bandwidth.

Figure 10. Spectrum comparison among Unimodular Noise, SCAN and BLASA SISO: B = 1 MHz, T =
Spectrum
Figure 10. Figure comparison
10.MSpectrum
4096 μs, = 2.
among
comparison amongUnimodular Noise,
Unimodular Noise, SCAN
SCAN and SISO:
and BLASA BLASA MHz, TB= = 1 MHz,
B = 1SISO:
T = 4096 µs, Mμs,
4096 = 2.M = 2.

Figure 11. Autocorrelation of BLASA Multiple-Input Multiple-Output (MIMO): B = 1 MHz, T = 256

Figure 11.Figure
M =11.
2. Autocorrelation
μs, Autocorrelation ofofBLASA
BLASA Multiple-Input
Multiple-InputMultiple-Output (MIMO): B =(MIMO):
Multiple-Output 1 MHz, T =B256= 1 MHz,
μs, M = 2.
T = 256 µs, M = 2.

The limited length of the suppressed zone represents a valid tool to mitigate the clutter effect,
especially in the proximity of the mainlobe, i.e., the target’s closest range cells. Because of the narrow
Aerospace 2017, 4, 15 12 of 14

Aerospace 2017, 4, 15
−3 dB bandwidth (see Figure 12), the mainlobe is four times wider than the expected12150 of 14
m, which
corresponds toThe a 1limited
MHz length
bandwidth.
of the suppressed zone represents a valid tool to mitigate the clutter effect,
Nevertheless,
especially inthis behaviour
the proximity is mainlobe,
of the deterministic and can
i.e., the target’s be range
closest overcome by properly
cells. Because choosing the
of the narrow
−3 dB bandwidth
signals’ occupied bandwidth.(see Figure
The12),
bigthe mainlobe isof
advantage four
thetimes wider MIMO
BLASA than the pseudorandom
expected 150 m, whichsignals with
corresponds to a 1 MHz bandwidth.
respect to the deterministic HNLFM is the possibility to coherently average the range sidelobes in
Nevertheless, this behaviour is deterministic and can be overcome by properly choosing the
azimuth. In fact, occupied
signals’ if the transmitted waveform
bandwidth. The changes
big advantage of theeach
BLASAWRTMIMOwithin the dwellsignals
pseudorandom time,with
the averaged
compressed pulse
respect to presents a sidelobe
the deterministic HNLFM level reduced
is the possibilitybytothe quantity
coherently [24]: the range sidelobes in
average
azimuth. In fact, if the transmitted waveform changes each WRT within the dwell time, the averaged
compressed pulse presents a sidelobe∆SLlevel =
reduced by the
10 · log ( Lquantity
), [24]: (8)
10
Δ = 10 ⋅ log ( ) , (8)
where ∆SLwhere
is the lowering
ΔSL in the
is the lowering in sidelobe level
the sidelobe level with respect
with respect to to
the the not-averaged
not-averaged case andcase and L is the
L is the
number ofnumber
coherently integrated
of coherently returns.
integrated returns.
|X(f)| [dB]

Figure
Figure 12. 12. Spectrum
Spectrum of BLASAMIMO:
of BLASA MIMO: BB==1 MHz, T = 256
1 MHz, T =μs,
256M µs,
= 2. M = 2.

Summing up all the considered waveforms, Table 1 shows their comparison. Each algorithm is
checked
Summing uptoall
verify
the whether it allows
considered a MIMO version,
waveforms, Table the level oftheir
1 shows the sidelobes, the frequency
comparison. Each algorithm
occupancy and its capability to be coherently integrated.
is checked to verify whether it allows a MIMO version, the level of the sidelobes, the frequency
occupancy and its capability
Tableto be coherently
1. Comparison integrated.
of algorithms (BTOT denotes the swept frequency).

MIMO Amplitude
Algorithm Sidelobe Level and Suppression Interval
Table 1. Comparison
Version of algorithms (BTOT denotes the swept frequency).
Modulation
Pseudo trapezoidal
HNLFM No −67 dB at BT = 4096 400%
AM
MIMO
Pure Unimodular BTOT Amplitude
Algorithm No Sidelobe Level and
−23 dB at BTSuppression
= 4096 Interval
100% Unimodular
B−3 dB
Noise Version Modulation
SCAN No −45 dB at BT = 4096 (within 15%) 100% Unimodular
Unimodular Pseudo
HNLFM BLASA SISO No No
−−50
67dBdBat at
BT = 4096
BT = 4096 100%
400%
−30 dB at BT = 256 (within 12%) if trapezoidal AM
BLASA MIMO Yes 400% Unimodular
unimodular amplitude with M = 2
Pure Unimodular
No −23 dB at BT = 4096 100% Unimodular
Noise The column “MIMO version” refers to the capability of jointly generated M waveforms with a
cross-correlationNo
SCAN level equal to−the auto-correlation
45 dB level. The
at BT = 4096 (within frequency occupancy
15%) 100% givesUnimodular
the
information of how much the mainlobe is enlarged by the algorithm.
BLASA SISOThe SCAN No −50zone
sidelobe level refers to the dB at BT =15%
within 4096
of the whole length, 100% Unimodular
as shown in Figure
9, while the −30 dB of BLASA − MIMO
30 dBisatrelated to the
BT = 256 suppressed
(within 12%) area
if which is about 12% of the
BLASA MIMO Yes two signals (M = 2) are jointly generated with a unimodular amplitude.
whole length when 400% Unimodular
unimodular amplitude with M = 2

The column “MIMO version” refers to the capability of jointly generated M waveforms with
a cross-correlation level equal to the auto-correlation level. The frequency occupancy gives the
information of how much the mainlobe is enlarged by the algorithm.
The SCAN sidelobe level refers to the zone within 15% of the whole length, as shown in Figure 9,
while the −30 dB of BLASA MIMO is related to the suppressed area which is about 12% of the whole
length when two signals (M = 2) are jointly generated with a unimodular amplitude.
Aerospace 2017, 4, 15 13 of 14

4. Conclusions
The design of the transmitted signal is an increasingly important element in the development of
modern radars. The availability of stable and accurate waveform generators and of real-time, high
dynamic range digital correlators makes it possible to use, as radar signals, arbitrary waveforms
with a number of degrees of freedom, i.e., independent samples, of the order of a thousand and
even more, permitting pulse compression with a processing gain of the same order of magnitude.
In this paper, two classes of novel waveforms for pulse-compression (and continuous-wave) radars
have been discussed and evaluated: (i) a (deterministic) type of frequency modulated pulse with a
“trapezoidal-like” amplitude modulation and; (ii) a subset of noisy waveforms in the Noise Radar
Technology context. Both are designed in order to make the range sidelobes as low as possible, which
are the main issue in the matched filter approach, both for pulsed and continuous-wave radars. A set
of representative results has been shown in terms of sidelobe level, width of the main lobe and spectral
occupancy. The selection of the particular waveform depends, of course, on the application.

Author Contributions: Gaspare Galati proposed the basic ideas, gave the theoretical support and
completed/checked the paper. Gabriele Pavan contributed to the HNLFM analysis and gave the Matlab software
support together with Francesco De Palo who also contributed to the Noise Radar and signals generation.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Pace, P.E. Detecting and Classifying Low Probability of Intercept Radar; Artech House: Norwood, MA, USA, 2004.
2. Govoni, M.A. Low Probability of Interception of an Advanced Noise Radar Waveform with Linear-FM.
IEEE Trans. AES 2013, 1351–1356. [CrossRef]
3. Galati, G.; Pavan, G.; De Palo, F. Noise Radar Technology: Pseudorandom Waveforms and their Information
Rate. In Proceedings of the International Radar Symposium, Gdansk, Poland, 16–18 June 2014; pp. 123–128.
4. Harman, S. The Performance of a Novel Three-Pulse Radar Waveform for Marine Radar System.
In Proceedings of the 5th European Radar Conference, Amsterdam, The Netherlands, 30–31 October 2008;
pp. 160–163.
5. Blunt, S.D.; Gerlach, K. Adaptive Pulse Compression via MMSE estimation. IEEE Trans. AES 2006, 42,
572–584. [CrossRef]
6. De Witte, E.; Griffiths, H.D. Improved ultra-low range sidelobe pulse compression waveform design.
Electron. Lett. 2004, 40, 1448–1450. [CrossRef]
7. Sarkar, T.K. An Ultra-low Sidelobe Pulse Compression Technique for High Performance Radar Systems.
In Proceedings of the IEEE National Radar Conference, Syracuse, NY, USA, 13–15 May 1997; pp. 111–114.
8. Lohmeier, S.P. Adaptive FIR filtering of Range Sidelobes for Air and Spaceborne Rain Mapping.
In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, Toronto, ON, Canada,
24–28 June 2002; Volume 3, pp. 1801–1803.
9. Ackroid, M.H.; Ghani, F. Optimum mismatched filters for sidelobe suppression. IEEE Trans. AES 1973, 9,
214–218. [CrossRef]
10. Cook, C.E.; Bernfeld, M. Radar Signals—An Introduction to Theory and Application; Artech House: Norwood,
MA, USA, 1993.
11. Richards, M.A. Fundamentals of Radar Signal Processing; Section 4.6; McGraw Hill: New York, NY, USA, 2005.
12. Levanon, N.; Mozeson, E. Radar Signals; John Wiley & Sons: Hoboken, NJ, USA, 2004.
13. Millett, R.E. A Matched-Filter Pulse-Compression System Using a Nonlinear FM Waveform. IEEE Trans. AES
1970, AES-6, 73–77.
14. Doerry, A.W. Generating Nonlinear FM Chirp Waveforms for Radar; Sandia Report SAND2006-5856;
Sandia National Laboratories: Albuquerque, NM, USA; Livermore, CA, USA, 2006.
15. Doerry, A.W. SAR Processing with Non-Linear FM Chirp Waveforms; Sandia Report SAND2006-7729;
Sandia National Laboratories: Albuquerque, NM, USA; Livermore, CA, USA, 2006.
16. Klauder, J.R.; Price, A.C.; Darlington, S.; Albefsheim, W.J. The theory and design of chirp radars.
Bell Syst. Tech. 1960, 39, 745–808. [CrossRef]
Aerospace 2017, 4, 15 14 of 14

17. Yanovsky, F. Radar development in Ukraine. In Proceedings of the International Radar Symposium, Gdansk,
Poland, 16–18 June 2014.
18. 2009 Pioneer Award. IEEE Trans. Aerosp. Electron. Syst. 2010, 46, 2139–2141.
19. Galati, G.; van Genderen, P. Special Section on Some Less-Well Known Contribution to the Development of Radar:
From its Early Conception Until Just after the Second World War; URSI Radio Science Bulletin, No 358; URSI:
Ghent, Belgium, September 2016; pp. 12–108.
20. Papoulis, A. Signal Analysis; McGraw Hill: New York, NY, USA, 1977.
21. Collins, T.; Atkins, P. Nonlinear frequency modulation chirps for active sonar. Proc. IEE Radar Sonar Navig.
1999, 146, 312–316. [CrossRef]
22. Zhiqiang, G.; Peikang, H.; Weining, L. Matched NLFM Pulse Compression Method with Ultra-low Sidelobes.
In Proceedings of the 5th European Radar Conference, Amsterdam, The Netherlands, 30–31 October 2008;
pp. 92–95.
23. Russell, M.; Marinilli, A.; Sullivan, F. Millimeter wave, solid-state amplifiers in radar and communication
systems. In Proceedings of the IEEE MTT-S International Microwave Symposium Digest, San Francisco, CA,
USA, 17–21 June 1996; pp. 53–56.
24. Galati, G.; Pavan, G.; De Palo, F. Generation of Pseudo-Random Sequences for Noise Radar Applications.
In Proceedings of the International Radar Symposium, Gdansk, Poland, 16–18 June 2014; pp. 115–118.
25. He, H.; Li, J.; Stoica, P. Waveform Design for Active Sensing Systems—A Computational Approach; Cambridge
University Press: Cambridge, UK, 2012.
26. Galati, G.; Pavan, G. Mutual interference problems related to the evolution of Marine Radars. In Proceedings
of the 18th IEEE International Conference on Intelligent Transportation Systems, Las Palmas, Spain,
15–18 September 2015; pp. 1785–1790.
27. Galati, G.; Pavan, G.; De Palo, F. Compatibility problems related with pulse compression, solid-state marine
radars. IET Radar Sonar Navig. 2016, 10, 791–797. [CrossRef]
28. Baroncelli, A. NAVICO Rbu, Montespertoli, Florence, Italy. Personal communication, 1 February 2017.
29. Galati, G.; Pavan, G.; De Franco, A. Orthogonal Waveforms for Multistatic and Multifunction Radar.
In Proceedings of the 9th European Radar Conference, Amsterdam, The Netherlands, 31 October–2
November 2012; pp. 310–313.
30. Stove, A.; Galati, G.; Pavan, G.; De Palo, F.; Lukin, K.; Kulpa, K.; Kulpa, J.S.; Maślikowski, Ł. The NATO
SET-184 noise radar trials 2016. In Proceedings of the 17th International Radar Symposium, Krakow, Poland,
10–12 May 2016; pp. 1–6.
31. Galati, G. Coherent Radar. Patent PCT/IC2014/061454, 15 May 2014.
32. Galati, G.; Pavan, G.; De Palo, F.; Stove, A. Potential applications of noise radar technology and related
waveform diversity. In Proceedings of the 17th International Radar Symposium, Krakow, Poland,
10–12 May 2016; pp. 1–5.
33. Stove, A.; Galati, G.; De Palo, F.; Wasserzier, C.; Erdogan, Y.A.; Kubilay, S.; Lukin, K. Design of a Noise Radar
Demonstrator. In Proceedings of the 17th International Radar Symposium, Krakow, Poland, 10–12 May 2016;
pp. 1–6.
34. Galati, G.; Pavan, G.; De Palo, F. On the potential of noise radar technology. In Proceedings of the Enhanced
Surveillance of Aircraft and Vehicles (ESAV 2014), Rome, Italy, 15–16 September 2014; pp. 66–71.
35. De Palo, F.; Galati, G. Orthogonal Waveforms for Multiradar and MIMO Radar Using Noise Radar Technology.
In Proceedings of the Signal Processing Symposium, Debe, Poland, 10–12 June 2015; pp. 1–4.
36. Marsaglia, G.; Tsang, W.W. The ziggurat method for generating random variables. J. Stat. Softw. 2000.
[CrossRef]
37. Nuttall, H.A. Some windows with very good sidelobe behavior. IEEE Trans. Acoust. Speech Signal Process.
1981, 29, 84–91. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).