applied

sciences
Article
A Study on the Amplitude Comparison
Monopulse Algorithm
Minjeong Kim, Daseon Hong and Sungsu Park *
Department of Aerospace Engineering, Sejong University, Seoul 05006, Korea; [email protected] (M.K.);
[email protected] (D.H.)
* Correspondence: [email protected]; Tel.: +82-2-3408-3769




Received: 28 April 2020; Accepted: 4 June 2020; Published: 7 June 2020


Abstract: This paper presents two amplitude comparison monopulse algorithms and their covariance
prediction equation. The proposed algorithms are based on the iterated least-squares estimation
method and include the conventional monopulse algorithm as a special case. The proposed
covariance equation is simple but predicts RMS errors very accurately. This equation quantitatively
states estimation accuracy in terms of major parameters of amplitude comparison monopulse radar,
and is also applicable to the conventional monopulse algorithm. The proposed algorithms and
covariance prediction equations are validated by the numerical simulations with 100,000 Monte
Carlo runs.

Keywords: monopulse radar; amplitude comparison; covariance prediction; least-squares estimation;

performance analysis

1. Introduction
Monopulse radar operates by comparing signal returns of four squinted beams steered around the
expected target direction [1]. It is classified as amplitude comparison monopulse and phase comparison
monopulse, and amplitude comparison monopulse is the most commonly used method, which detects
azimuth and elevation deviation errors from the center of the track axis by comparing the amplitudes
of the signal returns.
Many methods have been proposed to improve the performance of monopulse algorithms in
the literature [2–9]. Hofstetter and DeLong [2] dealt with the problem of detecting and estimating
the unknown angular position of the target observed simultaneously by controlling the number
of antennas in the amplitude comparison monopulse radar, where the method is derived using a
generalized likelihood ratio test. Sim et al. [3] carried out a performance analysis according to antenna
spacing using the phase comparison monopulse algorithm and proposed an efficient nonlinear antenna
array based on the analysis results. Han et al. [4] showed the effect of the functional variables of
the monopulse radar on angular tracking performance. An and Lee [5] showed the performance of
amplitude comparison monopulse through simulation-based MSE (mean-square error) comparison
with analytic MSE obtained by linearly approximating the nonlinear estimation equations. Jacovitti [6]
carried out a performance analysis on the monopulse receivers in the secondary surveillance radar
where the internal sources of error are considered. This paper discussed the general performance of
the receiver of the monopulse radar where Gaussian noise is applied in the environment. Chen and
Sheng [7] analyzed the performance of an angle measurement of the monopulse radar with reception
non-consistency. This paper showed that the performance of the angle measurement depends not only
on the algorithm but also on the reception consistency. Mosca [8] discussed the estimation problem of
the angle of arrival in thermal noise-applied amplitude comparison monopulse radar in terms of the

Appl. Sci. 2020, 10, 3966; doi:10.3390/app10113966 www.mdpi.com/journal/applsci

Appl. Sci. 2020, 10, 3966 2 of 13
Appl. Sci. 2020, 10, x FOR PEER REVIEW 2 of 13

comparison
coherency ofmonopulse radar
pulse returns. Inin terms
[9], of the
a novel coherency ofthat
methodology pulse returns. In
overcomes the[9], a novel methodology
limitation of difference
that overcomes the limitation
beam aided detection was proposed. of difference beam aided detection was proposed.
While many
While many monopulse
monopulse algorithms
algorithms areare proposed
proposed so so far
far to
to improve
improve the the estimation
estimation performance,
performance,
most of
most of them
them dodo not
not have
have clear
clear relationships
relationships with
with the
the conventional
conventional algorithm,
algorithm, andand only
only few
few conduct
conduct
the performance analysis under the presence of measurement noise. Therefore,
the performance analysis under the presence of measurement noise. Therefore, it is difficult to it is difficult to
evaluate their
evaluate their performance
performance improvement
improvement over over the
the conventional
conventional algorithm
algorithm under
under the
the measurement
measurement
noise presence. Although some papers deal with formal noise analysis on the conventional algorithm,
noise presence. Although some papers deal with formal noise analysis on the conventional algorithm,
it is extremely complex, and no useful information comes
it is extremely complex, and no useful information comes from it. from it.
Motivated by these observations, this paper presents least-squares-based monopulse algorithms
Motivated by these observations, this paper presents least-squares-based monopulse algorithms
and a novel covariance prediction equation. The proposed algorithms have an iterative process that
and a novel covariance prediction equation. The proposed algorithms have an iterative process
incrementally improves the accuracy of the conventional algorithm. It is shown that the conventional
that incrementally improves the accuracy of the conventional algorithm. It is shown that the
monopulse algorithm is a special case of the proposed algorithms with the first iterative step. The
conventional monopulse algorithm is a special case of the proposed algorithms with the first iterative
proposed covariance equation is simple but predicts very accurately the covariance of the estimated
step. The proposed covariance equation is simple but predicts very accurately the covariance of the
target location under the presence of measurement noise. This equation is independent of the
estimated target location under the presence of measurement noise. This equation is independent of
measurement signal, and thus can be used to predict the performance of the monopulse algorithms
the measurement signal, and thus can be used to predict the performance of the monopulse algorithms
before the algorithms are actually performed. This equation is also applicable to conventional
before the algorithms are actually performed. This equation is also applicable to conventional
monopulse algorithms, and quantitatively states estimation accuracy in terms of the major
monopulse algorithms, and quantitatively states estimation accuracy in terms of the major parameters
parameters of amplitude comparison monopulse radar.
of amplitude comparison monopulse radar.
It is also important in in electronic warfare systems to predict the performance of amplitude
It is also important in in electronic warfare systems to predict the performance of amplitude
comparison monopulse in terms of major parameters since intercepted signals from different types
comparison monopulse in terms of major parameters since intercepted signals from different types of
of radars are very often noisy and change depending on the mission being carried out and the task
radars are very often noisy and change depending on the mission being carried out and the task being
being performed [10,11].
performed [10,11].
This paper is organized as follows. In Section 2, the conventional amplitude comparison
This paper is organized as follows. In Section 2, the conventional amplitude comparison
monopulse algorithm is explained. In Section 3, two least-squares-based monopulse algorithms are
monopulse algorithm is explained. In Section 3, two least-squares-based monopulse algorithms are
proposed. Section 4 discusses the relationships between the proposed and the conventional
proposed. Section 4 discusses the relationships between the proposed and the conventional algorithms.
algorithms. In Section 5, a novel covariance equation is proposed as a performance prediction
In Section 5, a novel covariance equation is proposed as a performance prediction measure. In Section 6,
measure. In Section 6, numerical simulations are performed to evaluate the proposed algorithm and
numerical simulations are performed to evaluate the proposed algorithm and covariance equation.
covariance equation. Conclusions are given in Section 6.
Conclusions are given in Section 6.

2. Conventional
2. Conventional Amplitude
Amplitude Comparison
Comparison Monopulse
Monopulse Algorithm
Algorithm
Measurement of
Measurement of the
the target
target direction
direction angle
angle or
or target
target location
location is
is required
required for
for target
target tracking.
tracking.
Amplitude comparison monopulse radar operates by comparing the amplitude of signal
Amplitude comparison monopulse radar operates by comparing the amplitude of signal returns of returns of
four squinted
four squinted sub-beams
sub-beams steered
steered symmetrically
symmetrically around
around the
the expected
expected target
target direction
direction as
as shown
shown inin
Figure 1 [1]. The sum, horizontal difference, and vertical difference channels of the incoming signal
Figure 1 [1]. The sum, horizontal difference, and vertical difference channels of the incoming signal are
are used
used to estimate
to estimate target location,
target location, which iswhich is represented
represented by the
by the azimuth andazimuth and
elevation elevation
angle angle
deviations in
deviations in
the track axis. the track axis.

Figure 1.
Figure 1. Beam
Beam pattern
pattern of
of aa monopulse
monopulse radar.
radar.

Each antenna pattern can be modeled near the center of the track axis as a squinted Gaussian
beam. Each one-way voltage gain function is given by
𝜃
𝑔(𝜃) = 𝑔 exp −2 log 2 (1)
𝜃
Appl. Sci. 2020, 10, 3966 3 of 13

Each antenna pattern can be modeled near the center of the track axis as a squinted Gaussian
beam. Each one-way voltage gain function is given by
 !2 
θ
g θ = g0 exp−2
 
( ) log 2 (1)
Appl. Sci. 2020, 10, x FOR PEER REVIEW θ3dB 3 of 13

where θ𝜃 isisthe
thetotal
totalangle
anglefrom
fromthe
thetarget
targetlocation
locationto
tothe
thebeam peak,θ𝜃3dB isisthe
beampeak, thebeam
beamwidth, andg0𝑔is
width,and
thethe
is gain at at
gain thethe
beam
beampeak [12].
peak [12].
angle, θ,
The total angle, 𝜃, can be computed from the target location defined in Figure 2, which presents
the field of view as seen from the monopulse radar.radar.

Figure 2. Target location in the track axis.

Figure 2. Target location in the track axis.
For a beam with its maximum in quadrant 1, the target location from the beam peak can be written
For a beam with its maximum in quadrant 1, the target location from the beam peak can be
in terms of azimuth and elevation angle deviations (θaz , θel ) as
written in terms of azimuth and elevation angle deviations (𝜃 , 𝜃 ) as
!2 !2
2 𝜃 θs 𝜃 θs
𝜃 θ= = − √ 𝜃 − θaz+ + − √ 𝜃 − θel (2)(2)
√2 2 √2 2
and
and the
the quadrant
quadrant 11 voltage
voltage gain
gain is
is then
then
 𝜃  2 𝜃 
⎡   −√
θ𝜃 + −
θs 𝜃
2  ⎤

⎢ ⎛ √2 2 − θaz + √2 2 − θel⎞ log
s
 2 ⎥
√
 
𝑔 (𝜃)g=(θ
𝑔) (𝜃 , 𝜃 ) = 𝑔 exp −2

= g1 (θaz , θel ) = g0⎢ exp⎜−2 ⎟  log 2⎥ (3)(3)
𝜃
 
1   
⎢   θ 2
3dB

 ⎥
⎣ ⎝ ⎠ ⎦
 

where
where θ 𝜃s isis the
the squint
squint angle
angle of of the
the antenna.
antenna. Using Using the the above
above gain gain functions
functions and and aa received
received signal
signal
voltage of amplitude A, the so-called pseudorange
voltage of amplitude A, the so-called pseudorange equations are given as follows:equations are given as follows:

𝑠  𝑔 (𝜃 , 𝜃 )

s1 𝑠  ⎡ g1 (θaz , θ⎤el ) 𝑛  n1 
= 𝐴 ⎢  (𝜃 , 𝜃 )⎥ + 𝑛 
𝑔
 
𝑠 ( θ θ ) 𝑛 (4)
 
 s 2  = A⎢ 
 𝑔 (𝜃g 2 , 𝜃az , )⎥ el  
 +  n2 
 𝑠 𝑛  (4)
 g3 (θ , 𝜃az ,)θ⎦el )   n3
s3 ⎣𝑔  (𝜃

  
 
  
  
s4 g4 (θaz , θel ) n4
where 𝑠 is the measurement signal of 𝑖-th quadrant, which is contaminated by the measurement
where𝑛si .isHere,
noise 𝑠 can be considered
the measurement signal of i-th as pseudorange
quadrant, which since it is a function
is contaminated byofthe
geometrical
measurementdistance
noise
weighted by the antenna gain.
ni . Here, si can be considered as pseudorange since it is a function of geometrical distance weighted by
The conventional
the antenna gain. amplitude comparison monopulse algorithm for estimating azimuth and
elevation
The angle deviations
conventional is brieflycomparison
amplitude explained inmonopulse the next fewalgorithm paragraphs for[4,5,12].estimating azimuth and
Assuming that 𝜃 + 𝜃 ≈ 0, the quadrant 1 voltage
elevation angle deviations is briefly explained in the next few paragraphs [4,5,12]. gain is simplified as follows:
𝜃 𝜃 +𝜃
𝑔 (𝜃) = 𝑔 (𝜃 , 𝜃 ) ≈ 𝑔(𝜃 ) exp 2√2 log 2 (5)
𝜃 𝜃
If a monopulse error slope coefficient is defined as
𝜃
𝑘 = 2√2 log 2 (6)
𝜃
then, voltage gain for the beam in quadrant 1 should be
Appl. Sci. 2020, 10, 3966 4 of 13

Assuming that θ2az + θ2el ≈ 0, the quadrant 1 voltage gain is simplified as follows:

√ θs θaz + θel
" ! ! #
g1 (θ) = g1 (θaz , θel ) ≈ g(θs ) exp 2 2 log 2 (5)
θ3dB θ3dB

If a monopulse error slope coefficient is defined as

√
θs
!
km = 2 2 log 2 (6)
θ3dB

then, voltage gain for the beam in quadrant 1 should be

θaz + θel
" !# " #
km
g1 (θaz , θel ) = g(θs ) exp km ≈ g0 1 + (θaz + θel ) (7)
θ3dB θ3dB
 θ +θ 
assuming km azθ el  1.
3dB
Under the assumption that all patterns are in phase, the gain functions of the antenna patterns
directed into the other quadrants are approximately computed as
" #
km
g1 (θaz , θel ) ≈ g0 1 + (θaz + θel ) (8)
θ3dB
" #
km
g2 (θaz , θel ) ≈ g0 1 + (−θaz + θel ) (9)
θ3dB
" #
km
g3 (θaz , θel ) ≈ g0 1 − (θaz + θel ) (10)
θ3dB
" #
km
g4 (θaz , θel ) ≈ g0 1 + (θaz − θel ) (11)
θ3dB
Then, the sum (Σ), horizontal difference (∆az ), and vertical difference (∆el ) voltages are calculated by

Σ = A[ g1 + g2 + g3 + g4 ] = 4Ag0 (12)

km
∆az = A[ g1 + g4 − ( g2 + g3 )] = 4Ag0 θaz (13)
θ3dB
km
∆el = A[ g1 + g2 − ( g3 + g4 )] = 4Ag0 θ (14)
θ3dB el
Therefore, the azimuth and elevation angle deviations can be estimated as

∆az θ3dB
θ̂caz = (15)
Σ km

∆el θ3dB
θ̂cel = (16)
Σ km
where θ̂caz and θ̂cel are the estimates of azimuth and elevation angle deviations, respectively,
which represent the unknown target location estimate.

3. Least-Squares-Based Amplitude Comparison Monopulse Algorithm

3.1. Full Measurement-Based Algorithm

We can solve the estimation problems of azimuth and elevation angle deviations by first linearizing
the pseudorange Equation (4), then using the familiar least-squares method.
Appl. Sci. 2020, 10, 3966 5 of 13

Let the vector of measured signals be S = [s1 s2 s3 s4 ]T ; then, Equation (4) can be written in matrix
form compactly as follows:
S = Ah f ull (θaz , θel ) + n (17)

where h f ull = [ g1 g2 g3 g4 ]T , and n = [n1 n2 n3 n4 ]T .

If we expand the pseudorange Equation (17) in Taylor series about some nominal values (θaz0 , θel0 ),
and ignore the second and higher order terms, then this equation becomes

∂h f ull ∂h f ull
S ≈ Ah f ull (θaz0 , θel0 ) + A (θaz − θaz0 ) + A (θel − θel0 ) + n (18)
∂θaz ∂θel

Note that the partial derivatives in the above expression are computed using nominal values
(Appl.
θaz0Sci.
, θel0 2020, 10, x FOR PEER REVIEW
). The residual signal δS is defined as the difference between the actual signal and the signal 5 of 13

computed using nominal values:

𝜕ℎ 𝜕ℎ
≈𝐴 𝛿𝜃 + 𝐴 𝛿𝜃 + 𝑛
𝜕𝜃 𝜕𝜃
∂h f ull ∂h f ull
δS = S − Ah f ull (θaz0 , θel0 )≈ A δθaz + A δθel + n (19)
where 𝛿𝜃 = 𝜃 − 𝜃 and 𝛿𝜃 = 𝜃 − 𝜃 . ∂θaz ∂θel
This equation can be written in four scalar equations in matrix form:
where δθaz = θaz − θaz0 and δθel = θel − θel0 .
This equation can be written 𝜃 𝜃
⎡ in four− 𝜃 scalar
𝑔 equations−in 𝜃 matrix
𝑔 ⎤ form:
√2 √2
⎢  𝜃 
𝜃√
  ⎥ 
𝛿𝑠  ⎢− √ θs
θ θs
θ ⎥  𝑛
4 log 2  ⎢  2 + 𝜃  𝑔− az g 1 −
2 𝜃 el 𝑔 ⎥ 
− g1
𝑛  n1
δs1 𝛿𝑠 √2 √2 𝛿𝜃
  
 ≈ 𝐴

θs

θs + (20)
⎢
𝜃 2  − 𝜃√2 + θaz g2 √ − θel g2 ⎥  𝑛 
   
𝛿𝑠 4 log 𝜃 𝛿𝜃
 
δs δθ
    " #  
− + 𝜃 𝑔 − 2 +𝜃 𝑔 ⎥  az n
2
⎢ 𝑛 2
  
 ≈ A  +  (20)
δs3 𝛿𝑠
   
 
 ⎢ − √2
 
− √2

θ23dB θ θ  δθel n3
 
√s + θaz g3 √s + θel g3 ⎥ 
  
 ⎢ 𝜃 2 𝜃 2
    
δs4  ⎥  n4
   
θ −𝜃 𝑔 − √ θs + 𝜃 𝑔 

√s − θaz g4 θ

⎣
√22 − √22 + el g4 ⎦

which can be written in symbolic form as

which can be written in symbolic form as
𝛿𝑆 ≈ 𝐻 𝛿𝜃 + 𝑛 (21)
δS ≈ H f ull δθ + n (21)
This equation expresses a linear relationship between the residual signals 𝛿𝑆 and the unknown
correction
This equation expresses a 𝛿𝜃.
to angle deviations Note
linear that each row
relationship of 𝐻 theisresidual
between shown signals
to be purely
δS anda gain-weighted
the unknown
direction vector to each of the beam peaks, as observed from the unknown target location in Figure
correction to angle deviations δθ. Note that each row of H f ull is shown to be purely a gain-weighted
3.
direction vector to each of the beam peaks, as observed from the unknown target location in Figure 3.

Figure 3. Gain-weighted direction vectors and the target.

Figure 3. Gain-weighted direction vectors and the target.
Let us consider a solution for the linearized pseudorange equation, denoted as δθ̂ f ull . The estimated
Let eus
residual consider
is defined as athesolution forbetween
difference the linearized pseudorange
the measured equation,
signal and Usingasthe𝛿𝜃linearized
denoted
its estimate. . The
estimated
form of theresidual 𝑒 is defined
pseudorange as the
equations anddifference
ignoringbetween
measurement the measured
noise, thesignal and itsresidual
estimated estimate.is Using
the linearized form of the pseudorange equations and ignoring measurement noise, the estimated
residual is e = δS − H f ull δθ̂ f ull (22)
𝑒 = 𝛿𝑆 − 𝐻 𝛿𝜃 (22)
The least-squares solution can be found by minimizing the following functional:
1
𝐽 𝛿𝜃 = 𝑒 𝑒 (23)
2
Appl. Sci. 2020, 10, 3966 6 of 13

The least-squares solution can be found by minimizing the following functional:

  1
J δθ̂ f ull = eT e (23)
2
Then, the solution is
 −1
δθ̂ f ull = HTfull H f ull HTfull δS (24)

This equation assumes that the inverse to HTfull H f ull exists, which is always true unless voltage
gains are zero because of the beam peak and unknown target location geometries.
Iterated least-squares estimation starts with setting the initial gradient of Equation (24) with
respect to θaz0 = 0 and θel0 = 0 and repeatedly performs the following update until convergence or up
to a certain number of iterations:
θ̂ f ull = θ0 + δθ̂ f ull
(25)
θ0 ← θ̂ f ull

3.2. Sum-Difference Measurement-Based Algorithm

We can solve the estimation problem of azimuth and elevation angle deviations using the sum
(Σ), horizontal difference (∆az ), and vertical difference (∆el ) signals, which are used in the conventional
monopulse algorithm. Utilizing these signals is more useful because the same circuitry as the
conventional one can be used.
The measurement equations can be constructed as follows:

Σ = A[ g1 + g2 + g3 + g4 ] + n1 + n2 + n3 + n4
∆az = A[ g1 + g4 − ( g2 + g3 )] + n1 + n4 − n2 − n3 (26)
∆el = A[ g1 + g2 − ( g3 + g4 )] + n1 + n2 − n3 − n4
h iT
Let the vector of the measured signals be Z = Σ ∆az ∆el ; then, Equation (26) can be
written in matrix form compactly as follows:

Z = Ahsd (θaz , θel ) + Bn (27)

where    
 g1 + g2 + g3 + g4   1 1 1 1 
   
hsd =  g1 + g4 − ( g2 + g3 ) ,
 B =  1 −1 −1 1 
 (28)
 
g1 + g2 − ( g3 + g4 ) 1 1 −1 −1
 

Similar to Section 3.1, by Taylor series expansion about some nominal values (θaz0 , θel0 ),
the linearized measurement equation becomes
 
   n1 
 δΣ   

   n2 
 δ∆az  ≈ B
   

   n3 
δ∆el
   

 
n (29)
     4   
θ θ θ θ
 
 √s − θaz ( g1 + g4 ) − √s + θaz ( g2 + g3 ) √s − θel ( g1 + g2 ) − √s + θel ( g3 + g4 ) 
 2  2  2  2

 δθaz
     
θ θ θ θ

4 log 2 
+ A √s − θaz ( g1 + g4 ) + √s + θaz ( g2 + g3 ) √s − θel ( g1 − g2 ) + √s + θel ( g3 − g4 )
  
θ23dB   2   2   2   2   δθel
 
θ θ θ θ

√s − θaz ( g1 − g4 ) − √s + θaz ( g2 − g3 ) √s − θel ( g1 + g2 ) + √s + θel ( g3 + g4 )
 
2 2 2 2

which can be written in symbolic form as

δZ ≈ Hsd δθ + Bn (30)
Appl. Sci. 2020, 10, 3966 7 of 13

The least-squares solution can be found by minimizing the following functional:

  1 T  
J δθ̂sd = δZ − Hsd δθ̂sd δZ − Hsd δθ̂sd (31)
2
The solution is  −1
δθ̂sd = Hsd
T
Hsd HsdT
δZ (32)

Similar to the previous method, iteration can be used to estimate the unknown target location.

4. Relationships with the Conventional Monopulse Algorithm

If we assume that the unknown target location is near the center of track axis, i.e., θ2az + θ2el ≈ 0,
 −1
then matrix HTfull H f ull of Equation (24) is approximately computed by

 −1 θ43dB
HTfull H f ull ≈ I2 (33)
32(log 2)2 θ2s ( g(θs ))2 A2

where I2 is a 2 × 2 identity matrix. Therefore, with the nominal values of θaz0 = 0 and θel0 = 0, the full
measurement-based algorithm given by Equation (24) becomes approximately

θ23dB
" #
f ull 1 −1 −1 1
δθ̂ ≈ √ δS (34)
8 2 log 2 θs g(θs )A 1 1 −1 −1

which means that the first iterative step in least-squares estimation gives the following target
location estimate:
θ23dB
 f ull 
 θ̂az  ∆az
" #
  ≈ √ (35)
 f ull 
θ̂el 8 2 log 2 θs g(θs )A ∆el
For the case of the sum-difference measurement-based algorithm, a similar analysis is possible.
 −1
With the assumption of θ2az + θ2el ≈ 0, the matrix Hsd TH
sd of Equation (32) is approximately
computed by
 −1 θ43dB
T
Hsd Hsd ≈ I2 (36)
128 (log 2)2 θ2s ( g(θs ))2 A2
With the nominal values of θaz0 = 0 and θel0 = 0, the sum-difference measurement-based
algorithm given by Equation (32) becomes approximately

θ23dB
" #
0 1 0
δθ̂sd ≈ √ δZ (37)
8 2 log 2 θs g(θs )A 0 0 1

which means that the first iterative step in least-squares estimation gives the following target
location estimate:
θ̂az θ23dB ∆az
" sd # " #
≈ √ (38)
θ̂sd
el 8 2 log 2 θs g(θs )A ∆el

Note that this equation is the same as Equation (35) of full measurement-based algorithm.
To compare the least-squares-based estimate algorithm with the conventional one, Equations (6)
and (12) are applied to Equations (15) and (16), and, then,

θ23dB
θ̂caz = √ ∆az
8 2 log 2θs g0 A
θ3dB
2 (39)
θ̂cel = √ ∆
8 2 log 2θs g0 A el
Appl. Sci. 2020, 10, 3966 8 of 13

Equation (39) is also the same as Equations (35) and (38), since g(θs ) ≈ g0 is already assumed
in Equation (7), which means that the conventional monopulse algorithm provides the same target
location estimate as the least-squares-based algorithms with the first iterative step. Eventually, it turns
out that the three monopulse algorithms are nominally the same.
We can expect that performing multistep iterations with least-squares-based algorithm can provide
better target location estimation than the conventional one, since the accuracy of the first-order Taylor
expansion is improved.

5. Performance Analysis
Any noise n, including in the measurement signal, results in estimation error, and this error takes
exactly the same linear form as Equation (24) in the case of the full measurement-based algorithm:
 −1
f ull
δθ̂error = HTfull H f ull HTfull n (40)

If the measurement errors are uncorrelated, with zero mean, and have approximately equal
variance, σ2 , then h i
E[n] = 0, Cn = E nnT = σ2 I4 (41)

where I4 is a 4 × 4 identity matrix. Note that there are no assumptions about the measurement noise in
Equation (40), except that it is uncorrelated and identically distributed.
The expected covariance of the estimates for the least-squares solution takes on a simple form:

f ull

f ull
 f ull T 
Cθ = E δθ̂error δθ̂error
" −1  −1 #
T T T T
= E H f ull H f ull H f ull nn H f ull H f ull H f ull (42)
 −1
= σ2 HTfull H f ull

 −1
This equation explains that the covariance of estimate error is scaled by HTfull H f ull to the
 −1
covariance of measurement noise, and matrix HTfull H f ull can therefore be used to quantify how the
level of measurement noise can be related to the expected level of errors in the target location estimate.
The covariance of the estimated target location can be written in terms of its components:

σ211 σ212
 −1 " #
f ull
Cθ =σ 2
HTfull H f ull =σ 2
(43)
σ212 σ22

If the measurement noise is at the level σ, the error in azimuth angle estimate would be at the
level of σ σ11 . The off-diagonal elements indicate the degree of correlation between estimates.
Note that H f ull depends only on the relative weighted geometry of the beam peak and the
unknown target location. Therefore, the covariance of the estimated target location depends on the
unknown target location itself. If we assume that the unknown target location is near the center of the
 −1
track axis, i.e., θ2az + θ2el ≈ 0, then matrix HTfull H f ull is approximately computed by Equation (33).
Therefore, the covariance of the estimated target location is approximately given by

f ull θ43dB
Cθ ≈ σ 2 I2 (44)
32(log 2)2 θ2s ( g(θs ))2 A2

This covariance equation explains that better accuracy estimates are found where θs is larger,
the gain at the track axis is larger, and the beam width is smaller in the presence of the measurement noise.
Appl. Sci. 2020, 10, 3966 9 of 13

For the case of sum-difference measurement-based algorithms, similar analysis is possible to

predict its performance. With the same assumption for the measurement noise, the expected covariance
of the estimates for the least-square solution takes on a simple form:
  T 
Csd
θ
= E δθ̂sderror δθ̂error
sd
 −1  −1 
= E Hsd TH T BnnT BT H HT H
Hsd (45)
sd sd sd sd
 −1
2 T
= 4σ Hsd Hsd
 −1
In this case, matrix Hsd TH is used to quantify how the level of measurement noise can be
sd
related to the expected level of errors in the target location estimate.
If we assume θ2az + θ2el ≈ 0, then covariance Csd
θ
is approximately computed by

θ43dB
Csd
θ ≈ σ2 I2 (46)
32 (log 2)2 θ2s ( g(θs ))2 A2

This is the same covariance equation to that of the full measurement-based algorithm, which means
that the sum-difference measurement-based algorithm has the same nominal performance as the full
measurement-based algorithm.
Formal noise analysis on the conventional algorithm of Equations (15) and (16) is extremely
complex, and no useful information comes from it. Instead, Equation (44) or (46) can be utilized to
quantify the level of estimation error, even for the conventional algorithm, because the conventional
algorithm is a special case of the proposed iterated least-squares-based algorithm with the first
iterative step.
It is worth noting that the covariance prediction Equations (44) and (46) do not depend on the
measurement itself. Thus, if monopulse radar parameters are known beforehand, these equations can
be used to predict the performance of the monopulse algorithms off-line before the algorithms are
actually performed. The covariance prediction equation quantitatively states estimation accuracy in
terms of major parameters of amplitude comparison monopulse radar.
The estimates from the monopulse algorithm, together with covariance predictions, can be passed
as measurements to a nonlinear filter such as an extended Kalman filter, which processes them further
to provide filtered estimates of the direction angle and angular rate of the unknown target.

6. Simulations
Simulations were performed to evaluate the proposed least-squares-based monopulse algorithms
and their performance prediction with various combinations of measurement noise level, beam width,
squint angle, and target direction angles. In the simulations, a track axis gain of g0 = 1 and a received
signal gain of A = 1 V were used. The estimates of the proposed least-squares-based algorithms are
given after four iterations.
Figures 4 and 5 and Tables 1–3 show the simulation results for the conventional and the proposed
methods. They show RMS (root-mean-square) errors of the target location estimate in terms of azimuth
and elevation angle deviations from the center of track axis. The simulation results are all based on
N = 100, 000 Monte Carlo runs. The experimental RMS error in the azimuth estimate is defined as
v
u
N 
t 2
1 X (i)
σaz = θaz − θ̂az (47)
N
i=1

where superscript (i) denotes the results from run i. A similar expression yields the RMS error in the
elevation estimate.
 Appl.Sci.
Appl. Sci.2020,
2020,10,10, x FOR PEER REVIEW
3966 10ofof1313
10
Appl. Sci. 2020, 10, x FOR PEER REVIEW 10 of 13

Estimate of Target Azimuth Angle

1033 Estimate of Target Azimuth Angle
10
conventional
conventional
full
full
1022 sum-diff
10 sum-diff
prediction
prediction

1011
10

1000
10

10-1
10-1

10-2
10-2
10-2 10-1 1000
10-2 10-1 10
measurement noise ( (deg))
measurement noise ( (deg))
Figure 4. Three position vectors and true anomalies.
Figure4.4.Three
Figure Threeposition
positionvectors
vectorsand
andtrue
trueanomalies.
anomalies.

Estimate of Target Elevation Angle

1033 Estimate of Target Elevation Angle
10

conventional
conventional
full
1022 full
sum-diff
10 sum-diff
prediction
prediction

1011
10

1000
10

10-1
10-1

10-2
10-2
10-2 10-1 1000
10-2 10-1 10
measurement noise ( (deg))
measurement noise ( (deg))
Figure5.5.Three
Threeposition
position vectorsand
and trueanomalies.
anomalies.
Figure 5. Three positionvectors
Figure vectors andtrue
true anomalies.

Table 41 and
Figures shows theRMS
5 show RMSerrors
errors in respect
with Figuresto4the
and 5 in tabular
various standardform. In Table
deviations (σ) of1,measurement
the values in
Table 1 shows the RMS errors in Figures◦4 and 5 in tabular form. In Table 1, the values in
parenthesis
noise in the
in the case θ3dB =and
of ‘Full’ ◦ θs = 1 , θaz = 0columns
◦
3 ,‘Sum-Difference’ , θel = are
◦
0 . analytically
RMS errors computed RMS errors
increase linearly with
with the
parenthesis in the ‘Full’ and ‘Sum-Difference’ columns are analytically computed RMS errors with
Equations
increase (43) and (45)
in measurement noise until σ ≈ 0.3
respectively, ◦ is reached,
and the values in the
which ‘Prediction’
corresponds column are
approximately standard
to an SNR
Equations (43) and (45) respectively, and the values in the ‘Prediction’ column are standard
deviations computed
(signal-to-noise ratio) ofwith Equation
5.13 dB. (44).for larger noise levels than σ ≈ 0.3◦ , the performances of all
However,
deviations computed with Equation (44).
three algorithms become severely worse, although the proposed least-squares-based algorithms are
better than the conventional one.
Note that all three algorithms provide almost the same estimate accuracy, and the proposed
covariance prediction Equations (44) and (46) determine very accurate RMS errors with low noise levels.
Therefore, the covariance prediction equation can be used to analytically calculate the RMS errors even
Appl. Sci. 2020, 10, 3966 11 of 13

for the conventional monopulse algorithm, and this equation could be considered a measure for the
performance analysis of amplitude comparison monopulse radar algorithms.

Table 1. RMS errors for noise levels: θ3dB = 3◦ , θs = 1◦ , θaz = 0◦ , θel = 0◦ .

σ Target Position Conventional Full (Analysis) Sum-Difference (Analysis) Prediction

Azimuth 0.0269 0.0269 (0.0268) 0.0260 (0.0268) 0.0268
0.01
Elevation 0.0268 0.0268 (0.0268) 0.0268 (0.0268) 0.0268
Azimuth 0.1338 0.1331 (0.1335) 0.1330 (0.1336) 0.1339
0.05
Elevation 0.1346 0.1339 (0.1335) 0.1339 (0.1336) 0.1339
Azimuth 0.2699 0.2595 (0.2656) 0.2592 (0.2660) 0.2678
0.1
Elevation 0.2689 0.2593 (0.2656) 0.2591 (0.2660) 0.2678
Azimuth 0.8450 0.7235 (0.8581) 0.6066 (0.8471) 0.8033
0.3
Elevation 0.8468 0.6959 (0.8528) 0.6066 (0.8447) 0.8033
Azimuth 6.2642 3.4008 (2.8086) 2.7784 (2.9192) 1.3388
0.5
Elevation 7.1466 3.4119 (2.8092) 2.7817 (2.9239) 1.3388
Azimuth 610.3174 8.3577 (11.4122) 8.8205 (21.0155) 2.6776
1
Elevation 251.8055 8.3743 (11.4167) 8.9047 (20.9119) 2.6776

Table 2. RMS errors for target locations: θ3dB = 3◦ , θs = 1◦ , σ = 0.1◦ .

Deviations Conventional Full (Analysis) Sum-Difference (Analysis) Prediction

θaz = 0◦ , Azimuth 0.2699 0.2595 (0.2656) 0.2592 (0.2660) 0.2678
θel = 0◦ Elevation 0.2689 0.2593 (0.2656) 0.2591 (0.2660) 0.2678
θaz = 1◦ , Azimuth 0.3579 0.2914 (0.2920) 0.2932 (0.2943) 0.2678
θel = −1◦ Elevation 0.3601 0.2926 (0.2923) 0.2945 (0.2947) 0.2678

Table 3. RMS errors for beam width and squint angle: θaz = 0◦ , θel = 0◦ , σ = 0.1◦ .

Deviations Conventional Full (Analysis) Sum-Difference (Analysis) Prediction

θ3dB = 3◦ , Azimuth 0.2699 0.2595 (0.2656) 0.2592 (0.2660) 0.2678
θs = 1◦ Elevation 0.2689 0.2593 (0.2656) 0.2591 (0.2660) 0.2678
θ3dB = Azimuth 0.9596 0.6343 (0.8404) 0.6270 (0.8412) 0.9542
6◦ , Elevation 0.9599 0.6350 (0.8403) 0.6275 (0.8412) 0.9542
θs = 1◦
θ3dB = 6◦ , Azimuth 0.5363 0. 5160 (0.5312) 0.5156 (0.5320) 0.5355
θs = 2◦ Elevation 0.5390 0. 5185 (0.5311) 0.5181 (0.5319) 0.5355

Table 1 shows the RMS errors in Figures 4 and 5 in tabular form. In Table 1, the values in
parenthesis in the ‘Full’ and ‘Sum-Difference’ columns are analytically computed RMS errors with
Equations (43) and (45) respectively, and the values in the ‘Prediction’ column are standard deviations
computed with Equation (44).
Table 2 shows RMS errors of target location estimates with respect to unknown target locations
(θaz , θel ) in the case of θ3dB = 3◦ , θs = 1◦ , σ = 0.05. It is shown that the target location itself affects the
accuracy of the target location estimate, and that the proposed least-squares-based algorithms provide
better estimates than the conventional one for the situation where the unknown target location is far
from the center of the track axis.
Table 3 shows RMS errors with respect to beam width and squint angle (θ3dB , θs ) in the case
of θaz = 0◦ , θel = 0◦ , σ = 0.1. As expected from Equation (44), it is clear that the accuracy of the
estimates gets better where θs is larger and beam width is smaller.
Appl. Sci. 2020, 10, 3966 12 of 13

7. Conclusions
In this paper, two least-squares-based algorithms and their covariance prediction equations have
been presented for amplitude comparison monopulse radar. The conventional monopulse problem is
reinterpreted as a target location estimation problem in the track axis with four pseudorange equations
in which geometrical distance is weighted by the antenna gain.
The full measurement-based and sum-difference measurement-based algorithms are proposed
to estimate the target location based on the iterated least-squares estimation method. While the
full measurement-based algorithm utilizes all four receiving signals from the four antennas,
the sum-difference measurement-based algorithm uses the same sum and difference signals as
the conventional algorithm.
By theoretical analysis with the same assumptions used in the conventional monopulse algorithm,
it has been shown that the full measurement-based and sum-difference measurement-based algorithms
are nominally the same, and the conventional monopulse algorithm is a special case of the proposed
least-squares-based algorithms with the first iterative step. Furthermore, it turns out that the covariance
prediction equation for the proposed least-squares-based algorithms can be utilized to determine the
level of estimation errors even for the conventional algorithm.
According to numerical simulation results with 100,000 Monte Carlo runs, the proposed
least-squares-based algorithms show similar estimation performance as the conventional ones in low
levels of noise, and better performance with high levels of noise and large deviation of an unknown
target from the center of the track axis. The proposed covariance prediction equation provides
very accurate RMS errors for both the conventional and least-squares-based algorithms, and thus
it can be used to predict the performance of the monopulse algorithms before the algorithms are
actually performed.

Author Contributions: Methodology and mathematical formulation, S.P.; software and simulation, M.K., D.H.
and S.P.; formal analysis, S.P.; writing—original draft preparation, M.K., D.H. and S.P.; writing—review and
editing, M.K., D.H. and S.P. All authors have read and agreed to the published version of the manuscript.
Funding: The authors gratefully acknowledge support from the Electronic Warfare Research Center (EWRC)
at Gwangju Institute of Science and Technology (GIST), originally funded by the Defense Acquisition Program
Administration (DAPA) and the Agency for Defense Development (ADD).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Issacand, A.; Zhang, X.; Willet, P.; Bar-Shalom, Y. A Particle Filter for Tracking Two Closely Spaced Objects
using Monopulse Radar Channel Signls. IEEE Signal Process Lett. 2006, 13, 357–360.
2. Hofstetter, E.; DeLong, D. Detection and Parameter Estimation in an Amplitude-Comparison Monopulse
Radar. IEEE Trans. Inf. Theory 1969, 15, 22–30. [CrossRef]
3. Kim, H.K.; Jung, M.A.; Kim, S.C. A performance analysis of phase comparison monopulse algorithm for
antenna spacing and antenna array. J. Korean Inst. Commun. Inf. Sci. 2015, 40, 1413–1419.
4. Han, Y.J.; Kim, J.W.; Park, S.R.; Noh, S.U. An investigation into the monopulse radar using tx-rx Simulator in
electronic warfare settings. J. Korean Inst. Commun. Inf. Sci. 2017, 1, 705–706.
5. An, D.J.; Lee, J.H. Performance analysis of amplitude comparison monopulse direction-of Arrival Estimation.
Appl. Sci. 2020, 10, 1246. [CrossRef]
6. Jacovitti, G. Performance Analysis of Monopulse Receivers for Secondary Surveillance Radar. IEEE Trans.
Aerosp. Electron. Syst. 1983, AES-19, 8848–8897. [CrossRef]
7. Chen, L.; Sheng, W. Performance Analysis for Angle Measurement of Monopulse Radar with Non-Consistent
Amplitude-Phase Features. In Proceedings of the International Conference on Microwave and Millimeter
Wave Technology, Chengdu, China, 5–8 May 2012; pp. 1382–1385.
8. Mosca, E. Angle Estimation in Amplitude Comparison Monopulse System. IEEE Trans. Aerosp. Electron.
Syst. 1969, 205–212. [CrossRef]
Appl. Sci. 2020, 10, 3966 13 of 13

9. Cai, F.; Fan, H.; Song, Z.; Fu, Q. Difference beam aided target detection in monopulse radar. Chin. J. Aeronaut.
2015, 28, 1485–1493. [CrossRef]
10. Matuszewski, J. The Radar Signature in Recognition System Database. In Proceedings of the 19th
Intenational Conference on Microwaves, Radar and Wireless Communications MIKON-2012, Warsaw,
Poland, 21–23 May 2012; pp. 617–622, art. no. 6233565.
11. Matuszewski, J.; Pietrow, D. Recognition of electromagnetic sources with the use of deep neural networks.
XII onference on Reconnaissance and Electronic Warfare Systems. Proc. SPIE 2019, 11055, 110550D.
12. Golden, A., Jr. Radar Electronic Warfare (AIAA Education Series); Amer Inst of Aeronautics: Washington, DC,
USA, 1987.

© 2020 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/).