QPSO for Synthesis of Scanned Linear

Array Antenna for Fixed Side Lobe Level

and First Null Beam Width Including
Wide Null Placement

Hemant Patidar and Gautam Kumar Mahanti

Abstract In this research paper, quantum particle swarm optimization (QPSO)

algorithm, an algorithm founded on the basic theory of particle cluster and prop-
erties of quantum mechanics, has been introduced for design of uniformly spaced
scanned linear array antennas with definite value of sidelobe level (SLL), first null
beam width (FNBW), and broad null depth. An effort is made such that these
parameters are made equal to their related specific values. This is usually done by
changing excitation current amplitude of the elements. The expression of array
factor is derived by using the property of linear array antennas. One example has
been presented with 35 isotropic antennas. Generated Pattern is scanned to an angle
of 30°. The obtained results have shown that this technique is able to find required
value of SLL, FNBW, and wide null depth. Although, the proposed method is
developed and applied to a linear array of isotropic antennas; however, the principle
can easily be applied to other types of arrays.

Keywords Linear array antenna First null beam width Quantum particle swarm



optimization Sidelobe level Wide null depth

1 Introduction

Antenna array finds many applications in wireless communication system, radar,

military, commercial, etc., where phased-array antennas are widely used in different
applications, from military systems to commercial cellular communications net-
works. They are generally used in wireless communication systems to steer signals

H. Patidar (&)
G.K. Mahanti

Department of Electronics and Communication Engineering,
National Institute of Technology, Durgapur 713209, India
e-mail: [email protected]
G.K. Mahanti
e-mail: [email protected]

© Springer India 2015 197

D. Mandal et al. (eds.), Intelligent Computing and Applications,
Advances in Intelligent Systems and Computing 343,
DOI 10.1007/978-81-322-2268-2_21
198 H. Patidar and G.K. Mahanti

from base stations to desired locations while creating nulls to suppress interference.
Linear array is a very commonly used antenna array, which can constitute suitable
narrow beam. Antenna arrays are intentioned to have a high directive beam;
therefore, major part of the power should be transmitted by the main beam. This
means that the array has to have a low radiation in the other directions. Nowadays,
we are doing accrescent use of phased-array radar systems on naval ships and
aircrafts. In a phased array, the elements are fed with non-uniform current excitations
and phase shifters of all the elements are used to scan the beam in desired direction in
free space. We can change the shape of pattern by controlling the array factor
property by changing the geometrical configuration such as linear, planar, and
rectangular and antenna parameters such as distance between elements, excitation of
the elements, phase, and pattern of the individual elements [1, 2]. To fix the sidelobe
level (SLL), first null beam width (FNBW), and wide null depth with scanned far-
field pattern, elements are fed with non-uniform current excitation. But, the synthesis
problem is more complex and difficult to solve with analytical methods. Several
antenna array synthesis techniques were discussed in the letter [3–11].
To maximize the signal-to-noise ratio performance per undesirable interference,
null steering system finds huge applications in radar and communication systems. It
is achieved by placing the nulls along the direction of interference for antenna
arrays [3–6]. This is done by synthesis of linear antenna array pattern with pre-
scribed broad nulls [3]; genetic algorithm has been used for adaptive nulling by
phase-only control method [4], a method for effects of null control on the radiation
pattern is described in [5]. Null synthesis by controlling the phase for synthesis of
antenna arrays [6] and nulling of a symmetrical pattern location by only the weight
of phase control method was described in [7].
Here, the author discussed imposing of nulls in the desired directions for a
scanned antenna radiation pattern. Comparison for phased-array synthesis by par-
ticle swarm optimization and genetic algorithms is described in [8]. A method for
phase-differentiated reconfigurable array antenna including minimum dynamic
range ratio was reported in [9]. The optimization relationship between beam width
and SLL for broadside arrays [10] and a method for linear antenna arrays synthesis
was discussed in [11].
In this research paper, the authors have used quantum particle swarm optimi-
zation (QPSO) algorithm [12–15] for synthesis of linear array of isotropic antennas.

2 Approach

An N-element linear array [1] is considered along the y-axis that is assumed
uncoupled and uniformly spaced a distance d apart as shown in Fig. 1. Total
generated field of the isotropic antenna array is calculated by vector summation of
the field radiated by individual element. The free space [1] far-field pattern FP(θ) in
the vertical plane is stated by (1):
QPSO for Synthesis of Scanned Linear Array Antenna … 199

Fig. 1 Geometry of an N-
element linear array antenna
along the y-axis

X
N
FPðhÞ ¼ Bn ei/n eiðn1Þkd sin h ð1Þ
n¼1

where n = number of element, λ = the wavelength, Bn = the excitations amplitude,

i = the imaginary unit, k = 2π/λ = the wave number, d = spacing between the elements,
/n ¼ ðn  1Þ kd sin ho = uniform progressive phase, θo = scan angle, and θ is the
angle, represent the far-field point measured from z-axis as shown in Fig. 1.
Far-field in dB can be expressed in (2) as follows:
 
jFPðhÞj
FPn ðhÞ ¼ 20 log10 ð2Þ
jFPðhÞjmax

The main objective is to obtain the set of excitation current amplitude of the
elements using QPSO that will satisfy our goal. The proposed algorithm (QPSO)
will minimize the given cost function to obtain the desired value of SLL, FNBW,
and wide null depth.
 
Cost ¼ vr1  F12 þ vr2  F22 þ vr3  F33 ð3Þ

where

SLLo  SLLd ; if ! SLLo [ SLLd
F1 ¼ ð4Þ
0; if ! SLLo  SLLd

FNBWo  FNBWd ; if ! FNBWo [ FNBWd
F2 ¼ ð5Þ
0; if ! FNBWo  FNBWd
200 H. Patidar and G.K. Mahanti


WNomax  WNdmax ; if ! WNomax [ WNdmax
F3 ¼ ð6Þ
0; if ! WNomax  WNdmax

The coefficients vr1, vr2, and vr3 determine the relative weight specified to each
term in (3).
The value of vr1 = 15, vr2 = 1 and vr3 = 1 has been assumed to generate the
power pattern. SLLd and SLLo are desired and obtained values of sidelobe level,
FNBWo and FNBWd are obtained and desired values of FNBW, and WNomax and
WNdmax are obtained and desired values of maximum wide null depth, respectively.

3 Quantum Particle Swarm Optimization (QPSO)

Algorithm

QPSO algorithm [12–15] proposed by Sun et al. in 2004 is a novel optimization

algorithm which is founded on the fundamental theory of particle swarm and
properties of quantum mechanics in which all particles have the characteristic of
quantum behavior. However, QPSO algorithm’s global optimization performance is
better than standard PSO algorithm because QPSO is stated only by the position
vector; there is no velocity vector in QPSO. In QPSO, the dynamic behavior of the
particle is different from that of the particle in standard PSO. According to
uncertainty principle of quantum world, each particle can appear at any position of
search space with certain probability, so the position and velocity of a particle
cannot be found simultaneously.
The several steps involved in QPSO are given below:
Step 1. Initialize a population of particles with random positions between the
maximum and the minimum operating limits of the generators in the D-
dimensional space.
Step 2. Find out the cost value of each particle.
Step 3. Now compare the personal best (pbest) of each particle with its present
cost value. If the present cost value is better, then refer the present cost
value to pbest and refer the present coordinates to pbest coordinates.
Step 4. Calculate the mean best position (mbest) of all K particles using the fol-
lowing equation given below:

1X K
mbest ¼ pbesti
K i¼1

Step 5. Determine the present best cost value in the entire population and its
coordinates. Compare the global best position (gbest) of the current par-
ticle population with the previous one. If the present best cost value is
QPSO for Synthesis of Scanned Linear Array Antenna … 201

better than global best (gbest), then refer the current best fitness value to
gbest and refer the current coordinates to gbest coordinates.
Step 6. Determine the vector local focus of the particle by the given equation:
 
fidt ¼ rand1tid  ðpbestid Þ þ 1  rand1tid  ðgbestÞ

Step 7. The dth dimension of the ith particle position (Xid) is updated by the
following equations:

Xidt ¼ fidt þ ð1Þceilð0:5þrand2id Þ  b

t

 mbest  Xidt1  loge ð1=rand3tid Þ

If Xidt \Xmin
d
, then
 d 
Xidt ¼ Xmin
d
þ 0:25  rand4tid  Xmax  Xmin
d
ð7Þ

If Xidt [ Xmax
d
; then

Xidt ¼ Xmax
d
 0:25  rand5tid  ðXmax
d
 Xmin
d
Þ ð8Þ

where t is the current iteration, rand1, rand2, rand3, rand4, and rand5 are
the uniformly distributed random numbers between 0 and 1. Equations (7)
and (8)have been entered along all dimensions to squeeze the position
within Xmaxd
; Xmin
d
if the particle tries to cross the desired domain of
interest. These clipping techniques are sometimes essential to stop particles
from eruption.
Step 8. The steps 2–7 above are repeated until a stop criterion is satisfied; ordi-
narily, the algorithm is stopped when there is no further updates in the best
fitness value or reached maximum number of generation.
The parameter f is the local attractor of each particle. β is the contraction and
expansion coefficient that control the convergence speed and performance of the
particle. In order to avoid premature convergence, mean best position (mbest) is
regarded as the barycenter of all particles. The value of β is 0.75. Ceil (A) rounds
the elements of A to the nearest integers greater than or equal to A.

4 Simulated Results

In this paper, a linear array of 35 isotropic antennas with element spacing of 0.5λ
between the elements has been considered on the y-axis with specified SLL of
−25 dB or less, specified FNBW of 10° or less and specified wide null depth (WN)
202 H. Patidar and G.K. Mahanti

Table 1 Desired and obtained results

Design parameters Desired value Obtained value
Sidelobe level (dB) −25 −25.12
First null beam width (°) 10 10.5
Maximum wide null depth in dB from (θ = 54° to 61°) −50 −50
Computation time (s) … 1,393.54

of −50 dB or less from θ = 54° to 61° to generate the free space [1] scanned far-field
pattern FP(θ) at θ = 30° in the vertical plane. Desired results are obtained by taking
non-uniform current excitation amplitude of the elements for scanned linear antenna
array.
QPSO is run for 600 iterations with a particle size of 60 to generate the far-field
pattern in vertical plane. A set of 35 excitation current amplitude values is obtained.
The program is written in MATLAB for synthesis of the array antenna. The
computation time is measured by a PC with Intel Core2 duo processor of clock
frequency 2.93 GHz and 4 GB of RAM. Table 1 demonstrates the desired values
and obtained results using QPSO. Figure 2 shows the best fitness value versus
iteration number obtained by QPSO. Figure 3 shows the normalized current
amplitude distribution to generate the free space far-field pattern with asymmetrical
excitations. Figure 4 shows the normalized power pattern (far-field pattern) for
scanned antenna array.

2500

2000
Best Fitness Value

1500

1000

500

0
0 100 200 300 400 500 600
Iteration Number

Fig. 2 Best fitness value versus iteration number obtained by QPSO

QPSO for Synthesis of Scanned Linear Array Antenna … 203

Fig. 3 Distribution of 1

Normalized Current Amplitude Distribution

normalized current excitation 0.9
versus number of element
obtained by QPSO 0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 5 10 15 20 25 30 35

Fig. 4 Normalized far-field 0

Normalized Power Pattern in dB

pattern in dB for linear array

-10
scanned at θ = 30° using
QPSO with prescribed wide -20
null depth of −50 dB from -30
(θ = 54° to 61°)
-40

-50

-60

-70
-80
-100 -50 0 50 100
θ in degree

5 Conclusions

This paper used QPSO algorithm for synthesis of a scanned linear array antennas
with uniform inter-element spacing to obtain fixed SLL, fixed FNBW, and fixed
wide null depth. The obtained results using simulation depict a great deal between
the desired and obtained specifications. The main benefit of using this technique is
that it finds the suitable excitation current amplitude distribution using QPSO to
obtain the desired values. Results obtained from synthesis of linear array antennas
have illustrated the efficiency of this introduced technique.

References

1. Balanis, C.A.: Antenna Theory: Analysis and Design, 2nd ed. Wiley (Asia), Singapore (2003)
2. Haupt, R.L., Werner, D.H.: Genetic algorithms in electromagnetics. IEEE Press, New York
(2007)
204 H. Patidar and G.K. Mahanti

3. Er, M.H.: Linear antenna array pattern synthesis with prescribed broad nulls. IEEE Trans.
Antennas Propag. 38(9), 1496–1498 (1990)
4. Haupt, R.L.: Phase-only adaptive nulling with a genetic algorithm. IEEE Trans. Antennas
Propag. 45(6), 1009–1015 (1997)
5. Steyskal, H., Shore, R.A., Haupt, R.L.: Methods for null control and their effects on the
radiation pattern. IEEE Trans. Antennas Propag. 34(3), 404–409 (1986)
6. Vescovo, R.: Null synthesis by phase control for antenna arrays. Electron. Lett. 36(3), 198–
199 (2000)
7. Shore, R.A.: Nulling a symmetric pattern location with phase-only weight control. IEEE
Trans. Antennas Propag. 32(5), 530–533 (1984)
8. Boeringer, D.W., Werner, D.H.: Particle swarm optimization versus genetic algorithms for
phased array synthesis. IEEE Trans. Antennas Propag. 52(3), 771–779 (2004)
9. Mahanti, G.K., Chakraborty, A., Das, S.: Design of phase-differentiated reconfigurable array
antennas with minimum dynamic range ratio. IEEE Antennas Wirel. Propag. Lett. 5(1), 262–
264 (2006)
10. Dolph, C.L.: A current distribution for broadside arrays which optimizes the relationship
between beam width and sidelobe level. Proc. IRE Waves Electrons 34, 335–348 (1946)
11. Murthy, P.K., Kumar, A.: Synthesis of linear antenna arrays. IEEE Trans. Antennas Propagat.
24(6), 865–870 (1976)
12. Sun, J., Feng, B., Xu, W.: Particle swarm optimization with particles having quantum
behavior. In: Proceedings of Congress on Evolutionary Computation, Portland, Oregon, USA,
pp. 325–331(2004)
13. Sun, J., Xu, W.B., Feng, B.: A global search strategy of quantum behaved particle swarm
optimization. In: Proceedings of IEEE Conference on Cybernetics and Intelligent Systems,
Singapore, pp. 111–116 (2004)
14. Mikki, S.M., Kishk, A.A.: Quantum particle swarm optimization for electromagnetics. IEEE
Trans. Antennas Propag. 54(10), 2764–2775 (2006)
15. Ismail, T.H., Hamici, Z.M.: Array pattern synthesis using digital phase control by quantized
particle swarm optimization. IEEE Trans. Antennas Propag. 58(6), 2142–2145 (2010)