applied

sciences
Article
An Enhanced Genetic Algorithm for Parameter
Estimation of Sinusoidal Signals
Chao Jiang 1,2 , Pruthvi Serrao 2 , Mingjie Liu 1 and Chongdu Cho 3, *
1 School of Automation, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
chiaojiang@inha.edu (C.J.); liumj@cqupt.edu.cn (M.L.)
2 Department of Mechanical Engineering, Graduate School, Inha University, Incheon 22212, Korea;
serrao@inha.edu
3 Department of Mechanical Engineering, Inha University, Incheon 22212, Korea
* Correspondence: cdcho@inha.ac.kr; Tel.: +82-32-860-7321




Received: 17 June 2020; Accepted: 23 July 2020; Published: 25 July 2020


Abstract: Estimating the parameters of sinusoidal signals is a fundamental problem in signal

processing and in time-series analysis. Although various genetic algorithms and their hybrids
have been introduced to the field, the problems pertaining to complex implementation, premature
convergence, and accuracy are still unsolved. To overcome these drawbacks, an enhanced genetic
algorithm (EGA) based on biological evolutionary and mathematical ecological theory is originally
proposed in this study; wherein a prejudice-free selection mechanism, a two-step crossover (TSC),
and an adaptive mutation strategy are designed to preserve population diversity and to maintain a
synergy between convergence and search ability. In order to validate the performance, benchmark
function-based studies are conducted, and the results are compared with that of the standard genetic
algorithm (SGA), the particle swarm optimization (PSO), the cuckoo search (CS), and the cloud
model-based genetic algorithm (CMGA). The results reveal that the proposed method outperforms
the others in terms of accuracy, convergence speed, and robustness against noise. Finally, parameter
estimations of real-life sinusoidal signals are performed, validating the superiority and effectiveness
of the proposed method.

Keywords: sinusoidal signal; parameter estimation; genetic algorithm (GA); benchmark function

1. Introduction
Many practical signals, such as voice and audio signals, power system transient signals, and
response signals of sensors, are recognized as signals of the sum of sinusoidal signals. The parameter
estimation of sinusoidal signals from the periodic time series data is a classical problem of ongoing
interest in signal processing. In recent years, it has drawn considerable attention of scientists and
researchers from various fields including speech analysis [1], electrical power and energy system [2],
and sensor signal analysis [3], among others. There are some methods addressing this problem;
in general, these can be classified into two categories, classical techniques [3–10], and intelligent
algorithms [11–20]. The classical techniques, when compared with intelligent ones, are easier
to implement and are computationally more efficient. However, they do have limitations, including
frequency resolution, low accuracy, mathematical demands, and so on. On the other hand, the
intelligent algorithms and hybrids are favorably employed due to their fewer requirements about the
problems and excellent resolution. A few recent applications include atomic physics [21], electrical
machines [22], big data [23], traveling salesman problem [24], and fuel cell [25]. Thus, intelligent
techniques have extended significantly to provide effective and reliable solutions for parameter
estimations of sinusoidal signals [11–20,26–29] accordingly.

Appl. Sci. 2020, 10, 5110; doi:10.3390/app10155110 www.mdpi.com/journal/applsci

Appl. Sci. 2020, 10, 5110 2 of 16

In a literature [11], Spavieri et al., used a particle swarm optimization (PSO)-based approach for the
parameterization of a power capacitor model fed by harmonic voltages in power distribution systems.
For parameter estimation of exponentially damped sinusoidal signals, Xiao et al. [15] presented a
specific artificial neural network (ANN) based estimator, by which the parameters of each exponentially
damped sinusoidal signal component could be directly calculated with a high precision. However,
the performance strongly relies on the converged weights of the ANN, and hence, an absence of
optimal weights or insufficient training will result in a poor performance. Based on a genetic
algorithm (GA), Coury et al. [18,20] presented an efficient estimator to evaluate the frequency
and phase for a phasor management unit. In comparison to the phase-locked loop (PLL) and
the discrete Fourier transform (DFT) based methods, the GA revealed the fastest response to a sufficient
precision. In order to minimize bias and improve estimation performance, researchers generally
combine different methods to overcome the individual drawbacks. In another article, Djurović
et al. [17] presented a hybrid genetic algorithm (HGA) fused with a maximum likelihood approach
to estimate the phase parameters of frequency modulated signals. Zang et al. [30] proposed a cloud
model-based genetic algorithm (CMGA) for optimization problems. Benefiting from the cloud model,
the CMGA continuously produces new individuals, and hence, the exploration and randomness
can be significantly improved, showing a better optimization when compared to other methods.
In the literatures [31–33], Raja et al., introduced some hybrid techniques with ANN, GA, and other
algorithms for the dynamics of nonlinear singular heat conduction model of the human head, nonlinear
Painlevé II systems, and nonlinear singular Thomas–Fermi systems. Comparisons between the
proposed schemes and the standard numerical solutions as well as analytical methods revealed the
feasibility and effectiveness of the proposed schemes. Similarly, a hybrid approach using a differential
evolution (DE) and GA was proposed by Ali et al. [34], where a DE-based multi-parent crossover
operation was introduced to enhance the search ability and to avoid premature convergence by
exploring more solutions in the problem search space. In combination with a local search optimization
algorithms, Zaman et al. [35] proposed an HGA to estimate the amplitude, frequency, range, elevation
angle, and azimuth angle of near field sources; the proposed schemes achieved good results in terms
of accuracy, convergence rate, and robustness, but the complexity increased a lot.
In the light of these facts, authors are motivated in this work to design an enhanced genetic
algorithm (EGA), through a prejudice-free selection mechanism for preserving population diversity,
diversified individuals with a two-step crossover (TSC) operator, and a good cooperation of fast
convergence and searching performance with an adaptive mutation strategy, for estimating the
parameters of sinusoidal signals. The prominent features of the proposed method are briefly
summarized as follows:
• Validation through the results of a comparative analysis in terms of performance-monitoring
metrics based on the mean, coefficients of determination (R2 ), and the sum of squared residuals
value (SSR) of each algorithm;
• An intelligible concept with easy realization and a worry-free tradeoff between global and local
search in comparison to general hybrid techniques.
The rest of the work is organized as follows. The mathematical ecological theory foundation of the
proposed approach is presented in Section 2. Section 3 describes the proposed EGA in detail, including
the prejudice-free selection, the two-step crossover operator, and the adaptive mutation operator.
In Section 4, benchmark function based numerical analyses are carried out to verify the validity of the
EGA. The parameter estimations of real-life sinusoidal signals are performed and discussed in Section 5.
At last, we summarize our work and highlight the main contributions in Section 6.

2. Mathematical Ecological Theory Foundation

In ecology, the number of offspring restricted to be equal to or less than that of parents is not
common. Even though this may happen for reasons such as disease, food, water, and other factors, the
Appl. Sci. 2020, 10, 5110 3 of 16

species may become extinct or be restricted due to genetic competition as well. When the number of
offspring is more than two, the population size grows rapidly with more competitive individuals [24],
resulting in an increased chance for producing better offspring. Thus, it is desirable to have a larger
number of individuals in genetic algorithms.
Population dynamics were first studied by Verhulst [36] in 1838, who describes the size and age
composition of populations as dynamical systems driven by biological and environmental processes.
The population dynamics model of a species can be mathematically expressed by:

dP P
= rP · (1 − ), (1)
dt K
where P represents population size, t is time, r defines the growth rate, and K is a constant denoting
the environment carrying capacity. The solution to Equation (1) with an initial population P0 can be
written as:
KP0 · ert
P(t) = . (2)
K + P0 · (ert − 1)
Taking the birth and mortality rate into account, Equation (2) can be rewritten as:

KP0 · e(r1 −r2 )t

∆P(t) = , (3)
K + P0 · [e(r1 −r2 )t − 1]

where ∆P is the total population size, and r1 and r2 are birth rate and mortality rate, respectively.
Over time, the population dynamics presents different consequences in accordance with the relations
between r1 and r2 .
1. When the birth rate is higher than the mortality rate, r1 > r2 , the final population size for an
infinite amount of time, t, can be simplified as:

KP0 · e(r1 −r2 )t

lim ∆P(t) = lim =K (4)
t→∞ t→∞ K + P0 · [e(r1 −r2 )t − 1]

This means that a species can exist forever.

2. When the birth rate is equal to the mortality rate, r1 = r2 , the exponential terms in the numerator
and denominator become 1, so Equation (3) is given as follows:

lim ∆P(t) = P0 . (5)

t→∞

Here, as time approaches infinity, the final population size will equal to the initial value.
What if the initial population size equals to 1 or 2? Although this species can live for a long time,
the problem of self-reproduction and inbreeding results in an extinction.
3. When the birth rate is smaller than the mortality rate, r1 < r2 , it is clear that e− x → 0 as t → ∞.
Hence, one can get the following expression:

KP0 · e−(r2 −r1 )t

lim ∆P(t) = lim =0 (6)
t→∞ t→∞ K + P0 · [e−(r2 −r1 )t − 1]

Obviously, this equation represents an inevitable extinction of a species.

According to the analysis above, on the one hand, it can be concluded that the extinction of
a species occurs if the birth rate is equal to or less than the mortality rate; only when r1 > r2 can
the species survive forever. On the other hand, it is found that the final population size strongly
depends on both the growth rate and the time when the initial population size is fixed. Thus, in order
to increase the number of offspring and get excellent individuals within a limited time, it is very
Appl. Sci. 2020, 10, 5110 4 of 16

necessary to improve the reproduction rate of a species [24]. In the case of the standard genetic
algorithm (SGA), as the procreation is artificially controlled to be unchangeable, the extinction of
the population of SGA is not allowed. Hence, neither is the evolutionary law consistent with the
natural law, nor will it have a chance to acquire more excellent individuals. In this paper, based on
the biology theory and mathematical ecological theory foundation, a genetic algorithm with several
enhancements is proposed in order to achieve a higher accuracy, and faster convergence, and hence,
to further improve the parameter estimation results of sinusoidal signals.

3. The Proposed EGA

3.1. Prejudice-Free Selection

Elitism-based simple selection strategies [37] are popularly used in GAs, which favor the higher
fitness chromosomes (individuals) and discriminate against the lower ones. Consequently, outstanding
individuals dominate in iterations leading to a monotonous population diversity and breaking
up a healthy competition. Besides, in the SGA the production mechanism of offspring using a
random pairing through two individuals will reduce the population quality, because an outstanding
chromosome may combine with a poor one and lose its superior genes. To settle this, in the EGA the
individuals are divided into two groups according to their fitness values: the best half individuals
with high fitness values are in a benign group, while the remaining ones are assigned to a malignant
group; the evolution in each group takes place independently, and a new selection is always executed
to redistribute them in every iteration. So, offspring may stay in their original group or migrate
to the other in accordance with their new fitness values. Compared to traditional selections, this
operation not only reduces the overhead of selection resulting from a non-bias selection procedure [38],
but also it allows a parallel computation providing a faster convergence [39]. Thanks to this, the
population diversity and efficient convergence can be guaranteed.

3.2. Two-Step Crossover (TSC) Operator

In GAs, crossover operation simulates the processes of sexual reproduction, through which
offspring inherit parts of genetic information from their parents. However, conventional crossover
operators just follow the genetic concept, weakening the search ability [40,41]. For instance,
in a single-point crossover (SPC), two individuals are first selected as the parent chromosomes,
then a randomly-generated point is used to truncate the chromosomes and get two sub-sequences; the
fragments behind the point are exchanged with each other and coupled with the front ones to generate
two new offspring. The principle of a SPC operation is shown in Figure 1.

Random crossover point

Parent
01001010 10110100 11100101 Ă 00100100 11101101 00100001 01101110 Ă 11011011
individuals
x1 x2 x3 Ă xn x1 x2 x3 Ă xn
Offspring 01001010 10110001 01101110 Ă 11011011 11101101 00100100 11100101 Ă 00100100
x1 x'2 x3 Ă xn x1 x'2 x3 Ă xn

Figure 1. Single-point crossover operation for binary sequence.

For a given population of a size of N × M, the number of the newly generated variables after
a SPC operation can be calculated by [ NPc2 ], where N is the number of individuals, M denotes the
length of each chromosome, M = n × l, n is the number of the variables defined by the chromosomes,
l is the coding length of each variable, and Pc2 is the crossover probability. The symbol [·] rounds the
number toward positive infinity. It is clear that a SPC can produce new individuals and change the
population diversity. However, only the variables whose sequences are trimmed are updated during
the operation, and hence, the population diversity is still restricted. Although other types of crossover
Appl. Sci. 2020, 10, 5110 5 of 16

operations such as two-point crossover and multi-point crossover can mitigate this effect, the similar
operation mechanism dooms the same outcome.
In this regard, a two-step crossover strategy that comprehensively considers the information
exchange among entities and variables is designed in this work. The details are described as follows:
1. Dividing each chromosome into segments based on the number of the variables to be solved,
and gathering all the specific segments for a certain variable to form a variable set for the
related variable;
2. Randomly selecting some variable sets with a probability Pc1 ;
3. Implementing crossover operation on each elected set to update the variables with a
probability Pc2 ;
4. Once the crossover is completed, reassembling the variable sets together to get the
new individuals.
For a better understanding, a SPC is employed in the TSC to illustrate the mechanism, as shown
in Figure 2. Here, the number of the newly produced variables by the TSC can be calculated by
[nPc1 ] × [ NPc2 ], where Pc1 is the first crossover probability, and the other parameters are the same as
the ones mentioned above. The difference between the newly generated variables between the TSC and
SPC is mathematically expressed as [nPc1 ] × [ NPc2 ] − [ NPc2 ]. With some mathematical manipulations,
this expression can be simplified as [nNPc2 ] · ( Pc1 − 1/n). It can be known that a positive value
is always achieved if ( Pc1 − 1/n) > 0. So, by simply adjusting Pc1 , the number of the updated variables
yielded by the TSC can be much more than that of the SPC, which amounts to indirectly increase the
number of the offspring [24]. Hence, the population diversity can be significantly improved. Besides,
notice that the applied SPC in the TSC can also be replaced by other types of crossover operations; for
the case of multi-point crossover, a considerable amount of new variables can be obtained.

x1 x2 x3 Ă xn
Individual 1 7.2256631 0.6141592 -0.689756 Ă 4.2338761

Individual 2 7.9482294 0.6911503 -0.046752 Ă 4.5444923

Ă

Ă
Ă

Individual N 6.3585835 0.6220353 0.3120234 Ă 4.7041438

x1 x3 Ă xn-2 Select variables

using Pc1
Individual 1 7.2256631 -0.689756 Ă 0.0038361

Individual 2 7.9482294 -0.046752 Ă 0.0044137

Ă

Individual N 6.3585835 0.3120234 Ă -0.004173

x1 xn-2
Individual 1 7.2256631 0.0038361

Individual 2 7.9482294 Ă 0.0044137

Ă

Ă
Ă

Individual N 6.3585835 -0.004173

Crossover operation
x1` xn-2` using Pc2
Individual 1` 7.2585835 0.0038036

Individual 2` 7.9911287 Ă 0.0044173

Ă

Individual N` 6.3256631 -0.004099

Figure 2. The two-step crossover operation for floating sequence.

3.3. Adaptive Mutation Operator

The mutation operator has a strong influence on population diversity and convergence speed,
playing a significant role in GAs. Usually, mutation probability, Pm , keeps constant throughout the
whole iteration in the SGA and HGAs [16,19,36,38,41]. Although a constant mutation probability
Appl. Sci. 2020, 10, 5110 6 of 16

provides GAs a relatively stable performance, the problem of ever-increasing similar individuals
cannot be effectively solved. To address this, an adaptive mutation mechanism that considers the
evolutionary features of GAs at different stages is proposed in this work.
At the beginning of an evolution, a small Pm is allocated for boosting the convergence [42].
As the processing progresses, the number of similar individuals increases rapidly, which causes the
optimization to fall back to local optima repeatedly. In order to overcome this, a larger mutation
probability is assigned at the intermediate stage so that the evolution can easily get rid of predicaments.
In the final stage, a small probability is distributed again to ensure a global convergence [37].
Inspired also by the characteristics of population dynamics, a combination of a general Logistic
function [36] and its mirror is employed to support the proposed idea in its entirety. The function
is expressed as follows:
b
(
a+ ,g ≤ G
Pm = 1 + e − c ( g − g0 ) (7)
b
a+ + c ( g − g0 ) , g > G
1+ e

where a denotes the initial mutation probability, b is the maximum value of the original Logistic curve,
e is the natural logarithm base, c describes the steepness of the curve, g and g0 are the evolution
generation and the curve’s midpoint, respectively. G is the total number of evolutionary iterations.
Figure 3 shows the curves of the functions.

Figure 3. Adaptive mutation probability as the change of iteration.

3.4. EGA Procedures

To solve a problem using the proposed EGA, N individuals are randomly generated by a
float encoding method as the initial population Init_Pop. Each individual’s objective value and
fitness value are calculated, and the best individual BestX_Old in the current population is obtained.
In order to keep population diversity and promote a healthy competition, individuals are assigned
to a Benign group or Malignant group with the same size of N/2, respectively, according to their
fitness values. As the individuals in the Benign group have higher fitness values, performing a
crossover operation in this group is helpful to generate better descendants. Therefore, the TSC
operation with probabilities Pc1 and Pc2 is run for them only. Instead of being discarded directly,
in the Malignant group, the remaining individuals with a low fitness but rich population diversity go
through the adaptive mutation. Once the crossover and mutation are complete, the fitness values of
the individuals in the new population New_Pop will be evaluated. The individuals in the Malignant
group that have a high fitness will migrate to the Benign group, and a reverse movement is also
carried out synchronously for the ones in the Benign group having a low fitness. Again, the best
individual Best_New in the current population will be picked up against the Best_Old, and the better
one is saved as Best_Temp. As the optimum may hide near Best_Temp, some random individuals
(e.g., 10) are generated in this area, and their fitness values are calculated. If someone is better than
Best_Temp, they will be reassigned as Best_Temp; else, Best_Temp keeps its original value. At the end
of each iteration, BestX_Old and Old_Pop are updated by Best_Temp and New_Pop, respectively.
Appl. Sci. 2020, 10, 5110 7 of 16

The termination criterion will be triggered when the maximum iteration MaxGen reaches. A more
clear procedure of the EGA is shown in Figure 4 and Algorithm 1.

Start

Initialize population

Calculate fitness values

Benign group Prejudice-free selection Malignant group

Adaptive
TSC operation
mutation

Migration and generating new

population

Random search

No
Termination criterion

Yes

Stop

Figure 4. Flow chart of the proposed enhanced genetic algorithm (EGA).

Algorithm 1 The pseudo-code of the enhanced genetic algorithm.

Input: N : size of initial population
MaxGen : maximum number of iteration
[ Pc1 , Pc2 ] : probabilities of the two-step crossover operation
[ Pm1 , Pm2 ] : probabilities of the initial and final adaptive mutation operation
num : number for generating new individuals around the best solution in each
iteration
f ( x ) : objective function
Output: Trace : the best solution and corresponding objective value in each iteration
function EGA(N, MaxGen, [ Pc1 , Pc2 ], [ Pm1 , Pm2 ], num, f ( x ))
Init_Pop ← random individuals with a size o f N;
Old_Pop ← Init_Pop;
for i ≤ N do
Calculate the f itness value o f each individual in Old_Pop;
end for
gen ← 1;
while gen ≤ MaxGen do
Best_Old ← the best individual in Old_Pop;
Selection:
Benign ← N/2 high f it individuals in Old_Pop;
Malignant ← N/2 low f it individuals in Old_Pop;
Crossover:
Carry out the TSC operation to the individuals in the Benign group;
Appl. Sci. 2020, 10, 5110 8 of 16

Mutation:
Apply an adaptive mutation operation to the individuals in the Malignant group;
New_Pop ← combination o f the processed individuals f rom the Benign and Malignant groups;
for i ≤ N do
Calculate the f itness value o f each individual in New_Pop;
end for
BestX_New ← the best individual in New_Pop;
BestX_Temp ← ( BestX_New < BestX_Old) ? BestX_New : BestX_Old;
BestX_Temp ← (min(rand(num)) < BestX_Temp) ? min(rand(num)) : BestX_Temp;
BestX_Old ← BestX_Temp;
Old_Pop ← New_Pop;
Trace ← the best solution and corresponding objective value in current generation;
gen ← gen + 1;
end while
Return Trace;
end function

4. Benchmark Function Study

Benchmark functions [14,30,34,39,43] are widely adopted to demonstrate the performance
of algorithms. Thus, in the present work, benchmark based numerical experiments are performed for
the EGA as well. Here, eight unimodal and multimodal functions [40] are selected (see Appendix A).
As a comparison, studies using the SGA [44] as well as the PSO [11,29], the CS [28,29], and
the CMGA [30] are also performed for the functions; the corresponding parameter configurations
are listed in Table 1. Without favoritism, all the algorithms are assigned the same size initial population,
N = 50, and the maximum iteration number, MaxGen = 200. The optimization results are plotted
in Figures 5 and 6. To have an insight into how well these algorithms performed, all the vertical
coordinates are displayed on a logarithmic scale.

Table 1. Parameter configurations of different algorithms for benchmark function optimization.

SGA PSO CS CMGA EGA

N = 50 N = 50 N = 50 N = 50 N = 50
MaxGen = 200 MaxGen = 200 MaxGen = 200 MaxGen = 200 MaxGen = 200
Pc = 0.7 C1 = 2, C2 = 2 Pa = 0.25 m1 · t1 = 1/600, n1 = 0.1 [ Pc1 , Pc2 ] = [0.6, 0.7]
Pm = 0.01 w1 = 0.9, w2 = 0.4 − m2 · t2 = 1/600, n2 = 0.1 [ Pm1 , Pm2 ] = [0.01, 0.1]
− − − m3 = 1/3, n3 = 0.1 −

From Figures 5 and 6, it can be seen directly that the performance of the EGA is superior to that
of the SGA, PSO, CS, and CMGA for most of the benchmark functions, except for the Zakharov
function ( f 2 ( x )). Examining the optimization results of f 2 ( x ) shown in Figure 5b, it is noted that
the evolutionary curve of the SGA displays in part. This is caused by the estimating value of zero
(equaling to the optimum), and the logarithmic y-axis. In this regard, the SGA performs better than the
EGA. However, when compared to other results, the proposed EGA shows overwhelming advantages
against others, especially for Yang’s No. 5 function ( f 8 ( x )). Different from other functions, f 8 ( x )
contains uniform distribution terms, which increases not only the number of local optima but also the
uncertainty, and hence, the optimization becomes much more difficult. Taking a look at the results
in Figure 6d, frequent vibrations are observed in the evolution curves of the SGA and PSO, while the
EGA shows an unremitting evolution trend, demonstrating its robustness against noise.
Appl. Sci. 2020, 10, 5110 9 of 16

(a) (b)

(c) (d)

Figure 5. Evolutionary curves of unimodal benchmark functions. (a) Sphere function. (b) Zakharov
function. (c) Rosenbrock function. (d) Ackley function.

(a) (b)

Figure 6. Cont.
Appl. Sci. 2020, 10, 5110 10 of 16

(c) (d)

Figure 6. Evolutionary curves of multimodal benchmark functions. (a) Griewank function.

(b) WaveDrop function. (c) Xin-she Yang’s No.2 function. (d) Xin-she Yang’s No.5 function.

To further examine the performance, the obtained best and mean solutions as well as the
convergence speed of each function were compared and listed in Tables 2–4, respectively. All the results
were obtained over 50 Monte-Carlo simulations. Through Table 2, it is evident that the EGA
outperforms the SGA, PSO, CS, and CMGA for most cases. Looking at the optimization results
for f 2 ( x ), although only the SGA achieves the global optimum, the EGA still prevails over the others
when comparing the rest optimizations. The results listed in Table 3 also confirm this point. In terms
of the convergence speed, the maximum, the minimum, and the average termination iterations were
measured, where the termination criterion was defined by Oopt − Oobt < 10−4 , Oopt is the optimum
of a given problem, and Oobt is the obtained optimized result, respectively. From Table 4, one can find
that the convergence speed of the proposed EGA is fast, presenting a strong competitiveness over its
counterparts. From these comparisons, it can be concluded that the EGA shows an overwhelming
superiority over the competitors in terms of accuracy, convergence speed, and robustness against noise.

Table 2. Comparison of the best optimal solutions of standard genetic algorithm (SGA), particle swarm
optimization (PSO), cuckoo search (CS), cloud model-based genetic algorithm (CMGA), and EGA.

SGA PSO CS CMGA EGA

Function Optimum
Fbest Fbest Fbest Fbest Fbest
f1 (x) 0 4.7684 × 10−11 2.8466 × 10−10 2.7727 × 10−14 4.7684 × 10−11 2.7305 × 10−22
f2 (x) 0 0 1.0950 × 10−8 3.2559 × 10−12 9.7157 × 10−12 4.5721 × 10−22
f3 (x) 0 2.8240 × 10−6 2.6240 × 10−7 7.1331 × 10−8 5.4599 × 10−9 4.9467 × 10−10
f4 (x) 0 1.2212 × 10−4 0.0020 3.1397 × 10−5 6.4976 × 10−5 3.3660 × 10−10
f5 (x) 0 2.4573 × 10−7 0.0027 1.1657 × 10−4 2.4573 × 10−7 1.1102 × 10−16
f6 (x) −1 −1 a −1 a −1 a −1 a −1 b
f7 (x) 0 1.1984 × 10−5 1.2576 × 10−4 9.3775 × 10−5 1.4512 × 10−9 3.6105 × 10−11
f8 (x) 0 3.1936 × 10−8 0.0363 4.1578 × 10−7 1.0680 × 10−7 4.5682 × 10−13
a the practical optimized value is very close to −1. b the practical optimized value is −1.
Appl. Sci. 2020, 10, 5110 11 of 16

Table 3. Comparison of the average optimal solutions of SGA, PSO, CS, CMGA, and EGA.

SGA PSO CS CMGA EGA

Function Optimum
Fmean Fmean Fmean Fmean Fmean
f1 (x) 0 4.7684 × 10−11 2.4852 × 10−6 6.6205 × 10−7 1.8619 × 10−9 2.4070 × 10−13
f2 (x) 0 0 1.1387 × 10−5 2.3998 × 10−8 2.3486 × 10−10 9.8656 × 10−10
f3 (x) 0 0.0625 7.9783 × 10−5 1.6277 × 10−5 3.3094 × 10−7 1.1525 × 10−5
f4 (x) 0 1.2212 × 10−4 0.0331 4.2460 × 10−5 7.5088 × 10−5 1.0119 × 10−8
f5 (x) 0 0.0539 0.0239 0.0113 1.9530 × 10−6 0.9103 × 10−6
f6 (x) −1 −0.9319 −0.9719 −0.9864 −0.9955 −0.9998
f7 (x) 0 0.2374 0.0330 0.0431 6.0128 × 10−5 9.1184 × 10−8
f8 (x) 0 0.6746 0.5803 0.6875 1.6018 × 10−4 1.2212 × 10−5

Table 4. Comparison of the convergence speeds of SGA, PSO, CS, CMGA, and EGA.

SGA PSO CS CMGA EGA

Function
Gmax(min) Gmean Gmax(min) Gmean Gmax(min) Gmean Gmax(min) Gmean Gmax(min) Gmean
f1 (x) 45(7) 22.94 136(26) 64.10 41(10) 34.66 38(6) 28.16 36(5) 18.28
f2 (x) 54(8) 26.40 190(9) 101.86 65(27) 37.84 40(23) 35.26 44(12) 25.84
f3 (x) 200(13) 189.88 200(108) 135.40 187(30) 148.54 159(10) 121.63 139(7) 108.44
f4 (x) 200(200) 200.00 200(200) 200.00 200(120) 141.40 200(138) 120.33 156(51) 100.72
f5 (x) 200(200) 188.34 200(200) 200.00 200(120) 167.75 200(30) 69.93 200(51) 182.08
f6 (x) 200(48) 180.98 200(200) 189.58 200(108) 165.04 181(44) 140.71 169(6) 124.98
f7 (x) 200(39) 160.36 200(200) 200.00 200(105) 178.25 161(37) 153.24 153(39) 119.04
f8 (x) 200(19) 161.82 200(200) 200.00 200(42) 124.76 200(62) 133.08 200(18) 140.10

5. Parameter Estimation of Sinusoidal Signals

According to the numerical analyses above, the optimization performance of the EGA has already
been demonstrated. To evaluate its performance for parameter estimations of sinusoidal signals,
two real-life cases are selected, the voice data from singing the vowel ’ooh’ [45], and the Circadian
Rhythms [46]. As with the numerical analyses, the SGA, PSO, CS, and CMGA were also employed.

5.1. The Voice Dataset

The voice dataset gives the magnitudes of the voice when the vowel ‘ooh’ was sung at a pitch of
290 Hz. The frequencies and amplitudes found in the signal are used to determine the phonetic vowel,
and are of interest in voice synthesis, therapy, and training. In the present work, a ten-parameter
model is utilized for the signal as expressed as follows:

3
y(t) = K + ∑ Ai sin(ωi t + ϕi ), (8)
i =1

where K is the offset, A denotes the amplitude of sinusoid signal, ω and ϕ are the frequency and phase,
respectively. The parameter initialization is set as K ∈ [−1, 1], Ai ∈ [0, 100], ωi ∈ [−π/2, π/2], and
ϕi ∈ [−π/2, π/2].
The corresponding parameter configurations for the algorithms were the same as the ones
in Table 1, except for the crossover and mutation parameters of the EGA. The new values were set as
[ Pc1 , Pc2 ] = [0.8, 0.7], and [ Pm1 , Pm2 ] = [0.05, 0.22], respectively. The same N = 200, and MaxGen = 500
were assigned for all the algorithms as well.
Figure 7 shows the original data as well as the fitting curves, from which one can observe that
all the curves follow the pattern well. For having an insight into the difference of the algorithms, the
estimated parameters, and the coefficients of determination R2 were examined, listed in Table 5.

As can be seen from the table, the estimated parameters are very close, it is difficult to arrive at
a conclusion. However, by comparing the values of the R2 , the performance of the EGA can be
determined, which is the best among the algorithms. In other literature, Smyth et al. [47] considered
Appl. Sci. 2020, 10, 5110 12 of 16

the same dataset for frequency estimation, their reported result was 0.2299, 0.3408, and 0.1134, which
are very close to the results in this work.

Figure 7. Plots of voice data and fitting curves.

Table 5. Estimation results of SGA, PSO, CS, CMGA, and EGA for voice dataset.

Parameter SGA PSO CS CMGA EGA

Offset K −0.8437 0.1416 −0.9918 −0.9774 −0.9423
A1 22.1220 22.7440 20.9413 22.1339 22.5130
Amplitude A2 25.2844 26.5268 24.8365 24.9434 25.0975
A3 65.8691 66.8565 65.3737 64.6844 66.8589
ω1 0.2299 0.2320 0.2315 0.2293 0.2300
Frequency ω2 0.3404 0.3395 0.3418 0.3407 0.3407
ω3 0.1134 0.1126 0.1120 0.1133 0.1135
ϕ1 −0.0569 −0.1467 −0.1465 −0.0216 −0.0609
Phase ϕ2 −0.2428 −0.2082 −0.3285 −0.2656 −0.2611
ϕ3 0.1771 0.1797 0.2008 0.1409 0.1135
R2 0.9909 0.9912 0.9916 0.9922 0.9931

5.2. The Circadian Rhythms

The Circadian Rhythms was an experiment that recorded the temperature of a long-tail pocket mouse
every two minutes over three months. As some problems arose during the experiment, a proportion
of outliers remained in the dataset. Though these outliers make the estimation quite difficult,
they are meaningful and necessary for verifying the performance of an algorithm, because noise
is ubiquitous in real-life signals. In this work, a 20 min temperature sample averaged from the dataset
was extracted, and a four-component sinusoid model was used for the estimation, expressed as
Equation (9):
y(t) = K + A sin(ωt + ϕ), (9)

where K is the offset, A denotes the amplitude of sinusoid signal, ω and ϕ are the frequency and
phase, respectively.
The parameter initialization for the estimation was set as K ∈ [ym − 50, ym + 50], A ∈ [−100, 100],
ω ∈ [0, π ], and ϕ ∈ [0, π ]. Parameter ym is the median of the real data. The parameter configurations
for the algorithms were the same as those set in Section 5.1, except the ones for the EGA, whose new
Appl. Sci. 2020, 10, 5110 13 of 16

settings were: [ Pc1 , Pc2 ] = [0.7, 0.7], and [ Pm1 , Pm2 ] = [0.15, 0.35]. The corresponding evaluated results
are illustrated in Figure 8 and Table 6.
Figure 8 displays the original data and the best fitting curve. From this graph, it is evident to see
that the dataset contains a large number of outliers, but the fitting curve successfully ignores them and
follows nicely the periodic trend. In addition, the less evident outliers close to the fitting curve did
not distort the data fitting, either. To examine the results in Table 6, we found that the sum of squared
residuals (SSR) of the EGA is less than that of the others, indicating that the performance of the EGA
surpasses its competitors. Considering the same dataset, it was reported that the fitted frequency by
the elemental set method [48] was 0.0873, which once again is close to our frequency estimation.

Figure 8. Plots of Circadian Rhythms data and fitting curve.

Table 6. Estimation results of SGA, PSO, CS, CMGA, and EGA for Circadian Rhythms.

Parameter SGA PSO CS CMGA EGA

Offset (K ) 369.8530 370.1633 370.0567 370.1083 369.4427
Amplitude ( A) 23.1916 23.2972 23.3108 23.2656 22.5831
Frequency (ω ) 0.0886 0.0891 0.0874 0.875 0.0872
Phase ( ϕ) 0.8699 0.9205 0.9460 0.8877 0.9457
SSR 1.0902 × 103 1.0908 × 103 1.0912 × 103 1.0898 × 103 1.0194 × 103

6. Conclusions
This work presents an EGA for estimating the parameters of sinusoidal signals. The proposed
algorithm takes into account the preservation of population diversity, the balance between convergence
speed and accuracy, and the implementation complexity. In contrast to other genetic algorithms,
the main features of the EGA are:
• a prejudice-free selection mechanism for preserving population diversity;
• a TSC operation for enhancing information exchange among individuals and variables;
• An adaptive mutation strategy to avoid premature convergence and stagnation scenarios.
To evaluate the performance of the proposed method, studies based on benchmark functions
are conducted. The results indicate that the EGA not only achieves a higher accuracy and faster
convergence speed, but also it provides a good robustness against noise, showing an overwhelming
superiority over the SGA, the PSO, the CS, and the CMGA. At last, parameter estimations of real-life
sinusoidal signals are performed, which demonstrates the superiority and validity of the proposed
algorithm in its entirety.
The EGA has many issues worth studying further, such as the relationship between time
consumption and number of offspring, the applications where it excels and fails, and so on.
Appl. Sci. 2020, 10, 5110 14 of 16

Author Contributions: Conceptualization, C.J.; formal analysis, M.L.; software, C.J.; Supervision, C.C.;
visualization, P.S.; writing—original draft, C.J. and P.S.; writing—review and editing, P.S. and C.C. All authors
have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of interest.

Appendix A
The employed benchmark functions in this work:

2
Sphere: f 1 ( x ) = ∑ xi2 (A1)
i =1
xi ∈ [−5.12, 5.12] , x1∗ = (0, 0), f 1∗ = 0
!2 !4
2 2 2
Zakharov: f 2 ( x ) = ∑ xi2 + ∑ 0.5ixi + ∑ 0.5ixi (A2)
i =1 i =1 i =1
xi ∈ [−5, 10], x2∗ = (0, 0), f 2∗ = 0
1   2 
Rosenbrock: f 3 ( x ) = ∑ 100 xi+1 − xi + ( xi − 1)
2 2
(A3)
i =1
xi ∈ [−2.048, 2.048], x3∗ = (0, 0), f 3∗ = 0
s !
2 
2

−0.2 0.5 ∑ xi2 0.5 ∑ cos 2πxi
i =1
Ackley: f 4 ( x ) = −20 · e −e i =1 + 20 + e (A4)
xi ∈ [−32, 32], x4∗ =0= (0, 0), f 4∗
nx2 n
 
x
Griewank: f 5 ( x ) = 1 + ∑ i − ∏ cos √i (A5)
i =1
4000 i =1 i
xi ∈ [−600, 600], x5∗ = (0, 0), f 5∗ = 0
 q 
1 + cos 12 x12 + x22
WaveDrop: f 6 ( x ) = − (A6)
2 + 0.5 x12 + x22

xi ∈ [−5.12, 5.12], x6∗ = (0, 0), f 6∗ = −1

n
 
n − ∑ sin ( xi2 )
X. S. Yang’s No.2 fct.: f 7 ( x ) = ∑ | xi | · e i =1 (A7)
i =1
xi ∈ [−2π, 2π ], x7∗ = (0, 0), f 7∗ = 0
n
 
− ∑ Ui xi2 n
X. S. Yang’s No.5 fct.: f 8 ( x ) = 1 − e i =1 + ∑ Ui sin2 (2πnxi ) (A8)
i =1

Ui ∼ Uni f [0, 1], xi ∈ [−2π, 2π ], x8∗ = (0, 0), f 8∗ = 0

References
1. Drugman, T.; Stylianou, Y. Maximum Voiced Frequency Estimation: Exploiting Amplitude and Phase Spectra.
IEEE Signal Process. Lett. 2014, 21, 1230–1234. [CrossRef]
2. Beltran-Carbajal, F.; Silva-Navarro, G. A fast parametric estimation approach of signals with multiple
frequency harmonics. Electr. Power Syst. Res. 2017, 144, 157–162. [CrossRef]
3. Liu, B.; Zhang, C.; Ji, X.; Chen, J.; Han, T. An Improved Performance Frequency Estimation Algorithm for
Passive Wireless SAW Resonant Sensors. Sensors 2014, 14, 22261–22273. [CrossRef] [PubMed]
4. Stoica, P.; Li, H.; Li, J. Amplitude estimation of sinusoidal signals: Survey, new results, and an application.
IEEE Trans. Signal Process. 2000, 48, 338–352. [CrossRef]
Appl. Sci. 2020, 10, 5110 15 of 16

5. Duda, K.; Barczentewicz, S. Interpolated DFT for sinα ( x ) Windows. IEEE Trans. Instrum. Meas. 2014, 63,
754–760. [CrossRef]
6. Djukanović, S. An Accurate Method for Frequency Estimation of a Real Sinusoid. IEEE Signal Process. Lett.
2016, 23, 915–918. [CrossRef]
7. Zhang, J.; Wen, H.; Teng, Z.; Martinek, R.; Bilik, P. Power system dynamic frequency measurement based on
novel interpolated STFT algorithm. Adv. Electr. Electron. Eng. 2017, 15. [CrossRef]
8. Ye, S.; Kocherry, D.L.; Aboutanios, E. A novel algorithm for the estimation of the parameters of a real
sinusoid in noise. In Proceedings of the 23rd European Signal Processing Conference (EUSIPCO), Nice,
France, 31 August–4 September 2015; pp. 2271–2275. [CrossRef]
9. Karimi-Ghartemani, M.; Iravani, M.R. Robust and frequency-adaptive measurement of peak value.
IEEE Trans. Power Deliv. 2004, 19, 481–489. [CrossRef]
10. Samarah, A. A comparative study of single phase grid connected phase looked loop algorithms. Jordan J.
Mech. Ind. Eng. 2017, 11, 185–194.
11. Spavieri, G.; Ferreira, R.T.; Fernandes, R.A.; Lage, G.G.; Barbosa, D.; Oleskovicz, M. Particle Swarm
Optimization-based approach for parameterization of power capacitor models fed by harmonic voltages.
Appl. Soft Comput. 2017, 56, 55–64. [CrossRef]
12. Martinez-Ayala, E.; Ayala-Ramirez, V.; Sanchez-Yanez, R.E. Noisy signal parameter identification using
Particle Swarm Optimization. In Proceedings of the IEEE 21st International Conference on Electrical
Communications and Computers (CONIELECOMP), San Andres Cholula, Mexico, 28 February–2 March
2011; pp. 142–146. [CrossRef]
13. Wang, Y.; Wang, Z.; Zhao, B.; Xu, L. Parameters estimation of sinusoidal frequency modulation signal with
application in synthetic aperture radar imaging. J. Appl. Remote Sens. 2016, 10, 020502. [CrossRef]
14. Kiran, M.S. Particle swarm optimization with a new update mechanism. Appl. Soft Comput. 2017, 60, 670–678.
[CrossRef]
15. Xiao, X.; Lai, J.H.; Wang, C.D. Parameter estimation of the exponentially damped sinusoids signal using a
specific neural network. Neurocomputing 2014, 143, 331–338. [CrossRef]
16. Mitra, A.; Kundu, D.; Agrawal, G. Frequency estimation of undamped exponential signals using genetic
algorithms. Comput. Stat. Data Anal. 2006, 51, 1965–1985. [CrossRef]
17. Djurović, I.; Simeunović, M.; Lutovac, B. Are genetic algorithms useful for the parameter estimation of
FM signals? Digit. Signal Process. 2012, 22, 1137–1144. [CrossRef]
18. Silva, R.P.M.d.; Delbem, A.C.B.; Coury, D.V. Genetic algorithms applied to phasor estimation and frequency
tracking in PMU development. Int. J. Electr. Power Energy Syst. 2013, 44, 921–929. [CrossRef]
19. Mitra, S.; Mitra, A.; Kundu, D. Genetic algorithm and M-estimator based robust sequential estimation of
parameters of nonlinear sinusoidal signals. Commun. Nonlinear Sci. Numer. Simul. 2011, 16, 2796–2809.
[CrossRef]
20. Coury, D.V.; Silva, R.P.M.d.; Delbem, A.C.B.; Casseb, M.V.G. Programmable logic design of a compact
Genetic Algorithm for phasor estimation in real-time. Electr. Power Syst. Res. 2014, 107, 109–118. [CrossRef]
21. Raja, M.A.Z.; Zameer, A.; Khan, A.U.; Wazwaz, A.M. A new numerical approach to solve Thomas–Fermi
model of an atom using bio-inspired heuristics integrated with sequential quadratic programming.
SpringerPlus 2016, 5, 1400. [CrossRef]
22. Raja, M.A.Z.; Niazi, S.A.; Butt, S.A. An intelligent computing technique to analyze the vibrational dynamics
of rotating electrical machine. Neurocomputing 2017, 219, 280–299. [CrossRef]
23. Dong, H.; Li, T.; Ding, R.; Sun, J. A novel hybrid genetic algorithm with granular information for feature
selection and optimization. Appl. Soft Comput. 2018, 65, 33–46. [CrossRef]
24. Wang, J.; Ersoy, O.K.; He, M.; Wang, F. Multi-offspring genetic algorithm and its application to the traveling
salesman problem. Appl. Soft Comput. 2016, 43, 415–423. [CrossRef]
25. Sun, Z.; Wang, N.; Bi, Y.; Srinivasan, D. Parameter identification of PEMFC model based on hybrid adaptive
differential evolution algorithm. Energy 2015, 90, 1334–1341. [CrossRef]
26. Renczes, B. Numerical Problems of Sine fitting Algorithms. Ph.D. Thesis, Budapest University of Technology
and Economics, Budapest, Hungary, 2017.
27. Draa, A.; Bouzoubia, S.; Boukhalfa, I. A sinusoidal differential evolution algorithm for numerical
optimisation. Appl. Soft Comput. 2015, 27, 99–126. [CrossRef]
Appl. Sci. 2020, 10, 5110 16 of 16

28. Zhang, X.M. Parameter estimation of shallow wave equation via cuckoo search. Neural Comput. Appl.
2017, 28, 4047–4059. [CrossRef]
29. Merkle, D.; Middendorf, M. Swarm intelligence and signal processing [DSP Exploratory]. IEEE Signal
Process. Mag. 2008, 25, 152–158. [CrossRef]
30. Zang, W.; Ren, L.; Zhang, W.; Liu, X. A cloud model based DNA genetic algorithm for numerical optimization
problems. Future Gener. Comput. Syst. 2018, 81, 465–477. [CrossRef]
31. Raja, M.A.Z.; Umar, M.; Sabir, Z.; Khan, J.A.; Baleanu, D. A new stochastic computing paradigm for the
dynamics of nonlinear singular heat conduction model of the human head. Eur. Phys. J. Plus 2018, 133, 364.
[CrossRef]
32. Raja, M.A.Z.; Shah, Z.; Manzar, M.A.; Ahmad, I.; Awais, M.; Baleanu, D. A new stochastic computing
paradigm for nonlinear Painlevé II systems in applications of random matrix theory. Eur. Phys. J. Plus
2018, 133, 254. [CrossRef]
33. Sabir, Z.; Manzar, M.A.; Raja, M.A.Z.; Sheraz, M.; Wazwaz, A.M. Neuro-heuristics for nonlinear singular
Thomas-Fermi systems. Appl. Soft Comput. 2018, 65, 152–169. [CrossRef]
34. Ali, M.Z.; Awad, N.H.; Suganthan, P.N.; Shatnawi, A.M.; Reynolds, R.G. An improved class of real-coded
Genetic Algorithms for numerical optimization. Neurocomputing 2018, 275, 155–166. [CrossRef]
35. Zaman, F.; Qureshi, I.M. 5D parameter estimation of near-field sources using hybrid evolutionary
computational techniques. Sci. World J. 2014, 2014. [CrossRef]
36. Bacaër, N. A Short History of Mathematical Population Dynamics; Springer Science & Business Media: Berlin,
Germany, 2011. [CrossRef]
37. Sarmady, S. An investigation on genetic algorithm parameters. Sch. Comput. Sci. Univ. Sains Malays.
2007, 126.
38. Sathya, S.S.; Kuppuswami, S. Analysing the migration effects in nomadic genetic algorithm. Int. J. Adapt.
Innov. Syst. 2010, 1, 158–170. [CrossRef]
39. Sathya, S.S.; Radhika, M. Convergence of nomadic genetic algorithm on benchmark mathematical functions.
Appl. Soft Comput. 2013, 13, 2759–2766. [CrossRef]
40. Umbarkar, A.; Sheth, P. Crossover operators in genetic algorithms: A review. ICTACT J. Soft Comput. 2015, 6.
[CrossRef]
41. Lim, S.M.; Sultan, A.B.M.; Sulaiman, M.N.; Mustapha, A.; Leong, K. Crossover and mutation operators of
genetic algorithms. Int. J. Mach. Learn. Comput. 2017, 7, 9–12. [CrossRef]
42. Wang, K.; Wang, N. A protein inspired RNA genetic algorithm for parameter estimation in hydrocracking of
heavy oil. Chem. Eng. J. 2011, 167, 228–239. [CrossRef]
43. Jamil, M.; Yang, X.S. A literature survey of benchmark functions for global optimisation problems. Int. J.
Math. Model. Numer. Optim. 2013, 4, 150–194. [CrossRef]
44. Nabaei, A.; Hamian, M.; Parsaei, M.R.; Safdari, R.; Samad-Soltani, T.; Zarrabi, H.; Ghassemi, A. Topologies
and performance of intelligent algorithms: A comprehensive review. Artif. Intell. Rev. 2018, 49, 79–103.
[CrossRef]
45. Oliver, W.; Yu, J.; Metois, E. The Singing Tree:: Design of an Interactive Musical Interface. In Proceedings
of the 2nd Conference on Designing Interactive Systems: Processes, Practices, Methods, and Techniques
(DIS ’97), Amsterdam, The Netherlands, 18–20 August 1997; pp. 261–264. [CrossRef]
46. Andrews, D.; Herzberg, A. A Collection of Problems from Many Fields for the Student and Research Worker;
Springer: Berlin, Germany, 1985. [CrossRef]
47. Smyth, G.K. Employing Symmetry Constraints for Improved Frequency Estimation by Eigenanalysis
Methods. Technometrics 2000, 42, 277–289. [CrossRef]
48. Smyth, G.K.; Hawkins, D.M. Robust Frequency Estimation Using Elemental Sets. J. Comput. Graph. Stat.
2000, 9, 196–214. [CrossRef]

c 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/).