Selective Mapping scheme based on Modified Forest Optimization Algorithm for PAPR reduction in FBMC system

Abstract

Filter Bank Multicarrier(FBMC)is considered as one of the most standardized waveform for fifth generation (5G) mobile communication system application FBMC endures lot of nonlinear effects which occurs because of high Peak Average Power Ratio (PAPR). High value of PAPR due to the large dynamic range of multicarrier signal is one of the most significant issues in FBMC multicarrier based modulation technique. This paper presents one investigated PAPR reduction technique named as Selected Mapping (SLM) to minimize high PAPR by utilizing the complex signal divide into real and imaginary parts and then select minimum PAPR signal based on Modified Forest Optimization Algorithm (MFOA)to achieve good PAPR which can maintain the FBMC based system performance with a required Bit Error Rate (BER). The associated method was produced with the aim of optimize the phase factors so that the phase rotation operation is accomplished to minimize PAPR by fixing the MFOA into the conventional SLM. The simulation results demonstrate that the proposed technique gives better performance in terms of BER and PAPR compared to other SLM based optimization techniques.

Keywords

Bit error rate filter bank multicarrier modified forest optimization algorithm selected mapping peak average power ratio

1 Introduction

Rapid improvement in mobile communication based devices and services in wireless communication technology have already become an essential part of world over the past decades made an incredible progress leads to various prospect from the fifth generation (5 G) [1]. One can easily understand this tremendous development by observing the advancement of mobile phones from heavy pushbutton cellular phones designed solely for talking and texting to sleek smart phone devices with numerous advanced features [2]. With the introduction of two innovations into our lives known as the internet of things (IoT), the competency of high bandwidth, which should indeed requisite to arrangement with the increasing mobile connections, will no longer be sufficient. High availability, low latency, energy efficiency, increased scalability and high connectivity will all be important requirements in the coming years. As a result, it was unavoidable to build innovative waveforms with good skills for 5 G waveform as a potential substitute to the Orthogonal Frequency Division Multiplexing (OFDM) technique, whose flaws had begun to become intolerable with the intended path in next-generation wireless communication [3]. While the applicable features of the OFDM waveform are retained in this new waveform, the advanced features eliminate the shortcomings of the OFDM waveform is implemented. The use of nonlinear high PAPR is the predominant issue that should be addressed in multicarrier schemes transmission systems with high-power amplifiers (HPAs) [4]. It operates when the input signal exceeds a certain threshold level at that time the linear amplification process functionality is dropped by the amplifiers, severe signal degradation occurs in the input signal. It must be operated in this range to perform linear amplification without degradation. Since, these amplifiers frequently fail due to high-PAPR signal the only way to maintain linear amplification ranges signals from becoming distorted, reduce their PAPR [5]. Several multicarrier modulation schemes have developed for cognitive radio application instead of using multicarrier OFDM system. FBMC has drawn much attention due to its low Adjacent Channel Leakage Ratio (ACLR) substitute to OFDM mainly in Cognitive Radio (CR) applications. The most suitable scheme for PAPR reduction has been investigated with extremely low ACLR is FBMC with advanced feature that is a promising waveform contender for 5 G and future wireless systems. The most efficient PAPR reduction technique includes Partial Transmit Sequence (PTS), Selected Mapping (SLM) and Tone Reservation (TR). SLM is the most significant approach without any data loss when transmitting the multicarrier signal for PAPR reduction.

1.1 Related works

In FBMC PAPR problem is a new topic that has only recently begun to be addressed because the FBMC waveform is a advanced 5 G waveform contender introduced to the scientific world not offer enough time for researchers to conduct a large number of studies on the PAPR issue in the FBMC waveform. Several optimization techniques were applied to improve the system performance [6]. The mostly utilized optimization technique for PAPR reduction, phase factor search and BER performance improvement in both SLM and PTS techniques such as Ant Colony Optimization (ACO), Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Artificial Bee Colony Optimization (ABC). One disadvantage is to control parameters such as population size and number of iterations, additional parameter is required for better generation performance by using the above optimization algorithms. When conducting studies for dealing with the PAPR concern in the FBMC system, it is clear that there is a significant gap needs to be proved in this area. The following are some notable papers on the subject [7–13]:

In [7], the author suggested an effective PAPR reduction technique in FBMC multicarrier scheme. This paper proposed PTS based Ant Colony Optimization to reduce PAPR signal.PTS associated with an optimization approach reduces the PAPR to a larger extent by significantly reducing the PAPR of each data block and optimizing the use of each unit. The clipping technique is being used to compress the alerts, and the bit error rate (BER) is calculated at the receiver section. In [8], reported modified clipping technique based on biasing and precoding techniques were combined to reduce PAPR without degrading the BER performance. The simulation result shows better trade-off between PAPR and BER performance with FBMC based VLC system. In [9], the author presented Filter bank multicarrier with offset quadrature amplitude modulation (FBMC/OQAM) introduced efficient PAPR reduction based on the number of subcarriers gives a suitable solution of nonlinear distorted output. In [10], the author presented PTS based optimization technique to reduce PAPR. One of the main drawbacks in conventional PTS technique is overlapping of adjacent data that cannot directly applied to FBMC/OQAM system so that sub-optimal solution depends on discrete PSO is introduced. In [11], nonlinear companding technique is proposed to overcome the degradation of BER performance and PAPR issue. It was determined that the logarithmic rooting combined with linear codes offers better PAPR reduction gain of nearly 5 dB and coding gain around 2 dB. In [12], FBMC and Universal Filtered Multicarrier (UFMC) waveform performance analysis have been studied based on number of subcarrier and various types of modulation techniques to reduce the PAPR problem and compare the simulation results with OFDM technique. In [13], FBMC used nonlinear companding technique such as logarithmic rooting, tangential rooting, A-law companding, μ-law companding and error function. Nonlinear companding is one of the most efficient technique to reduce PAPR. The result shows that μ-law provides better result interms of BER and PAPR when compared to other nonlinear companding technique in FBMC signal waveform.

When the literature is investigated, it is found that the SLM based on Modified Forest Optimization Algorithm (MFOA) method has been successfully applied used to solve the high PAPR problem of FBMC signals and also to increase the system performance.

1.2 Motivation

Though every PAPR reduction approach does seem to have distinct benefits in different aspects, the SLM is a very actual and standard being used in various applications. Among the most important technique SLM is a straightforward, dependable, and distortion-free phase factor optimization scheme for lowering PAPR. The standard SLM determining the best phase factors possible rotation that reduces PAPR significantly, but one of the complicated issue is identification of lowest PAPR. This method creates by multiplying the input signals with a pair of signals by twisting phase factors ranging from [0, 2]. These resulting signals are just the similar signal data, and with the lowest PAPR one of the value is preferred [14]. Furthermore, an extensive verification of all allowed phase sequence combinations is required, which adds to the search complexity. Advanced techniques have recently applied to the phase factors search complexity problem in order to achieve minimal computational effort for the PAPR reduction 5 G applications. The main objective is to make the SLM a more sophisticated scheme capable of reaching better solutions with fewer searches, initiated a unique modified forest optimization algorithm(MFOA) to optimize the sequence of phase factors. By using MFOA instead of random blind experiments to find the better phase sequence achieve high PAPR reduction performance and BER improvement with the proposed technique called SLM-MFOA [15].

A new technique based on the Isosceles Trapezoidal Distribution Transform (ITDT) is proposed to reduce the PAPR of FBMC/OQAM transceiver by transforming original Gaussian distributed FBMC signals into an Isosceles Trapezoidal, while maintaining a constant average output power level. However, it is limited mainly by its high peak to average power ratio (PAPR), which can significantly reduce system efficiency [16].

A PAPR reduction scheme that improves the Partial Transmit Sequence (PTS) to an enhanced PTS (EPTS), then employs a low complexity search to determine the optimal combination of phase factors. Despite its impressive advantages over Orthogonal Frequency Division Multiplexing (OFDM), the FBMC/OQAM system exhibits high Peak-to-Average Power Ratio (PAPR) and due to its high sensitivity to timing errors, OFDM scheme becomes less robust in handling huge synchronization scenarios, which could plague its performance if it is not properly managed [17].

Further motivating factor for conducting this research is the occurrence of a significant gap in the literature regarding BER improvement and PAPR reduction studies conducted for the FBMC waveform. This paper will make an important contribution toward closing this gap. Motivated by the chance of trying to eliminate the above-mentioned SLM deficit caused by the random search strategy, this paper developed a novel modified version of SLM scheme called selective mapping based modified forest optimization algorithm (SLM-MFOA) for the FBMC signal.

1.3 Contributions

The main contributions ensured by this paper to the literature survey are given below:

For the FBMC system, the SLM schemes are utilized to realize an efficient PAPR reduction for 5 G application. Further PAPR reduction is achieved on employing the optimization techniques as well as conventional reduction of PAPR improvement techniques

SLM based MFOA technique applied on the FBMC power spectral density (PSD), PAPR reduction performance and BER performance achievements the simulations were used to test the waveform.

By using MFOA-based phase optimization,the phase rotation operation is accomplished to minimize PAPR by fixing the MFOA into the conventional SLM so that better solutions can be found without conducting so many more investigations in the SLM scheme.

To determine PSD performance, PAPR and BER performance, the proposed SLM-MFOA procedure outperforms the existing SLM scheme and also various optimization techniques such as Ant Colony Optimization (ACO), Binary Bat Algorithm (BBA), and Discrete Particle Swarm Optimization (DPSO).

1.4 Novelties

The following are the novelties of this paper:

Our paper’s first innovation is the creation of a novel modified variant of the standard forest optimization algorithm.

This research to the need for DFOA is different, which is previously not applied for the performance betterment in either SLM or any traditional PAPR reduction strategies.

Another innovative aspect of this paper is the first incorporation of SLM based MFOA scheme to a FBMC transmitter in order to reduce the PAPR of the FBMC signal.

Paper Organization: In Section.2 states the PAPR formulation of FBMC signal. Section.3 describes the implementation of SLM based PAPR reduction of FBMC. The proposed modified forest optimization algorithm is explained in Section.4. In Section.5 illustrates the proposed SLM based MFOA for PAPR reduction is described. Finally Section.6 briefs the simulation results and Conclusion of the paper is stated in Section.7

2 Papr formulation of the FBMC signal

The FBMC waveform is being considered as a candidate waveform for 5 G wireless transmissions. It has a more efficient spectral design or a well localised spectrum. The proposed block schematic in Fig. 1 depicts the transmitter operations performed to generate the FBMC signal. The signal generation process begins with the conversion of binary numbers into modulated symbols [18]. The resulting symbol sequence from the Quadrature Amplitude Modulation (QAM) is then converted into a set of FBMC sub-bands as follows, where X_l denotes input symbol, N₁ is the subcarrier. The typical baseband continuous-time model of the FBMC-OQAM modulated signal is provided by the following formula:

Fig. 1

Block schematic of FBMC transmitter.

$X_{l} = [X_{l}^{1}, X_{l}^{2}, \dots . X_{l}^{N_{1}}]$ (1)

Where $, X_{l}^{n}$ is the lth symbol complex input on the nth subcarrier which can be expressed as, $X_{l}^{n} = α {X_{l}^{n}} + j β {X_{l}^{n}}, 1 ⩽ n ⩽ N_{1}, 1 ⩽ l ⩽ L,$ (2)

Where α{ . } and β{ . } are the real and imaginary parts respectively.

In the time domain, the real and imaginary parts of the symbols are separated by half of the symbol duration T₁. Following that the symbols is passed through the prototype filter, the transmitted block of data is obtained by adding all of the filters subcarriers [19]. The ‘L’ data output time-domain signal x (t₁) usually forms as blocks can be written as, $\begin{matrix} x (t_{1}) = \sum_{l = 1}^{L} \sum_{n = 1}^{N_{1}} \\ {α {X_{t}^{n}} h (t_{1} - {lT}_{1}) + j β {X_{t}^{n}} h (t_{1} - {lT}_{1} - \frac{T_{1}}{2})} e^{jn θ l, n} \end{matrix}$ (3)

Where, $[θ_{l, n} = \frac{π}{2} (l + n)]$ is a phase parameter

By using the prototype filter, inter symbol interference(ISI) is significantly suppressed and side lobe is minimised with the help of a filter bank. The proportion of the filter impulse response time period and duration of multicarrier symbol T₁ is said as oversampling factor K₁.

After all, the PAPR of multicarrier signals is defined as the ratio of the maximum and mean power of the transmit power over the data block period. $PAPR (dB) = 10 \log_{10} \frac{\begin{matrix} maximum \\ 0 ⩽ n ⩽ N_{1} \end{matrix} {| x (t_{1}) |}^{2}}{E [{| x (t_{1}) |}^{2}]}$ (4)

The most commonly used criterion for evaluation of PAPR reduction performance is the Complementary Cumulative Distribution Function (CCDF). The associated function can be stated in the following manner: $\begin{matrix} CCDF (PAPR) = P_{r} {PAPR (x (t_{1})) > {PAPR}_{0}} \\ = 1 - P_{r} {PAPR (x (t_{1})) > {PAPR}_{0}} = 1 - {(1 - e^{- γ})}^{N} \end{matrix}$ (5)

Where, γ = PAPR₀.

3 SLM technique based PAPR reduction

The basic principle of the SLM technique is shown in Fig. 2 to generate a number of different FBMC signal representations using data blocks, and the sequence with the lowest possible PAPR will be chosen for transmission [20]. To be more specific, using array multiplication data block ‘b’ is multiplied by U′ different phase sequence vectors in the first step. Let u′ represent the uth phase vector, where [u′ = 1, 2, … U′].

Fig. 2

Block schematic of SLM technique.

The u’th FBMC candidate vector is denoted by x (u′), which is numerically represented as, and by modifying the original data block it can be generated with phase sequence vector u′.

The input dataset is transferred into many sub-blocks and converted to parallel by the converter to select the smallest blocks, which can be denoted as, $S (k) = {[S_{0} (k), S_{1} (k), S_{2} (k), \dots ., S_{N - 1} (k)]}^{T}$ (6)

The input symbol sequence is created by multiplying the input data sequence by the phase sequence, Where, Pf^u′ is the rotation factor that rotates the phase sequence. ${Rf}^{u^{'}} = IFFT (R \otimes P^{(f μ^{'})})$ (7)

Where, $P^{μ^{'}} = {[{pf}_{0}^{μ^{'}}, {pf}_{1}^{μ^{'}}, {pf}_{3}^{μ^{'}}, \dots \dots \dots . ., {pf}_{N - 1}^{μ^{'}}]}^{T}$ (8)

In the order of phases: $| {pf}_{0}^{μ^{'}} | = 1, (n^{'} = 0, 1, N - 1) .$ (9)

To avoid the complexity of sophisticated multiplication or to incorporate untreated and treated data, ±1 is commonly used. The provided data is then multiplied by the phase shift U. The following is the order of Phase U: $R^{μ^{'}} = [R_{0} {pf}_{μ^{'}, 0,} R_{1} {pf}_{μ^{'}, 1,}, R_{2} {pf}_{μ^{'}, 2,} \dots . . R_{N - 1} {pf}_{μ^{'}, N - squo 1}]$ (10) where (μ′ = 0, 1, U– 1).

After comparing the U′ data sequences x (u′) the most ideally mapped x with the lowest PAPR is selected. $\hat{r} = \underset{0 ⩽ U^{'} ⩽ U^{'}}{argmin} [PAPR (R^{(f μ^{'})})]$ (11)

Following the completion of the last FBMC procedure, the transmission signal obtained from the FBMC block’s outcome will have a reduced PAPR [21].

4 Proposed scheme for PAPR reduction based on SLM-MFOA

The algorithm states how the MFOA algorithm can be utilized to investigate optimum phase factors based on SLM system as shown in Fig. 3.

Fig. 3

Flowchart for proposed SLM based MFOA.

The growing and seeding process of trees in the forest inspired this FOA algorithm. This FOA consists of three major stages:

Local tree seeding operation,

Population limiting operation and

Global tree seeding.

FOA, like other evolutionary algorithms, begins with a trees population, with each tree representing a solution to the problem. Age is a numerical variable at the beginning stage the age of the tree is fixed to zero attached to each tree and characterizes the age of the specific tree;. Following tree initialization, the local seeding operator will produce new trees from the tree of age is equal to zero. The time of life of the tree is one year older than that of other newly planted trees.

During the operation of population limiting, trees that increase the forest’s area limit are accumulated to form the applicant population for the global seeding phase. It introduces a unique potential solution from the selected population to obtain the optimum solution. The trees in the forest are now consistently rated based on fitness values [22].

4.1 Proposed SLM based modified forest optimization algorithm

In this section, the PAPR reduction of FBMC signal based on SLM using MFOA is elaborately discussed below. The generated phase sequences for the phase rotation process at random manner in the SLM scheme consist of 1 and – 1, the MFOA, which is useful for binary optimization space, to our problem by treating 0 and 1 as minus one as well as one [23]. The proposed SLM based MFOA strategy includes rather than phase vectors generating random to attain the optimal solution. From the phase sequences, SLM based on MFOA optimization process is used to find the best one that are represented by the following tree positions:

$\begin{matrix} M^{(s)} (k) = [a^{(s)} (0), a^{(s)} (1), \dots \dots, \\ a^{(s)} (N - 1), {age}^{(s)}], s = 1, 2, 3, \dots . . S \end{matrix}$ (12)

Here, a^(s) (k)∈ { - 1, 1 }

Where, S- is a variable parameter and specifies the current population of trees in the forest.

age^(s)-is the age value of the Sth tree in the tree population, denoted by M^(s) (k).

A detailed explanation of how to optimize the phase vectors in the SLM scheme using the MFOA is provided in the following steps [24]:

The tree population is initially initialized by creating an area limit number of random trees. As a result, the area limit is equal to the initial value of ‘S’. For each of the initial trees, the value of age^(s) is then set to 0.

Fitness value is determined based on this below equation

\begin{matrix} Fitness (f^{(s)} (k)) = \\ \begin{matrix} \max \\ 0 ⩽ n ⩽ HN - 1 \end{matrix} {| \frac{1}{\sqrt{N}} \sum_{k = 0}^{N - 1} S (k) . M^{(s)} (k) . e^{\frac{j 2 π kn}{HN}} |}^{2}, s = 1, 2, \dots . S \end{matrix}

(13)

For, M^(s) (k) ifage^(s) = 0 the LSC can be expressed as, $\begin{matrix} H^{(φ, s)} (k) = \begin{matrix} flip \\ 0 ⩽ k ⩽ N - 1 \end{matrix} (M^{(s)} (k), 1), \\ φ = 1, 2, \dots . LSC, s = 1, 2, \dots . . S \end{matrix}$ (14)

Where, $\begin{matrix} flip \\ 0 ⩽ k ⩽ N - 1 \end{matrix} (M^{(s)} (k)$ reverse the sign of (M^(s) (k)) in the range 0 ⩽ k ⩽ N - 1. The overall majority of young trees created in the local seeding phase for the parent tree ‘L’ with the forest ‘0’ age would be equal to L . LSC. In Equation (14). H^(φ,s) (k) denotes the tree population.

Fitness calculation of new trees can be expresses as follows [23],

\begin{matrix} Fitness (H^{(φ, s)} (k)) = \begin{matrix} \max \\ 0 ⩽ n ⩽ HN - 1 \end{matrix} \\ {| \frac{1}{\sqrt{N}} \sum_{k = 0}^{N - 1} S (k) . H^{(φ, s)} (k) . e^{\frac{j 2 π kn}{HN}} |}^{2}, \\ φ = 1, 2, \dots LSC; s = 1, 2, \dots S \end{matrix}

(15)

The newly created young trees, denoted by H^(φ,s) (k), are added to the population. As a result, S = (arealimit) + L . LSC is the forest tree number.

The old people to whom the ages exceed the ‘life time’ parameter are retrieved from the prevailing tree population to form a candidate population subset.

Depending on the fitness abilities the residual population individuals with ages or even less identical to the life time variable are selected. In forest, the of number of trees is adjusted in accordance with S ⩽ arealimit. To form candidate population remaining section, if outside of the trees available of the area limit after trying to sort, the associated trees are removed from the forest and the maximum population size is set to the area limit [26].

Let us consider, B^(γ) (k) is the preferred trees and ‘G’ be the number of selected trees then the uses of GSC of B^(γ) (k) can be expressed as,

D^{(γ)} (k) = \begin{matrix} flip \\ 0 ⩽ k ⩽ N - 1 \end{matrix} (B^{(γ)} (k), GSC), γ = 1, 2, \dots . G

(16) where,

\begin{matrix} flip \\ 0 ⩽ k ⩽ N - 1 \end{matrix} (B^{(γ)} (k), GSC)

, reverse the sign of GSC in the range 0 ⩽ k ⩽ N - 1 in γth candidate is expresses as,

B^{(γ)} (k) = [b^{(γ)} (0), b^{(γ)} (1), \dots \dots, b^{(γ)} (N - 1), {age}^{(γ)}]

The related dimensions are chosen at random from the range ⩽k ⩽ N - 1.

Where, D^(γ) (k) denotes the γth global seeding phase new solution. The newly generated trees are then incorporated into the foremost population. As a result, the highest number of trees in the forest becomes arealimit + G. The most recent trees population number equals S ⩽ arealimit ⩽ G. Finally, for newly added trees set the age values of ‘G’ to 0 [27].

Based on the fitness function final population is sorted. To avoid discarding population solution in next iteration of optimization method the current best tree its age is set to 0 after identifying the best solution.

The average number of fitness determinations (NFD) is computed as end criteria in SLM based MFOA then the algorithm is terminated. Else, the optimization process continues from the LSC step.

This section discusses the proposed MFOA pseudo code for PAPR reduction of FBMC signal [28, 29].

Input: Local Seeding Changes (LSC), Global Seeding Changes (GSC), lifetime, area limit and transfer rate.

Output: The best tree is displayed, along with the order of changeable structure.

Forest Initialization

Initialize by tree populating the forest with random trees by producing n-dimensional vectors.

For a D-dimensional problem, each tree is noted as a vector to keep its age value

Set the “Age” of each tree to “0.”

If the stop condition is not met, perform do.

Local Seeding

Perform local seeding operation to the age 0 trees

For i = 1:LSC

Find a value at random from the selected tree.

Alter the tree’s selected variable at random from 1 to 0

End for

Increase the individual age value of all trees by 1excluding the newly generated trees.

Population Limiting

Remove trees older than the “lifetime” parameter and add the extracted trees to the candidate population.

Trees are sorted based on their fitness value.

Eliminate extra trees from the forest’s edge that exceed the “area limit” parameter and add the extra trees to the population of candidates

Global seeding

Select “transfer rate” as a fraction of the candidate population.

For each chosen individual do

From the chosen dimension select the “GSC” variable randomly

Replace the value of each variable from 1 to 0

End for

Update of recent best tree

Sorting trees based on their fitness value

Evaluate the best fitness value

Reset the age of recent greatest tree to 0

End while

Show the best

The flowchart of MFOA for PAPR reduction in FBMC signal using SLM technique is depicted in the above Fig. 3.

5 Simulation results

In this section to express the effectiveness of proposed SLM based MFOA method is compared with the existing SLM techniques based on PAPR and BER performance through the simulation results.. It is established through the Matrix Laboratory (MATLAB) simulations that, significant PAPR improvements are obtained through the MFOA-SLM scheme in the FBMC signal. In the simulations, SLM based MFOA the performance improvement guaranteed by the phase optimization is compared with SLM based Ant Colony Optimization (ACO), SLM based Artificial Bee Colony (ABC), SLM based Genetic Algorithm (GA) to optimize the phase factors in point of their BER performance and PAPR reduction achievements comparison are considered in this research work. In addition, the performance analysis of FBMC system for the proposed schemes is evaluated based on BER FBMC signal degradations and reduction of PAPR. The simulation parameters for FBMC system PAPR reduction applied in this paper are mentioned clearly in Table 1.

Table 1
Simulation Parameters for FBMC system PAPR reduction

Parameters Values

FFT length 1024

Modulation type 16-QAM

Number of subcarriers(N) 128

Channel type Additive White Gaussian Noise (AWGN)

Oversampling factor parameter(L) 8

Sampling frequency 15.48MHz

Filter model Dolph-Chebyshev

Nonlinear HPA type SSPA

Search number (SN) 512

Cyclic prefix length(l_cp) 64

Parameters	Values
FFT length	1024
Modulation type	16-QAM
Number of subcarriers(N)	128
Channel type	Additive White Gaussian Noise (AWGN)
Oversampling factor parameter(L)	8
Sampling frequency	15.48MHz
Filter model	Dolph-Chebyshev
Nonlinear HPA type	SSPA
Search number (SN)	512
Cyclic prefix length(l_cp)	64

In Fig. 4, shows the CCDF curves of proposed SLM based MFOA plot in order to determine PAPR performance of the FBMC waveform. Here, the PAPR performance of proposed SLM based MFOA technique is compared with FBMC original signal, FBMC with SLM technique, SLM based ACO, SLM based GA and SLM based ABC method. By comparing the PAPR reduction performance the FBMC original signal obtained the value of 11.2 dB, FBMC with SLM technique achieved 8.5 dB. For SLM-ACO, SLM-GA and SLM-ABC obtained the value of 8.2 dB, 7 dB and 6 dB respectively. The PAPR obtain great reduction by using proposed SLM based MFOA technique to satisfy the future 5 G requirements.

Fig. 4

Comparison of proposed SLM based MFOA with other existing methods in terms of PAPR reduction performance.

As seen in Fig. 5, the BER curves of proposed SLM based MFOA plot in order to determine PAPR performance of the FBMC waveform. Here, the BER performance of proposed SLM based MFOA technique is compared with FBMC original signal, FBMC with SLM technique, SLM based ACO, SLM based GA and SLM based ABC method. The BER reduction performance of the FBMC original signal obtained the value of 18 dB, FBMC with SLM technique achieved 15.9 dB. For SLM-ACO, SLM-GA and SLM-ABC obtained the value of 15 dB, 14.2 dB and 14 dB respectively.

Fig. 5

Comparison of proposed SLM based MFOA with other existing methods in terms of BER reduction performance.

In Fig. 6, the SLM based MFOA method was analyzed for six area limit values starts from 5 to 30 obtained by determining the value of the area limit parameter values of 5, 10, 15,20, 25 and 30. The PAPR values achieved for area limit assessments at CCDF = 10–3 are equivalent to 5.08 dB, 5.12 dB, 5.23 dB, 5.3 dB, 5.32 dB and 5.45 dB respectively. Better performance is achieved based on the values of area limit in SLM based MFOA proposed technique which offers far PAPR reduction.

Fig. 6

Proposed SLM based MFOA performance for different values of area limit.

The result of the transfer rate parameter on proposed SLM based MFOA performance was evaluated in Fig. 7 by setting specific values towards the related parameter. In this Figure, Eight transfer rate values were tested in order to obtain better result in which value absolutely suitable for the better performance based on the transfer rate by using proposed technique. The increase in transfer rate starts from 5 to 40 results great reduction in PAPR reduction. The SLM based MFOA obtain the PAPR values of 5.06 dB, 5.12 dB, 5.18 dB, 5.23 dB, 5.27 dB, 5.33 dB, 5.38 dB, and 5.42 dB, under the transfer rate 5, 10,15, 20, 25, 30, 35 and 40 respectively.

Fig. 7

CCDF performance of SLM based MFOA performance for different values of transfer rate.

The PAPR reduction performance on proposed SLM based MFOA performance was evaluated in Fig. 8 by setting specific values towards the related parameter evaluated based on different values of LSC. The CCDF curves for associated figure is thoroughly investigated can be seen that the various LSC obtained values are separated with a distinct difference in the horizontal direction. The LSC values greater than 3 is set to category one group and the values from 2 to 6 is set to the category two obtained the values 5.12 dB, 5.19 dB, 5.27 dB, 5.38 dB, 5.59 dB, and 5.74 dB, under the LSC values 1, 2, 3,4, 5 and 6 respectively.

Fig. 8

CCDF performance of Proposed SLM based MFOA performance for the effect of LSC.

The PAPR reduction performance of SLM based MFOA was evaluated in Fig. 9 based on different values of GSC. The associated figure is thoroughly investigated can be seen that the CCDF curves for different GSC obtained values are separated with a distinct similarities in horizontal direction. The GSC values increased from 1 leads to worse performance and the values from 2 to 6 leads to slight degradation performance obtained the values of 5.06 dB, 5.11 dB, 5.17 dB, 5.23 dB, 5.27 dB and 5.32 dB respectively.

Fig. 9

Proposed SLM based MFOA performance for different values of GSC.

In Fig. 10, clearly depicts the parameter of different values of life time ranging from one to eight to analyze the PAPR reduction performance difference in the proposed SLM based MFOA technique. To be more specific, how far the performance of SLM based MFOA can be improved the life time parameter value was determined. The CCDF curve obtained the values of 5.06 dB, 5.12 dB, 5.18 dB, 5.23 dB, 5.27 dB, 5.33 dB, 5.38 dB, and 5.74 dB, under the transfer rate 1, 2, 3, 4, 5, 6, 7 and 8 respectively.

Fig. 10

CCDF performance of Proposed SLM based MFOA performance for lifetime parameter.

Figure 11 illustrates the PAPR reduction of real signal and PAPR reduced signal using SLM based optimization technique. Figure 12 displays the BER improvements for SLM-ACO, SLM-GA and SLM-ABC obtained the value of 7 dB, 8 dB and 9 dB IBO values.

Fig. 11

FBMC shows the PAPR reduction.

Fig. 12

BER improvement of proposed scheme for various IBO with existing comparison performance.

Table 2 compares the search complexities of the SLM-based procedures under consideration. Furthermore, the expressions for search complexity given in the PAPR values obtained by the search number column in this table adjusting the associated search complexities to search number SN = 512.

Table 2

SLM based techniques search complexity performance study

Techniques	Search number	PAPR(dB)
FBMC Original [24]	0	11.2
SLM [25]	K=512	8.5
SLM-GA [22]	Gen. no*pop size = 16x32 = 512	7
SLM-ACO [7]	Max. iteration no.*pop size = 40x30 = 1200	8.2
SLM-ABC [23]	Max. iteration no*pop size = 30x30 = 900	6
SLM-MFOA(Proposed)	NFD=512	5.2

6 Conclusion

In this paper developed MFOA optimization technique and combined the related optimization technique to the SLM to obtain a modified version as SLM based MFOA for the effective PAPR reduction of FBMC signal waveform. In this technique phase sequence optimization in the SLM method is utilized to enhance the conventional SLM PAPR reduction performance method. In the simulation result the proposed SLM based MFOA technique performed better performance in terms of BER and PAPR compared with other existing techniques such as SLM based ACO, SLM based GA and SLM based ABC method and it clearly proved that the SLM – MFOA is a promising technique for PAPR reduction to be employed in the FBMC waveform with its powerful capabilities for 5 G application is considered in this paper.

References

Moon

Ji-Hyun

, Nam

Ye-Ri

and Kim

Jeong-Ho

, PAPR reduction in the FBMC-OQAM system via segment-based optimization, IEEE Access 6 (2018), 4994–5002.

Van der Neut, Nuan, Maharaj

Bodhaswar T.J.

, De Lange

Frederick

, González

Gustavo J.

, Gregorio

Fernando

and Cousseau

Juan

, PAPR reduction in FBMC using an ACE-based linear programming optimization, EURASIP Journal on Advances in Signal Processing 2014(1) (2014), 1–21.

Zongmiao

, Zhou

Lingyu

, Chen

Yiou

and Ling

Xiang

, Low-complexity PTS scheme for PAPR reduction in FBMC-OQAM systems, IEEE Communications Letters 22(11) (2018), 2322–2325.

Wang

Han

, Wang

Xianpeng

, Xu

Lingwei

and Du

Wencai

, Hybrid PAPR reduction scheme for FBMC/OQAM systems based on multi data block PTS and TR methods, IEEE Access 4 (2016), 4761–4768.

Daiming

, Lu

Shixian

and Jiang

Tao

, Multi-block joint optimization for the peak-to-average power ratio reduction of FBMC-OQAM signals, IEEE Transactions on Signal Processing 61(7) (2013), 1605–1613.

Kaiming

Liu

, Jundan

Hou

, Peng

Zhang

and Yuan’an

Liu

, PAPR reduction for FBMC-OQAM systems using P-PTS scheme, The Journal of China Universities of Posts and Telecommunications 22(6) (2015), 78–85.

Kumar, Ch, Karpurapu

Amit Bindaj

and Singh

Yash Pal

, Reduction of PAPR for FBMC-OQAM system using Ant Colony Optimisation technique, Soft Computing 26(9) (2022), 4295–4302.

Salama

Gerges M.

, Abdalla

Haitham F.

, Mohamed

Amira A.

, Hassan

Emad S.

, Dessouky

Moawad I.

, Khalaf

Ashraf A.M.

, El-Emary

Atef

and Elsafrawey

Amir S.

, PAPR reduction technique for FBMC based visible light communication systems, IET Communications (2022).

Lei

, Xue

Lunsheng

, Chen

Xihong

and Yuan

Dizhe

, Partial transmit sequence based on discrete particle swarm optimization with threshold about PAPR reduction in FBMC/OQAM system, IET Communications 16(2) (2022), 142–150.

10.

Mobin

Anam

and Ahmad

Anwar

, PAPR and BER analysis of coded FBMC-OQAM system with non-linear companding techniques, Engineering Research Express (2022).

11.

Vamsi

T. Sairam

, Terlapu

Sudheer Kumar

and Krishna

M. Vamshi

, PAPR Analysis of FBMC and UFMC for 5G Cellular Communications, In Intelligent Data Engineering and Analytics: Proceedings of the 9th International Conference on Frontiers in Intelligent Computing: Theory and Applications (FICTA 2021), vol. 266, p. 351, Springer Nature, 2022.

12.

Shaheen

Imad A.

, Zekry

Abdelhalim

, Newagy

Fatma

and Ibrahim

Reem

, Performance evaluation of PAPR reduction in FBMC system using nonlinear companding transform, ICT Express 5(1) (2019), 41–46.

13.

Wang

Han

, A hybrid PAPR reduction method based on SLM and multi-data block PTS for FBMC/OQAM systems, Information 9(10) (2018), 246.

14.

Cheng

Xing

, Liu

Dejun

, Shi

Wenzhe

, Zhao

Yang

, Li

Yang

and Kong

Dejin

, A novel conversion vector-based low-complexity SLM scheme for PAPR reduction in FBMC/OQAM systems, IEEE Transactions on Broadcasting 66(3) (2020), 656–666.

15.

van der Neut

Nuan

, Maharaj

Bodhaswar T.

, de Lange

Frederick H.

, Gonzalez

, Gregorio

and Cousseau

, PAPR reduction in FBMC systems using a smart gradient-project active constellation extension method, In 2014 21st International Conference on Telecommunications (ICT), pp. 134–139. IEEE, 2014.

16.

Shaheen

Imad

and Zekry

Abdelhalim

, Efficient PAPR Reduction Method for FBMC based Systems Using Isosceles Trapezoidal Distribution Transforms.

17.

Ango

Mahaman Noura Issaka

, Mwangi

Elijah

and Idrissa

Moctar Mossi

, Minimizing the PAPR for FBMC/OQAM Signals Using Enhanced-PTS with Low Complexity Search, Journal of Communications 17(6) (2022).

18.

Deng

Honggui

, Ren

Shuang

, Liu

Yan

and Tang

Chengying

, Modified PTS-based PAPR reduction for FBMC-OQAM systems, In Journal of Physics: Conference Series, 910(1) (2017), 012057. IOP Publishing.

19.

Jirajaracheep

Panya

, Mata

Tanairat

and Boonsrimuang

Pisit

, In PAPR reduction in FBMC-OQAM systems using trellis-based D-SLM with ABC algorithm, 2020 17th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), pp. 506–509. IEEE, 2020.

20.

Dongjun

and Choi

Kwonhue

, PAPR reduction scheme for FBMC-OQAM without side information, In 2019–2019 IEEE International Conference on Communications (ICC) , pp. 1–6. IEEE, 2019.

21.

Rujiprechanon

Latthawit

, Boonsrimuang

Pisit

, Sanpan

Sunisa

and Boonsrimuang

Pornpawit

, Proposal of sub-optimum algorithm for trellis-based SLM reducing PAPR of FBMC-OQAM signals, In 2018 international workshop on advanced image technology (IWAIT), pp. 1–4. IEEE, 2018.

22.

Taşpınar

Necmi

and Şimşir

Şakir

, Advanced SLM scheme based on discrete forest optimization algorithm for PAPR minimization in UFMC waveform, Wireless Networks 27(2) (2021), 1353–1368.

23.

Roberge

Jean-Michel

, Öhman

Karin

, Lämås

Tomas

, Felton

Adam

, Ranius

Thomas

, Lundmark

Tomas

and Nordin

Annika

, Modified forest rotation lengths: Long-term effects on landscape-scale habitat availability for specialized species, Journal of Environmental Management 210 (2018), 1–9.

24.

Siying

, Zhao

Junhui

, Yang

Lihua

and Li

Qiuping

, Genetic algorithm based bilayer PTS scheme for peak-to-average power ratio reduction of FBMC/OQAM signal, IEEE Access 8 (2020), 17945–17955.

25.

Sidiq

Sumina

, Sheikh

Javaid Ahmad

, Mustafa

Farhana

, Malik

Bilal Ahmad

and Sofi

Ishfaq Bashir

, PAPR minimization of FBMC/OQAM scheme by hybrid SLM and PTS using artificial: bee-Colony phase— Optimization, Arabian Journal for Science and Engineering 46(10) (2021), 9925–9934.

26.

Jirajaracheep

Panya

, Mata

Tanairat

and Boonsrimuang

Pisit

, A PAPR Reduction in FBMC-OQAM System via Overlapping-PTS with Artificial Bee Colony Phase Optimization, International Journal of Intelligent Engineering and Systems 12(5) (2019), 249–256.

27.

Prasad

D. Rajendra

, Tamil

and Chourasia

Bharti

, PAPR Reduction for FBMC-OQAM Signals Using PSO-Based JPTS Scheme, In Innovations in Electronics and Communication Engineering, pp. 29–38. Springer, Singapore, 2022.

28.

Bulusu

S.S. Krishna Chaitanya

, Shaiek

Hmaied

and Roviras

Daniel

, Potency of trellis-based SLM over symbol-by-symbol approach in reducing PAPR for FBMC-OQAM signals, In 2015 IEEE International Conference on Communications (ICC), pp. 4757–4762. IEEE, 2015.

29.

Ming-Wei

, Xu

Dong-Yang

, Geng

Jing

and Hong

Wei-Chiang

, A ship motion forecasting approach based on empirical mode decomposition method hybrid deep learning network and quantum butterfly optimization algorithm, Nonlinear dynamics 107(3) (2022), 2447–2467.