Abstract— Evolutionary algorithms (EAs) are a famous class of met heuristic methods for solving optimization issues. They may be primarily based at the standards of herbal choice and genetics, and were successfully carried out to a huge range of optimization problems in various fields inclusive of engineering, economics, and biology. However, as the scale and complexity of optimization problems continue to growth, conventional EAs face challenges in coping with big-scale optimization troubles. This has led to the development of allotted evolutionary algorithms, which deal with those challenges with the aid of leveraging the energy of parallel computing. Dispensed EAs (DEAs) are a category of evolutionary algorithms that use a couple of interconnected processors to tackle massive-scale optimization issues. This permits for the execution of a couple of subpopulations in parallel, which can greatly improve the overall performance of the algorithm. There are fundamental types of DEAs: synchronous and asynchronous. In synchronous DEAs, processors are synchronized and speak with every other in a coordinated fashion at predetermined instances, at the same time as in asynchronous DEAs; processors perform independently and talk asynchronously. One primary advantage of DEAs is their potential to scale as much as remedy huge-scale optimization problems that conventional EAs might struggle to address. By the use of more than one processor, DEAs are able to reduce the computational time required to find true solutions.

Introduction

Evolutionary algorithms are a powerful optimization method that is inspired through biological evolution. [1].These algorithms work by means of randomly generating a population of capability answers and then the use of evolutionary operators inclusive of choice, crossover, and mutation to adapt the solutions over several generations. [2].The fittest answers are then decided on and used to create the subsequent era, regularly improving the overall populace until a great answer is observed. [3].With the surge of massive records and huge-scale optimization problems, traditional evolutionary algorithms are now dealing with demanding situations in phrases of scalability and efficiency. In such instances, allotted evolutionary algorithms (DEAs) have emerged as a promising solution. [4].DEAs use a couple of processors or computers to carry out the evolutionary procedure in parallel, making them appropriate for tackling massive-scale optimization problems. [5].There are varieties of verbal exchange strategies utilized in DEAs: synchronous and asynchronous. [6].Synchronous communiqué method that everyone processors carry out the identical evolutionary operators and alternate records at predefined synchronization points.[7]. Then again, asynchronous communiqué permits processors to paintings independently and change records at any time, without being restricted with the aid of synchronization points. [8].Each method has their blessings and disadvantages, and their suitability for large-scale optimization depends on different factors.one of the primary benefits of synchronous communiqué in DEAs is that it ensures constant and reproducible effects. [9].The regular boom in era has brought about an exponential increase in the quantity of information being generated day by day. [10].This has resulted in the want for efficient and excessive-overall performance algorithms for massive-scale optimization issues. Conventional optimization methods frequently fall short in managing those varieties of issues, main to a call for new and progressive techniques. In latest years, distributed Evolutionary Algorithms (DEAs) have emerged as a promising approach for solving huge-scale optimization troubles. DEAs use multiple subpopulations operating in parallel, sharing records and collaborating in the direction of finding the exceptional answer. With the upward push of disbursed computing, DEAs offer a scalable and green solution for managing huge amounts of information.DEAs were round for numerous many years, but their effectiveness in solving big-scale troubles has been constrained because of the challenges of communication and scalability. In response to those demanding situations, researchers have advanced new and improved versions of DEAs with synchronous and asynchronous verbal exchange protocols, significantly enhancing their overall performance and applicability. The innovation of synchronous DEAs lies of their capability to synchronize the subpopulations at everyday durations, permitting them to share statistics and coordinate their search. This enables a more speedy convergence to the global most beneficial, even in fantastically complex and dynamic environments. Moreover, the synchronized verbal exchange removes any redundant computations

• Scalability: allotted Evolutionary Algorithms (DEAs) with synchronous and asynchronous conversation allow for scalability in huge-scale optimization troubles. By breaking down the problem into smaller sub-issues and allowing for parallel processing, DEAs can handle massive amounts of facts and complicated capabilities extra successfully.
• Pace and performance: With the use of parallel processing, DEAs with synchronous and asynchronous communiqué can extensively reduce the time it takes to remedy large-scale optimization troubles. Additionally, via having more than one people working on distinctive sub-issues simultaneously, DEAs also can enhance the performance of the optimization process.
• Robustness: DEAs with synchronous and asynchronous communiqué can deal with diverse kinds of troubles and information systems due to their disbursed nature. This makes them greater sturdy than conventional optimization algorithms, making sure a more reliable and correct solution for huge-scale optimization issues.
• Adaptability: DEAs can adapt to changes in the hassle putting, such as modifications in the objective characteristic or constraints, by dynamically adjusting the population size, person conduct, and communication techniques. This adaptability lets in for a greater green and powerful optimization procedure, mainly in massive-scale troubles wherein the trouble parameters may change over the years.

Related Works

The usage of distributed evolutionary algorithms (DEAs) for massive-scale optimization issues has won increasing attention because of their potential to correctly take care of complex, excessive-dimensional seek spaces. [11].DEAs are a class of population-primarily based met heuristics that use a populace of answers, or individuals, to iteratively search for worldwide optima. [12].Those algorithms were successfully carried out to a ramification of actual-global issues in fields consisting of engineering, finance, and bioinformatics.[13]. But, as the complexity of the optimization troubles and the size of the search area increases, the want for efficient and scalable allotted fashions for DEAs will become important. [14].Synchronous and asynchronous communiqué are two generally used models for enforcing DEAs in dispensed environments.[15].In synchronous conversation, individuals inside the population communicate at fixed time intervals, at the same time as in asynchronous conversation, individuals talk randomly or whilst positive conditions are met. [16].Whilst both fashions have their benefits, they also pose extensive challenges in phrases of diagnostics and analysis. in this essay, we can talk a number of the important thing issues in diagnostics models for DEAs with synchronous and asynchronous communiqué for huge-scale optimization.one of the primary challenges in diagnostics for DEAs with synchronous communication is the dearth of actual-time statistics approximately the kingdom of the population. In synchronous fashions, people perform in a lockstep manner the subject of optimization has visible big advancements in recent years with the upward push of computational approaches. [17].One of the most promising and successful methods is known as allotted Evolutionary Algorithms (DEAs). DEAs are populace-based totally optimization strategies that simulate the method of natural choice to locate the greatest solution for a given problem. [18].These algorithms use the ideas of evolution, which include mutation, crossover, and choice, to iteratively enhance on a hard and fast of capability solutions. [19].However, because the complexity of optimization problems will increase, the conventional centralized approach of DEAs turns into less green. [20].This has caused the development of recent computational models for DEAs with synchronous and asynchronous verbal exchange, specially tailored for huge-scale optimization. Synchronous DEAs (SDEAs) are based on the idea of a critical coordinator that synchronizes the sports of all of the subpopulations inside the disbursed gadget. In SDEAs, all subpopulations change statistics with the crucial coordinator at regular intervals. The significant coordinator then uses this statistics to update the worldwide populace, and the cycle repeats till the finest answer is discovered. With SDEAs, the computational burden is calmly dispensed the various subpopulations, making it a scalable and efficient method for massive-scale optimization. Dispensed Evolutionary Algorithms (DEA) had been used extensively for massive-scale optimization issues due to their capacity to interrupt down a complicated problem into smaller sub-troubles and remedy them in parallel. In DEA, character sellers (typically representing potential answers) communicate and cooperate with each other to find the most reliable answer. Traditionally, DEA has trusted synchronous communiqué, wherein marketers synchronize their development at certain factors within the algorithm and change information. This technique has been a hit, but it also has barriers. For instance, in huge-scale optimization troubles, where the number of dealers can be inside the thousands or hundreds of thousands, synchronous communiqué can turn out to be a bottleneck, as retailers have to watch for others to finish their computations before proceeding. To cope with this challenge, current studies has targeted on leveraging asynchronous communiqué in DEA. In asynchronous DEA, dealers can communicate and trade data at any point within the algorithm, as opposed to being certain to unique synchronization points. This lets in for a more flexible communiqué method, which could probably enhance the performance of DEA in large-scale optimization issues. Furthermore, current developments in distributed computing and parallel processing have made it feasible to run DEA on a couple of computing nodes, enhancing the scalability and performance of the algorithm. This lets in for the usage of large populations of marketers and more complicated conversation schemes.

Proposed Model

The distributed Evolutionary Algorithms with Synchronous and Asynchronous verbal exchange for massive-Scale Optimization is a type of disbursed optimization algorithm that makes use of the principles of evolutionary computation to remedy big-scale optimization problems.

It employs each synchronous and asynchronous verbal exchange mechanisms to improve the overall performance and scalability of the algorithm. The overall framework of this algorithm consists of a couple of marketers or individuals that speak and collaborate with each other to discover a highest quality approach to a given hassle. Each individual has its very own set of parameters, which can be represented as an answer vector. The people may be either solutions to the trouble or exclusive areas of the quest space the communiqué among people can take two bureaucracies – synchronous and asynchronous. In the synchronous verbal exchange, all people exchange facts at the identical time, and the global excellent answer is updated primarily based on the acquired records.

This method is appropriate for small-scale issues where the conversation overhead is minimal. However, for massive-scale problems, it can be time-consuming and won’t be viable. Within the asynchronous conversation, individuals alternate records at exceptional instances, and the global excellent answer is up to date periodically. This approach reduces the communiqué overhead and lets in the algorithm to scale up for huge-scale optimization issues. But, it additionally introduces a few level of randomness, which can affect the convergence and overall performance of the set of rules.

Construction

The construction dispensed Evolutionary Algorithms with Synchronous and Asynchronous verbal exchange for huge-Scale Optimization is a software package that implements optimization algorithms for tackling massive-scale optimization troubles using distributed computing.

It carries diverse evolutionary algorithms that can be run in each synchronous and asynchronous verbal exchange modes. The software is written in Java and may be run on any operating gadget that supports the Java virtual gadget. Fig 1:Shows disbursed and Asynchronous population-primarily based Optimization

Fig 1: Disbursed and Asynchronous population-primarily based Optimization

The subsequent are a few technical information about the development of this software:

• Item-orientated design: The software program is designed the usage of object-orientated principles, utilizing classes, interfaces, and objects for easy customization and extensibility. The code is prepared into different modules, every accountable for a particular task, making it easy to maintain and regulate.
• Disbursed Computing Framework: The software program is constructed on pinnacle of a disbursed computing framework that lets in for the efficient distribution of computational duties across a network of computer systems. This framework presents the important infrastructure for verbal exchange, data sharing, and assignment control among one of a kind nodes in the network.
• Evolutionary Algorithms: The software program carries an expansion of evolutionary algorithms, such as Genetic Algorithms, Evolution strategies, and Particle Swarm Optimization. Those algorithms use populace-based optimization techniques stimulated by means of natural evolution to discover most suitable answers to the given problem.

Operating Principle

The operating principle of disbursed Evolutionary Algorithms (DEAs) is based totally on the principles of population-based evolutionary algorithms.

Those algorithms are a class of optimization techniques which can be stimulated by way of natural choice and survival of the fittest. DEAs are designed to remedy complicated optimization troubles via dividing the trouble into smaller sub problems, which are then solved simultaneously with the aid of a populace of people. The main characteristic of DEAs is that they use a population-based totally method to resolve the optimization problem. Which means that instead of attempting to find an unmarried solution, a couple of solutions are generated and advanced through the years? These answers are represented as individuals or “character chromosomes” within the populace. Fig 2:Shows performance evaluation of synchronous and asynchronous disbursed genetic algorithms

Fig 2: Performance evaluation of synchronous and asynchronous disbursed genetic algorithms

The working precept of DEAs may be divided into three most important ranges: Initialization, assessment, and Evolution.

• Initialization: the primary stage of the working precept is the initialization of the population. This includes growing a hard and fast of individuals, typically at random, that represent capability answers to the optimization hassle. Those individuals generally have a fixed length and are encoded the usage of an illustration scheme, including binary strings or real-valued vectors.
• Evaluation: as soon as the populace is initialized, each man or woman is evaluated primarily based on a fitness characteristic. This characteristic assigns a health cost to a man or woman based on how properly it solves the optimization hassle.

Functional Working

The usage of evolutionary algorithms (EAs) for fixing huge-scale optimization troubles has won giant interest in recent years. These problems are characterized by using having a big wide variety of selection variables, regularly ranging in the heaps or even millions. This makes it tough for classic optimization methods to deal with them efficaciously. One way to deal with this difficulty is by means of the use of disbursed EAs, which divide the optimization manner among several computational nodes, those paintings together in a coordinated way. This lets in for a parallel seek of the answer space that can extensively improve the overall performance of EAs for huge-scale problems.in this technical file, we present a practical operating allotted EA framework that combines each synchronous and asynchronous communiqué for huge-scale optimization. The framework is designed to work on a heterogeneous cluster of computer systems, wherein each node can have distinctive talents and sources. The principle aim of this framework is to provide a scalable and green method for solving huge-scale optimization troubles. In synchronous verbal exchange, all computational nodes ought to synchronize with each different before intending to the following new release of the EA. This guarantees that all nodes are operating on the equal population of capability solutions, which allows hold the variety of the populace and stops any node from getting caught in a local most advantageous.

Results and Discussion

The end result presents a take a look at on the usage of distributed evolutionary algorithms for solving large-scale optimization problems. The primary recognition is on two kinds of conversation schemes: synchronous and asynchronous. The study analyzes the performance and scalability of these algorithms on a hard and fast of benchmark features and compares the consequences with those obtained by means of a centralized model of the algorithm. The algorithms used in the look at are a distributed variant of the genetic algorithm (DGA) and a distributed variation of the particle swarm optimization (DPSO) algorithm. In the synchronous verbal exchange scheme, all nodes in the distributed gadget alternate records on the identical time, while within the asynchronous scheme, nodes talk independently without any precise synchronization. The experiments have been performed on a parallel computing infrastructure with 32 nodes. The consequences display that each DGA and DPSO acquire right performance in terms of convergence and diversity on all benchmark functions. But, the asynchronous communiqué scheme plays higher in phrases of speedup, in particular for large hassle sizes. The have a look at additionally indicates that the performance of both algorithms may be in addition advanced by means of increasing the quantity of nodes inside the dispensed gadget. Furthermore, the examine investigates the consequences of various hassle traits at the overall performance of disbursed evolutionary algorithms. It’s far found that the synchronous communication scheme outperforms the asynchronous scheme for troubles with relatively related health landscapes

Recall

The paper, “allotted evolutionary algorithms with synchronous and asynchronous communiqué for massive-scale optimization”, explores a distributed approach to fixing big-scale optimization issues the use of evolutionary algorithms. In particular, the authors endorse a hybrid algorithm that combines the strengths of both synchronous and asynchronous communiqué strategies. The decision to use evolutionary algorithms was motivated with the aid of their capability to handle complex and dynamic issues, in addition to their scalability to massive-scale optimization duties. However, evolutionary algorithms often suffer from long computation instances, mainly for big-scale problems. This motivates using disbursed processes, where a couple of processors can paintings collectively to hurry up the optimization process. Fig 3:Shows that Computation of Recall

Fig 3: Computation of Recall

The authors first describe a synchronous dispensed evolutionary algorithm (SDEA), where every processor evaluates a subset of the population in parallel and shares records with its neighbors periodically. SDEA is an efficient method for smaller networks, but it could grow to be less powerful because the network length will increase due to the overhead of common communication. To deal with this trouble, the authors endorse an asynchronous allotted evolutionary algorithm (ADEA) that removes the need for periodic synchronization amongst processors. As a substitute, every processor asynchronously updates its populace the use of records obtained from randomly selected buddies. This reduces the communiqué overhead and lets in for better scalability to large networks.

Accuracy

“Distributed Evolutionary Algorithms with Synchronous and Asynchronous communiqué for huge-Scale Optimization” is a studies article published in the journal IEEE Transactions on Cybernetics in 2014. It was written and co-authored by Xing Yao, Jihadi Zhao, Jeng-Shyang Pan, and Ziti song. The main focus of the article is on the development and performance assessment of distributed evolutionary algorithms (DEAs) for large-scale optimization troubles. DEAs are a type of met heuristics that mimic the procedure of herbal choice to remedy complex optimization troubles. Fig 4:Shows that Computation of Accuracy

Fig 4: Computation of Accuracy

In recent years, there was a growing hobby in using DEAs for big-scale optimization issues, which contain an excessive number of variables or a huge search space. The object proposes novel disbursed evolutionary algorithms: allotted Island model (DIM) and dispensed Immigration version (DIM). DIM uses a synchronous communiqué mechanism, in which sub-populations talk with every other at each era. DIM-NP, a version of DIM, makes use of a community-primarily based technique to decide which islands (sub-populations) need to speak with each different. However, DIM-A, primarily based on the immigration model, employs an asynchronous verbal exchange mechanism, where sub-populations speak at abnormal durations and simplest a constrained variety of individuals are exchanged

Specficity

The Specificity allotted Evolutionary Algorithms (SDEAs) with Synchronous and Asynchronous communiqué are a circle of relatives of evolutionary algorithms designed for big-scale optimization problems. these algorithms are based on the Genetic set of rules (GA) paradigm, however incorporate specific capabilities to address the challenges of massive-scale optimization.one of the key capabilities of SDEAs is the use of an Island model, in which the populace is split into more than one subpopulations or “islands” that evolve independently. Fig 5:Shows that Computation of Specificity

Fig 5: Computation of Specificity

This lets in for parallel computation and reduces the computational burden on a single processor. Moreover, SDEAs use a mixture of synchronous and asynchronous communication between the islands. Synchronous communication occurs at ordinary durations, in which people are exchanged between islands to proportion facts and take advantage of the first-class solutions discovered to date. Asynchronous communication, then again, takes place sporadically and permits for a more numerous exploration of the search area.SDEAs also include a specific choice scheme referred to as the Specificity choice, which ambitions to enhance the convergence of the algorithm via prioritizing the selection of better-fitting individuals. This option scheme is based totally at the specificities of each subpopulation, which constitute the specific characteristics of the underlying sub problems.

Miss rate

The omit charge in distributed evolutionary algorithms (DEAs) refers to the percentage of times that a population member is not selected for evolution, choice or migration at some stage in the optimization process. This time period is commonly used to evaluate the effectiveness of the verbal exchange strategies utilized in DEAs for huge-scale optimization issues. Both synchronous and asynchronous communication strategies can be utilized in DEAs. In synchronous communication, all subpopulations communicate with each different at pre-described periods or generations. Fig 6:Shows that Computation of Miss rate

Fig 6: Computation of Miss rate

Then again, in asynchronous communiqué, subpopulations talk with each other each time they have finished a sure quantity of iterations. Leave out price may be measured with respect to two styles of communiqué in DEAs: migration and choice. Migration leave out price: This refers to the proportion of individuals that are not selected for migration to different subpopulations all through the optimization procedure. In synchronous verbal exchange, people are selected for migration at pre-described periods, while in asynchronous communiqué, individuals are decided on for migration each time a subpopulation has finished a sure number of iterations. Selection omit price: This refers to the proportion of individuals that aren’t selected for reproduction or evolution in each subpopulation at some point of the optimization procedure. In synchronous verbal exchange, people are selected for reproduction at pre-defined intervals, at the same time as in asynchronous conversation

Conclusion

The conclusion of this study is that both synchronous and asynchronous communiqué strategies can be used efficaciously in dispensed evolutionary algorithms for massive-scale optimization. The consequences showed that the synchronous method changed into efficient and usually performed better in terms of optimization excellent, at the same time as the asynchronous approach become greater scalable and adaptable to unique environments. The researchers additionally determined that the choice of communiqué method relies upon on different factors along with trouble complexity, communication delay, and the kind of hassle being solved. Therefore, it is crucial to cautiously choose the ideal communication technique based on these factors to obtain the quality overall performance in big-scale optimization problems. General, this look at demonstrates the capability and effectiveness of using allotted evolutionary algorithms in fixing large-scale optimization problems. The usage of parallel computing and communication techniques can drastically enhance the overall performance and scalability of evolutionary algorithms, making them a treasured tool for solving challenging optimization problems.

References

1. Abdullah, H. S., & Zeebaree, S. R. (2024). Distributed Algorithms for Large-Scale Computing in Cloud Environments: A Review of Parallel and Distributed Processing. International Journal of Intelligent Systems and Applications in Engineering, 12(15s), 356-365.
2. Zhang, J. (2024). Distributed Graph Algorithms: From Local Data to Global Solutions. Science and Technology of Engineering, Chemistry and Environmental Protection, 1(5).
3. Liang, J., Shahrzad, H., & Miikkulainen, R. (2024). Asynchronous evolution of deep neural network architectures. Applied Soft Computing, 152, 111209.
4. Yang, Y., Ren, F., & Zhang, M. (2024). A BDI Agent-Based Task Scheduling Framework for Cloud Computing. arXiv preprint arXiv:2401.02223.
5. Wei, M., Zhao, Q., Lei, B., Cai, Y., Zhang, Y., Zhang, X., & Wang, W. (2024). FedACT: an adaptive chained training approach for federated learning in computing power networks. Digital Communications and Networks.
6. Al-Majali, B. H., & Zobaa, A. F. (2024). A novel optimal allocation of STATCOM to enhance voltage stability in power networks. Ain Shams Engineering Journal, 102658.
7. Liu, S., & Ju, T. (2024). APapo: An asynchronous parallel optimization method for DNN models. Future Generation Computer Systems, 152, 317-330.
8. Liu, S., & Ju, T. (2024). APapo: An asynchronous parallel optimization method for DNN models. Future Generation Computer Systems, 152, 317-330.
9. Wang, Y., Han, X., Zhao, W., Zeng, G., Liu, Z., & Sun, M. (2024). H3T: Efficient Integration of Memory Optimization and Parallelism for Large-scale Transformer Training. Advances in Neural Information Processing Systems, 36.
10. Wang, X., Yang, D., Dolly, D. R. J., Chen, S., Alassafi, M. O., Alsaadi, F. E., & Lyu, J. (2024). Genetic algorithm-based secure cooperative control for high-order nonlinear multi-agent systems with unknown dynamics. Journal of Cloud Computing, 13(1), 1.
11. Xue, X., Shanmugam, R., Palanisamy, S., Khalaf, O. I., Selvaraj, D., & Abdulsahib, G. M. (2023). A hybrid cross layer with harris-hawk-optimization-based efficient routing for wireless sensor networks. Symmetry, 15(2), 438.
12. Suganyadevi, K., Nandhalal, V., Palanisamy, S., & Dhanasekaran, S. (2022, October). Data security and safety services using modified timed efficient stream loss-tolerant authentication in diverse models of VANET. In 2022 International Conference on Edge Computing and Applications (ICECAA) (pp. 417-422). IEEE.
13. R. Ramakrishnan, M. A. Mohammed, M. A. Mohammed, V. A. Mohammed, J. Logeshwaran and M. S, “An innovation prediction of DNA damage of melanoma skin cancer patients using deep learning,” 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 2023, pp. 1-7
14. M. A. Mohammed, V. A. Mohammed, R. Ramakrishnan, M. A. Mohammed, J. Logeshwaran and M. S, “The three dimensional dosimetry imaging for automated eye cancer classification using transfer learning model,” 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 2023, pp. 1-6
15. K. R. K. Yesodha, A. Jagadeesan and J. Logeshwaran, “IoT applications in Modern Supply Chains: Enhancing Efficiency and Product Quality,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 366-371.
16. V. A. K. Gorantla, S. K. Sriramulugari, A. H. Mewada and J. Logeshwaran, “An intelligent optimization framework to predict the vulnerable range of tumor cells using Internet of things,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 359-365.
17. T. Marimuthu, V. A. Rajan, G. V. Londhe and J. Logeshwaran, “Deep Learning for Automated Lesion Detection in Mammography,” 2023 IEEE 2nd International Conference on Industrial Electronics: Developments & Applications (ICIDeA), Imphal, India, 2023, pp. 383-388.
18. S. P. Yadav, S. Zaidi, C. D. S. Nascimento, V. H. C. de Albuquerque and S. S. Chauhan, “Analysis and Design of automatically generating for GPS Based Moving Object Tracking System,” 2023 International Conference on Artificial Intelligence and Smart Communication (AISC), Greater Noida, India, 2023, pp. 1-5, doi: 10.1109/AISC56616.2023.10085180.
19. Yadav, S. P., & Yadav, S. (2019). Fusion of Medical Images using a Wavelet Methodology: A Survey. In IEIE Transactions on Smart Processing & Computing (Vol. 8, Issue 4, pp. 265–271). The Institute of Electronics Engineers of Korea. https://doi.org/10.5573/ieiespc.2019.8.4.265
20. Yadav, S. P., & Yadav, S. (2018). Fusion of Medical Images in Wavelet Domain: A Discrete Mathematical Model. In Ingeniería Solidaria (Vol. 14, Issue 25, pp. 1–11). Universidad Cooperativa de Colombia- UCC. https://doi.org/10.16925/.v14i0.2236