genetic algorithm

With the expansion of the Internet network and public access to this network, the amount of information and data exchange increases day by day; Therefore, there will be a possibility of unauthorized access to the information that is exchanged. On the other hand, almost all programs that run on the Internet, such as social networks, have full access to images and content stored on the host device; Therefore, there is a possibility of unauthorized access and theft of personal information. Therefore, the security and accuracy of information must be guaranteed. In order to maintain the confidentiality of data, information encryption methods, such as image encryption algorithms, can be used. In the proposed method of this article, the encryption algorithm key is created with the help of information extracted from the person's face, image abstract, and public key so that the encryption algorithm will be highly sensitive to the change of any of the information used in key generation. Lorenz chaos mapping is used for image encryption in the substitution phase. Each color channel of the image is divided into four parts and each part is encrypted using a separate pseudo-random sequence and the encrypted image is produced. In the permutation phase, to achieve the best-encoded image, the genetic heuristic algorithm (GA) is used to bring the pixel correlation value of the encoded image to the minimum possible value by choosing the optimal parameters for the Arnold chaos mapping. According to the obtained results, the correlation values of pixels in all three horizontal, vertical, and diagonal directions are much smaller than other similar presented methods and the encryption algorithm has been able to significantly reduce the correlation and connection between pixels. Also, according to the results of NPCR and UACI differential analysis, it is higher than 99.6% and 33.4% for all the tested images, respectively. Therefore, the proposed method will have high resistance to statistical and differential attacks. Also, for the process of decoding the image, two-step authentication and safe storage of the encryption key are used in the blockchain network.

The current research aims to provide an algorithm using artificial intelligence calculations to satisfy all the regulations of Mabhas-6, Mabhas-10, and standard 2800 of the national building regulations while minimizing the weight of the structure. In previous research, the control of constraints and the optimal design of steel moment frames were done in two dimensions and only with older algorithms. One of the most important tasks of this research is the completion of the new restrictions of regulations to control all the restrictions and optimal design of the widely used types of short and intermediate steel structures. All the regulations and controls necessary for the three main types of steel structures widely used in the country, such as 1) braced frames and 2) frames with shear walls, and 3) a dual system of moment frames and shear walls, have been implemented. Finally, the results of this algorithm were validated with a Tabriz hospital project. One of the valuable results of the work is to facilitate the safe design and control process of the steel buildings by providing a graphic panel of the model input and displaying all the essential outputs of the structure in the form of text and diagrams. The work process is intelligently programmed to be performed automatically by just inputting a standard ETABS file into the optimal seismic design process algorithm. In the ETABS file, each design group (e.g., columns, beams, braces, and walls) is defined as a list of usable sections so that the artificial intelligence algorithm can choose the best arrangement. For the three structural examples and the hospital project, we showed that in addition to design safety, significant savings between 11 to 30 percent could be made on the total steel weight, which would take several years to achieve these results for a computer without the use of artificial intelligence.

In today's digital era, security issues and cyber attacks have become a serious and attention-needed concern as they hamper secured and vital information relating to organizations or individuals. Accordingly, timely detection of these vulnerabilities made by intruders is essential, wherein the cornerstone of security ensures the user's data privacy as an intrusion detection system (IDS). On the other hand, with the rapid development of machine learning (ML) and deep learning (DL) methods in the data world, one of their significant applications is dedicated to IDS using state-of-the-art classification algorithms, which has been the subject of numerous research to enhance accuracy and reliability in recent years. As a consequence, this paper presents a hybrid model integrating feature selection, classification, and hyper-parameters optimization. First, the initial massive features are subjected separately to the modified mutual information (MMI), genetic algorithm (GA), and Anova F-value approaches, followed by extracting the common outputs as optimal and reduced final features. Subsequently, a compound CNN and LSTM classifier (CNN-LSTM) is employed, where its hyper-parameters will be determined through a random switch grey wolf-whale optimization algorithm (RS-GWO-WOA) instead of a time-consuming trial and error manual process. Ultimately, to analyze the suggested scheme, a comparison with other strategies in terms of accuracy, precision, recall, F1 score, and periods of time on the NSL-KDD dataset has been accomplished, confirming the superiority of the developed approach.

Conventional Analysis of Reinforced Concrete Slabs under blast load requires a long process. This process, in addition to being time consuming, does not determine the extent to which the geometric parameters of the slab are loaded. To overcome this problem, in this study, by Analyzing 351 different concrete slabs under blast load by LS-DYNA software, the maximum Deflection, its corresponding arrival time and Slab Stiffness were recorded. Then, by using the genetic algorithm method, by processing the data related to the maximum Deflection and its time, a direct relationship between the calculation of these parameters was extracted. Considering the nature of the relationship extracted from the Response Surface Method, the coefficients of the relationship variables presented for stiffness determined the importance of each variable in the stiffness of the slab. Finally, the cosine amplitude method has been used to determine the maximum sensitivity of the slab to geometric parameters and loading. The results of this study show that the explosive distance parameter (R) has the most effect and the cross-sectional variable of steel (AS) has the least effect on the maximum slab deflection. It has the maximum effect at the time of reaching the maximum deflection.

The Intrusion detection system (IDS) manages a massive volume of data that comprises irrelevant and redundant features, leading to more significant resource consumption, long-time training and testing procedures, and lower detection rate. Hence, Feature selection is a crucial phase in intrusion detection. The aim of this paper is to introduce an intersection-based strategy that optimally selects the features for classification. This feature selection involves an intersection of simultaneous mutual information based on the transductive model (MIT-MIT), Anova F-value, and Genetic Algorithm (GA) methods. A benchmark dataset, named NSL-KDD, is applied to evaluate the effectiveness of the proposed approach. This study includes accuracy, precision, recall, and F1 score as evaluation metrics for IDS, which analyzes the proposed method with state-of-the-art classifiers. The evaluation results confirm that our feature selection algorithm provides more essential features for IDS to achieve high accuracy, overwhelming the other comparative algorithms.

Software fault prediction methods are used to predict fault-prone modules in the early stages of software development. Machine learning techniques are the most common techniques used in software fault prediction. Data dimensionality is one of the problems that affect the performance of machine learning algorithms. Data dimensionality means the existence of irrelevant or redundant features that may mislead the learning algorithm hence decrease its accuracy. Low accuracy of software fault prediction causes late detection of some faulty modules and as a result increases the effort and cost of fixing faults abnormally. Therefore, solving data dimensionality problem is necessary to increase the accuracy of software fault prediction. Researchers use feature selection algorithms for dimensionality reduction. Feature selection algorithms are divided into two types of filter-based feature selection and wrapper-based feature selection algorithms. Wrapper-based algorithms lead to higher accuracy prediction models. In these algorithms we can use different methods to search for the best solutions that best of them is metaheuristic search. Each of the metaheuristic algorithms has some strengths and weaknesses that researchers use combination of these algorithms to address these weaknesses. In this research, to address the weaknesses of each metaheuristic algorithm, a combination of genetic, ant colony and whale optimization algorithm is used as wrapper feature selection. Obviously, the application of early software faults prediction methods before the actual test is one of the effective passive defense techniques in reducing software systems development costs. 19 software projects are used to evaluate proposed method. Comparison of the results with other methods shows that the proposed method outperforms other methods.

Adaptive steganography methods use variable embedding capacity according to the uniformity or edges of image areas. ALSBMR is an adaptive method with two main stages: Selecting suitable pixels, and embedding them using the LSBMR method. This method utilizes two adaptive keys between the sender and the receiver to determine the block rotation angle and select the embedding path. In the original method, the keys are randomly selected by the sender with no specific criteria and then sent to the receiver. The proposed method models key selection as an optimization problem and uses Genetic Algorithm (GA) and Teaching-Learning-Based Optimization (TLBO) to find the optimal keys. Two fitness functions are used to further evaluate the difference as well as the histogram difference between the cover and stego images. The results show that the image embedded with the proposed method has improved quality and security compared to the base method. Since all steganography methods require embedding keys, intelligent key selection can improve the performance of existing steganography methods.

This paper presents a new method for determining the middle band of discrete cosine transform based on the combination of genetic algorithm and tabu search. As embedded watermark is visible in low frequencies, and not resistant against attacks in high frequencies, the most suitable band for embedding is the middle band. In the Fourier domain, it can be determined by some filters but such a method does not exist in the discrete cosine domain. In discrete cosine transform the arrangement of frequency components does not have the orderliness of the Fourier transform itself, so, it is tried to find the middle band of image by Fourier transform and then find the exact location of the  middle band of discrete cosine transform by genetic algorithm and tabu search. In this method, we minimize the difference between the image that is obtained by applying the middle band of discrete cosine transform mask and the image obtained by the Fourier transform mask. Also, the proposed method can be used for low and high band of discrete cosine transform masks. The implementation and tests for low, middle and high bands resulted in 0.0062, 0.0149 and 0.0061 averages errors, respectively. These results for the test images are satisfactory.

Absorbents are used in RCS reduction to prevent reflection of the wave and not interfere with the test wave.By reducing the amount of reflection in a wave using materials used in adsorbents, they reduce the reflection.The things that are to be considered in relation to these reflections are the effects of these waves, which are termed ”pure environmental reflections“, the error caused by not considering the pure reflections of the environment and other problems caused by weaknesses. The main wave of the note is noteworthy. In this research, the optimum thickness of adsorbent material has been determined using genetic and Particle Swarm Optimization (PSO) algorithms with the aim of determining the minimum amount of electromagnetic wave reflection. The genetic algorithm is used to adjust the plate resistance values as well as the PSO algorithm to determine the optimal screen thickness. The results show that determination of microwave absorption coating parameters has a significant effect on the absorption rate of the wave. The applied method provides the conditions for achieving optimal coverage.

As computer networks grow, so attacks and intrusions to these networks are increased. In order to have a fully secure computer network, one needs ‘intrusion detection systems’ (IDS) on top of firewalls. The goal of using an IDS is to supervise the abnormal activities and differentiate between normal and abnormal activities in a host system or in a network. An efficient IDS has high detection rate while keeping a low false alarm rate. In this paper, a new approach to classify KDD-Cup-99 data set using a combination of random forest method and genetic algorithm is presented. The purpose is to increase the speed of learning and test phases while improving the accuracy. Random forest is an ensemble learning method based on decision trees. Due to its relatively simple structure and good performance, it is used in many supervised learning applications.  However, like all tree based machine learning algorithms, having too many categorical features, can be a problem both for the speed and accuracy. This is exactly the case with the problem in hand, i.e. intrusion detection; many of the features are in the form of categorical data. For example, in R language, the maximum number of definable categorical features for random forest is 53. The contribution of this work is resolving this issue with the aid of Genetic Algorithm (GA). In this research information gain as a measure of importance is defined and the number of features is reduced based on genetic algorithm.

The Closed-Circuit Television (CCTV) system is effective in a variety of applications, such as traffic monitoring, crime prevention and the safety of public sites. Therefore, the area coverage of CCTV which increases security and maintains cost reduction is a challenging issue. The typical camera insertion techniques, often use design techniques and trial and error experience, which require more time and cannot determine the optimal location of the cameras. In this paper, we have proposed a new method for optimal camera placement based on computer graphics, harmony search and genetic algorithms. The proposed method can enhance visual coverage, and can also increase environment safety and reduce implementation costs. In the proposed method, the map of the building is received as an input, then the proposed algorithm increases the coverage area by calculating and changing the angles of the cameras and finding the best location in the sensing area. Single point crossover and random mask crossover are used in the proposed method; single point crossover method improving the time of search and the random mask crossover method providing nearly optimal coverage. Also, the population diversification in random mask crossover method leads to an optimal global answer, which has been the main problem in the previous methods. The proposed method increases visual area coverage up to 40%.

Throughout the history of software engineering, the existence of software defects at the heart of a system and lack of proper treatment before operational use has always led to serious personal and financial      disasters. A test that can provide an appropriate coverage at the code-level of software can prevent many of these incidents. The basis path test is considered as the strongest coverage criterion in the white software box test. The prerequisite for a basis path testing is to have a set of test paths. The greater the number of test paths to be scanned, the greater the amount of software source code that will be covered and so more software holes will be discovered. As a result, a basic challenge before running a software path test is to produce the maximum test paths that can be scrolled. So far, some work has been done to maximize the number of scrollable test paths, including the GSO method, but the results indicate that the number of test paths can be greater than currently achieved. In this paper we have proposed a method to achieve this goal by a hybrid solution based on two evolutionary genetic and birds algorithms. The results of evaluations show that using the proposed solution has led to an increase in the number of scrollable test paths up to 91% comparing with the GSO method.

Joint optimization of spectrum sensing and spectrum access parameters of a cognitive radio sensor network (CRSN) leads to a higher sum throughput of secondary users (SUs) while the interference introduced to primary users (PUs) is kept under certain tolerable level. In this work, first, by using the concept of probabilistic spectrum access, joint spectrum sensing and power allocation is performed in a multiband CRSN. The considered optimization problem is formulated with the aim of maximizing the average opportunistic secondary data rate under constraints on the interference introduced to PU and limited power budget of SU. The considered system model leads to a non-convex optimization problem which is converted into a convex problem. Based on using genetic algorithms, optimal solution of this problem is obtained using two different approaches: i) Lagrange multipliers method and ii) Linear programming method. We provide several numerical simulation results to evaluate the performance of our proposed methods in terms of achievable CR data rate, interference introduced to the PU and convergence properties of the proposed algorithms.

Using the mathematical and optimization models has significant impact in military strategic decision making problems such as finding location of domestic fire launching site of hard and soft kill. In this paper, an integer linear programming model is developed for location problem of fire launching sites with goal of maximizing the expected value of the target accessibility and protection of strategic realms. Also, two metaheuristic algorithms based on genetic algorithm and particle swarm optimization algorithm have been designed to solve the problem. The computational results of these methods have been compared to exact answers from modeling. It is revealed that with time limit of 60 seconds, the developed genetic algorithm and particle swarm optimization have 0.16% and 0.07% average deviation from optimal solutions, indicating they perform efficiently.

Using the mathematical and optimization models has significant impact in military strategic decision making problems such as finding location of domestic fire launching site of hard and soft kill. In this paper, an integer linear programming model is developed for location problem of fire launching sites with goal of maximizing the expected value of the target accessibility and protection of strategic realms. Also, two metaheuristic algorithms based on genetic algorithm and particle swarm optimization algorithm have been designed to solve the problem. The computational results of these methods have been compared to exact answers from modeling. It is revealed that with time limit of 60 seconds, the developed genetic algorithm and particle swarm optimization have 0.16% and 0.07% average deviation from optimal solutions, indicating they perform efficiently.

In the combat management systems, mathematical and optimization models have significant impact to find good solutions for fire allocation and scheduling problems. In this paper, a linear integer programming model has been developed for a fire allocation and scheduling problems the aim of which was to maximize the expected value of the target distruction and strategic realms protection and efficient use of weapons by considering the operational constraints for weapon allocation. Since the available operation research solvers can not find the optimal solution of this problem in the large scale sizes, two metaheuristics based on genetic algorithm and chaotic particle swarm optimization was developed. Finally, based on randomnly generated test instances and extensive computation results, the performance of the developed algorithms was evaluated. The computational experiments reveal that the developed chaotic particle swarm optimization algorithm is more efficient especially in the limited and short CPU run time

In this paper, generalized Group Method of Data Handling (GMDH)-type neural network has been successfully used for modeling concrete core testing including reinforcing bars based on various data obtained experimentally. Genetic Algorithm (GA) and Singular Value Decomposition (SVD) techniques are deployed for optimal design of such model. A set of input-output data for the training and testing the evolved models are employed in which core diameter, length-to-diameter ratio, number of reinforcing bars, distance of bar axis from nearer end of core as well as strength of cores, with or without reinforcing bars, are considered as inputs and standard cube strength of concrete is regarded as the output variables. The comparison of results obtained experimentally in this work with the proposed GMDH model depicted that this model has a great ability for prediction of the concrete compressive strength on the basis of core testing. Finally, sensitivity analysis was performed on the models obtained by GMDH-type neural network to study the influence of input parameters on model output. The sensitivity analysis reveals that the output variable (standard cube strength) is significantly changed by core strength and number of rebars in comparison with other input variables.

In recent years, video-based applications have increased. Therefore, researchers are trying to make video coding techniques more effective and efficient. So, several methods are proposed to improve the quality of video against the channel error. The aim of this article is increasing video quality at the receiver. Basis of the proposed method is as follows; in a fixed rate, channel coding rate increases and using it to increase channel error robustness. Channel encoder rate increases with increasing compression rate and decreasing source data. Since the motion estimator block is unable to minimize the variance of the information frames correctly; in this paper, the secondary estimator is proposed which is applied to the information and it causes to increase source compression rate. So the combination of wavelet and neural network with genetic algorithm are used in this secondary estimator. The secondary estimator significantly reduces the information frames variance. Thus, fewer bits are needed to send information. So in this method, channel encoder rate increases without increasing transmitted data for each frame and video frames could be more robustness against to channel error. To evaluate the proposed method, we have tested different source coding rates with several SNRs and compared the results by other methods

Each channel of the phased array antenna intrinsically has the phase errors. This is because of the position and orientation error of antenna elements due to assembling tolerance or mechanical distortion, and the electric length error of each RF channel. These errors can decrease the antenna performance, for instance the gain reduction, sidelobe level enhancement, and inaccurate beam direction. In order to improve the performance of the antenna in the presence of phase errors, a phase correction method using Genetic Algorithm (GA) is proposed. By using the proposed method, the antenna overall radiation pattern is recovered close to ideal radiation pattern without error. By applying the simulation data to a large array of 32 40 elements at the frequency of S-band the effectiveness of the proposed method is verified.

Automatic detection and tracking of small target in infrared images is the great importance in the modern world. Top-Hat transformation is the important class of nonlinear transformation morphology that has wide usages in the detection and tracking the target in infrared images. The most important problem in improving the efficiency of conversion of Top-Hat transformation, the use of structural elements according to the SNR of each image. Since the images have different clutter and targets, the use of structural elements with fixed shape and dimations for images with different SNR cannot lead to accurate detection. Therefore, to improve the detection, adaptive structural elements should be used. For this purpose, in this paper, we use continuous genetic algorithm to achieve Top-Hat transformation adaptive structural elements, in order to identify more precisely target point. Qualitative and quantitative evaluation results on real images show that the proposed method compared with other methods based on morphology with fixed structural element has a better performance in target detection in infrared image and produce fewer false alarms.