Usage Of Networks Research Articles

Self-Presentation of Teachers in Social Networks

This study was undertaken to investigate the use and motivations behind teachers' engagement in social networks, while exploring their self-presentations and attitudes within these digital communities. The aim was to shed light on the relationship between real-life and digital communities by focusing on teachers as a specific demographic group. To address the research questions, namely the demographic characteristics of teachers using social networks, the reasons behind teachers' usage of social networks, the role of these platforms in shaping teachers' self-perception and professional identity, a mixed-method approach was employed. The findings of the study indicate that social networks provide an ideal platform for individuals to present themselves, and teachers exhibit distinct patterns of behavior compared to other groups. Teachers express themselves more openly and assertively when their identities are anonymous. They actively seek attention to expand their social network, yet refrain from making political posts to avoid jeopardizing their online presence. They strive to portray an idealized image of themselves and make efforts to distinguish themselves from others. Based on the study's conclusions, it is recommended to develop a digital platform that caters to the interests, needs, and inclinations of teachers.

Recurrent Neural Networks based Online Behavioural Malware Detection Techniques for Cloud Infrastructure

: With the increased basis of applications and services on Cloud, the dependence on cloud technologies has expanded rapidly. This transition eases some of the hardware management troubles, but it also exposes broad security threats, especially in regard to malware. Malware is a major threat to cloud computing systems mainly in IaaS dominated environments. In this paper we explore the usage of Recurrent Neural Networks, RNNs in malware detection in cloud VMs. In particular, we examine two of the most popular RNN models; Long short term Memory, LSTM and Bidirectional RNNs, BIDI, that try to understand malware patterns by watching the activity of systems and the use of CPU, Memory and Disk. On the dataset of 40,680 samples 40,680 malware and benign samples, the models were trained on process level features from real malware in a free and open cloud environment. Both models achieved detection rates of over 99% in average key performance metrics. Moreover, wealso investigate the impact of different representations of input data on the performance of the models. The results of the study clearly support the viability of RNNs for online malware detection in cloud infrastructure, and allow for high accuracy and real time detection.

Application of feedforward and recurrent neural networks for model-based control systems

AbstractIn this paper, a new study concerning the usage of artificial neural networks in the control application is given. It is shown, that the data gathered during proper operation of a given control plant can be used in the learning process to fully embrace the control pattern. Interestingly, the instances driven by neural networks have the ability to outperform the original analytically driven scenarios. Three different control schemes, namely perfect, linear-quadratic, and generalized predictive controllers were used in the theoretical study. In addition, the nonlinear recurrent neural network-based generalized predictive controller with the radial basis function-originated predictor was obtained to exemplify the main results of the paper regarding the real-world application.

Diagnosing skin cancer using social spider optimization (SSO) and error correcting output codes (ECOC) with weighted hamming distance

Skin cancer is a common disease resulting from genetic defects, and early detection is critical to improve treatment outcomes. Diagnostic programs that use computer aid especially those that use supervised learning are very useful in early diagnosis of skin cancer. This research therefore presents a new approach that integrates optimization methods with supervised learning to improve skin cancer diagnosis using machine vision approach. The presented method is initiated by data pre-processing that involves the removal of unnecessary data. Then, to segment the images, a combination of K-means clustering and social spider optimization technique is employed. The region of interest is then extracted from the segmented image and from this region a convolutional neural network extracts the significant features. To enhance the classification performance as compared with the standard classifiers, this research introduces a new concept of error correcting output codes coupled with a weighted Hamming distance in the group of gamma classifiers. The ability of the proposed approach in segmentation of skin lesions and classifying them was tested using samples from the ISIC-2017 and ISIC-2016 databases. The introduced method obtained state-of-the-art accuracy on both datasets (ISIC-2016: 97.10%, ISIC-2017: 95.17%). In particularly, the accuracy of the introduced approach for both these databases is at least 1.17% higher than the compared methods. This proves the high performance of the suggested method based on the usage of the convolutional neural networks for feature extraction and gamma classifiers with error correcting output codes for classification in skin cancer detection.

Shielding networks: enhancing intrusion detection with hybrid feature selection and stack ensemble learning

The frequent usage of computer networks and the Internet has made computer networks vulnerable to numerous attacks, highlighting the critical need to enhance the precision of security mechanisms. One of the most essential measures to safeguard networking resources and infrastructures is an intrusion detection system (IDS). IDSs are widely used to detect, identify, and track malicious threats. Although various machine learning algorithms have been used successfully in IDSs, they are still suffering from low prediction performances. One reason behind the low accuracy of IDSs is that existing network traffic datasets have high computational complexities that are mainly caused by redundant, incomplete, and irrelevant features. Furthermore, standalone classifiers exhibit restricted classification performance and typically fail to produce satisfactory outcomes when dealing with imbalanced, multi-category traffic data. To address these issues, we propose an efficient intrusion detection model, which is based on hybrid feature selection and stack ensemble learning. Our hybrid feature selection method, called MI-Boruta, combines mutual information (MI) as a filter method and the Boruta algorithm as a wrapper method to determine optimal features from our datasets. Then, we apply stacked ensemble learning by using random forest (RF), Catboost, and XGBoost algorithms as base learners with multilayer perceptron (MLP) as meta-learner. We test our intrusion detection model on two widely recognized benchmark datasets, namely UNSW-NB15 and CICIDS2017. We show that our proposed IDS outperforms existing IDSs in almost all performance criteria, including accuracy, recall, precision, F1-Score, false positive rate, true positive rate, and error rate.

The Potential of Artificial Intelligence in Teaching a Foreign Language at a University

The rapid development of artificial intelligence, neural networks, naturally gives impetus to the development and modernization of the education sector. New technologies practically affect the processes of teaching. The purpose of the study is to test the practical effect of the technology of interaction between a teacher, students and artificial intelligence. The task is solved to develop and describe the techniques, algorithm and results of the work of a teacher and students with the integration of the artificial intelligence technologies into the educational process. This shows the theoretical and practical significance of this study. To eliminate negative situations of uncontrolled usage of neural networks in the educational process, to prevent the situations of academic deception, unethical usage of artificial intelligence, it is proposed to develop the techniques of a teacher’s work with a study group that will transform negative factors of the usage of artificial intelligence by students into positive ones, bringing situations of recourse to artificial intelligence into a teacher-controlled format of its use. The technique is proposed that prevents academic deception: the use of artificial intelligence for educational purposes through a teacher-supervision when using artificial intelligence by a student. This is the novelty of the study presented. The research methods are: the analysis of special literature and methodological foundations of teaching AI, as well as verbal-communicative and empirical methods. As a result of the research, the author proposed the list of principles of using artificial intelligence in the educational process is proposed, two functions of artificial intelligence while learning and teaching foreign languages are identified. It is stated, that the phenomenon of the emergence of a new subject of educational process – artificial intelligence – can be observed and the triad of the subjects of educational process has been defined – “Teacher – Artificial Intelligence – Student”. Moreover, the author describes the experience of testing the use of artificial intelligence technologies and gives practical recommendations on their use in the educational process at the university, describes the techniques and formats of tasks given for students. This will further enhance the tools and functionality of digital platforms used in the educational process based on artificial intelligence in general and on the mechanism of neural networks in particular. Keywords: artificial intelligence, English language, principles of teaching, techniques of teaching, artificial intelligence functions, the triad of the educational process

Integrating Cost-Effective Measurements and CFD Modeling for Accurate Air Quality Assessment

Assessing air quality in urban areas is vital for protecting public health, and low-cost sensor networks help quantify the population’s exposure to harmful pollutants effectively. This paper introduces an innovative method to calibrate air-quality sensor networks by combining CFD modeling with dependable AQ measurements. The developed CFD model is used to simulate traffic-related PM10 dispersion in a 1.6 × 2 km2 urban area. Hourly simulations are conducted, and the resulting concentrations are cross-validated against high-quality measurements. By offering detailed 3D information at a micro-scale, the CFD model enables the creation of concentration maps at sensor locations. Through regression analysis, relationships between low-cost sensor (LCS) readings and modeled outcomes are established and used for network calibration. The study demonstrates the methodology’s capability to provide aid to low-cost devices during a representative 24 h period. The precision of a CFD model can also guide optimal sensor placement based on prevailing meteorological and emission scenarios and refine existing networks for more accurate urban air quality representation. The usage of cost-effective air quality networks, high-quality monitoring stations, and high-resolution air quality modeling combines the strengths of both top-down and bottom-up approaches for air quality assessment. Therefore, the work demonstrated plays a significant role in providing reliable pollutant monitoring and supporting the assessment of environmental policies, aiming to address health issues related to urban air pollution.

DEVELOPMENT OF LOW-LATENCY WIRELESS COMMUNICATION SYSTEMS FOR AUTONOMOUS VEHICLES

This research aims to explore the effectiveness of the L3 protocol in improving the V2X communication systems in key performance areas such as low communication delay, high-speed protocols and the effectiveness under real-life conditions. According to the L3 protocol simulation of the research, which employs platforms including COOJA and NS-3, the latency of the networks is lowered compared to that of the conventional DSRC/LTE system at below one millisecond levels adopting the time slots and the capture effect. These challenges observed on other systems such as delays experienced due to high-speed data transfer are addressed in the protocol. The performed measurements prove that the L3 protocol of the system exhibits reliability and scalability depending on the number of vehicles on the road as well as the network load, and retains its workability even with the number of 225 automobiles. These results conform with prior works that underlined the necessity of using sophisticated and low-latency relative communication systems for the usage of an autonomous vehicle network. In doing so, the study shows a future of L3 protocol in enhancing the current V2X communication systems in aspects of low latency, fast data transfer and competency in various conditions. It is suggested that future research include extensive field trials, the increase of security measures, the analysis of the scalability and integration of the new technologies, and the solution of the interoperability issues for the further investigation of the possibilities of the application of the L3 protocol.

A Multiobjective Genetic Algorithm to Evolving Local Interpretable Model-Agnostic Explanations for Deep Neural Networks in Image Classification

Deep convolutional neural networks have become a dominant solution for numerous image classification tasks. However, a main criticism is the poor explainability due to the black-box characteristic, which hurdles the extensive usage of deep convolutional neural networks. To address this issue, this paper proposes a new evolutionary multi-objective based method, which aims to explain the behaviours of deep convolutional neural networks by evolving local explanations on specific images. To the best of our knowledge, this is the first evolutionary multi-objective method to evolve local explanations. The proposed method is model-agnostic, i.e. it is applicable to explain any deep convolutional neural networks. ImageNet is used to examine the effectiveness of the proposed method. Three well-known deep convolutional neural networks -VGGNet, ResNet, and MobileNet, are chosen to demonstrate the modelagnostic characteristic. Based on the experimental results, it can be observed that the local explanations are understandable to end-users, who need to check the sensibility of the evolved explanations to decide whether to trust the predictions made by the deep convolutional neural networks. Furthermore, the local explanations evolved by the proposed method improves the confidence of deep convolutional neural networks making the predictions. Lastly, the pareto front and convergence analyses indicate that the proposed method can form a good set of nondominated solutions.

CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization.

Deep convolutional neural networks are shown to be overkill with high parametric and computational redundancy in many application scenarios, and an increasing number of works have explored model pruning to obtain lightweight and efficient networks. However, most existing pruning approaches are driven by empirical heuristics and rarely consider the joint impact of channels, leading to unguaranteed and suboptimal performance. In this article, we propose a novel channel pruning method via c lass-aware t race r atio o ptimization (CATRO) to reduce the computational burden and accelerate the model inference. Utilizing class information from a few samples, CATRO measures the joint impact of multiple channels by feature space discriminations and consolidates the layerwise impact of preserved channels. By formulating channel pruning as a submodular set function maximization problem, CATRO solves it efficiently via a two-stage greedy iterative optimization procedure. More importantly, we present theoretical justifications on convergence of CATRO and performance of pruned networks. Experimental results demonstrate that CATRO achieves higher accuracy with similar computation cost or lower computation cost with similar accuracy than other state-of-the-art channel pruning algorithms. In addition, because of its class-aware property, CATRO is suitable to prune efficient networks adaptively for various classification subtasks, enhancing handy deployment and usage of deep networks in real-world applications.

Hybrid quantum architecture for smart city security

Currently and in the near future, Smart Cities are vital to enhance urban living, address resource challenges, optimize infrastructure, and harness technology for sustainability, efficiency, and improved quality of life in rapidly urbanizing environments. Owing to the high usage of networks, sensors, and connected devices, Smart Cities generate a massive amount of data. Therefore, Smart City security concerns encompass data privacy, Internet-of-Things (IoT) vulnerabilities, cyber threats, and urban infrastructure risks, requiring robust solutions to safeguard digital assets, citizens, and critical services. Some solutions include robust cybersecurity measures, data encryption, Artificial Intelligence (AI)-driven threat detection, public–private partnerships, standardized security protocols, and community engagement to foster a resilient and secure smart city ecosystem. For example, Security Information and Event Management (SIEM) helps in real-time monitoring, threat detection, and incident response by aggregating and analyzing security data. To this end, no integrated systems are operating in this context. In this paper, we propose a Hybrid Quantum-Classical Architecture for bolstering Smart City security that exploits Quantum Machine Learning (QML) and SIEM to provide security based on Quantum Artificial Intelligence and patterns/rules. The validity of the hybrid quantum-classical architecture was proven by conducting experiments and a comparison of the QML algorithms with state-of-the-art AI algorithms. We also provide a proof of concept dashboard for the proposed architecture.

Evaluating the performance of multilingual models in answer extraction and question generation

Multiple-choice test generation is one of the most complex NLP problems, especially in languages other than English, where there is a lack of prior research. After a review of the literature, it has been verified that some methods like the usage of rule-based systems or primitive neural networks have led to the application of a recent architecture, the Transformer architecture, in the tasks of Answer Extraction (AE) and Question Generation (QG). Thereby, this study is centred in searching and developing better models for the AE and QG tasks in Spanish, using an answer-aware methodology. For this purpose, three multilingual models (mT5-base, mT0-base and BLOOMZ-560 M) have been fine-tuned using three different datasets: a translation to Spanish of the SQuAD dataset; SQAC, which is a dataset in Spanish; and their union (SQuAD + SQAC), which shows slightly better results. Regarding the models, the performance of mT5-base has been compared with that found in two newer models, mT0-base and BLOOMZ-560 M. These models were fine-tuned for multiple tasks in literature, including AE and QG, but, in general, the best results are obtained from the mT5 models trained in our study with the SQuAD + SQAC dataset. Nonetheless, some other good results are obtained from mT5 models trained only with the SQAC dataset. For their evaluation, the widely used BLEU1-4, METEOR and ROUGE-L metrics have been obtained, where mT5 outperforms some similar research works. Besides, CIDEr, SARI, GLEU, WER and the cosine similarity metrics have been calculated to present a benchmark within the AE and QG problems for future work.

Breast cancer relapse disease prediction improvements with ensemble learning approaches

Diagnosis and prognosis are especially difficult areas of medical research related to cancer due to the high incidence of breast cancer, which has surpassed all other cancers in terms of female mortality. Another factor that has a substantial influence on the quality of life of cancer patients is the fear that they may experience a relapse of their disease. The objective of the study is to give medical practitioners a more effective strategy for using ensemble learning techniques to forecast when breast cancer may recur. This research aimed to investigate the usage of deep neural networks (DNNs) and artificial neural networks (ANNs) in addition to machine learning (ML) based approaches, including bagging, averaging, and voting, to enhance the efficacy of breast cancer relapse diagnosis on two breast cancer relapse datasets. Results from the empirical study demonstrate that the proposed ensemble learning-enabled approach improves accuracies by 96.31% and 95.81%, precisions by 96.70% and 96.15%, sensitivities by 98.88% and 98.68%, specificities by 84.62% in both, F1-scores by 97.78% and 97.40%, and area under the curve (AUCs) of 0.987 and 0.978, with University Medical Centre, Institute of Oncology (UMCIO) and Wisconsin prognostic breast cancer (WPBC) datasets respectively. Consequently, these improved disease outcomes may encourage physicians to use this model to make better treatment choices.

Model of Shape Memory Alloy Actuator with the Usage of LSTM Neural Network.

Shape Memory Alloys (SMAs) are used to design actuators, which are one of the most fascinating applications of SMA. Usually, they are on-off actuators because, in the case of continuous actuators, the nonlinearity of their characteristics is the problem. The main problem, especially in control systems in these actuators, is a hysteretic loop. There are many models of hysteresis, but from a control theory point of view, they are not helpful. This study used an artificial neural network (ANN) to model the SMA actuator hysteresis. The ANN structure and training method are presented in the paper. Data were generated from the Preisach model for training. This approach allowed for quick and controllable data generation, making experiments thoroughly planned and repeatable. The advantage and disadvantage of this approach is the lack of disturbances. The paper's main goal is to model an SMA actuator. Additionally, it explores whether and how an ANN can describe and model the hysteresis loop. A literature review shows that ANNs are used to model hysteresis, but to a limited extent; this means that the hysteresis loop was modelled with a hysteretic element.

The Impact of Foreign Stock Market Indices on Predictions Volatility of the WIG20 Index Rates of Return Using Neural Networks

AbstractThe paper investigates the issue of volatility of stock index returns on the Warsaw Stock Exchange (WIG20 index returns volatility). The purpose of this review is to compare how other stock market indexes as HANG SENG, NIKKEI 225, FTSE 250, DAX, S&P 500 and NASDAQ 100 influance the volatility of WIG20 index returns. The innovation of this work is the usage of a new neural network with three different activation functions to predict future volatility of WIG20 index returns. The input for this network is the last 3 values of WIG20 index returns volatility and the last 3 values of one of the considered foreign index returns volatility. As measurements for the best forecasting performance of neural networks are taken common used forecast errors: ME (mean error), MPE (mean percentage error), MAE (mean absolute error), MAPE (mean absolute percentage error), RMSE (root mean square error). The study shows that the Polish stock market is mainly influenced by the European and US markets.

Random forest with differential privacy in federated learning framework for network attack detection and classification

Communication networks are crucial components of the underlying digital infrastructure in any smart city setup. The increasing usage of computer networks brings additional cyber security concerns, and every organization has to implement preventive measures to protect valuable data and business processes. Due to the inherent distributed nature of the city infrastructures as well as the critical nature of its resources and data, any solution to the attack detection calls for distributed, efficient and privacy preserving solutions. In this paper, we extend the evaluation of our federated learning framework for network attacks detection and classification based on random forest. Previously the framework was evaluated only for attack detection using four well-known intrusion detection datasets (KDD, NSL-KDD, UNSW-NB15, and CIC-IDS-2017). In this paper, we extend the evaluation for attack classification. We also evaluate how adding differential privacy into random forest, as an additional protective mechanism, affects the framework performances. The results show that the framework outperforms the average performance of independent random forests on clients for both attack detection and classification. Adding differential privacy penalizes the performance of random forest, as expected, but the use of the proposed framework still brings benefits in comparison to the use of independent local models. The code used in this paper is publicly available, to enable transparency and facilitate reproducibility within the research community.

Theory for Equivariant Quantum Neural Networks

Quantum neural network architectures that have little to no inductive biases are known to face trainability and generalization issues. Inspired by a similar problem, recent breakthroughs in machine learning address this challenge by creating models encoding the symmetries of the learning task. This is materialized through the usage of equivariant neural networks the action of which commutes with that of the symmetry. In this work, we import these ideas to the quantum realm by presenting a comprehensive theoretical framework to design equivariant quantum neural networks (EQNNs) for essentially any relevant symmetry group. We develop multiple methods to construct equivariant layers for EQNNs and analyze their advantages and drawbacks. Our methods can find unitary or general equivariant quantum channels efficiently even when the symmetry group is exponentially large or continuous. As a special implementation, we show how standard quantum convolutional neural networks (QCNNs) can be generalized to group-equivariant QCNNs where both the convolution and pooling layers are equivariant to the symmetry group. We then numerically demonstrate the effectiveness of a SU(2)-equivariant QCNN over symmetry-agnostic QCNN on a classification task of phases of matter in the bond-alternating Heisenberg model. Our framework can be readily applied to virtually all areas of quantum machine learning. Lastly, we discuss about how symmetry-informed models such as EQNNs provide hopes to alleviate central challenges such as barren plateaus, poor local minima, and sample complexity. Published by the American Physical Society 2024

Emerging applications and ongoing challenges in WSN assisted IoT networks

In current scenario, Wireless Sensor Networks (WSN) based on Internet of Things (IoT) have seen a significant growth in research area. IoT is a network that enables sensors, and other events to interact with each other without human interference. In addition, wireless network is an essential part of the Internet of Things, generating an abundance of real-time applications. WS Ns can be recognized by their distributed sensor nodes and application fields including environmental monitoring, healthcare, agriculture, industrial automation, and smart cities. Therefore, an extensive usage of such networks is the outcome of their adaptability in variety of fields. The present study digs into the immense scope of WSN applications and the daunting challenges that come alongside. However, these applications face a slew of obstacles, including those related to energy, scalability, security, quality of service, network dependability, and data management. Further, the study focuses on gathering insights into the changing WSN technology environment by examining these applications and the issues they encounter, allowing to satisfy the demands of both existing and forthcoming deployments. As WSN technology is improving, the present study contributes by investigating prospective solutions, such as machine learning integration, energy harvesting, and standardization efforts.

Improved segmentation of brain tumors with data augmentation using MU-Net

Semantic segmentation is crucial for the treatment and prevention of brain cancers. Several neural network–based strategies were rapidly presented by research groups to enhance brain tumor thread segmentation. The tumor’s uneven form necessitates the usage of neural networks for its detection. Therefore, improved patient outcomes may be achieved with precise segmentation of brain tumor. Brain tumors can range widely in size, form, and position, making diagnosis difficult. Thus, this work offers a Multi-level U-Net (MU-Net) approach for analyzing the brain tumor data augmentation for improved segmentation. Therefore, a significant amount of data augmentation is employed to successfully train the recommended system, removing the problem of a lack of data when using MR images for the diagnosis of multi-grade brain cancers. Here, we presented the “Multi-Level Pyramidal Pooling (MLPP)” component, where a new pyramidal pool will be employed to capture contextual data for augmentation. The “High-Grade Glioma” (HGG) datasets from the Kaggle and BraTs2021 were used to assess the proposed MU-Net. Overall Tumor (OT), Enhancing Core (EC), and Tumor Core (TC) were the three main designations to be segmented. The dice score was used to contrast the results empirically. The suggested MU-Net fared better than most existing methods. Researchers in the fields of bioinformatics and medicine might greatly benefit from the high-performance MU-Net.

Digital Profiler Based On A Low-Cost 3D-Scanner To Evaluate The Hydraulic Performance Of Homogeneous Low-Crested Structures

In order to evaluate the hydraulic performance of breakwaters, mechanical profilers were first used in wave flumes. Vertical bars and rolling wheels have been used to track the breakwater shape. Once surveyed, useful hydraulic parameters were able to be measured (e.g., envelope shape and eroded area). Nevertheless, the methodologies based on mechanical profilers have some limitations: they are intrusive, require specific equipment and are time-consuming. On the other hand, non-intrusive laser scanners can also be used; recent studies proved the feasibility of non-intrusive fast methods of surveying with 3D-scanning, (see Musumeci et al., 2018). These instruments are fast and reliable at capturing the shape of the breakwater in real-time. However, this method did not consider the distortion caused by light refraction. Laser scanners usually require to measure models in dry conditions, which is not efficient in time and resources. Regardless of the drawbacks when using non-intrusive laser scanners as digital profilers, the distortion can be corrected. The data suits for the usage of Neural Networks (NN) that can be trained to correct distortion. This study is focused on Homogeneous Low-Crested Structures (HLCS) which is a new typology of Low-Crested Structures (LCS), made out large quarry stones or concrete armor units that are suitable to reduce shoreline erosion on degraded coastlines and natural environments. HLCS can be classified as reef-type breakwaters and are considered an environmentally-friendly solution that fulfils two purposes: to protect beaches nearby and to enhance coral regeneration and marine colonization, (see Medina et at., 2020). For the correct design of HLCS or conventional LCS, it is required to evaluate both the hydraulic stability and wave transmission (see van der Meer and Daemen, 1994). HLCS are structures without core which may erode under intense wave attack. Breakwater damage and crest freeboard must be estimated depending on incident wave conditions. The breakwater envelope evolves and finds an equilibrium state related to the incident wave conditions in which the crest freeboard and the transmitted wave energy levels may change. The goal of this study is to provide a low-cost non-intrusive methodology based on 3D-scanning to study the hydraulic performance of mound breakwaters, specifically for Cubipod HLCS. This surveying procedure allows to obtain the real-magnitude shape of the envelope of the HLCS, even with a certain water level. The methodology mimics previous surveying techniques and provides relevant hydraulic and structural parameters (crest freeboard, damage, envelope shape, etc.) that can be then used to study the hydraulic performance of breakwaters.