Babylonian Journal of Networking

Using Artificial Intelligence to Evaluating Detection of Cybersecurity Threats in Ad Hoc Networks

Rasha Hameed Khudhur Al-Rubaye , AYÇA KURNAZ TÜRKBEN — Tue, 30 Apr 2024 00:00:00 +0000

This paper is devoted to the use of AI managed to contribute to security of the MANETs (Mobile Ad-hoc Networks), decentralized and mobile wireless networks, that are fully dynamic in nature. The intention of the research is to audit the dangers of cyber and to spot the variety of cyber threats types, including Distributed Denial of Service (DDoS) attacks, malware intrusions, leakages or data breaches, or unauthorized access attempts, using AI-powered algorithms and models. The purpose is to obtain higher degree of veracity of defining and classifying these threats and as result puts more security and reliability to MANET networks. Anomaly detection addressed as a secondary line of defense specific for MANET hardware and network traffic. The monitoring method is needed here to find abnormal behavior that might anyhow signify the possible security flaws or the attacks of the MANET environments. This ultimate goal is penetrated with the timely detection Peculiarities, which makes possible to reinforce MANET security capabilities that require to be well-developed against cyber threats. Experimental results reveal a clear trend of Fleet Grid Algorithm Improvements along with Detection Accuracy (Digital Signals and Anomaly) by means of training AI models (CNN and RF) with algorithms like Random forest and Convolutional neural networks. The machine learning based algorithms often present remarkable results comprising efficiency in detecting and effectively categorizing different cyber threats existing such as DDoS attacks, malware infiltrations and attempted unauthorized access. This method of anomaly detection is able to accurately detect robot anomalies and malicious activities in network traffic in addition to we preventing system vulnerabilities or threats from occurring prematurely. Besides, the findings of this study wide relatively efficient AI-based cybersecurity systems for dynamic decentralized MANET systems, which are developed for street-view switching and path finding, self healing and self configuration.

Survey on Neural Networks in Networking: Applications and Advancements

Jenan Ayad — Fri, 02 Aug 2024 00:00:00 +0000

The integration of neural networks into networking has paved the way for substantial advancements in network performance, security, and management. This survey delves into the diverse applications of neural networks within the networking domain, highlighting their impact on traffic prediction, anomaly detection, network optimization, and security enhancements. By leveraging the inherent capabilities of neural networks to model complex and non-linear relationships, significant improvements in efficiency and reliability can be achieved. This survey provides a thorough examination of current methodologies, key advancements, and practical implementations, offering insights into the future potential of neural networks in the field of networking.

Mining Utilities Itemsets based on social network

Sara salman Qasim, Lubna Mohammed Hasan — Sun, 03 Mar 2024 00:00:00 +0000

Mining utility item sets based on social network data involves extracting meaningful patterns and associations from user interactions. In this paper, the process begins by collecting and preprocessing data from platforms like Facebook, Twitter, or LinkedIn. Utility measures are defined based on frequency of occurrence, user engagement metrics, or other domain-specific criteria. Itemsets that meet certain thresholds are identified using techniques like frequent itemset mining or advanced algorithms like Apriori or FP-growth. Additional analyses, such as association rule mining, uncover relationships between different itemsets or user segments, providing valuable insights for personalized recommendations, targeted advertising, and decision-making processes.

Integrating Behavioral Analytics and Intrusion Detection Systems to Protect Critical Infrastructure and Smart Cities

Mon, 01 Jul 2024 00:00:00 +0000

In an age notable by growing digitization and relatedness, protecting critical infrastructure and smart cities against cyber threats is a biggest obstacle. This abstract examines the combination of Behavioural Analytics (BA) and Intrusion Detection Systems (IDS) as a active and best plan to boost cybersecurity defences. Behavioural Analytics uses machine learning algorithms and statistical models to notice usual entities behaviour patterns inside networks, empowering the identification of anomalies that indicate possible security infringements. This approach is improved through modern techniques that includes Statistical Anomaly Detection, which measures divergence and Long Short-Term Memory (LSTM) networks, skilled at grabbing temporal dependencies in data flow of network pursuit. Cross-Event Correlation methodologies and approaches improve the abilities of IDS by finding similarity between disparate events, giving a broad aspect of possible threats across inter related systems. Entity Behaviour Analysis (EBA) enhance these works by building thorough behaviour profiles and allocating risk scores based on divergence, improving targeted response plans. Network-Based IDS (NIDS) lengthen defence by observing whole networks for unusual activities, while Cyber Threat Intelligence (CTI) devices gives findings into progressing threats, enabling defensive security scales. Convolutional Neural Networks (CNNs) plays a part in removing complicated attribute from network data, improving anomaly detection. The results shows enhancements in threat detection accuracy, with a drop in false positives by 30% and an rise in anomaly detection precision to 95%. The Sensor Data (Units) changes from 80 to 90 units over monitored time periods.

Advancements in Time Series-Based Detection Systems for Distributed Denial-of-Service (DDoS) Attacks: A Comprehensive Review

Sara salman Qasim, Sarah Mohammed NSAIF — Sat, 20 Jan 2024 00:00:00 +0000

Distributed denial-of-service assaults, often known as DDoS attacks, pose a significant danger to the stability and security of the internet, particularly in light of the increasing number of devices that are linked to the internet. Intelligent detection systems are absolutely necessary in order to lessen the impact of distributed denial of service assaults. In this study, a comprehensive overview of recent research on intelligent approaches, such as Machine Learning (ML), Deep Learning (DL), and Artificial Intelligence (AI), is presented. The review focuses on the application of these techniques in the detection of Distributed Denial of Service (DDoS) assaults. In addition to providing a taxonomy and conceptual framework for DDoS mitigation, the study places particular emphasis on the application of time series data analysis for the detection of distributed denial of service attacks. A number of different intelligent techniques are investigated in this paper. Some of these techniques include clustering, deep reinforcement learning, graph neural networks, support vector machines, and others. For the purpose of performance evaluation, real datasets are utilized, and prospective future research areas in this area are explored.

Secure Routing and Reliable Packets Transmission In MANET Using Fast Recursive Transfer Algorithm

Sat, 15 Jun 2024 00:00:00 +0000

Mobile Ad-hoc Network (MANET) autonomous operation can be multi-hop it is the infrastructure-less wireless network. Security is one of the biggest challenges in Mobile Adhoc Network. The MANETs security there are considerations must be so that the routing protocol in order to protect the secure data transmission. In the routing and security that is an important aspect for in a MANETs, existing method routing protocol, however, is not enough to security requirements. The proposed method using Fast Recursive Transmission Algorithm (FRTA) used designed to maximize the data security, routing optimization, minimizing the impact of malicious attack using Collision Detection Avoid Algorithm (CDAA) activity over the MANET and select the best path. Table-driven routing protocols, also known as proactive routing protocols, mandate that every node on the network keep up-to-date routing data. To maintain consistent routing information for network nodes, these protocols propagate frequent updates to the routing table throughout the network, necessitating changes to the network topology. There's a lot of overhead from these upgrades. The suggested FRTA algorithm lowers node data loss rates and increases network energy efficiency. Table-driven routing protocols, also known as proactive routing protocols, mandate that every node on the network keep up-to-date routing data. To maintain consistent routing information for network nodes, these protocols propagate frequent updates to the routing table throughout the network, necessitating changes to the network topology. There's a lot of overhead from these upgrades. The suggested FRTA algorithm lowers node data loss rates and increases network energy efficiency. The proposed method shows high performance than other existing evaluations of the most advanced security and routing energy, end-to-end delay, packet transfer rate, packet loss

A Novel Deep Learning Approach for Detecting Types of Attacks in the NSL-KDD Dataset

HADEEL M SALEH SALEEH, Hend Marouane , Ahmed Fakhfakh — Sun, 01 Sep 2024 00:00:00 +0000

The growing prevalence of Internet intrusions poses significant threats to the security, privacy, and reliability of systems and networks. Denial-of-service (DoS) attacks are a cause for concern as they aim to disrupt access to network resources, posing major risks. Traditional intrusion detection systems (IDS) face challenges in detecting attacks because of the evolving nature of these attacks. Therefore, advanced techniques are necessary to accomplish accurate and timely detection. This study introduces a novel approach that combines Deep learning techniques, specifically the CNN algorithm, with Principal Component Analysis (PCA) and Singular Value Decomposition (SVD) for the purpose of feature selection. The effectiveness and efficiency of our method are shown by rigorous testing on DDoS datasets. We present a novel Fast Hyper Deep Learning Model that attains a remarkable accuracy of 99%, along with perfect recall and F1-measurement scores of 100%. This model surpasses existing methodologies by a significant margin. The NSL-KDD data set allows for achieving a level of precision of100%.

Performance Analysis of Hybrid Optical OFDM Schemes in terms of PAPR, BER and Spectral Efficiency using Various PAPR Reduction Methods

Karthik Kumar Vaigandla, Mounika Siluveru, Nanda Dharavath, RadhaKrishna Karne — Thu, 01 Aug 2024 00:00:00 +0000

Orthogonal Frequency Division Multiplexing (OFDM) is an extremely important technique for transmitting data in both optical wireless and optical fiber communication. Optical wireless systems are limited to transmitting real and positive values to the optical transmitter, since they rely only on the intensity of a signal to convey information. Consequently, traditional OFDM is not suitable for direct implementation in optical systems. Various altered OFDM schemes have been examined to counteract multipath distortion, such as DC-biased optical OFDM (DCO-OFDM), asymmetrically clipped optical OFDM (ACO-OFDM), asymmetrically clipped DC biased optical OFDM (ADO-OFDM) and layered asymmetrically-clipped optical OFDM (LACO-OFDM) technique to improve the bit error rate (BER) performance and spectral efficiency of optical system. This article explores ways for reducing Peak-to-Average Power Ratio (PAPR). A mathematical model is formulated to account for clipping and Selective Mapping (SLM), as well as channel noise, in all of these approaches for received signal. LACO-OFDM has been determined to be a superior option when compared to other techniques. LACO-OFDM, a recently developed technique, demonstrates excellent performance in terms of PAPR, BER, and spectral efficiency. The simulation results demonstrate that the LACO-OFDM approach has greatly enhanced the spectral efficiency in comparison to previous approaches.

Robust Color Image Encryption Using 3D Chaotic Maps and S-Box Algorithms

Jenan Ayad , Mustafa A Jalil — Mon, 05 Aug 2024 00:00:00 +0000

Chaos' key qualities, such as initial state sensitivity and unpredictability, make it an ideal contender for cryptography applications. This study presents an encryption scheme for efficient and secure image encryption. The encryption scheme includes two ciphering stages and a substitution stage. In this work, an algorithm for key generation has been proposed. The design of a Pseudo-random number generator that used for key generation is based on chaotic algorithms. The chaotic map will be utilised in encryption systems due to its high security. To evaluate the proposed PRNG randomness, NIST tests are used for these sequences. In the following subsection, security analysis of the proposed image encryption technique has been made to know the efficiency of the proposed technique. The statistical analysis now reaffirms that the technique is secure and efficient for and encrypting simple or complex images, be it in shades of black and white or, colorful. By a comparison with past chaotic investigations, demonstrating that our algorithm is competitive with earlier work.

Using a Fuzzy Approach as an Assessment Method to Extend the Lifespan of Wireless Sensor Networks using the LEACH Protocol

Janan Farag Yonan, Ayser Hadi Oleiwi — Sun, 10 Mar 2024 00:00:00 +0000

Wireless sensor network is the term used to describe a network where network nodes are wirelessly configured to collect data from the real world. Node sensors depend on finite energy sources, such as batteries, because of the wireless configuration they have. If the battery-operated sensor of the node is not charged, it will be unable to carry out its intended function. If a specific amount of nodes fail, the network will cease to function. Several energy-efficient protocols were developed for Wireless Sensor Networks (WSN), including the LEACH Protocol. The LEACH protocol demonstrates a single cluster-based protocol by dividing available sensor nodes into sets and interacting with each set individually. The shape of an energy can be altered by compressing or expanding it, based on the cluster's configuration. We are comparing the network lifespans of three distinct versions of the LEACH protocol that utilize fuzzy techniques for cluster selection with the lifespan of WSNs generated by a previous version of the protocol.

Mapping Web Service Characteristics to Queueing Theory Models for Performance Analysis

Sat, 20 Jul 2024 00:00:00 +0000

The article presents a comprehensive analysis of mapping the characteristics of Web services to queueing theory models, specifically the M/M/1 queueing model. The authors aim to establish a theoretical foundation for modeling the performance of Web services and deriving response time formulae. The key points covered in the article are as follows: 1. Analyzing the characteristics of Web services: The article examines the six fundamental characteristics of queueing systems, including arrival patterns, service patterns, queue discipline, system capacity, and the number of servers, and maps them to the Web service environment. 2. Representation using Kendall's notation: The Web service characteristics are represented using Kendall's notation, resulting in the M/M/1/∞/FCFS model, where M denotes exponential distributions for arrival and service patterns, 1 represents a single server, ∞ represents infinite system capacity, and FCFS represents the first-come, first-served queue discipline. 3. Derivation of response time formulae: The article provides a detailed derivation of the response time formulae for the M/M/1 model using stochastic processes, Markov chains, and the birth-death process. The derivation incorporates steady-state conditions and results in a formula that calculates the probability of completing a request within a user-specified response time. 4. Practical applications: The derived response time formula can be used by service providers to determine the probability of completing user requests within specified Service Level Agreements (SLAs). By setting a threshold value (filter value), the service provider can select requests with a higher probability of completion for further processing. Overall, the article contributes to the understanding of Web service performance modeling by establishing a theoretical foundation based on queueing theory. The mapping of Web service characteristics to the M/M/1 model and the derivation of response time formulae provide valuable insights for service providers to analyze and optimize the performance of their Web services while adhering to SLAs

A ELECTRIC VEHICLE BLOCKCHAIN: PROBLEMS AND OPPORTUNITIES

sahar Mohammed, Thaaer kh.Asman, Hadeel M Salih, Alaa Mohammed Mahmood — Mon, 19 Feb 2024 00:00:00 +0000

These days, we are observing a very rapid spread of the electric vehicle
industry. This means a significant increase in the data and energy exchanged between
these vehicles. The existing centralized approach is less secure and more vulnerable
to data destruction and manipulation by intruders. Therefore, it became necessary to
search for an alternative that provides excellent protection for this massive amount
of data and energy. Although blockchain technology and cryptocurrencies are closely
associated, they also have many other potential applications in fields including energy
and sustainability, the Internet of Things (IoT), smart cities, smart mobility, and
more. In the Internet of Vehicles (IoV) idea, blockchain can provide security for
electric vehicle (EV) transactions, enabling electricity trading to be carried out in
a decentralized, transparent, and secure manner. . This paper will explain the use of
blockchain in this field and how it can handle the trade of transmitted and received
energy between electric vehicles. The advantages of using blockchain with electric
cars and how it can secure the transactions of energy trading will be shown too. A
group of researchers in this field and the challenges that face this technology in energy
trading will be discussed too; the studies will be looked at, and recommendations for
investments and security will be made. Additionally, the future implications of various
blockchain technologies will be highlighted.

A Review of Underwater Wireless Sensor Networks Deployment Techniques and Challenges

Kadoke Marco, Sonia Goyal , Ranjit Kaur — Thu, 25 Jul 2024 00:00:00 +0000

Underwater Wireless Sensor Networks (UWSNs) have become increasingly important due to their critical roles in marine life monitoring, communication, ocean data collection, sampling, and military security operations. The success of UWSNs largely depends on efficient node deployment techniques that ensure optimal coverage, connectivity, cost-effectiveness, network lifetime, and energy utilization. This paper presents a comprehensive review of various node deployment types and techniques specifically designed for UWSNs. It covers depth-adjustment, movement-assisted, self-movement, and soft-computing techniques, highlighting their advantages, limitations, and application scenarios. Each technique is evaluated based on key performance metrics such as network coverage, connectivity, energy consumption, network lifetime, and deployment cost. Additionally, the paper discusses the challenges and identifies open research directions in the field, providing valuable insights for researchers and practitioners in selecting appropriate node deployment techniques for UWSNs.

Evaluation of OFDM system in terms of PAPR and BER using PAPR Reduction Techniques : Windowing and Clipping

Mounika Siluveru, Dharavath Nanda, Madhavi Kesoju — Wed, 10 Jan 2024 00:00:00 +0000

The issue of transferring signals in a non-intrusive and efficient manner is highly intriguing. Consequently, the utilisation of the Orthogonal Frequency Division Multiplexing (OFDM) transmission technology is employed due to its enhanced resilience against multipath fading and its ability to achieve better efficiency compared to alternative wireless communication methods. Although the PAPR issue offers certain benefits, it poses undesirable consequences for OFDM, leading to a significant decrease in data transmission rate. Consequently, the mitigation of peak-to-average power ratio (PAPR) issues leads to an improvement in the quality of service. The reduction of PAPR can be achieved by the implementation of clipping and windowing techniques. These approaches arrange the signal into discrete levels of similar magnitude and apply clipping to minimise the bit error rate (BER), resulting in a drop in PAPR. The proposed approach is implemented using the MATLAB software.

Underwater Wireless Sensors Increase Routing Performance using Impact Efficient localization-based Routing protocols

Mon, 03 Jun 2024 00:00:00 +0000

The Underwater Wireless Sensor Network (UWSN) is an organization used to perform observing of errands over a particular region; it is furnished with shrewd sensors and vehicles that are adjusted to convey helpfully through wireless connections. Remote sensor networks are enormous scope networks comprised of modest, reduced sensors with immense scope energy and settling limit that might be utilized in different unpredictable circumstances under factor conditions. The use of UWSNs is growing daily due to their significant contribution to several applications, including underwater surveillance and search. Wireless sensor networks submerged in water confront unique challenges. Therefore, particular routing protocols are needed from source to destination; security concerns that should be taken into consideration by routing protocols must be addressed by many UWSN applications. The proposed routing protocols influencing IHELBRP (Impact High Efficient Localization-Based Routing Protocols) and the UWSN architectural perspective. Reviewing and examining steering conventions concerning energy consumption, packet delivery rate, and packet delivery rate is finished. The advantages and disadvantages of each steering convention are recorded. To shield the correspondence medium here, a rundown of safety needs is incorporated alongside an investigation of safety worries in UWSN.

Securing Container Images through Automated Vulnerability Detection in Shift-Left CI/CD Pipelines

Arvind Kumar Bhardwaj, P.K. Dutta, Pradeep Chintale — Tue, 20 Aug 2024 00:00:00 +0000

Integrating shift-left security practices and automated vulnerability detection in container images is imperative for modern software development, given the dynamics and vulnerability landscape. This crucial methodology emphasizes security from the initial stages of integration in container-based environments like Docker and Kubernetes. The paper examines containerization security challenges, including image vulnerabilities, insecure configurations, runtime risks, weak orchestration security, and supply chain weaknesses, while stressing compliance with regulatory rules. It explores how this automated approach leverages vulnerability detection methods integrated into Continuous Integration/Continuous Deployment (CI/CD) pipelines through static and dynamic analyses, vulnerability databases, and policy-enforcement mechanisms. Beyond identifying vulnerabilities in CI/CD pipelines, the paper outlines methods to avoid policy violations, mitigate vulnerable images, and prevent recurring practices. Importantly, it underscores the continuous enforcement and remediation of policies and security standards. Security teams must invest efforts in developing policies, automated executions, and remediation procedures, fostering cross-departmental collaboration. In essence, this proactive stance aims to enhance software security, reduce risks, and improve adherence in containerized ecosystems, making it an indispensable component of modern software development.