Ensemble Voting based Intrusion Detection Technique using Negative Selection Algorithm
Abstract: This paper proposes an Intrusion Detection Technique (IDT) using an Artificial Immune System (AIS) based on Negative Selection Algorithm (NSA) to distinguish the self and non-self (intrusion) in computer networks. The novelties of the work are 1) use of Stacked Autoencoders (SAEs) and random forest for dimensionality reduction of data, 2) use of AIS to exploit its feature like self-learning, distributed, self-adaption, self-regulation with self and non-self-distinguishing capability, 3) implementation of two algorithms i.e., NSA based on Cosine Distance (NSA_CD) and NSA based on Pearson Distance (NSA_PD) to explore their intrusion detection capabilities, and iv) development of a new ensemble voting based Intrusion Detection Technique (IDT-NSAEV) to detect and test the anomalies in the system. The proposed IDT-NSAEV technique combines the power of NSA_CD, NSA_PD and NSA based on Euclidean distance (NSA_ED) algorithms to enhance the detection rate by reducing the false alarm rate. The performance of the proposed technique is tested on standard benchmark NSL-KDD dataset and the results are compared with the state-of-the-art techniques. The results are in the favour of the proposed technique.
Keywords: Artificial immune system, security, negative selection algorithm, anomaly detection.
Received December 27, 2020; accepted February 28, 2022
Image Segmentation with Multi-feature Fusion in Compressed Domain based on Region-Based Graph
Abstract: Image segmentation plays a significant role in image processing and scientific research. In this paper, we develop a novel approach, which provides effective and robust performances for image segmentation based on the region-based (block-based) graph instead of pixel-based graph. The modified Discrete Cosine Transform (DCT) is applied to obtain the Square Block Structures (DCT-SBS) of the image in the compressed domain together with the coefficients, due to its low memory requirement and high processing efficiency on extracting the block feature. A novel weight computation approach focusing on multi-feature fusion from the location, texture and RGB-color information is employed to efficiently obtain weights between the DCT-SBS. The energy function is redesigned to meet the region-based requirement and can be easily transformed into the traditional Normalized cuts (Ncuts). The proposed image segmentation algorithm is applied to the salient region detection database and Corel1000 database. The performance results are compared with the state-of-the-art segmentation algorithms. Experimental results clearly show that our method outperforms other algorithms, and demonstrate good segmentation precision and high efficiency.
Keywords: Image segmentation, region-based graph, multi-feature fusion, compressed domain.
Received January 8, 2021; accepted February 10, 2022
Malaria Parasite Detection on Microscopic Blood Smear Images with Integrated Deep Learning Algorithms
|
Christonson Berin Jones Computer Science and Engineering, Madurai Institute of Engineering and Technology, Tamil *This email address is being protected from spambots. You need JavaScript enabled to view it. |
Chakravarthi Murugamani Department of Information Technology, Bhoj Reddy Engineering College for Women,Telangana This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Malaria is a deadly syndrome formed by the Plasmodium parasite that spreads through the bite of infected Anopheles mosquitoes. There are several drugs to cure malaria but it is difficult to detect due to inadequate equipment and technology. Microscopic check-ups of blood smear images by experts help to detect malaria-infected parasites accurately. However, manual analysis is tedious and time-consuming as the experts have to deal with many cases. This paper presents computer assisted malaria parasite detection model by classifying the blood smear image with hybrid deep learning methods that have high accuracy for classification. In the proposed approach the blood smear images are pre-processed using bilateral filtering technique in which features are extracted with the convolutional neural network. These features are selected by the improved grey-wolf optimization, and image classification is performed with the support vector machine. To evaluate the efficiency of the proposed technique, the NIH malaria dataset is utilized and the results are compared with existing approaches in terms of accuracy, F-Measure, recall, precision, and specificity. The outcome reveals that the proposed scheme is accurate and can be more helpful to pathologists for reliable parasite detection.
Keywords: Blood smear images, image classification, image processing, malaria, plasmodium parasite.
Received February 2, 2021; accepted January 10, 2022
Mining Android Bytecodes through the Eyes of Gabor Filters for Detecting Malware
Abstract: One of the basic characteristics of a Gabor filter is that it provides useful information about specific frequencies in a localized region. Such information can be used in locating snippets of code, i.e., localized code, in a program when transformed into an image for finding embedded malicious patterns. Keeping this phenomenon, we propose a novel technique using a sliding Window over Gabor filters for mining the Dalvik Executable (DEX) bytecodes of an Android application (APK) to find malicious patterns. We extract the structural and behavioral functionality and localized information of an APK through Gabor filtered images of the 2D grayscale image of the DEX bytecodes. A Window is slid over these features and a weight is assigned based on its frequency of use. The selected Windows whose weights are greater than a given threshold, are used for training a classifier to detect malware APKs. Our technique does not require any disassembly or execution of the malware program and hence is much safer and more accurate. To further improve feature selection, we apply a greedy optimization algorithm to find the best performing feature subset. The proposed technique, when tested using real malware and benign APKs, obtained a detection rate of 98.9% with 10-fold cross-validation.
Keywords: Android bytecode, malware analysis and detection, sliding window, gabor filters, gabor features, machine learning.
Received February 14, 2021; accepted September 26, 2022
Coverless Data Hiding in VoIP based on DNA Steganography with Authentication
|
Deepikaa Soundararajan School of Computer Science and Engineering Vellore Institute of Technology, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Saravanan Ramakrishnan School of Computer Science and Engineering Vellore Institute of Technology, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Data hiding in Voice over Internet Protocol (VoIP) using coverless approach improves the undetectability by preserving the cover bits from modification. This paper focuses on hiding the secret message in VoIP streams using Deoxyribonucleic Acid (DNA) steganography approach. DNA steganography is known for its low cracking probability. The embedding process is done in two steps. The first step converts the VoIP sample, secret message and a user generated key (for Authentication) into m-RNA pattern during transcription and the second step converts the m-RNA to form a triplet during translation process to create a protein array, where the secret message is embedded. The secret message is extracted from the protein array by applying reverse translation and Transcription. The proposed approach improves the undetectability by leaving the cover bits unmodified with Perceptual Evaluation of Signal Quality (PESQ) values 84% comparatively greater than the state of art techniques.
Keywords: DNA, VoIP, coverless approach, data hiding, steganography.
Received March 27, 2021; accepted March 17, 2022
A Bayesian Network-based Uncertainty Modeling (BNUM) to Analyze and Predict Next Optimal Moves in Given Game Scenario
|
Vinayak Jagtap College of Engineering, Pune, Maharashtra, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Parag Kulkarni iKnowlation Research Labs Pvt Ltd, Tokyo International University, Japan This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: As machine learning emerged, it is being used in a variety of applications like speech recognition, image recognition, sequence modeling, etc., Sequence modeling is one type of application where resultant sequences are generated based on historical data inputs provided. These sequences are fairly work in an uncertain environment like games or sports. In the case of a game or a sport, there is a sequence of moves selected by multiple players. There is a statistical uncertainty observed for simple to more complex games. For example, while playing chess, a simple statistical modeled uncertainty would be enough to choose the next possible. This move selection is dependent on available free spaces of pieces or pawns. The sports like tennis, cricket, and other games need a more complex design for uncertainty modeling for next move selection. A Bayesian Network model will work if there is fairly less uncertainty in the selection of the next move. A Bayesian Network-based model will be best fitted if all possible moves are included before training any machine learning or deep learning model. This will be achieved with the usage of the Context-Li model. The proposed Bayesian Network-based Uncertainty Modeling (BNUM) is used to incorporate uncertainty, for next move selection. BNUM is a multi-variable, multi-level association to incubate uncertainty in learning. It helps to predict the next move in an uncertain gaming environment. Different case studies are incorporated to verify the hypothesis and the results are a sequence of moves represented in the context graph.
Keywords: Bayesian network, uncertainty modeling, deep learning, context graph, next move.
Received April 15, 2021; accepted June 12, 2022
Analysis of TCP Issues and Their Possible Solutions in the Internet of Things
|
Syed Zeeshan Hussain Department of Computer Science, Jamia Millia Islamia, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Sultana Parween Department of Computer Science, Jamia Millia Islamia, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: The Internet of Things (IoT) is widely known as a revolutionary paradigm that offers communication among different types of devices. The primary goal of this paradigm is to implement efficient and high-quality smart services. It requires a protocol stack that offers different service requirements for inter-communication between different devices. Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) are used as transport layer protocols in IoT to provide the quality of service needed in various IoT devices. IoT encounters many shortcomings of wireless networks, while also posing new challenges due to its uniqueness. When TCP is used in an IoT system, a variety of challenging issues have to be dealt with. This paper provides a comprehensive survey of various issues which arises due to the heterogeneous characteristics of IoT. We identify main issues such as Retransmission Timeout (RTO) algorithm issue, congestion and packet loss issue, header overhead, high latency issue, link layer interaction issue, etc., Moreover, we provide several most probable solutions to the above-mentioned issues in the case of IoT scenarios. RTO algorithm issue has been resolved by using algorithms such as CoCoA, CoCoA+, and CoCoA++. Apart from these, the high latency issue has been solved with the help of a long lived connection and TCP Fast open. Congestion and packet loss issue has been resolved by using several TCP variants such as TCP New Reno, Tahoe, Reno, Vegas, and Westwood.
Keywords: RTO, acknowledgment, round trip time.
Received June 4, 2021; accepted August 29, 2022
Highly Accurate Grey Neural Network Classifier for an Abdominal Aortic Aneurysm Classification Based on Image Processing Approach
Abstract: An Abdominal Aorta Aneurysm (AAA) is an abnormal focal dilation of the aorta. Most un-ruptured AAAs are asymptomatic, which leads to the problem of having abdominal malignancy, kidney damage, heart attack and even death. As it is ominous, it requires an astute scrutinizing approach. The significance of this proposed work is to scrutinize the exact location of the ruptured region and to make astute report of the pathological condition of AAA by computing the Ruptured Potential Index (RPI). To determine these two factors, image processing is performed in the retrieved image of aneurysm. Initially, it undergoes a process to obtain a high-quality image by making use of Adaptive median filter. After retrieving high quality image, segmentation is carried out using Artificial Neural Network-based segmentation. After segmenting the image into samples, 12 features are extracted from the segmented image by Gray Level Co-Occurrence Matrix (GLCM), which assists in extracting the best feature out of it. This optimization is performed by using Particle Swarm Optimization (PSO). Finally, Grey Neural Network (GNN) classifier is applied to analogize the trained and test set data. This classifier helps to achieve the targeted objective with high accuracy.
Keywords: Adaptive median filter, artificial neural network-based segmentation, GLCM, PSO, gray neural network classifier.
Received September 2, 2021; accepted January 23, 2022
Using MCDM and FaaS in Automating the Eligibility of Business Rules in the Decision-Making Process
Abstract: Serverless Computing, also named Function as a Service (FaaS) in the Azure cloud provider, is a new feature of cloud computing. This is another brick, after managed and fully managed services, allowing to provide on-demand services instead of provisioned resources and it is used to strengthen the company’s ability in order to master its IT system and consequently to make its business processes more profitable. Knowing that decision making is one of the important tasks in business processes, the improvement of this task was the concern of both the industry and the academy communities. Those efforts have led to several models, mainly the two Object Management Group (OMG) models: Business Process Model and Notation (BPMN) and Decision Model and Notation (DMN) in order to support this need. The DMN covers the decision-making task in business processes mainly the eligibility of business rules. This eligibility can be automated in order to help designers in the mastering of this important task by the running of an algorithm or a method such as the Multiple Criteria Decision Making (MCDM). This feature can be designed and implemented and deployed in various architectures to integrate it in existing Business Process Management Systems (BPMS). It could then improve supporting several business areas such as the Business Intelligence (BI) process. In this paper, our main contribution is the enrichment of the DMN model by the automation of the business rules eligibility through Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) using FaaS to further streamline the decision-making task in business processes. Results show to strengthen business-IT alignment and reduce the gap between the real world and associated IT solutions.
Keywords: Serverless computing, FaaS, BPMN, DMN, decision-making, business-rule, MCDM, TOPSIS.
Received November 8, 2020; accepted April 13, 2022
Credit-card Fraud Detection System using Neural Networks
|
Salwa Al Balawi Faculty of Computers and Information Technology, University of Tabuk, Saudi Arabia This email address is being protected from spambots. You need JavaScript enabled to view it. |
Njood Aljohani Faculty of Computers and Information Technology, University of Tabuk, Saudi Arabia This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Recently, with the development of online transactions, the credit-card transactions begun to be the most prevalent online payment methods. Credit-card fraud refers to the use fake Credit-Cards to purchase goods without paying. With the fast research and development in the area of information technology and data mining methods including the neural networks and decision trees, to advanced machine learning and deep learning methods, researchers have proposed a wide range of antifraud systems. Mainly, the Machine Learning (ML) and Deep Learning (DL) methods are employed to perform the fraud detection task. This paper aims to explore the existing credit-card fraud detection methods, and categorize them into two main categories. In addition, we investigated the deployment of neural network models with credit-card fraud detection problem, since we employed the Artificial Neural Network (ANN) and Convolutional Neural Network (CNN). ANN and CNN models are implemented and assessed using a credit-card dataset. The main contribution of this paper focuses on increasing the fraud-detection classification accuracy through developing an efficient deep neural network model.
Keywords: Credit-card, fraud detection, machine learning, deep learning, neural networks, classifications.
Received December 3, 2020; accepted August 31, 2021
A Cognitive Approach To Predict the Multi-Directional Trajectory of Pedestrians
|
Jayachitra Virupakshipuram Panneerselvam Department of Computer Technology, Anna University, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Bharanidharan Subramaniam Department of Computer Technology, Anna University, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Mathangi Meenakshisundaram Department of Computer Technology, Anna University, India This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Pedestrian detection is one of the important areas in computer vision. This work is about detecting the multi-directional pedestrian’s left, right, and the front movements. On recognizing the direction of movement, the system can be alerted depending on the environmental circumstances. Since multiple pedestrians moving in different directions may be present in a single image, Convolutional Neural Network (CNN) is not suitable for recognizing the multi-directional movement of the pedestrians. Moreover, the Faster R-CNN (FR-CNN) gives faster response output compared to other detection algorithms. In this work, a modified Faster Recurrent Convolutional Neural Network (MFR-CNN), a cognitive approach is proposed for detecting the direction of movement of the pedestrians and it can be deployed in real-time. A fine-tuning of the convolutional layers is performed to extract more information about the image contained in the feature map. The anchors used in the detection process are modified to focus the pedestrians present within a range, which is the major concern for such automated systems. The proposed model reduced the execution time and obtained an accuracy of 88%. The experimental evaluation indicates that the proposed novel model can outperform the other methods by tagging each pedestrian individually in the direction in which they move.
Keywords: Automated driving system, deep neural networks, faster recurrent-convolutional neural network, object recognition, pedestrian detection, pedestrian movement direction.
Received January 3, 2021; accepted March 20, 2022
Secure Blockchain-Based Electronic Voting Mechanism
|
Pin-Chang Su Department of Information Management, National Defense University, Taiwan This email address is being protected from spambots. You need JavaScript enabled to view it. |
Tai-Chang Su Department of Information Management, National Defense University, Taiwan This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Many countries have strived to popularise electronic voting (e-voting), but owing to various security concerns, large-scale elections are still invariably held using paper ballots. Electronic voting systems must find solutions to various issues with authentication, data privacy and integrity, transparency, and verifiability. On the other hand, Blockchain technology offers an innovative solution to many of these problems. In this study, we constructed a private blockchain network with a large number of nodes, which is only accessible to the relevant voters. Because of its decentralised design, the system is robust against attacks by malicious actors. The security of the system was enhanced using an elliptic curve discrete logarithm problem-based blind multi-document signcryption mechanism. As this mechanism can be used to blindly sign and encrypt multiple voting documents in a single pass, it will minimise redundant signing processes and thus improve efficiency. Furthermore, a self-certification mechanism was used in lieu of centralised certificate servers, so that the voters can participate in the computation of public and private keys. In summary, we designed an electronic voting mechanism that is sufficiently secure for practical purposes, which will improve trust in e-voting, and reduce the costs associated with vote checking.
Keywords: Blockchain, e-voting, blind multi-document signcryption, self-certification.
Received July 21, 2021; accepted November 7, 2022
A Brief Review of Massive MIMO Technology for the Next Generation
Imadeldin Elmutasim
Department of Electrical Engineering, University Malaysia Pahang, Malaysia
This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract: Massive Multiple Input Multiple Output (MIMO) is an evolving technology based on the principle of spatial multiplexing which consists in using at the same time the same radio frequencies to send different signals. The several transmitting antennas from a base station can transmit different signals and several receiving antennas from a device can receive and divide them simultaneously. Due to the physically difficult of installing antennas close to each other, standard MIMO networks generally limit four antenna-side transmitters and receivers for data transmission while it could be more. The study aims to review the traditional MIMO different types as well as investigates the Signal-to-Noise Ratio (SNR) between Single Input Single Output (SISO) and MIMO to ensure the best wireless connection functionality. In addition to that, a simple comparison to distinguish between SISO, SIMO, MISO, and MIMO in term of capacity and data rate to provide an indication for the quality of the wireless connection. The work's contribution is to illustrate technological benefits like MIMO, which boosts data speeds and increases the reliability of wireless networks. The outcome shows a SISO system would have a lower data rate than other systems because it only has one antenna at the transmitter and receiver, whereas a MISO system would typically have a greater SNR than a SISO or SIMO system because it uses several transmit antennas. MIMO, however, took advantage of all the positive characteristics and emerged as the best solution overall.
Keywords: MIMO, SISO, SIMO, MISO, 5G, SNR, antenna.
Received March 21, 2022; accepted January 12, 2023
Improved YOLOv3-tiny for Silhouette Detection Using Regularisation Techniques
Abstract: Although recent advances in Deep Learning (DL) algorithms have been developed in many Computer Vision (CV) tasks with a high accuracy level, detecting humans in video streams is still a challenging problem. Several studies have, therefore, focused on the regularisation techniques to prevent the overfitting problem which is one of the most fundamental issues in the Machine Learning (ML) area. Likewise, this paper thoroughly examines these techniques, suggesting an improved you Only Look Once (YOLO) v3-tiny based on a modified neural network and an adjusted hyperparameters file configuration. The obtained experimental results, which are validated on two experimental tests, show that the proposed method is more effective than the YOLOv3-tiny predecessor model. The first test which includes only the data augmentation techniques indicates that the proposed approach reaches higher accuracy rates than the original YOLOv3-tiny model. Indeed, Visual Object Classes (VOC) test dataset accuracy rate increases by 32.54 % compared to the initial model. The second test which combines the three tasks reveals that the adopted combined method wins a gain over the existing model. For instance, the labelled crowd_human test dataset accuracy percentage rises by 22.7 % compared to the data augmentation model.
Keywords: Silhouette/person detection, GPU, loss function, convolutional neural network，YOLOv3-tiny.
Received April 15, 2021; accepted December 14, 2022
In Loco Identity Fraud Detection Model Using Statistical Analysis for Social Networking Sites: A Case Study with Facebook
|
Shalini Hanok Electronics and Communication Engineering, ATME College of Engineering, Karnataka This email address is being protected from spambots. You need JavaScript enabled to view it. |
Shankaraiah Sri Jayachamarajendra College of Engineering (SJCE), JSS Science and Technology University, Karnataka This email address is being protected from spambots. You need JavaScript enabled to view it. |
Abstract: Rapid advancement in internet has made many Social Networking Sites (SNS) popular among a huge population, as various SNS accounts are interlinked with each other, spread of stored susceptible information of an individual is increasing. That has led to various security and privacy issues; one of them is impersonation or identity fraud. Identity fraud is the outcome of illegitimate or secret use of account owner’s identity to invade his/her account to track personal information. There are possibilities that known persons like parents, spouse, close friends, siblings who are interested in knowing what is going on in the account owner’s online life may check their personal SNS accounts. Hence an individual’s private SNS accounts can be invaded by an illegitimate user secretly without the knowledge of the account owner’s which results in compromise of private information. Thus, this paper proposes an in loco identity fraud detection strategy that employs a statistical analysis approach to constantly authenticate the authorized user, which outperforms the previously known technique. This strategy may be used to prevent stalkers from penetrating a person's SNS account in real time. The accuracy attained in this research is greater than 90% after 1 minute and greater than 95% after 5 minutes of observation.
Keywords: Continuous authentication, in loco, SNS, identity fraud, stalkers.
Received April 27, 2021; accepted September 22, 2021