Computer and Information Science
Image Steganalysis using Deep Convolution Neural Networks: A Literature Survey
Steganography is the technique of hiding data for secret communication in a public media format. The image in which the hidden data is stored is called a stego image. Steganalysis is the process of targeting the methods of steganography to identify remove destroy and exploit the secret data in stego images. The identification of embedded secret data in the image is the basis for steganalysis. The proper selection of the type and composition of cover files contributes to a better embedding. Several steganalysis techniques exist for detecting steganography in the images given. Because of the embedded data the performance of the steganalysis technique relies on the capacity to retrieve the feature representations to identify the statistical portion of the image. Steganalysis & steganography has experienced tremendous development in recent years with the emergence of Deep Convolution Neural Networks (DCNN). In this paper we explored the current state of research from the latest systems of image steganalysis based on deep learning. This paper presents different methodologies and frameworks of CNN the research being carried out on image steganalysis based on deep learning and implementation complexities and highlights the benefits and limitations of the existing techniques. This study also provides the direction for future research and may serve as a fundamental source for further research in deep learning-based image steganalysis.
Spectrum and Power Efficient Anti-jamming Approach for Cognitive Radio Networks Based on Reinforcement Learning
Spectrum scarcity spectrum efficiency power constraints and jamming attacks are core challenges that face wireless networks. While cognitive radio networks (CRNs) enable the sharing of licensed bands when they are unoccupied the spectrum should be used efficiently by the secondary user (SU) to ensure a high data rate transmission. In addition the mobility of the SUs makes power consumption a matter of concern in wireless networks. Because of the open environment the jamming attack can easily deteriorate the performance and disrupt the connections.
We aim to enhance the performance of CRN and establish more reliable connections for the SU in the presence of smart jammer by ensuring efficient spectrum utilization and extending the network lifetime.
To achieve our objectives we propose an anti-jamming approach that adopts frequency hopping. Our approach assumes that SUs observe spectrum availability and channel gain. Then SU learns the jammer behaviour and goes for the appropriate policy in terms of the number of data and control channels that optimize jointly spectrum efficiency and power consumption. Within the interaction between the SU and the jammer is modelled as a zero-sum stochastic game and we employ reinforcement learning (RL) to address this game.
SUs learn the optimal policy that maximizes the spectrum efficiency and minimizes the power consumption in the presence of a smart jammer. Simulation results show that the low channel gain leads the SU to select a high number of data channels. However when the channel gain is high the SU increases the number of control channels to guarantee a more reliable connection. Taking into account the spectrum efficiency SUs save their energy by decreasing the number of used channels. The proposed strategy achieves better performance in comparison with myopic learning and the random strategy.
Under a jamming attack considering the gain of utilized channels SUs select the appropriate number of control and data channels to ensure a reliable efficient and long-term connection.
Sub-1 GHz RF-based Energy-efficient Sensor Node for Secure Communication in Low-power IoT and Embedded Applications
The Internet of Things (IoT) devices consist of a microcontroller unit for data processing a low-power wireless radio module for data transmission and various sensors for data collection. The sensor nodes and processing devices used in the Internet of Things are resource-constrained with power consumption and security being the two most critical parameters.
This paper addresses the challenges of power consumption and security in IoT scenarios. It presents a low-power and secure heterogeneous multicore sensing architecture designed for low-power IoT and wireless sensor networks. The architecture comprises a sensing and control subsystem an information processing unit and a wireless communication module.
The architecture uses a microcontroller unit based on ARM Cortex M4 a low-power sub-1 GHz RF-compliant communication radio and a few sensors. The proposed architecture has been implemented and tested using the Contiki Operating System.
The implemented sensor node architecture demonstrated performance efficiency lower energy consumption and higher security.
By leveraging efficient power management data transmission strategies and cryptographic security the architecture contributes to developing energy-efficient and secure IoT devices.
Novel Energy-efficient Modified LEACH Routing Protocol for Wireless Sensor Networks
In wireless sensor networks (WSNs) hierarchical clustered routing protocols play a crucial role in minimizing energy consumption. The Low Energy Adaptive Clustering Hierarchy (LEACH) architecture is commonly employed for application-specific protocols in WSNs. However the LEACH protocol may lead to increased energy consumption within the network if the rotational distribution of cluster heads (CHs) is not considered.
A novel average energy residual energy-based modified LEACH (aerem-LEACH) routing protocol for improving the WSN’s energy efficiency is proposed. This approach simultaneously considers the average energy of the networks and the residual node energy for routing thereby reducing overall power consumption.
The suggested approach in aerem-LEACH accounts for optimal CHs numbers and nodes in close proximity to the sink are forbidden from participating in cluster formation in order to achieve sufficient performance in the form of reduced sensor node energy consumption. Furthermore a new threshold is employed in the proposed approach for selecting CHs for the network and the aerem-LEACH uses free space multiple hopping and a hybrid communicating model for an energy-efficient network.
The simulation result demonstrates that there is a substantial reduction in the consumption of energy in WSNs with the proposed aerem-LEACH routing protocol compared with existing routing protocols namely Stable Energy Efficient Network (SEEN) Energy Efficient LEACH (EE LEACH) Optical LEACH (O-LEACH) LEACH-Mobile (LEACH-M) LEACH-Centralized (LEACH-C) and LEACH for small-scale as well as large-scale sensor field.
ACKNOWLEDGEMENT TO REVIEWERS
Jamming Attacks Detection Based on IGWO for Optimization of Fast Correlation-Based Feature extraction in Wireless Communication
Wireless networks are essential communication technologies that prevent cable installation prices and burdens. Because of this technology's pervasive usage wireless network safety is a significant problem. Owing to distributed and open wireless medium aspects attackers might use different jamming methods to exploit physical and MAC layer protocol vulnerabilities. In addition jamming attacks require to be accurately grouped so that suitable countermeasures can be considered. Given the potential severity of such attacks precisely identifying and classifying them is critical for implementing effective responses. The motivation for this paper is the need to improve the detection and categorization of jamming signals using modern machine learning algorithms consequently enhancing wireless network security and reliability.
In this paper we compare some machine learning models' efficiency for diagnosing jamming signals.
Such algorithms refer to support vector machine (SVM) and k-nearest neighbors (KNN). We checked the signal features that recognize jamming signals. After the jamming attack model the developed grey wolf optimizer version known as IGWO (improved grey wolf optimizer) has been discussed for feature extraction of software usability. Four separate metrics were employed as features to detect jamming attacks in order to evaluate the machine learning models. This novel feature extraction method is crucial for improving the accuracy of jamming detection.
The measurements of these parameters were gathered through a simulation of a real setting. And generated a large dataset using these parameters.
The simulation results illustrate that the KNN algorithm based on jamming detection could diagnose jammers having a minimal likelihood of false alarms and a high level of accuracy.
Malignancy Detection in Lung and Colon Histopathology Images by Transfer Learning with Class Selective Image Processing
Due to its ferocity enormous metastatic potential and variability cancer is responsible for a disproportionately high number of deaths. Cancers of the lung and colon are two of the most common forms of the disease in both sexes worldwide. The excellence of treatment and the endurance rate for cancer patients can be greatly improved with early and precise diagnosis.
We suggest a computationally efficient and highly accurate strategy for the rapid and precise diagnosis of lung and colon cancers as a substitute for the standard approaches now in use. The training and validation procedures in this work made use of an enormous dataset consisting of lung and colon histopathology pictures. There are 25000 Histopathological Images (HIs) in the dataset split evenly among 5 categories (mostly lung and colon tissues). Before training it on the dataset a pretrained neural network (AlexNet) had its four layers fine-tuned.
The study enhances malignancy detection in lung and colon histopathology images by applying transfer learning with class-selective image processing. Instead of enhancing the entire dataset a targeted contrast enrichment was applied to images from the underperforming class improving the model's accuracy from 92.3% to 99.2% while reducing computational overhead. CONCLUSION: This approach stands out by emphasizing class-specific enhancements leading to significant performance gains. The results meet or exceed established CAD metrics for breast cancer histological images demonstrating the method's efficiency and effectiveness.
A Passenger Ship Emergency Evacuation Line Efficiency Evaluation Operator Considering Safety Capacity and its Application
Aiming at evaluating the efficiency of passenger ship emergency evacuation routes this study proposes an evaluation operator for passenger ship emergency evacuation routes and then conducts simulation research on the passenger ship evacuation process.
The main innovations of this study are as follows. Firstly this study proposes an evaluation operator for passenger ship emergency evacuation lines based on consideration of the passage capacity ship heeling effect traffic flow obstruction personnel evacuation mood accident location and other factors. Secondly this study discretizes the emergency evacuation path of passenger ships into a three-dimensional topological network structure and uses simulation data to calculate the traffic capacity between nodes. By comparing the calculated value with the simulated value a risk assessment method for emergency evacuation is given. Thirdly this study takes the three-story ro-ro passenger ship in the European “SAFEGUARD” project as an example and gives three passenger ship evacuation simulation processes under different passenger flow densities.
The results of the calculation example show that the risk-prone areas are mainly concentrated in the passenger ship stairs and the risk value increases with the passenger flow.
The calculation results verify the effectiveness of the emergency evacuation line efficiency evaluation operator proposed in this study.
Revolutionizing Patient Safety: Machine Learning and AI for the Early Detection of Adverse Drug Reactions and Drug-Induced Toxicity
Adverse drug reactions and drug-induced toxicity provide significant issues in drug research jeopardizing patient safety and driving up healthcare costs. Toxicity has a greater potential impact than infectious diseases since it is less visible. Early diagnosis of these difficulties is critical to determining a drug's safety and viability profile. The combination of machine learning and artificial intelligence has marked a watershed moment in the identification of early adverse drug reactions and toxicity. These computational approaches enable rapid extensive and precise prediction of likely adverse drug reactions and toxicity even before practical drug manufacture preclinical testing and clinical trials. This paradigm change strives to create more efficient and safe drugs lowering the likelihood of drug withdrawal. This comprehensive review investigates the critical role of machine learning and artificial intelligence in quickly detecting adverse drug reactions and toxicity including approaches from data mining to deep learning. It lists essential databases modelling techniques and software that may be used to model and predict a wide range of toxicities and adverse drug reactions. This review provides a comprehensive overview outlining recent developments and projecting future opportunities in machine learning and artificial intelligence-driven rapid identification of adverse drug reactions and drug-induced toxicity. It highlights the capabilities of these technologies and their enormous potential to improve patient safety and revolutionize medication discovery.
Utilizing AspectJ for Defense Against Evasive Malware Attacks in Android System
Mobile devices have become an integral part of our digital lives facilitating various tasks and storing a treasure trove of sensitive information. However as more people utilize mobile devices sophisticated cyber threats are emerging to elude traditional security measures.
The use of evasion techniques by malicious actors presents a significant challenge to mobile security necessitating creative solutions. In this work we investigate the potential critical role that the aspect-oriented programming paradigm AspectJ can play in strengthening mobile security against evasion attempts. Evasion techniques cover a wide range of tactics including runtime manipulation code obfuscation and unauthorized data access.
These tactics usually aim to bypass detection and avoid security measures. In order to address the aforementioned issues this paper uses AspectJ to give developers a flexible and dynamic way to add aspects to their coding structures so they can monitor intercept and respond to evasive actions. It illustrates how AspectJ can improve mobile security and counteract the long-lasting risks that evasion techniques present in a dynamic threat landscape.
Consequently this work proposes a novel defense mechanism incorporating AspectJ into a significant paradigm of security against evasion with 91.33% accuracy and demonstrates the successful detection of evasive attacks.
MCDM Method for Managing the Water Resources Based on Possibility Theory under Bipolar Fuzzy Environment
Uncertainty is a common factor in every real-life decision-making problem. Possibility theory is one uncertainty theory in Fuzzy sets (FS). The possibility-based decision-making under a fuzzy environment is a significant multi-criteria decision-making (MCDM) method.
A bipolar FS is an extension of a fuzzy set. With the bipolarity concept we can handle both positive and negative thoughts. In this study we have provided a possibility mean of a bipolar fuzzy number. We have developed a ranking method for bipolar fuzzy numbers using this possibility concept. A novel possibility MCDM method is suggested for solving the water resources management (WRM) in the Nagpur area Maharashtra State India.
The MCDM technique is an effective tool for solving WRM problems in an area. Many uncertainties and bipolarities occur together in the Nagpur water resources systems WRM technique with fuzzy is one approach that can be used to solve the area's water problem. We have used the proposed MCDM to address the water-related issues of this district. With this proposed MCDM method numerically we employed water resource problems under a bipolar fuzzy environment.
The Nagpur area is covered by Basaltic rock and faces water shortage. The district is experiencing severe water shortages. Groundwater surface water and rainfall are three water resources considered as alternatives. According to the proposed MCDM technique in Nagpur district Groundwater is the best water source from three flanks: quality of water affordability and availability.
Biofuels Policy as the Indian Strategy to Achieve the 2030 Sustainable Development Goal 7: Targets, Progress, and Barriers
The use of 20% blended biofuels to fossil fuels is one of the important targets of the Government of India to address the impacts of Climate Change energy-related environmental pollution and illnesses due to air pollution.
The National Policy on Biofuels 2018 (NPB 2018) is in place to boost the emerging production of biofuels and therefore respond to different international agreements including the Sustainable Development Goals (SDGs) and the Paris Agreement. Hence this article examined the production of biofuels in India in line with Agenda 2030 to project the share to be taken by biofuels as its contribution to the country’s energy needs.
The results were compromising; it was observed that the data from 2000 up to 2017 were not on the side of realizing the targets of production and consumption of biofuels in India whereas the data from 2018 up to now showed a hope of achieving 2030 set goal of E20 petrol in 2025-26 and E5 diesel in 2030. It was clear that the production of bioethanol was boosting compared to its sibling biodiesel and renewable energy will continue to have a hard take a good share in the total annual energy used in India.
It is recommended to share data between different stakeholders to promote more research as the low performance in achieving the targets was due to poor communication and missing technology rather than the lack of feedstock or unavailability of production facilities.
Data Security and Privacy Preservation in Cloud-Based IoT Technologies: an Analysis of Risks and the Creation of Robust Countermeasures
The Internet of Things (IoT) is a revolutionary technology being used in many different industries to improve productivity automation and comfort of the user in the cloud and distributed computing settings. Cloud computing is essential because it makes data management and storage more effective by automatically storing and examining the enormous amounts of data generated by Internet of Things applications. End users companies and government data are consequently migrating to the cloud at an increasing rate. A survey of the literature however reveals a variety of issues including data integrity confidentiality authentication and threat identification that must be resolved to improve data security and privacy. To effectively address contemporary data security concerns the existing approaches need to be improved. Ensuring secure end-to-end data transmission in a cloud-IoT situation requires innovative and dependable protocol architecture. New technologies that address some of the issues related to cloud data include edge computing fog blockchain and machine learning. This paper provides a thorough examination of security risks classifying them and suggesting possible defenses to safeguard cloud-IoT data. It also highlights innovative approaches such as blockchain technology and machine learning applied to privacy and data security. The paper also explores existing issues with respect to data privacy and security in today's cloud-IoT environments. It suggests possible future directions including the need for end-user authentication enhanced security and procedures for recovering data in the event of an attack.
Multiple Criteria-Based Intelligent Techniques for Efficient Handover in Next-Generation Networks
Several wireless technologies are assumed to be operating in cooperation for next-generation networks. These networks offer various services to mobile users; however efficient handover between different networks is the most challenging task in high mobility scenarios. The traditional signal strength-based handover algorithms are not able to cope with mobile users' high-quality requirements.
In this paper multiple criteria-based intelligent techniques are proposed to deal with inefficiencies related to handover. This technique makes use of artificial neural networks that take multiple parameters as inputs in order to predict the degradation of parameters. These predictions are further used to design rules for initiating handover procedures prior to service quality degradation.
Based on the prediction results obtained by deep neural networks the handover decisions are recommended according to the type of application: conversational or streaming.
The simulation results demonstrate the efficacy of the proposed method as there is an improvement of up to 40% and 25% in terms of handover rate and service disruption time respectively with an acceptable prediction error of 0.05.
Data Collection and Recharging of Sensor Node by Mobile Sink in Wireless Sensor Network
The wireless sensor network (WSN) has limited battery and storage capacity. The main challenge in static sink nodes is the lack of energy and data transfer to the base station (BS). A mobile sink (MS) is an excellent way to handle these difficulties in WSN. The use of an MS solution not only ensures long-term network functionality but also improves network performance. In this paper the MS-based data collection and recharging mobile device is used for data collection from sensor nodes and recharging the sensor nodes throughout the network whenever required. There are two types of MS-based solutions described as (i) Mobile devices for data collection from sensor nodes in the network. (ii) Mobile device for collecting data and recharging sensor nodes from the network. Finally in this paper we presented the advantages disadvantages and other criteria of both types. Also a potential framework is presented for data collection and recharging of sensor nodes by MS in WSN. The core contribution of this paper is present the state of art and future roadmap for data collection and recharging of sensor nodes by MS in WSN. The identification of the main open challenges and future direction in this research area are also highlighted and discussed.
Optimization of Graphene-based Antenna using Metasurface for THz Applications using Ensemble Learning models
These days communication is governed by scaled-down devices with extremely high data transfer rates. The days of data transfer rates in megabytes are long gone and the systems are touching data speeds of hundreds of gigabytes. To cater to this requirement the evolution of metasurface-based antennas working in the THz frequency range is gaining pace.
Metasurface layers in antennas have unique mechanical and electrical properties that make them feasible. Graphene is a preferred metasurface material when designing antennas. In this paper two novel designs of graphene-based metasurface antennas are proposed. These patch antennas with graphene metasurface layers have two configurations for slotted patches. Both graphene and gold-slotted patches are tested in this paper and results are presented and validated for various performance parameters like return loss peak gain peak directivity and radiation efficiency.
The design of the proposed Graphene Patch Antennas (GPA) is done in HFSS and once the design is complete the data is collected for variations in antenna parameters and is further processed via machine learning for better convergence in terms of return loss using ensemble learning. Optimal tuning of antenna components like substrate length patch length slot width etc. is done using two regressor algorithms viz Histogram-based Gradient Boosting Regression Tree (HBDT) and Random Forest Tree (RFT) in machine learning. The radiation efficiencies of the two designs presented in this work are 89.04% 97.54% an average radiation efficiency of of 93.29%. The bandwidth of the two antenna designs is 8.63 THz and 8.69 THz respectively.
Experimental results indicate that the proposed designs are achieved better bandwidth and radiation efficiencies compared to state-of-art designs.
Deep Learning-Based Network Security Situation Prediction for Sensor-Enabled Networks
Network security detection has become increasingly complex due to the proliferation of Internet nodes and the ever-changing nature of network architecture. To address this a multi-layer feedforward neural-network has been employed to construct a model for security threat detection which has enhanced network security protection.
Improving prediction accuracy and real-time performance this research suggests an optimal strategy based on Clockwork Recurrent-Neural-Networks(CW-RNNs) to handle nonlinearity and temporal dynamics in network security circumstances. We get the model to pick up on both the short-term and long-term temporal aspects of network-security situations by using the clock-cycle RNN. To further improve the network security scenario prediction model we tune the network hyperparameters using the Grey-Wolf-Optimization(GWO) technique. By incorporating a clock-cycle for hidden units the model can improve its pattern recognition capabilities by learning short-term knowledge from high-frequency update modules and preserving long-term memory from low-frequency update modules.
The optimized clock-cycle RNN achieves better prediction accuracy than competing network models when it comes to extracting nonlinear and temporal characteristics of network security scenarios according to the experimental data.
In addition our method is perfect for tracking massive amounts of data transmitted by sensor networks because of its minimal time complexity and outstanding real-time performance.
Evaluation of the Critical Success Factors for Household Product Sustainability
Sustainability and sustainable development have received growing attention in both industry and academia due to concerns regarding the rapid decrease in natural resources and increase in carbon emissions.
In this study we focus on the determination evaluation and analysis of the critical success factors in product sustainability by specifically focusing on the household goods industry. In the first phase of the study we determine the critical success factors by referring to the existing literature and opinions of the experts who have experience in the household goods industry. Next we use a trapezoidal type-2 fuzzy AHP algorithm to rank the determined criteria and discuss the main findings from a practical point of view.
Computational results bring several important managerial insights. First we observe that all three aspects of sustainability (economic environmental and social) should be considered to ensure product sustainability. Second the analysis reveals that cost (economic) quality (economic) generated waste and emission during the life cycle (environmental) energy and water consumption during the life cycle (environmental) and occupational health and safety (social) are among the highly ranked criteria.
In order to increase product sustainability the companies should determine ways to decrease water usage energy usage carbon emission and waste without neglecting the cost and quality of the product and without ignoring occupational health and safety.
BER Analysis of Underlay Cooperative Cognitive Radio-based NOMA System
The integration of Cooperative Communications (CC) Cognitive Radio (CR) technology and Non-Orthogonal Multiple Access (NOMA) techniques termed Cooperative Cognitive Radio used NOMA (CCR-NOMA) systems has emerged as a promising solution to address spectrum scarcity and connectivity challenges anticipated in sixth generation (6G) networks.
This studyaims to investigate the Bit Error Rate (BER) performance of underlay CCR-NOMA systems.
We derive precise closed-form expressions for BER at distant users under perfect and imperfect Channel State Information (CSI) conditions. These mathematical formulations are validated through Monte Carlo simulations.
Our results indicate that the near user 𝐶𝑈2 exhibits superior performance compared to the far user 𝐶𝑈2. Additionally distant users utilizing the CCR-OMA protocol demonstrate better BER performance than those employing the CCR-NOMA protocol.
The presence of imperfect CSI adversely affects BER performance. Moreover the derived closed-form expressions for BER in the investigated system align well with Monte Carlo simulations. These findings provide valuable insights for optimizing the performance of CCR-NOMA systems in real-world scenarios.
Evaluation of the Critical Success Factors for Household Product Sustainability
Sustainability and sustainable development have received growing attention in both industry and academia due to concerns regarding the rapid decrease in natural resources and increase in carbon emissions.
In this study we focus on the determination evaluation and analysis of the critical success factors in product sustainability by specifically focusing on the household goods industry. In the first phase of the study we determine the critical success factors by referring to the existing literature and opinions of the experts who have experience in the household goods industry. Next we use a trapezoidal type-2 fuzzy AHP algorithm to rank the determined criteria and discuss the main findings from a practical point of view.
Computational results bring several important managerial insights. First we observe that all three aspects of sustainability (economic environmental and social) should be considered to ensure product sustainability. Second the analysis reveals that cost (economic) quality (economic) generated waste and emission during the life cycle (environmental) energy and water consumption during the life cycle (environmental) and occupational health and safety (social) are among the highly ranked criteria.
In order to increase product sustainability the companies should determine ways to decrease water usage energy usage carbon emission and waste without neglecting the cost and quality of the product and without ignoring occupational health and safety.