Automated Summarization of Legal Texts: Evaluating Individual and Ensemble NLP Models
Dhull M., Dewangan A., Khalsa I.K., Gupta V., Biswas C., Carie A.
Conference paper, 3rd International Conference on Advancements in Smart, Secure and Intelligent Computing, ASSIC 2025, 2025, DOI Link
View abstract ⏷
The complexity and length of legal documents make it challenging to quickly identify key information. This study examines how natural language processing (NLP) techniques can be used to automate the summarization of legal texts. We implemented six advanced summarization models - Legal Pegasus, BART, Legal LED, Law2Vec, LSA, and T5 - and evaluated their performance separately. To boost accuracy, we created four ensemble models by merging these techniques and assessed their effectiveness. The results reveal the promise of NLP-driven summarization in streamlining legal document analysis and provide a comparative look at the strengths of individual versus ensemble approaches.
Blockchain and AI for Educational Data Analytics in the Modern Education System
Book chapter, Blockchain and AI in Shaping the Modern Education System, 2025,
View abstract ⏷
This chapter explores the transformative potential of integrating blockchain and artificial intelligence (AI) technologies within educational data analytics. It begins by examining blockchain's capacity to enhance data security, streamline record-keeping, and ensure transparent credential verification. Concurrently, it analyzes AI's role in enabling adaptive learning, predictive modeling, and insightful data analysis to improve student outcomes and optimize educational strategies. The chapter further evaluates the synergistic benefits of combining blockchain and AI, proposing a robust framework to address prevalent challenges in the education sector, including data privacy, security, and personalized learning. By securing student records through blockchain's immutability and enhancing personalized learning experiences via AI-driven analytics, the chapter presents a comprehensive approach to modernizing educational systems. Additionally, it addresses technical challenges such as scalability and interoperability, alongside ethical considerations like data privacy, consent, and algorithmic bias. The chapter concludes with a call for collaborative efforts among educators, technologists, and policymakers to leverage these technologies, navigate their challenges, and fully realize their potential in revolutionizing education.
Deep Learning Approaches for Intelligent Cyber Threat Detection in Modern Education Systems
Book chapter, Blockchain and AI in Shaping the Modern Education System, 2025,
View abstract ⏷
In the ever-evolving landscape of modern education systems, the integration of technology has become ubiquitous, opening new avenues for teaching and learning. However, this increased reliance on digital platforms has also given rise to unprecedented cybersecurity challenges, necessitating advanced detection mechanisms to safeguard sensitive educational data. This book chapter explores the application of deep learning approaches for intelligent cyber threats detection in the context of the modern education system. The chapter begins by providing a comprehensive overview of the evolving cyber threat landscape within educational institutions, highlighting the diverse range of attacks targeting student records, intellectual property, and critical infrastructure. It emphasizes the need for proactive and adaptive cybersecurity measures to counteract these threats effectively. Subsequently, the chapter delves into the foundational principles of deep learning, elucidating its capacity to autonomously learn intricate patterns and anomalies from vast datasets. Various deep learning architectures, such as convolutional neural networks and recurrent neural networks are discussed in the context of their applicability to cybersecurity in education. The practical implementation of deep learning models for cyber threats detection is then explored. Case study that illustrate how these models can analyze detect malware, and identify suspicious activities, thereby fortifying the resilience of educational systems against cyber threats. In conclusion, this book chapter provides a comprehensive exploration of deep learning approaches as a potent tool for intelligent cyber threats detection in modern education systems.
Machine learning and deep learning with Swarm algorithms
Book chapter, Swarm Intelligence: Theory and Applications in Fog Computing, Beyond 5G Networks, and Information Security, 2025, DOI Link
Disentangling Proximate and Distal Controls on Methane Variability in South Asian Rivers Using Machine Learning
Article, ACS ES and T Water, 2025, DOI Link
View abstract ⏷
The ubiquitous oversaturation of methane (CH4) in fluvial environments is hypothesized to be sustained by both proximate and distal factors, which reflect their respective roles in modulating CH4metabolism through water chemistry as well as hydrological, morphological, and landscape features. Yet, efforts to disentangle their complex interplay in regulating riverine CH4variability remain limited. Herein, we used machine learning (ML) approach to examine drivers of CH4variability in five South Asian river basins where CH4concentration varied widely (0.01–455.75 μmol L–1). Among proximate variables, dissolved oxygen (DO) emerged as the strongest predictor of CH4concentration, explaining 61–75% of CH4variability explained by different ML models, followed by total phosphorus (TP) and dissolved organic carbon. Conversely, the extent of built-up area (%) was the key predictor among the distal variables. When combining proximate and distal variables in ML analysis, proximate factors emerged as the dominant drivers, whereas distal factors had only a marginal impact suggesting that local biogeochemical conditions outweigh broader landscape features in determining fluvial CH4variability. Our ML analysis reveals that while remote-sensing-derived distal variables can assist in predicting CH4concentrations, they offer limited mechanistic insights. Therefore, integrating proximate factors with landscape variables is important in deriving a comprehensive and mechanistic understanding of CH4dynamics in river networks.
Double Layered Blockchain-based trust model for secure interest and data forwarding in Vehicular Information Centric Network
Article, Array, 2025, DOI Link
View abstract ⏷
Vehicular Information Centric Networks(V-ICNs) which is an alternate to the traditional Vehicular Adhoc Networks (VANETs) is proposed to enable the content-based addressing instead of IP based data access to improve the efficiency of the Vehicular Network. V-ICN is more susceptible to security attacks from several sources because of its wireless, heterogeneous connection style and highly dynamic architecture. In contrast to entity-based security authentication, it is crucial to think about how to safeguard the data's security. Although reputation based quantification has been used in state-of-the-art research to assess the dependability of interactive data, there are still some problems with the design of safe reputation management systems. This result in low efficiency, inadequate security and unreliable administration. The Double Layered Blockchain (DLB) technique for communication security in vehicle Information-Centric Networks will be presented in this study. It will take into account of both the worldwide reputation of vehicle chain and the one-day local information chain. In this, each vehicle's activities that are documented in the Local One-day Message Blockchain (LOMB) will be used to update the reputation score of the vehicles on a regular basis. In addition, the proposed work would also include a Secured Neighbourhood Recognition Protocol (SNRP) to introduce flexible connectivity among the vehicle nodes and the blockchain network. Based on the Experimental Analysis, the proposed methodology is outperformed in comparison with the existing models in terms of network throughput, interest success rate, varying interest generation rate and the interest success rate.
Blockchain-enabled SDN in resource constrained devices
Book chapter, Blockchain-based Cyber Security: Applications and Paradigms, 2024,
Enhancing Agricultural Decision-Making Through Machine Learning-Based Crop Yield Predictions
Marapelli B., Anamalamudi L., Potluri C.S., Carie A., Anamalamudi S.
Conference paper, Lecture Notes in Networks and Systems, 2024, DOI Link
View abstract ⏷
Food production through Agriculture plays an important role in keeping the world’s population hunger-free and nations economically secure. The continuous change in land minerals, weather situation, and pesticide usage affect the yield of the crops. Farmers can choose successful crops for the season with the help of machine learning algorithms used for crop yield prediction. In this study, we forecasted agricultural production using numerous kinds of machine learning models while considering several factors that affect crop yields, such as rainfall, temperature, and pesticide use. By merging multiple separate model predictions, ensemble machine learning models improve the performance of the machine learning models. We have worked with individual models and ensemble models like SVR, RandomForestRegressor, LinearRegressor, and DecisionTreeRegressor to predict crop yield and found an ensemble solution that combines the strengths of both the stacked generalization model and the gradient boost algorithm which can provide improved accuracy and robustness in crop yield prediction. According to the findings, the ensemble solution provided an R2 score of 98 percent, which is higher than the R2 scores of 96 percent obtained using the Decision Tree Regressor and 89 percent obtained using the Gradient Boosting Regressor.
Monitoring and enhancing the co-operation of IoT network rhrough scheduling function based punishment reward strategy
Article, PLoS ONE, 2024, DOI Link
View abstract ⏷
The Internet of Things (IoT) has revolutionized the connectivity of physical devices, leading to an exponential increase in multimedia wireless traffic and creating substantial demand for radio spectrum. Given the inherent scarcity of available spectrum, Cognitive Radio (CR)- assisted IoT emerges as a promising solution to optimize spectrum utilization through cooperation between cognitive and IoT nodes. Unlicensed IoT nodes can opportunistically access licensed spectrum bands without causing interference to licensed users. However, energy constraints may lead to reduced cooperation from IoT nodes during the search for vacant channels, as they aim to conserve battery life. To address this issue, we propose a Punishment-reward-based Cooperative Sensing and Data Forwarding (PR-CSDF) approach for IoT data transmission. Our method involves two key steps: (1) distributing sensing tasks among IoT nodes and (2) enhancing cooperation through a reward and punishment strategy. Evaluation results demonstrate that both secondary users (SUs) and IoT nodes achieve significant utility gains with the proposed mechanism, providing strong incentives for cooperative behaviour.
Performance Modeling and Analysis of Internet of Things Enabled Healthcare Monitoring Systems
Conference paper, Proceedings - 2023 15th IEEE International Conference on Computational Intelligence and Communication Networks, CICN 2023, 2023, DOI Link
View abstract ⏷
The rise of the Internet of Things (IoT) has brought about significant changes in the healthcare sector, leading to the development of more advanced healthcare monitoring systems. To ensure that these systems are dependable and effective, it is crucial to conduct performance modeling and analysis. This research is focused on evaluating the performance of healthcare monitoring systems enabled by IoT. It takes into account various factors, including IoT network infrastructure, communication protocols, data processing, and storage. Moreover, the study utilizes machine learning technique that is Artificial Neural Network to assess healthcare data collected through IoT networks. By employing simulationbased methods, this investigation models how the system behaves and evaluates its performance using metrics like response time, throughput, and reliability. The findings from this study provide valuable insights into the performance of these systems, pinpointing areas where improvements can enhance the overall efficiency of healthcare monitoring systems. As a result, this research makes a substantial contribution to the improvement of efficient IoT-enabled healthcare monitoring systems, which, in turn, offer dependable and cost-effective healthcare solutions.
Crop Prediction System with a Convolution Neural Network Model
Marapelli B., Balcha B.S., Nega A.T., Carie A.
Conference paper, Smart Innovation, Systems and Technologies, 2023, DOI Link
View abstract ⏷
Most of the countries in this world are dependent on agriculture, and a lot of the population depends on their predator’s knowledge in choosing the crop selection. However, as the years pass, change in agricultural land makes it difficult for the farmers in getting the expected yield. To improve the yield from the agricultural land, a change in the crop is recommended. Instead of using predator’s knowledge for crop selection, we can make it easy and productive by using precision agriculture a new farming technique that recommends crops based on nutrition values in the land and weather conditions. In our work, we build a deep learning model using convolutional neural networks (CNNs) to predict a crop for selected land based on nutrition values in the land and weather conditions. The CNN model is trained and tested using research data with nutrition values and weather conditions that predict suitable crops for that land. Our model achieved an accuracy score of 93.69% on training data and 94.20% on testing data. The CNN model is compared with other baseline models.
Software Defect Prediction Using ROS-KPCA Stacked Generalization Model
Marapelli B., Carie A., Islam S.M.N.
Conference paper, Smart Innovation, Systems and Technologies, 2023, DOI Link
View abstract ⏷
Software quality assurance is an area that deals with software defect prediction also. Identifying and eliminating defects is a crucial task that helps organizations deliver quality software products to customers. Machine learning approaches help in identifying software modules that are defective and which are not defective. The existing software defect prediction datasets contain data with features that could classify projects are defective or not. The machine learning model’s performance will be degraded with the existence of noisy attributes and class imbalance problems. In this work, we propose a ROS-KPCA-SG model (Random Over Sampling-Kernel Principal Component Analysis-Staked Generalization Model) model to solve the noisy dataset and class imbalance problems and to improve the software defect prediction accuracy. The performance of the ROS-KPCA-SG model is compared with individual models with different combinations of sampling techniques. The results show the proposed ROS-KPCA-SG model solves the problems and gives better performance than other models. The AUC-ROC score is between 0.9 and 1 for the ROS-KPCA-SG model on all the datasets, and the accuracy is near to 90% and above which is a higher value than other models. The proposed model gives accuracy on datasets CM1 is 98%; JM1 is 89%; PC1 is 98%; KC1 is 92% and with KC2 dataset is 89%.
Secure Multi-Level Privacy-Protection Scheme for Securing Private Data over 5G-Enabled Hybrid Cloud IoT Networks
Budati A.K., Vulapula S.R., Shah S.B.H., Al-Tirawi A., Carie A.
Article, Electronics (Switzerland), 2023, DOI Link
View abstract ⏷
The hybrid cloud is a secure alternative for enterprises to exploit the benefits of cloud computing to overcome the privacy and security concerns of data in IoT networks. However, in hybrid cloud IoT, sensitive items such as keys in the private cloud can become compromised due to internal attacks. Once these keys are compromised, the encrypted data in the public cloud are no longer secure. This work proposes a secure multilevel privacy-protection scheme based on Generative Adversarial Networks (GAN) for hybrid cloud IoT. The scheme secures sensitive information in the private cloud against internal compromises. GAN is used to generate a mask with the input of sensory data-transformation values and a trapdoor key. GAN’s effectiveness is thoroughly assessed using Peak Signal-to-Noise Ratio (PSNR), computation time, retrieval time, and storage overhead frameworks. The obtained results reveal that the security scheme being proposed is found to require a negligible storage overhead and a 4% overhead for upload/retrieval compared to the existing works.
A novel approach to minimize the Black Hole attacks in Vehicular IoT Networks
Conference paper, ACM International Conference Proceeding Series, 2023, DOI Link
View abstract ⏷
Vehicular Ad-hoc IoT Networks (VA-IOT) have gained significant attention due to their ability to enable the distributed data transmission between vehicles to vehicle. However, VA-IOT are susceptible to various security threats, including the Black Hole attack. With Black Hole attacks, an intruder or malicious node attracts the internet traffic by broadcasting fake messages and drops all the received packets, which can significantly impact the network's performance. To mitigate, this paper presents an new mechanism to minimize the Black Hole attack on VANETs by combining two techniques: A trust management system and an intrusion detection system. The proposed approach involves assigning trust values to each vehicle based on their past behavior and routing packets through only trusted nodes. Additionally, an intrusion detection system is used to identify malicious nodes that violate the trust threshold and to take appropriate measures. The performance of the proposed approach is outperformed in terms of achievable end-To-end throughput and minimized network delays.
Redundant Transmission Control Algorithm for Information-Centric Vehicular IoT Networks
Article, Computers, Materials and Continua, 2023, DOI Link
View abstract ⏷
Vehicular Adhoc Networks (VANETs) enable vehicles to act as mobile nodes that can fetch, share, and disseminate information about vehicle safety, emergency events, warning messages, and passenger infotainment. However, the continuous dissemination of information from vehicles and their one-hop neighbor nodes, Road Side Units (RSUs), and VANET infrastructures can lead to performance degradation of VANETs in the existing host-centric IP-based network. Therefore, Information Centric Networks (ICN) are being explored as an alternative architecture for vehicular communication to achieve robust content distribution in highly mobile, dynamic, and error-prone domains. In ICN-based Vehicular-IoT networks, consumer mobility is implicitly supported, but producer mobility may result in redundant data transmission and caching inefficiency at intermediate vehicular nodes. This paper proposes an efficient redundant transmission control algorithm based on network coding to reduce data redundancy and accelerate the efficiency of information dissemination. The proposed protocol, called Network Cording Multiple Solutions Scheduling (NCMSS), is receiver-driven collaborative scheduling between requesters and information sources that uses a global parameter expectation deadline to effectively manage the transmission of encoded data packets and control the selection of information sources. Experimental results for the proposed NCMSS protocol is demonstrated to analyze the performance of ICN-vehicular-IoT networks in terms of caching, data retrieval delay, and end-to-end application throughput. The end-to-end throughput in proposed NCMSS is 22% higher (for 1024 byte data) than existing solutions whereas delay in NCMSS is reduced by 5% in comparison with existing solutions.
Corrections to “A Human-in-the-Loop Probabilistic CNN-Fuzzy Logic Framework for Accident Prediction in Vehicular Networks†[Jul 21 15496-15503]
Usman M., Chettupally A.C., Marapelli B., Bedru H.D., Biswas K.
Erratum, IEEE Sensors Journal, 2022, DOI Link
View abstract ⏷
IN THE above article [1], the surname Chettupally of the author Anil Carie Chettupally was inadvertently omitted from the authors list when the article was originally published. To avoid any ambiguity for readers, the more standard references [2] and [3] have been added as [35] and [36] before (1) on page 15498 and in step 5 on page 15500. Furthermore, the dataset sources [4], [5], and [6] are identified as [37], [38], and [39], instead of [22].
A Bi-Level Stochastic Model with Averse Risk and Hidden Information for Cyber-Network Interdiction
Li M.C., Dong W., Zheng X., Carie A., Tian Y.
Conference paper, Lecture Notes in Networks and Systems, 2022, DOI Link
View abstract ⏷
This paper proposes a method to enable a risk-averse and resource-constrained network defender to deploy security countermeasures in an optimal way to prevent multiple potential attackers with uncertain budgets. To solve the problem of information asymmetry between the attacker and the defender, a fake countermeasure (FC) is placed on the arc, and the situation of multiple attackers is also taken into consideration. This method is based on the risk aversion bi-level stochastic network interdiction model on the attack graph, which can easily map the path of attackers. Meanwhile, our method can minimize the weighted sum of all losses and minimize the risk of the defender’s key nodes being destroyed. At the same time, in order to prevent the key node of the defender from being destroyed, the risk condition value measurement is taken into account in the stochastic programming model. We design a SA-CPLEX algorithm to provide a high-quality approximate optimal solution. And computational results suggest that our method provides better network interdiction decisions than traditional deterministic and risk-neutral models.
Applications of Game Theory in Vehicular Networks: A Survey
Sun Z., Liu Y., Wang J., Li G., Anil C., Li K., Guo X., Sun G., Tian D., Cao D.
Article, IEEE Communications Surveys and Tutorials, 2021, DOI Link
View abstract ⏷
In the Internet of Things (IoT) era, vehicles and other intelligent components in an intelligent transportation system (ITS) are connected, forming vehicular networks (VNs) that provide efficient and safe traffic and ubiquitous access to various applications. However, as the number of nodes in an ITS increases, it is challenging to satisfy a varied and large number of service requests with different quality of service (QoS) and security requirements in highly dynamic VNs. Intelligent nodes in VNs can compete or cooperate for limited network resources to achieve the objective for either an individual or a group. Game theory (GT), a theoretical framework designed for strategic interactions among rational decision makers sharing scarce resources, can be used to model and analyze individual or group behaviors of communicating entities in VNs. This paper primarily surveys the recent developments of GT in solving various challenges of VNs. This survey starts with an introduction to the background of VNs. A review of GT models studied in the VNs is then introduced, including the basic concepts, classifications, and applicable vehicular issues. After discussing the requirements of VNs and the motivation of using GT, a comprehensive literature review on GT applications in dealing with the challenges of current VNs is provided. Furthermore, recent contributions of GT to VNs that are integrated with diverse emerging 5G technologies are surveyed. Finally, the lessons learned are given, and several key research challenges and possible solutions of applying GT in VNs are outlined.
Software Effort Estimation with Use Case Points Using Ensemble Machine Learning Models
Marapelli B., Carie A., Islam S.M.N.
Conference paper, International Conference on Electrical, Computer, and Energy Technologies, ICECET 2021, 2021, DOI Link
View abstract ⏷
Software development process includes estimating effort as a crucial task. The Use Case Point Analysis (UCPA) is a well know size metric that can be used to calculate effort. The software size is measured using use case diagrams in the UCPA method, using the calculated software size the effort required to complete the project is estimated. The traditional effort estimation with statistical methods is not accurate when compared to the real effort. In-accurate estimation of effort leads to problems with cost calculation and human resources calculation and it might lead to project failure. Machine learning techniques based on regression might help in estimating the effort with accuracy. In this study we proposed a method to identify best performing regression based machine learning model using two data sets Dataset1, Dataset2. Ensembles of different Machine learning methods is created and compared with individual methods and other ensemble methods to find an accurate estimation model. The results shows the individual model SVR and ensemble model GBR gives the best performance with a regression score above 98% with both data sets.
A Human-in-the-Loop Probabilistic CNN-Fuzzy Logic Framework for Accident Prediction in Vehicular Networks
Usman M., Carie A., Marapelli B., Bedru H.D., Biswas K.
Article, IEEE Sensors Journal, 2021, DOI Link
View abstract ⏷
The vehicle accident prediction methods are designed to improve the vehicular safety and reduce the rescue response time in the case of an accident. The existing accident prediction methods, however, do not involve Human-in-the-Loop, i.e., do not consider the emotional state of a driver to predict the likelihood of an accident. We propose a Probabilistic Convolutional Neural Network (CNN)-Fuzzy Logic framework that involves Human-in-the-Loop and takes into account the multiple input streams of sensor generated data, i.e., human emotions and traffic data. The features extracted from the CNN model are fed to our designed probabilistic graph-based inference model to determine the accident probability. The probability is then mapped with accident severity through fuzzy membership functions for accident prediction. The experiment results show the promising performance of our proposed framework, i.e., 93.1% accuracy of face expressions, 76.2% accuracy of heartbeat, and 76.9% accuracy of traffic inputs and predicts the accident likelihood with 90% accuracy. The comparison, with related works, shows that the proposed framework can predict accidents with higher probabilities.
Structural relational inference actor-critic for multi-agent reinforcement learning
Zhang X., Liu Y., Xu X., Huang Q., Mao H., Carie A.
Article, Neurocomputing, 2021, DOI Link
View abstract ⏷
Multi-agent reinforcement learning (MARL) is essential for a wide range of high-dimensional scenarios and complicated tasks with multiple agents. Many attempts have been made for agents with prior domain knowledge and predefined structure. However, the interaction relationship between agents in a multi-agent system (MAS) in general is usually unknown, and previous methods could not tackle dynamical activities in an ever-changing environment. Here we propose a multi-agent Actor-Critic algorithm called Structural Relational Inference Actor-Critic (SRI-AC), which is based on the framework of centralized training and decentralized execution. SRI-AC utilizes the latent codes in variational autoencoder (VAE) to represent interactions between paired agents, and the reconstruction error is based on Graph Neural Network (GNN). With this framework, we test whether the reinforcement learning learners could form an interpretable structure while achieving better performance in both cooperative and competitive scenarios. The results indicate that SRI-AC could be applied to complex dynamic environments to find an interpretable structure while obtaining better performance compared to baseline algorithms.
Non-cooperative game to balance energy and security in resource constrained IoT networks
Shah S.B.H., Wang L., Reddy P., Carie A.
Conference paper, IEEE INFOCOM 2020 - IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2020, 2020, DOI Link
View abstract ⏷
The restricted resources in IoT networks such as limited battery have resulted in strict requirements to prolong the network life time. To improve the communication, IoT nodes attempt to optimize the available energy in the sensor network, this makes them vulnerable to the malicious attacks from adversaries because of open scenario. In addition, enhancing the security level will consume the energy and decreases network life time. In order to balance energy and security in the network game theory concept is used. We design a non cooperative game between energy and security where the utilities of both energy and security players are maximized by controlling the number of nodes transmitting and hash length. We consider complete and incomplete information game and determine Nash equilibrium. Extensive simulation have been performed to examine Nash equilibrium. We obtained Nash equilibrium for both energy and security players.
RNN-CNN MODEL: A bi-directional long short-term memory deep learning network for story point estimation
Marapelli B., Carie A., Islam S.M.N.
Conference paper, CITISIA 2020 - IEEE Conference on Innovative Technologies in Intelligent Systems and Industrial Applications, Proceedings, 2020, DOI Link
View abstract ⏷
In recent years, an increased interest in the adaption of agile software development by companies. Using iterative methodology enables them to do issue-based estimation and respond quickly to changes in the requirements. Agile methodology adopts Story Point Approach to estimate the effort that involves a user story or a resolving issue. Unlike traditional estimation, Agile Methodology focuses on individual programming task estimation instead of whole project estimation. In this work, we approach story point estimation using the RNN-CNN model. We consider the contextual information in a user story in both forward and backward directions to build the RNN-CNN model. The proposed model adopts a Bi-directional Long Short-Term Memory (BiLSTM), a tree-structured Recurrent Neural Network (RNN) with Convolutional Neural Network (CNN), tries to predict a story point for a user story description. Here, BiLSTM forward and backward feature learning will make network preserve the sequence data and CNN makes feature extraction accurate. The experimental results show the improvement in estimating the story points with a user story as an input using the proposed RNN-CNN. Furthermore, the analysis shows that the proposed RNN-CNN model outperforms the existing model and gives 74.2 % R2 Score on the Bamboo data set.
Efficient data transfer in clustered IoT network with cooperative member nodes
Seema B., Yao N., Carie A., Shah S.B.H.
Article, Multimedia Tools and Applications, 2020, DOI Link
View abstract ⏷
Wireless Sensor Network (WSN) is composed of numerous tiny smart sensors nodes integrated with Internet of Things (IoT) play a crucial role in many applications. The IoT connects physical devices to form a network which consist of software, sensor for exchange of information. Clustering is most common technique for efficient energy utilization in WNS. Sensor nodes when they have data, forwards it to Cluster Head (CH) and CH transfers the received data from the sensor nodes to the sink. When the sink nodes are far away from CH, long-haul transmission consumes higher power. In this paper we propose efficient data transfer mechanism for clustered IoT network through the cooperation of member nodes. First, we use greedy algorithm to select cooperative sensor nodes to act as relay for long distance transmission. Then, to encourage sensor nodes in data forwarding, cluster head uses priority buffers to prioritize assisting sensor nodes data. Simulation results show that, the proposed approach conserves energy and increases the life-time of clustered IoT network.
Lifetime improvements of smart sensors maintenance protocol in prospect of IoT-based rampal power plant
Shah S.B.H., Wang L., Haque M.E., Islam M.J., Carie A., Kumar N.
Conference paper, Proceedings - 2020 16th International Conference on Mobility, Sensing and Networking, MSN 2020, 2020, DOI Link
View abstract ⏷
In the 21st century, the power quality and availability with customer demands to the society is the main challenging factor right now. Therefore, the gird monitoring system become a vital issue to monitor power grid system. The current smart grid system mainly focuses on smart metering system and improving the customer utility communication system. On customer management side, although those advancement provides an extra benefit, in spite of, the management of a grid system is one of the major dominating field in the era of Internet of Thing (IoT). From the field of industry and academia researches, Wireless sensor networks (WSNs) is getting to much popularity for monitoring power grid. Moreover, saving node energy enhance the lifespan of whole monitoring network. Due to lack of energy making policy, unnecessary activating all participate nodes consume node energy drastically, which is the main reason for shortening the lifetime of monitoring system. To solve this issue, maintenance technology provides the best opportunity to preserve node energy. This study investigates the issues that are associated with energy consumption using maintenance protocols in prospect of Rampal, Bangladesh power plant data. The modelling data has been collected through literature survey. Extensive simulation work has done for monitoring Rampal power using WSN. Finally, a comparative study of maintenance protocols were performed to maintain optimal network correction and thereafter extending the lifetime of monitoring network.
A novel network user behaviors and profile testing based on anomaly detection techniques
Tahir M., Li M., Zheng X., Carie A., Jin X., Azhar M., Ayoub N., Wagan A., Aamir M., Jamali L.A., Imran M.A., Hulio Z.H.
Article, International Journal of Advanced Computer Science and Applications, 2019, DOI Link
View abstract ⏷
The proliferation of smart devices and computer networks has led to a huge rise in internet traffic and network attacks that necessitate efficient network traffic monitoring. There have been many attempts to address these issues; however, agile detecting solutions are needed. This research work deals with the problem of malware infections or detection is one of the most challenging tasks in modern computer security. In recent years, anomaly detection has been the first detection approach followed by results from other classifiers. Anomaly detection methods are typically designed to new model normal user behaviors and then seek for deviations from this model. However, anomaly detection techniques may suffer from a variety of problems, including missing validations for verification and a large number of false positives. This work proposes and describes a new profile-based method for identifying anomalous changes in network user behaviors. Profiles describe user behaviors from different perspectives using different flags. Each profile is composed of information about what the user has done over a period of time. The symptoms extracted in the profile cover a wide range of user actions and try to analyze different actions. Compared to other symptom anomaly detectors, the profiles offer a higher level of user experience. It is assumed that it is possible to look for anomalies using high-level symptoms while producing less false positives while effectively finding real attacks. Also, the problem of obtaining truly tagged data for training anomaly detection algorithms has been addressed in this work. It has been designed and created datasets that contain real normal user actions while the user is infected with real malware. These datasets were used to train and evaluate anomaly detection algorithms. Among the investigated algorithms for example, local outlier factor (LOF) and one class support vector machine (SVM). The results show that the proposed anomaly-based and profile-based algorithm causes very few false positives and relatively high true positive detection. The two main contributions of this work are a new approaches based on network anomaly detection and datasets containing a combination of genuine malware and actual user traffic. Finally, the future directions will focus on applying the proposed approaches for protecting the internet of things (IOT) devices.
Cognitive radio assisted WSN with interference aware AODV routing protocol
Carie A., Li M., Marapelli B., Reddy P., Dino H., Gohar M.
Article, Journal of Ambient Intelligence and Humanized Computing, 2019, DOI Link
View abstract ⏷
Software configurable radio with dynamic spectrum support is the inherent property of Cognitive radio. Interoperability of Cognitive radio with wireless sensor network would enable the sensor nodes to access and transmit the application data in licensed PU free channels. Since wireless sensor nodes operate in heavily crowded ISM bands (902 MHz/2.4 GHz), there will be severe performance degradation during channel saturation and increased collision rate. With opportunistic spectrum access, enhanced performance can be achieved by minimizing the channel access collisions and control message overhead delays. To achieve, this paper proposes a novel approach to integrate the sensor nodes with cognitive radio (CR) nodes to route sensor data to sink using licensed channels opportunistically. We further extend our novel approach to cluster sensor node and CR nodes to achieve energy efficiency. Simulation results show that the proposed solution with dynamic spectrum access enhances the throughput, minimizes energy consumption and reduces the delay compared with existing solutions.
Hybrid Directional CR-MAC based on Q-Learning with Directional Power Control
Carie A., Li M., Liu C., Reddy P., Jamal W.
Article, Future Generation Computer Systems, 2018, DOI Link
View abstract ⏷
In this paper, we investigate the Hybrid Directional CR-MAC based on Q-Learning with Directional Power Control in cognitive radio (CR) systems. In CR systems, nodes can switch to heterogeneous non-overlapping channels opportunistically which offer higher achievable throughput. However, the random channel selection policy in existing CR-MAC protocol has problems like delay, packet collisions, and quality of service. The proposed channel selection scheme which is quite different from the traditional scheme is adopted by nodes to achieve context awareness and intelligence for adaptive channel selection. The nodes select a channel based on the results learned by interactions with the other nodes and channels. The directional transmission power control scheme allows the nodes to reuse the channels subject to interference constraints. The simulation results show that nodes using the proposed algorithm can select channels adaptively and optimal transmission power which helps to achieve high throughput and minimized power consumption.
An internet of software defined cognitive radio ad-hoc networks based on directional antenna for smart environments
Carie A., Li M., Anamalamudi S., Reddy P., Marapelli B., Dino H., Khan W., Jamal W.
Article, Sustainable Cities and Society, 2018, DOI Link
View abstract ⏷
In opportunistic radio access based Software defined Cognitive-MAC protocols, Common Control Channel (CCC) plays a significant role in providing node synchronization, legitimate channel access and exchanging cognitive control message. State of the art research in existing CR-MAC protocols design CCC in overlay based In-band or out-of-band spectrum with Omni-directional mode of control and data transmission. However, directional antenna based MAC protocol for software defined CR Ad-hoc networks offer great potential to reduce interference, efficiently reuse spatial spectrum, conserve node transmits power consumption, and extend directional communication range. To achieve above-mentioned features, this paper proposes a “Directional antenna Hybrid CCC based CR-MAC protocol for Smart Environments”. The present study proposes Directional Hybrid control channel with GPS which is used to exchange cognitive control information whereas directional opportunistic PU free channel is used for application data transmission. Experimental results reveal that proposed Directional antenna based Hybrid CCC-CR-MAC protocol has increased link throughput and reduced link delays with minimized node power consumption when compared with Omni-directional antenna based software defined CR-MAC protocols.
An extended framework for recovering from trust breakdowns in online community settings
Binmad R., Li M., Wang Z., Deonauth N., Anil Carie C.
Article, Future Internet, 2017, DOI Link
View abstract ⏷
The violation of trust as a result of interactions that do not proceed as expected gives rise to the question as to whether broken trust can possibly be recovered. Clearly, trust recovery is more complex than trust initialization and maintenance. Trust recovery requires a more complex mechanism to explore different factors that cause the decline of trust and identify the affected individuals of trust violation both directly and indirectly. In this study, an extended framework for recovering trust is presented. Aside from evaluating whether there is potential for recovery based on the outcome of a forgiveness mechanism after a trust violation, encouraging cooperation between interacting parties after a trust violation through incentive mechanisms is also important. Furthermore, a number of experiments are conducted to validate the applicability of the framework and the findings show that the e-marketplace incorporating our proposed framework results in improved efficiency of trading, especially in long-term interactions.
Recognizing the languages in WebPages – A framework for NLP
Rajesh S., Vandana L., Carie C.A., Marapelli B.
Conference paper, 2013 IEEE International Conference on Computational Intelligence and Computing Research, IEEE ICCIC 2013, 2013, DOI Link
View abstract ⏷
In this paper we describe an experimental system using java programming language which demonstrates a variety of application level tradeoffs available to distributed NLP applications. In this paper, we proposed language identification system with N-gram-based matching for document retrieval. By using a well known N-gram based algorithm for automatic language identification, we construct a system that dynamically adds language labels for whole documents or text fragments. © 2013 IEEE.
