مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726Linearity Enhanced Noise Cancelling Low Noise Amplifier for Ultra-Wideband ApplicationLinearity Enhanced Noise Cancelling Low Noise Amplifier for Ultra-Wideband Application779010.61882/jist.45666.13.50.77enNileshkumar KPatelGujarat Technological Universityhttps://orcid.org/0000-0003-3762-9542HASMUKH PKORINGAGovernment Engineering College Rajkot Gujarat Indiahttps://orcid.org/0000-0002-2521-3972202421The Low Noise Amplifier (LNA) stands as a crucial element RF receiver chain, demanding a delicate interplay of characteristics such as high gain, low noise figure (NF), superior linearity, and an extensive dynamic range. De-signing an ultrawideband (UWB) LNA poses a complex challenge as engineers grapple with intricate trade-offs inherent in these parameters. To address these challenges, noise cancellation techniques have emerged as valuable tools, revolutionizing the design of UWB LNAs by relaxing the traditional trade-off between bandwidth and input matching. This innovative approach not only enhances bandwidth but also effectively cancels out the un-desirable noise and nonlinearities from the input MOSFET. Despite the advancements afforded by noise cancellation, the broad bandwidth of UWB LNAs presents a significant hurdle. If the linearity is insufficient, the UWB LNA faces performance degradation due to increase in-band interference. In response, this article proposes an inventive linearization technique, a combination of Noise Cancelling (NC) and complementary derivative super-position (CDS), aiming to increase the linearity of UWB LNAs. Through meticulous simulations conducted using Cadence Virtuoso with GPDK090 library, the proposed LNA showcases impressive performance metrics across the UWB spectrum. Notably, it achieves a gain ranging from 12.5 dB to 15.5 dB, a noise figure within the range of 3.9 dB to 5.26 dB, and an IIP3 spanning from 6.3 dBm to 8.8 dBm. Remarkably, this innovative LNA accomplishes these feats while operating with a modest power consumption of 11.36 mW from a 1.2 V supply. This groundbreaking technique holds promise for significantly enhancing the efficiency and overall performance of UWB LNAs within contemporary RF receiver systems.

http://jist.ir/en/Article/Download/45666

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726Optimizing Hyperparameters for Customer Churn Prediction with PSO-Enhanced Composite Deep Learning TechniquesOptimizing Hyperparameters for Customer Churn Prediction with PSO-Enhanced Composite Deep Learning Techniques9111010.61882/jist.48088.13.50.91enMohammadSedighimaneshDepartment of Computer Engineering, Pooyesh Institute of Higher Education Qom, Iran0000-0003-3361-1097AliSedighimaneshDepartment of Computer Engineering, Pooyesh Institute of Higher Education Qom, Iran000000029224607XHessam Zandhessami Department of Management and Economics, Science and Research branch, Islamic Azad University, Tehran, Iran0000-0002-8815-915X2024924For Telecom operators, customer churn, i.e., the event when the customers leave a service provider, becomes a critical concern, studies have shown that acquiring new customers cost five times more than to retain them. In competitive markets, where is increasingly important, to sustain growth as well as profitability correctly predicting the tendencies for customer churn is important. Traditional predictive fashions frequently underperform due to the complex nature of client behavior. In this examine, we introduce a unique composite deep mastering framework whose hyperparameters are optimized the usage of the Particle Swarm Optimization (PSO) set of rules. Our method integrates a couple of neural community architectures to effectively capture each spatial and temporal patterns in client interactions. The PSO set of rules systematically first-rate-tunes parameters including activation functions, regularization techniques, gaining knowledge of rates, optimizers, and neuron counts—ensuing in a model that demonstrates robust overall performance. We evaluated our approach the usage of key metrics consisting of accuracy, precision, recollect, F1 score, and ROC AUC on a numerous purchaser dataset. Comparative analyses were conducted in opposition to established deep studying fashions (LSRM_GRU, LSTM, GRU, CNN_LSTM) in addition to other conventional methods (KNN, XG_BOOST, DEEP BP-ANN, BiLSTM-CNN, and Decision Tree). Experimental results stompy that our PSO-enhanced composite deep learning model stands out significantly compared with conventional models. Comparing the ROC-AUC scores of 0.932 and 0.93, F1 scores of 0.90 and 0.895, and accuracy rates of 83.2% and 93% on both Cell2Cell and IBM Telco datasets. it is indeed effective for practical churn prediction use incitements efficiencies. Var The experimental results demonstrate that our PSO express tree model outperforms conventional methods, achieving better performance with ROC totter score above 0.932 and 0.93, F 1 scores above 0.90 and 0.895 as well as accuracy rates in excess of 83.2% (% paper) and 93% (on the Telco data set) for Cell2Cell and IBM Telco respectively. This is further confirmation of its effectiveness and promise for practical churn prediction applications.

http://jist.ir/en/Article/Download/48088

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726A Holistic Approach to Stress Identification: Integrating Questionnaires and Physiological Signals through Machine LearningA Holistic Approach to Stress Identification: Integrating Questionnaires and Physiological Signals through Machine Learning11112110.61882/jist.48271.13.50.111enMrunalFatangareSchool of Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India HemlataOhalSchool of Engineering and Technology, Dr. Vishwanath Karad MIT World Peace University, Pune, India 000900014016833X20241014This research project presents a comprehensive methodology for stress identification by combining subjective self-report data and objective physiological signals. The proposed system employs a carefully designed questionnaire, tailored to different age groups, to enhance accuracy in stress assessment. Subjects respond to the questionnaire, providing valuable insights into their emotional well-being. Subsequently, physiological data is collected using an infrared (IR) sensor positioned beneath the wrist, close to the artery. The pulse data obtained is meticulously converted into a CSV file, allowing for efficient preprocessing. The preprocessing phase ensures the integrity of the data, preparing it for machine learning (ML) analysis. The study harnesses ML techniques, specifically SVM (Support Vector Machines) & KNN (K-Nearest Neighbors), to classify stress levels based on the pre-processed data. Through feature extraction, relevant patterns are identified, contributing to the accurate characterization of stress states. This integrative approach offers a robust framework for stress assessment, taking into account both subjective and physiological dimensions. Results demonstrate promising accuracy levels: Support Vector Machine (SVM) Reached a level of precision of 0.98 (+/- 0.20), Decision Tree showed 0.93 (+/- 0.30), and K-Nearest Neighbors (KNN) reached 0.88 (+/- 0.44). It also implements the voting classifier for improved performance of 98.6% of accuracy. These findings underscore the effectiveness of the proposed methodology in accurately identifying stress levels. Integrating subjective insights with objective physiological data not only enhances stress identification but also offers a comprehension of the intricate correlation between mental states and physiological reactions. This comprehensive strategy holds substantial implications across diverse domains such as healthcare, psychology, and human-computer interaction.

http://jist.ir/en/Article/Download/48271

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726Transmission Parameter-based Demodulation in Visible Light Communications using Deep LearningTransmission Parameter-based Demodulation in Visible Light Communications using Deep Learning12212910.61882/jist.47404.13.50.122enSarahAyashmDepartment of Electrical and Computer Engineering, Urmia University, Iran0000000344044489Seyed SadraKashefDepartment of Electrical and Computer Engineering, Urmia University, Iran0000-0002-1374-6915MortezaValizadehDepartment of Electrical and Computer Engineering, Urmia University, Iran0000-0003-2204-3303HastiAkhavanDepartment of Electrical and Computer Engineering, Urmia University, Iran2024718This paper proposes an innovative approach by employing a one-dimensional Convolutional Neural Network (CNN) for demodulation in VLC systems. The used Data-set is real and available online, providing a robust foundation for analysis. It encompasses modulated signals in seven different modulation types, with 29 transmission distances ranging from 0 to 140 centimeters. By accounting for the varying distances between the transmitter and receiver, the model can more accurately interpret the received signals. Additionally, the study suggests that utilizing memory to learn previous symbols, which is essential for mitigating the effects of inter-symbol interference (ISI), can significantly improve demodulation accuracy. Our results of memory-based demodulation show a better performance in contrast to the previous one (AdaBoost).

http://jist.ir/en/Article/Download/47404

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726Designing a Hybrid Algorithm that Combines Deep Learning and PSO for Proactive Detection of Attacks in IoT NetworksDesigning a Hybrid Algorithm that Combines Deep Learning and PSO for Proactive Detection of Attacks in IoT Networks13013810.61882/jist.48455.13.50.130enZahraBakhshaliDepartment of Information Technology Management, SRC, Islamic Azad University, Tehran, Iran0009-0008-2037-206XAlirezaPourebrahimiDepartment of Industrial Management, Karaj Branch, Islamic Azad University, Alborz, Iran0000-0001-5741-0260AhmadEbrahimiDepartment of Industrial and Technology Management, SRC, Islamic Azad University, Tehran, Iran0000-0002-5373-7466Nazanin PilevariDepartment of Industrial Management, West Tehran Branch, Islamic Azad University, Tehran, Iran0000-0002-0312-42312024112As a result, with the establishment of Internet of Things (IoT) at a booming&ensp;pace, the demand for effective, green security systems to detect cyber-attacks is escalating. Despite thorough investigation in this domain, the heterogeneous nature and multifaceted&ensp;characteristic of IoT data make successful attack detection a challenging task. This paper introduces a new method&ensp;for enhancing IoT attack detection through a hybrid deep learning model (CNN-GRU-LSTM) integrated with Particle Swarm Optimization (PSO) for hyperparameter optimization. This methodology consists of different steps, starting with a&ensp;CSV (Comma Separated Values) file to use it as the dataset, performing different data science operations like feature selection, calculating weights to balance the class for learning the model, etc. A hybrid CNN-GRU-LSTM model is subsequently established and trained with the integration of the merit of each algorithm: CNN for spatial feature abstraction, GRU for effectiveness in managing&ensp;the sequential information, and LSTM for discovering the long-range dependencies. The hyperparameters of the PSO algorithm are optimized to find the best combination&ensp;of features/parameters to improve detection performance and efficiency.  The results show remarkable accuracy and efficiency improvements over&ensp;traditional methods. H. PSO for Optimizing Hybrid Deep Learning Architecture The gainful approach&ensp;to building deep neural networks for IoT frameworks is through PSO based improvements. The results help to advance a realm of research work in IoT security and&ensp;lay a grouped foundation for further work in optimizing attack detection models with different machine learning algorithms and optimization approaches.

http://jist.ir/en/Article/Download/48455

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726NeuroIS: a State-of- the- Art Analysis NeuroIS: a State-of- the- Art Analysis 13915310.61882/jist.47292.13.50.139en Nahid EntezarianDepartment of Management, Faculty of Economic and Administrative Science, Ferdowsi University of Mashhad, Mashhad, Iran0009000615445100MohammadMehraeenDepartment of Management, Ferdowsi University of Mashhad, Mashhad, Iran0000-0002-4154-8975202477NeuroIS, the interdisciplinary field merging neuroscience and information systems, has recently garnered significant attention for its potential to enhance our understanding of human behavior in the tech context. This analysis delves into the current NeuroIS research landscape, examining key trends, methodologies, and discoveries in the field. By synthesizing recent research, the aim is to shed light on potential applications of NeuroIS across various domains and identify future research directions in this rapidly evolving field. Currently an emerging area within information systems, NeuroIS has a limited number of studies available. To aid researchers entering NeuroIS, we have analyzed 244 articles and summarize their findings to give more details of NeuroIS studies. This examination of literature reveals various avenues for future NeuroIS exploration, including influencing factors, measurement tools, and subject areas. We believe that our work will offer valuable insights for upcoming NeuroIS studies. The fusion of neuroscience and information systems holds immense potential for uncovering profound insights into human-computer interaction, decision-making processes, cognitive responses to technology, and enhancement of user experiences. As the field progresses, researchers are increasingly exploring innovative methods such as functional magnetic resonance imaging (fMRI), electroencephalography (EEG), and eye-tracking to unravel the complex mechanisms underlying human cognition in the digital age. By delving into the neurobiological basis of technology-mediated interactions, NeuroIS presents opportunities for designing more intuitive, efficient, and user-centric systems. With numerous uncharted research paths ahead, the future of NeuroIS looks promising, signaling a potential shift in how we understand and utilize information systems to impact human behaviors and decisions.

http://jist.ir/en/Article/Download/47292

مرکز منطقه ای اطلاع رسانی علوم و فناوریJournal of Information Systems and Telecommunication (JIST) 2322-143713502025726Credit Risk Prediction: An Application of Federated LearningCredit Risk Prediction: An Application of Federated Learning15416410.61882/jist.49000.13.50.154en SaraHoushmandDepartment of Industrial Engineering, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran0009-0007-9965-9348AmiralbadviDepartment of Industrial Engineering, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran0000-0002-7758-992020241227Credit risk is one of the major challenges faced by all financial institutions. Different institutions apply various techniques and models to reduce the risks associated with lending and other financial activities. However, due to the sensitivity of financial data and the diversity of modeling approaches, sharing data among institutions is extremely difficult, often impossible. As a result, improvements in credit risk prediction models typically occur in isolation, hindering collective progress toward higher accuracy and broader effectiveness. Federated learning offers a promising solution by allowing institutions to collaboratively train models without exposing or transferring sensitive data. In this research, we present a federated learning architecture for credit risk prediction that ensures privacy throughout the entire training process. Our results indicate that this approach not only protects data confidentiality but also maintains high predictive accuracy over numerous training rounds, offering a reliable and efficient framework for institutional adoption. The core contribution of this work is the development of a decentralized federated learning (FL) architecture tailored to heterogeneous, non-IID financial data. This framework enhances privacy, scalability, and regulatory compliance, and demonstrates performance advantages over traditional methods. In this article, we demonstrate that using five real-world credit risk datasets, the decentralized FL architecture significantly improves model accuracy (ranging from 71% to 99%) compared to traditional machine learning methods, especially in scenarios where privacy and communication efficiency are essential. While centralized FL achieves the highest average accuracy (up to 83%), the decentralized model provides a strong trade-off between performance and privacy-aware collaboration.

http://jist.ir/en/Article/Download/49000