International Journal of Information Technology最新文献

The smart world currently employs smart devices that are inextricably linked with everyday life. These smart devices are lightweight due to their small size, low memory capacity, low-power batteries, and limited computational capability.Conventional cryptographic algorithms aren’t applicable there. This demand leads to the development of lightweight cryptography (LWC). In various literature surveys, many researchers analysed lightweight ciphers in terms of area, throughput, latency, power consumption, energy dissipation, encryption-decryption time, etc. However, no single paper includes a variety of statistical cryptanalysis of LWCs. Such a type of bit-level data analysis checks the vulnerabilities of algorithms against different kinds of attacks. It ensures the difficulties of cryptanalysis. This paper has included seven classical lightweight ciphers PRESENT, SIMON, TEA, SPECK, CLEFIA, MICKEY2.0, and GRAIN V1, for statistical data analysis. The analysis includes non-homogeneity, avalanche ratio, entropy, floating frequency, frequency distribution, auto-correlation, periodicity, 4-gram pattern analysis.Moreover, four randomness like frequency, serial, run, and poker tests are also added. Finally, a comparative and compact discussion has made on ciphers’ efficiency. This study makes a trade-off, which proves the uniqueness of this work. It opens a new window for the upcoming researchers to search their work area.

Computer-assisted diagnosis (CAD) is preferred for cancer identification across the globe, which relies on computerized image processing. The creation of previous CAD instruments involved a semi-automated approach that employed traditional deep learning techniques. Such techniques are not well versed with CAD instruments in terms of accuracy. Therefore, the given manuscript presents a convolutional neural network normalized architecture embedded with Bayesian optimization and long short-term memory (CNN-BO-LSTM) for the identification of liver cancer. Early pre-processing is done on the input magnetic resonance imaging (MRI) scan images to improve clarity. Next, we used dynamic binary classification to apply the accurate and region of interest (ROI) extracting approach. It is followed by automatic retrieval of CNN-based appearances from the ROI approach. For classification purposes, LSTM is used, which categorizes the images as benign or malignant. The proposed design’s testing outcomes, which combine characteristics with CNN-based ROI extraction and LSTM classification, surpassed the current state-of-the-art techniques.

The existing limitations in extractive text summarization encompass challenges related to preserving contextual features, limited feature extraction capabilities, and handling hierarchical and compositional aspects. To address these issues, the RoBERTa-BiLSTM-CNN-Attention Extractive Text Summarization, i.e., the RBCA-ETS model, is proposed in this work. RoBERTa word embedding is used to generate contextual embeddings. Parallelly connected CNN and BiLSTM layers extract textual features. CNN focuses more on local features, and BiLSTM captures long-range dependencies that extend across sentences. These two feature sets are concatenated and forwarded to the attention layer, highlighting the most relevant features. In the output layer, a fully connected layer receives the attention vector and calculates sentence scores for each sentence. This leads to the generation of the final summary. The RBCA-ETS model has demonstrated superior performance on the CNN-Daily Mail (CNN/DM) dataset compared to many state-of-the-art methods, and it has also outperformed existing state-of-the-art techniques when tested on the out-of-domain DUC 2002 dataset.

Lymphoma, a malignancy originating from the aberrant proliferation of lymphocytes in the lymphatic system, becomes one of the prime concerns in health care across the world. The traditional, mono-therapeutic interventions using chemotherapeutic agents and radiation showed limitations in relation to disease control, suboptimal response, and development of intractable drug resistance. This study thus opens the reasons for combination therapies in detail by explaining the intrinsic rationale and revealing the intrinsic advantages of lymphoma treatment with an initiative underpinned by scrupulous analysis. Furthermore, the broad analysis had comparisons between combination therapies, were derived from studies, clinical trials, and actual instances. Key performance metrics like as overall response rates, progression-free survival, and overall survival, together with a thorough review of safety profiles, are used to determine the efficacy and long-term tolerability of various combination regimens. Using the PROMETHEE-II technique, this study has attempted to offer a comprehensive knowledge of the complex therapeutic landscape by addressing various issues and factors such as drug-disease interactions, toxicities, cost-effectiveness, and accessibility. This study avails practitioners and researchers with indelible understanding that seeks to enhance the optimum therapeutic trajectory and evidence-based decision-making toward an improved outcome of treatment in lymphoma.

Secure multi-keyword search for outsourced cloud data has gained popularity, especially for scenarios involving multiple data owners. This work proposes a method for secure multi-keyword searches across encrypted cloud data from various owners, utilizing an administrator or trustworthy third party for key management. The approach employs the TF-IDF model and Vector Space Model (VSM) to generate index and query vectors, and uses an advanced Attribute-Based Encryption (ABE) algorithm for data privacy. The Improved Dingo Optimization Algorithm (IDOA) optimally selects the key value. Encrypted emails are sent to a cloud server, and a Grouped Balanced Binary (GBB) tree structure is used for index construction. The Greedy Depth-First Search (GDFS) method efficiently searches the index, grouping relevant documents to exclude unnecessary ones. Performance is evaluated based on memory usage, execution time, encryption and decryption times, and search times, implemented in Java and tested on a cloud simulator.

The main objective of present work is to analyse the reliability measures of a juice clarification unit in sugar plants under the concepts of various redundancy polices. The juice clarification unit (JCU) is a complex system having eight subsystems configured according to various redundancies internally. It is assumed that all the associated random variables follow exponential distribution. The Markov birth–death process and basic reliability arguments are used to derive Reliability, Availability, Maintainability, and Dependability (RAMD) measures as well as Runge–Kutta forth order is used to derive transient solution. The numerical results for reliability, availability, maintainability, Mean Time Between Failure (MTBF), Mean Time to Repair (MTTR), and dependability are derived to identify the most sensitive component. The system attains transient availability 0.995881 at 5-time units. To highlight the importance of the study numerical results of transient availability are derived by making 10% and 50% variation in failure and repair rates. The derived numerical results of the proposed model draw significant insights that can be directly used in the design and maintenance process of juice clarification unit. It will improve the plant efficiency. It will give designers a solid plan to use resources intelligently and make strategies that make juice clarification unit live to work better.

Ensuring the secure transmission of data between smart devices and the cloud is paramount in the context of Blood Pressure Monitoring Sensors. In the realm of “Securing Data Transmission Channel between Smart Devices and Cloud using Homomorphic Encryption for Blood Pressure” “Monitoring sensors play a crucial role in various applications.” The incorporation of homomorphic encryption is considered essential to guarantee the privacy and security of health data. In smart cities, where mobile healthcare networks play a vital role, concerns arise over the transmission of sensitive health information to untrusted cloud servers. Our focus lies in the development of a specialized architecture explicitly designed for the unique requirements of Blood Pressure Monitoring Sensors. Within this paper, a concise yet comprehensive framework, for mobile healthcare networks is outlined, emphasizing three secure medical calculations encompassing tasks such as computing average heart rate detecting long QT syndrome, and performing Chi-square tests are integral aspects of healthcare. Through this methodology, a distinctive and innovative solution is provided, tailored specifically to the challenges of securing data transmission channels for Blood Pressure Monitoring Sensors.

Noise figure (NF) and linearity at higher frequencies are challenging tasks in Low Noise Amplifiers (LNAs) design. The existing LNAs impose limits on these parameters resulting in enhanced NF. In this paper, two ultra-wide band (UWB) small signal Sziklai pair based LNAs are proposed with improved linearity and minimum NF for high frequency wireless communication. Theses LNAs use NMOS Sziklai and PMOS Sziklai pairs as amplifying components with bandwidth boosting arrangement in common source mode. The simulation results indicate that NMOS LNA exhibits bandwidth of 580.554 GHz with noise figure of 1.66 dB at 63.09 GHz, whereas PMOS LNA shows bandwidth of 299.956 GHz, with noise figure of 11.71 dB at 283.09 GHz, at 180 nm CMOS technology. Furthermore, NMOS and PMOS LNAs show high linearity with input 1 dB compression point (P1dB) of 0.883 dBm and 0.822 dBm while input third order intercept point (IIP3) are 10.22 dBm and 6.72 dBm respectively. Compared to CMOS LNA, bulk CMOS LNA, GaAs pHEMT LNA, and SiGe BiCMOS LNA, the proposed LNAs significantly improve power gain, bandwidth, noise figure, linearity, and area. To further expand the bandwidth, three stage broadband distributed amplifiers (DAs) are also configured using the designed LNAs which measures the bandwidth of 0–7.5 GHz and 10,000–800,000 THz. The performance of these LNAs is verified through designing D-band microwave tuneable second order band reject filter (BRF) which exhibits ultra-wide band (UWB) of 110.362 GHz, notch depth of 10.3 GHz while consuming DC power of 68.32 pW at a supply voltage of ± 3 V.

Autonomous vehicles face significant challenges in accurately identifying vehicles, objects, and traffic signals under adverse weather conditions and poor lighting. To address these issues, we introduce a novel detection system utilizing automatic white balance techniques, specifically the Adaptive Retinex algorithm, to restore visibility and enhance color. This system is integrated with a Faster R-CNN framework enhanced by non-maximum suppression to improve the accuracy of object detection. Employing a combination of three datasets—Dawn, MCWRD, and Indian Roads Dataset (IRD)—our method includes over 6000 augmented images representing diverse environmental conditions. We also implement an optimized version of Differentiable ARchiTecture Search (DARTS) to dynamically fine-tune the architectural parameters of our detection model. This approach has successfully achieved a detection accuracy of 97.43% with a minimal loss rate, demonstrating significant potential for enhancing navigation safety in autonomous vehicles across various challenging environments.

This study addresses the challenge of enhancing network security by proposing a novel intrusion detection system using Generative Adversarial Networks. Traditional intrusion detection system often fail to keep up with rapidly evolving cyber threats. Our approach integrates Generative Adversarial Networks to dynamically learn and adapt to both genuine and adversarial network traffic patterns. Using the KDD Cup 1999 dataset for validation, we design a sophisticated Generative Adversarial Network architecture with a generator and discriminator to improve the resilience of intrusion detection system. Our experimental results demonstrate the model’s effectiveness, evaluated through metrics such as F1 score, accuracy, precision, and recall. This research advances the state-of-the-art in cybersecurity by showcasing the potential of Generative Adversarial Networks to fortify intrusion detection system against evolving threats, underscoring the necessity for adaptive defense mechanisms in modern network security.