IEEE Access最新文献

{"title":"Windowed Octonion Quadratic Phase Fourier Transform: Sharp Inequalities, Uncertainty Principles, and Examples in Signal Processing","authors":"Manish Kumar;Bhawna","doi":"10.1109/ACCESS.2024.3473298","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473298","url":null,"abstract":"In this paper, we define the Windowed Octonion Quadratic Phase Fourier Transform (WOQPFT) and derive its inversion formula, including its essential properties, such as linearity, anti-linearity, parity, scaling, modulation, shifting, and joint time-frequency shifting, as well as its link to Octonion Quadratic Phase Fourier Transform (OQPFT). Additionally, we derive the Riemann-Lebesgue lemma using this transform. Following the present analysis, we formulated Sharp Pitt’s and Sharp Hausdorff-Young’s inequalities. Further, Logarithmic, Heisenberg’s, and Donoho-Stark’s uncertainty principles are also formulated. The practical application of WOQPFT and the five elementary examples of signal theory are discussed, and their particular cases are analyzed through graphical visualization, including interpretation.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"146771-146794"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704622","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142447154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transformer-Based Fault Detection Using Pressure Signals for Hydraulic Pumps","authors":"A. Ran Kim;Ha Seon Kim;Sun Young Kim","doi":"10.1109/ACCESS.2024.3472750","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3472750","url":null,"abstract":"In this paper, modified transformer-based fault detection for a hydraulic pump is performed using the pressure signals of the hydraulic pump. The pump is considered a swash plate axial piston pump used in the excavator. Additionally, the outlet pressure data of the pump are extracted based on Amesim. The proposed transformer is a modified transformer, which allows fast fault detection by modifying the transformer and reducing the size of this model. The classes are normal and 6 fault types, and comparison models are long short-term memory (LSTM) and its family models, which are representative time series models. Unlike comparison models, the modified transformer has an average accuracy of 100% and a detection time of 0.00271 s, which is a slight difference of 0.00036 s from the single LSTM that showed the shortest operation time among the models. We also perform fault detection by changing data points and show a stable high accuracy of 99.93% for all data points of 500, 1,000, and 1,500 without any optimization. Various external noises are added because excavators are construction equipment used in rough terrain. Therefore, we conduct detection performance analysis at different additional noise levels with Gaussian noise with zero mean. As a result, we confirm that the modified transformer showed a high detection accuracy of over 98.08% up to standard deviation 4, where data characteristics were well maintained, unlike other time series models. Through the various analyses above, we confirm that fast and accurate fault detection is possible based on the modified transformer.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"145795-145808"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704636","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142408943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"APTracker: A Comprehensive and Analytical Malware Dataset Based on Attribution to APT Groups","authors":"Mohamad Erfan Mazaheri;Alireza Shameli-Sendi","doi":"10.1109/ACCESS.2024.3473021","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473021","url":null,"abstract":"Malware poses a significant threat to organizations, necessitating robust countermeasures. One such measure involves attributing malware to its respective Advanced Persistent Threat (APT) group, which serves several purposes, two of the most important ones are: aiding in incident response and facilitating legal recourse. Recent years have witnessed a surge in research efforts aimed at refining methods for attributing malware to specific threat groups. These endeavors have leveraged a variety of machine learning and deep learning techniques, alongside diverse features extracted from malware binary files, to develop attribution systems. Despite these advancements, the field continues to beckon further investigation to enhance attribution methodologies. The basis of developing an effective attribution systems is to benefit from a rich dataset. Previous studies in this domain have meticulously detailed the process of model training and evaluation using distinct datasets, each characterized by unique strengths, weaknesses, and varying number of samples. In this paper, we scrutinize previous datasets from several perspectives while focusing on analyzing our dataset, which we claim is the most comprehensive in the realm of malware attribution. This dataset encompasses 64,440 malware samples attributed to 22 APT groups and spans a minimum of 40 malware families. The samples in the dataset span the years 2020 to 2024, and their developer APT groups originate from Russia, South Korea, China, USA, Nigeria, North Korea, Pakistan and Belarus. Its richness and breadth render it invaluable for future research endeavors in the field of malware attribution.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"145148-145158"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704627","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142408967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Real-Time Software-Defined Radio Platform for Sub-Terahertz Communication Systems","authors":"Hussam Abdellatif;Viduneth Ariyarathna;Arjuna Madanayake;Josep Miquel Jornet","doi":"10.1109/ACCESS.2024.3473615","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473615","url":null,"abstract":"Wireless communication in the sub-terahertz and terahertz (THz) bands (broadly from 100 GHz to 10 THz) is a critical building block of the future generations of telecommunication and networking due to the large available bandwidth at these frequencies and the opportunities it brings for ultra-broadband communication and sensing systems. Alongside the high data rates offered by this band, the huge bandwidth can be shared more generously across multiple users with hopes of reducing network congestion. With the recent improvements being made on the electronics side such as high-speed data converters and high-frequency oscillators, several testing platforms for experimental THz communication have been recently developed. However, these are mostly device technology demonstrators, channel sounders, or physical-layer testbeds, which do not support real-time digital signal processing (DSP). Such platforms have supported the large body of THz research focused on studying the channel or developing physical layer solutions. However, the lack of real-time DSP capabilities prevents the testing of upper networking protocols, on which the research community is only now starting to focus. While real-time networking platforms, namely, software-defined radio (SDR) platforms, developed for lower frequency systems could be utilized, their very low bandwidth misses the point of moving to the sub-THz and THz bands. To fill the gap, in this paper, we design an SDR platform able to process multi-GHz of baseband bandwidth in real-time by leveraging the state of the art in radio-frequency systems on chip (RFSoC), a custom frequency-multiplexing analog network and a multi-phase implementation of an orthogonal frequency-division modulation (OFDM) physical layer. As an instantiation of the platform, we demonstrate a real-time link at 135 GHz with 8 GHz of bandwidth supporting a bit-rate of 33 Gbps when frequency-multiplexing four 2-GHz-wide channels, each with 64-sub-carrier OFDM. Finally, we identify immediate next steps and cross-layer challenges foreseen when implementing wireless communication and sensing systems at frequencies above 100 GHz.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"146315-146327"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704668","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142447016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"End-to-End Autonomous Driving in CARLA: A Survey","authors":"Youssef Al Ozaibi;Manolo Dulva Hina;Amar Ramdane-Cherif","doi":"10.1109/ACCESS.2024.3473611","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473611","url":null,"abstract":"Autonomous Driving (AD) has evolved significantly since its beginnings in the 1980s, with continuous advancements driven by both industry and academia. Traditional AD systems break down the driving task into smaller modules—such as perception, localization, planning, and control– and optimizes them independently. In contrast, end-to-end models use neural networks to map sensory inputs directly to vehicle controls, optimizing the entire driving process as a single task. Recent advancements in deep learning have driven increased interest in end-to-end models, which is the central focus of this review. In this survey, we discuss how CARLA-based state-of-the-art implementations address various issues encountered in end-to-end autonomous driving through various model inputs, outputs, architectures, and training paradigms. To provide a comprehensive overview, we additionally include a concise summary of these methods in a single large table. Finally, we present evaluations and discussions of the methods, and suggest future avenues to tackle current challenges faced by end-to-end models.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"146866-146900"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704612","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142447017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating Multiscale Linear Attention and Focal Loss for Robust Pest Classification","authors":"Shulin Zhao;Hai Wang;Tailian Liu;Shulai Huang","doi":"10.1109/ACCESS.2024.3473536","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473536","url":null,"abstract":"Agricultural pests significantly impact crop yield and quality, threatening food security and causing economic losses. Therefore, the precise identification of pests is crucial for improving agricultural production. However, traditional pest classification methods struggle to capture the complex relationships among different parts of pest images and often lack strong generalization capabilities, resulting in poor performance. To address these issues, we propose an agricultural pest classification model based on a multi-scale linear attention mechanism and Focal Loss. This model employs a multi-scale linear attention module to capture local features at various scales, as well as the long-distance dependencies and global relationships among these local features. It utilizes an attention mechanism with linear time complexity to ensure computational efficiency. In addition, we use the Focal Loss function to alleviate the impact of sample imbalance in the dataset and explore the effects of various data augmentation techniques on the model’s generalization ability. Experimental results demonstrate that our model performs excellently across datasets of different scales.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"146610-146619"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704656","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142447231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced ProtoNet With Self-Knowledge Distillation for Few-Shot Learning","authors":"Mohamed El Hacen Habib;Ayhan Küçükmanisa;Oǧuzhan Urhan","doi":"10.1109/ACCESS.2024.3472530","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3472530","url":null,"abstract":"Few-Shot Learning (FSL) has recently gained increased attention for its effectiveness in addressing the problem of data scarcity. Many approaches have been proposed based on the FSL idea, including prototypical networks (ProtoNet). ProtoNet demonstrates its effectiveness in overcoming this issue while providing simplicity in its architecture. On the other hand, the self-knowledge distillation (SKD) technique has become popular in assisting FSL models in achieving good performance by transferring knowledge gained from additional training data. In this work, we apply the self-knowledge distillation technique to ProtoNet to boost its performance. For each task, we compute the prototypes from the few examples (local prototypes) and the many examples (global prototypes) and use the global prototypes to distill knowledge to the few-shot learner model. We employ different distillation techniques based on prototypes, logits, and predictions (soft labels). We evaluated our method using three popular FSL image classification benchmark datasets: CIFAR-FS, CIFAR-FC100, and miniImageNet. Our approach outperformed the baseline and achieved competitive results compared to the state-of-the-art methods, especially on the CIFAR-FC100 dataset.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"145331-145340"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10703064","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142408936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Learning Approach for Efficient 5G LDPC Decoding in IoT","authors":"Sivarama Prasad Tera;Ravikumar V. Chinthaginjala;Priya Natha;Shafiq Ahmad;Giovanni Pau","doi":"10.1109/ACCESS.2024.3472466","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3472466","url":null,"abstract":"The tremendous progress of 5G technology has transformed the landscape of the Internet of Things (IoT), allowing for fast data speeds, low delay, and widespread connection that is crucial for a variety of applications, including smart cities and industrial automation. In the context of 5G enabled IoT networks, colored noise introduces varying levels of interference across different frequency bands, which can significantly degrade the performance of 5G LDPC decoding. This paper presents a novel Deep learning approach for 5G channel LDPC code decoding tailored for next-generation IoT applications. The proposed method integrates an Iterative Normalized Min-Sum (NMS) algorithm with a Convolutional Neural Network (CNN) to enhance the performance of LDPC decoding in the presence of colored noise, a common interference in real-world communication channels. Through extensive simulations and analysis, our approach demonstrates a significant performance improvement, achieving a 3.8 dB enhancement at a Bit error rate of \u0000<inline-formula> <tex-math>$10^{-6}$ </tex-math></inline-formula>\u0000. This is achieved by accurately estimating and mitigating channel noise, thereby ensuring reliable data transmission for critical IoT applications. The findings indicate that our approach to decoding technique not only enhances error correction capabilities but also adapts to varying channel conditions, optimizing IoT network performance and efficiency. This research contributes a robust solution to the challenges posed by colored noise in 5G-enabled IoT networks, promoting the deployment of more reliable and efficient IoT systems.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"145671-145685"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10703047","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142408993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint Speech-Text Embeddings for Multitask Speech Processing","authors":"Michael Gian Gonzales;Peter Corcoran;Naomi Harte;Michael Schukat","doi":"10.1109/ACCESS.2024.3473743","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3473743","url":null,"abstract":"Devices that use speech as the communication medium between human and computer have been emerging for the past few years. The technologies behind this interface are called Automatic Speech Recognition (ASR) and Text-to-Speech (TTS). The two are distinct fields in speech signal processing that have independently made great strides in recent years. This paper proposes an architecture that takes advantage of the two modalities present in ASR and TTS, speech and text, while simultaneously training three tasks, adding speaker recognition to the underlying ASR and TTS tasks. This architecture not only reduces the memory footprint required to run all tasks, but also has performance comparable to single-task models. The dataset used to train and evaluate the model is the CSTR VCTK Corpus. Results show a 97.64% accuracy in the speaker recognition task, word and character error rates of 18.18% and 7.95% for the ASR task, a mel cepstral distortion of 4.31 and two predicted MOS of 2.98 and 3.28 for the TTS task. While voice conversion is not part of the training tasks, the architecture is capable of doing this and was evaluated to have 5.22, 2.98, and 2.73 for mel cepstral distortion and predicted MOS, respectively.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"145955-145967"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704626","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142409069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimize the Activity-on-Arc Network Planning Through the Structure Matrix and Genetic Algorithm","authors":"Gongyu Hou;Haoxiang Li;Qinhuang Chen;Yaohua Shao;Dandan Wang","doi":"10.1109/ACCESS.2024.3468313","DOIUrl":"https://doi.org/10.1109/ACCESS.2024.3468313","url":null,"abstract":"The resource-constrained project scheduling problem (RCPSP) poses several challenges for optimizing activity-on-arc network planning. The existing genetic algorithms for solving this problem have high computational complexity and low efficiency in terms of running and optimization. To address this issue, this study proposes an algorithm based on the structure matrix and genetic algorithm (SM-GA). First, information about activities in the activity-on-arc network was represented in the structure matrix (SM), and a data storage format and a coordinate-coded structure were constructed. Then, a chromosome correction operator and serial schedule generation scheme were designed based on the SM. Further, an adaptive probability operator, along with its related similar-uniform crossover operator and hybrid mutation operator, were designed based on the level of population diversity. Finally, Python programs were written in combination with a case study, and the efficiency of the algorithm was statistically analyzed from two aspects: data formats, and operators. SM-GA enhances the running efficiency by approximately 36 times compared to the genetic algorithm using a traditional data format. Compared to the genetic algorithm using traditional crossover and mutation operators, SM-GA improves the optimization efficiency by approximately four times. The results show that SM-GA could solve the RCPSP of activity-on-arc network planning more efficiently.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"12 ","pages":"143733-143744"},"PeriodicalIF":3.4,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10704997","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142430812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}